<%BANNER%>

Foreign Direct Investment and the Investment Climate of Developing Countries in the Western Hemisphere


PAGE 1

FOREIGN DIRECT INVESTMENT AND THE INVESTMENT CLIMATE OF DEVELOPING COUNTRIES IN THE WESTERN HEMISPHERE By ERIC T. BONNETT A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Eric T. Bonnett

PAGE 3

This dissertation is dedicated to my lovely wife, Rita.

PAGE 4

iv ACKNOWLEDGMENTS I would first like to thank the members of my supervisory committee for making my time as a graduate student at the Universi ty of Florida as enj oyable as it has been. During my years in the Food and Resource Ec onomics Department I became distinctly aware of how well respected Drs. Timothy Ta ylor, Gary Fairchild, P.J. van Blokland and Roy Crum are among those in the academic worl d. More recently, I have also come to realize the level of respect those outsid e of academia possess for these men. Their contributions to this project are greatly appreciated and will never be forgotten. Also I would be remiss if I did not mention my appreciation for the hours Dr. Ronald Ward spent assisting me with this research project. My parents deserve special r ecognition for instilling in me the confidence to pursue such an ambitious goal as writing a doctor al dissertation. While they have always encouraged me to aggressively pursue my goa ls, they also taught me the importance of being able to maintain a sound conscience. I will forever be in debt to my mother and father for passing these most admira ble personal traits along to me. As I find myself at the term inal point of this project, I am reminded of a very special individual who was taken from this wo rld in a most untimely manner. If it were not for Dr. Patrick Byrne, I most likely woul d have never attended graduate school. He was a source of inspiration for many students during his tenu re at the University of Florida, and his legacy liv es on with each of them.

PAGE 5

v Finally, I would like to thank my wife, Rita, to whom this dissertation is dedicated. Just as she is essential to my happiness, her unremitting encouragement and support were essential to the successful completion of th is project. I shall always treasure her commitment to our relationship, and I will fore ver carry the knowledge that I am a better person because of her.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT....................................................................................................................... ..x CHAPTER 1 INTRODUCTION........................................................................................................1 Impacts of FDI on Developing Host Countries............................................................3 Problematic Situation....................................................................................................6 Problem Statement........................................................................................................8 Objectives...................................................................................................................10 Scope.......................................................................................................................... .11 2 HISTORICAL CONTEXT.........................................................................................13 A Taxonomy of Capital Flows...................................................................................15 A Method of Characterizing Differe nt Types of Capital Flows.................................18 The Structure and Evolution of Capital Flows from 1970 to 2001............................22 The Andean Community.....................................................................................24 The Central American Common Market.............................................................28 The Caribbean Community and Common Market..............................................32 MERCOSUR.......................................................................................................35 Mexico and Chile................................................................................................37 Summary of Trends....................................................................................................40 3 THEORETICAL FOUNDATIONS...........................................................................42 Neoclassical Economic Theory vs. Strategic Management Theory...........................44 Dunnings OLI Framework........................................................................................48

PAGE 7

vii 4 EMPIRICAL CONSIDERATIONS...........................................................................56 The MNCs Motivations for Ma king Direct Investments..........................................56 Decisions on the Mode of Organization..............................................................60 Decisions on Location.........................................................................................63 The Empirical Model..................................................................................................68 The Host-Country Investment Climate...............................................................69 The Policy and Regulatory Environment............................................................70 Political risk..................................................................................................73 Factors that affect th e return on investment.................................................75 Control Variables........................................................................................................78 5 MODEL SPECIFICATION, ESTIMATION, AND EMPIRICAL ANALYSIS.......79 Data and Sources........................................................................................................79 Model Specification and Estimation...........................................................................81 Results of Estimation..................................................................................................85 MERCOSUR.......................................................................................................90 The Andean Community.....................................................................................96 CARICOM........................................................................................................101 The Central American Common Market...........................................................107 Mexico and Chile..............................................................................................113 6 SUMMARY AND CONCLUSIONS.......................................................................117 The Least Responsive Countries..............................................................................117 The Most Responsive Countries...............................................................................119 Policy-Related Conclusions......................................................................................120 Considerations for Future Research..........................................................................122 APPENDIX DEFINITIONS OF SELECTED VARIABLES........................................124 LIST OF REFERENCES.................................................................................................126 BIOGRAPHICAL SKETCH...........................................................................................133

PAGE 8

viii LIST OF TABLES Table page 4-1 Survey Results on the Importance of Investment Climate Factors..........................70 5-1 Descriptive Statisti cs by Country (1984-2001)........................................................80 5-2 Random Effects Regression Results........................................................................86 5-3 Investment Climate Variables MERCOSUR........................................................90 5-4 FDI Responsiveness MERCOSUR.......................................................................90 5-5 Investment Climate Variables Andean Community..............................................96 5-6 FDI Responsiveness Andean Community.............................................................96 5-7 FDI Responsiveness CARICOM........................................................................102 5-8 Investment Climate Variables CARICOM.........................................................103 5-9 Investment Climate Variables CACM................................................................108 5-10 FDI Responsiveness CACM...............................................................................108 5-11 Investment Climate Variables Mexico and Chile...............................................113 5-12 FDI Responsiveness Mexico and Chile..............................................................114

PAGE 9

ix LIST OF FIGURES Figure page 2-1 Williamsons Key Characteristics of Capital Flows................................................21 2-2 Capital Flows to Developing Economies in the Western Hemisphere....................23 2-3 Private Capital Flows to Developing Countries in the Western Hemisphere..........23 2-4 Capital Flows to the Andean Community................................................................25 2-5 ODA Flows to Members of the Andean Community..............................................26 2-6 Private Capital Flows to the Andean Community....................................................26 2-7 Capital Flows to CACM...........................................................................................29 2-8 ODA Flows to Members of CACM.........................................................................30 2-9 FDI Flows to Members of CACM...........................................................................31 2-10 Capital Flows to CARICOM....................................................................................34 2-11 Private Capital Flows to CARICOM.......................................................................34 2-12 Capital Flows to MERCOSUR................................................................................36 2-13 Private Capital Flows to MERCOSUR....................................................................37 2-14 Capital Flows to Mexico..........................................................................................38 2-15 Private Capital Flows to Mexico..............................................................................39 2-16 Private Capital Flows to Chile.................................................................................40 3-1 Modes of Internationalization..................................................................................51

PAGE 10

x Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FOREIGN DIRECT INVESTMENT AND THE INVESTMENT CLIMATE OF DEVELOPING COUNTRIES IN THE WESTERN HEMISPHERE By Eric T. Bonnett December 2004 Chair: Timothy G. Taylor Cochair: Gary F. Fairchild Major Department: Food and Resource Economics The structure of capital flows to devel oping countries around the globe has changed considerably since the 1970s. In the Western Hemisphere specifically, foreign direct investment (FDI) accounted for a much larger share of total inflows than any other type of capital by the end of the 1990s. This trend has not escaped recognition by economic theorists, and a larg e literature has emerged as researchers attempt to understand why some countries attract more FDI than others. This study examines the impact of changes in the investment climate on a countrys ability to attract FDI flows relative to other countries. Specific atte ntion is given to the developing countries of the Western Hemisphe re. The investment climate is broadly defined as (i) governmental policies and regula tions that affect the relative openness of the country to FDI, (ii) fact ors that impact the potential return on capital to foreign investors, and (iii) the leve l of political risk and co rruption in the host country.

PAGE 11

xi Analyzing the impact of changes in these types of factors over time revealed some interesting results. First, there is evidence to suggest that the state of the investment climate is an important consideration for fo reign investors. Second, FDI in large (in terms of GDP), relatively unsta ble economies tends to be the most responsive to small changes in the investment climate. Convers ely, in small economies that either receive substantial amounts of official developmen t assistance or are dominated by a single industry (e.g., the production of oil), FDI tends to be less re sponsive to changes in the investment climate. Finally, the results make clear that, acro ss the sample of c ountries, there is no single model that can explain all of the differe nces in the level of response to changes in the investment climate. In other words, the relative impact of investment climate variables on the level of FDI differs accord ing to the economic, political and social conditions inherent to each country. Thus, it is likely that future re search into the issue will yield more interesting resu lts if analysis is conducted on a country-by-country basis.

PAGE 12

1 CHAPTER 1 INTRODUCTION The volatility of capital flows to devel oping countries in Latin America and the Caribbean is well established. Investors in the region often hold short-term assets and withdraw their funds at the first sign of economic distre ss (Rojas-Suarez and Weisbrod 1996). This, in turn, contributes to the vul nerability of Latin American and Caribbean economies to external economic shocks such as changes in commodity prices and international interest rate s (Birdsall and Lozada 1996). The financial crises that struck many deve loping countries after the surge in capital flows during the 1990s led to increased skeptici sm about the benefits of attracting foreign capital (World Bank 2001). However, the liter ature makes clear that just as there are many different ways to define a financial crisis, there are eq ually as many theories of what causes them. Nevertheless, it is generally agreed that th e sharp rise in incidence of financial crises in the last 20 years is not independent of the observed increase in the magnitude and frequency of in ternational capital flows. From 1970 to 1992, Latin American and Caribbean economies were between two and three times as volatile as industriali zed economies (Hausmann and Gavin 1996). While much of the observed volatility is a pr oduct of inconsistent macroeconomic policy, other factors also had an influence. According to Hausmann and Gavin (1996): another reason for the volatility of La tin American [and indeed, Caribbean] macroeconomic outcomes is the large external disturbances that routinely buffet the region. The most important of these are sudden changes in the terms of trade and in international capita l flows. (Hausmann and Gavin 1996, p. 27)

PAGE 13

2 Large capital inflows cause rapid monetary expansion, inflationary pressures, real exchange rate appreciation and widening curr ent account deficits (Calvo et al. 1996). Consequently, when flows are interrupte d, the current account and exchange rate experience reverse adjustments (Hoti 2002). Fu rthermore, these terms-of-trade shocks have large and statistically significant effects on the varian ce of economic growth rates over time (Easterly et al. 1993). In many of the developing economies in the Western Hemisphere, FDI has become the dominant component of capital inflows, accounting for a much higher percentage of gross domestic product (GDP) in the 1990s than in the previous two decades. However, there are exceptions to this rule, and it is often those countries that are the least economically and politically stable that are th e least successful at at tracting FDI. Some have argued that economic instab ility leads to uncertainty for the firms and investors who make direct investments. As a result, investors choose to locate in less risky countries. The cross-country variation in FDI sin ce the early 1990s (discussed further in Chapter 2) serves as the inspiration for this study. While a large body of empirical literature addresses the global factors that ca use expansions and contractions in global FDI flows, less attention has been given to explaining the differences in FDI flows across countries. More specifically, there is a need for additional insight into th e factors that affect the distribution of FDI flows acro ss the developing countri es of the Western Hemisphere. Empirical research suggests that the packag e of assets that ac companies FDI brings many benefits to the host country including economic growth, development of domestic industries, increased employment and a hi gher standard of living (Parry 1973).

PAGE 14

3 Theoretical support for each of these conten tions has developed over time while there have been fewer attempts to argue against th e benefits of FDI. What is made clear by attempts to test the theory is that the be nefits realized by the hosts of FDI funds are conditional upon many factors. Impacts of FDI on Deve loping Host Countries The increase in foreign equity flows to developing countries in the Western Hemisphere reflects the fact that countries have increased the nu mber of investment opportunities by promoting the privatization of government enterprises and facilitating the development of deeper and more liquid financial markets (Moreno 2000). However, Wells, Jr. (1998) points out that although te nsions between foreign investors and developing host countries had clearly w eakened by the mid 1990s, there were few definite conclusions as to the net impact of FDI in the lesser devel oped countries (LDCs). Since then, a vast literatur e on the subject has emerged. Much of the research into the economic im pacts of FDI is aimed at examining the indirect or spillover effects of hosting di rect investment. It has been argued that indirect effects primarily take the form of technological spillovers and competition effects. In fact, Blomstrm and Kokko (2003) contend that the expectation that technological know-how will spillover from multinational corporations (MNCs) to indigenous firms has been strong enough for hos t countries to lower barriers to entry, open up new sectors to foreign investment and, in some cases, provide investment incentives to foreign firms. Nonetheless, em pirical research into the matter is fraught with conflicting results. For instance, in a study on manufacturing industries in Mexic o, Blomstrm (1986) found that the presence of foreign MNCs in an industry was positively correlated with

PAGE 15

4 structural efficiency. However, the results did not suggest that the increase in efficiency was due to the transfer of technology from the MNCs to domestic firms (whether through imitation or labor migration). Rather, increase s in competitive pressure appeared to have a more important effect. Aitken and Harrison (1999) found that foreign equity pa rticipation [in Venezuelan plants] is positively correlated with plant productivity . [but,] foreign investment negatively affects the productivity of domestically owne d plants (p.605). The authors indicate that in the presence of these offs etting effects, the net impact of FDI on the productive efficiency of the host economy as a whole is negligible. More recent research suggests that the extent to which foreign participation impacts the productivity of indigenous firms appears to depend upon th e type of linkages that are examined. While Aitken and Harrison (1999) examine th e effect of FDI on the productivity of all firms in the economy, Smarzynska (2002) li mits her analysis to the effect on upstream firms (i.e., the backward linkage between fore ign affiliates and thei r local suppliers). Pointing out that spillovers are more likely to be vertical rather than horizontal in nature1 (Smarzynska 2002, p.2), she states, a rise of ten percent in the foreign presence in downstream industries is associated with a 0.38 percent increas e in output of each domestic firm in the upstream sector (Smarzynska 2002, p.16). Furthermore, Smarzynska found no difference between the eff ect of wholly-owned foreign subsidiaries and joint ventures with both fo reign and domestic investors. 1 This is due to the multinational firms incentive to tr ansfer knowledge to its suppliers in order to increase the quality or decrease the price of its inputs, and prev ent the horizontal leakage of information that might enhance the performance of its competitors.

PAGE 16

5 It has been shown by some researchers that controlling for certain domestic conditions leads to more conclusive findings on the net impact of FDI. One such case is that of the relationship between FDI and ec onomic growth in developing host countries. Although empirical findings on the nature of this relationship do vary, the World Bank (2001) indicates that differences in host countries absorptive capacity2 accounts for a majority of the variance. In fact, where ab sorptive capacity is high, it is generally agreed that FDI has a positive impact on productivity and hence economic growth (World Bank 2001). The study by Borenzstein et al. (1998) serves to illustrate this point more clearly. Analyzing data on FDI flows to 69 developi ng countries over a peri od of 20 years, the authors found that FDI had a positive eff ect on economic growth. Furthermore, the magnitude of this effect was shown to de pend strongly upon the available stock of human capital, a variable commonly used as a meas ure of absorptive capacity. That is, higher levels of human capital caused FDI to have a larger positive impact on growth. Furthermore, with sufficient levels of huma n capital, FDI appeared to be more productive than, and complementary to domestic investment The authors also note that the positive impact of FDI persisted ev en after controlling for initial income, human capital, government consumption and the parallel ma rket premium for foreign exchange. Aside from the results of a handful of empi rical studies, it is generally agreed that FDI has the potential to facilita te the transfer of ideas from industrialized to developing countries, thus increasing productivity in the latter (World Bank 2001). In addition to the stability of FDI relative to other forms of foreign capital the potential for increased 2 Proxies for absorptive cap acity include but are not limited to ope nness to trade, the amount and quality of infrastructure and human capital, and inflation.

PAGE 17

6 productivity provides some rationale for w hy many developing countries have shown an appetite for FDI. Problematic Situation It has been suggested by the Economic Commission for Latin America and the Caribbean (ECLAC 2002) that the high volatility of net capital inflows (other than FDI) during the last ten years has undermined the st ability of economic growth in the region. When net capital inflows are high, domestic cr edit and liquidity grow too quickly, while the opposite is true in times of recession (ECLAC 2002). This phenomenon tends to amplify boom-bust cycles, thus destabilizing economic growth. As an example, Mishkin (2001) states The financial crises that struck Mexico in 1994 and the East Asian countries in 1997 led to a fall in the growth rate of GD P on the order of ten percentage points. The financial crises in Russia in 1998 and Ecuador in 1999 have had similar negative effects on real output. Not only did these crises le ad to sharp increases in poverty, but to political instabil ity as well. (Mishkin 2001, p.1) The link between capital flow volatil ity and economic growth has led many researchers to gain interest in the factors that drive capital flows. There is considerable debate over whether capita l flows to developing c ountries are driven by forces external to the countries themselves, or by domestic factors.3 The existing empirical literature suggests that the answer depe nds upon how the issue is addr essed and that analyzing the sum of all capital flows may be misleading. Af ter all, each form of capital has its own characteristics and set of circumstances (Lusinyan 2002). Foreign direct investment accounts for a much larger share of GDP than any other form of foreign capital infl ow for many of the developi ng countries in the Western 3 Moreno (2000) provides one of the more concise review s of the empirical literature addressing this issue.

PAGE 18

7 Hemisphere, and much of this investment activ ity has been driven, or at least encouraged, by the actions of host-country governments. In addition to privatizing government enterprises, developing countri es have adopted policies intended to foster direct investment by MNCs. It is therefore surpri sing that the empirica l literature does not contain more models that attempt to measure the effect of these policies. Gabel and Bruner (2003) report that multinational corporations and their subsidiaries employ nearly 200 million people, generate $1.5 trillion in wages and pay over $1.2 trillion in taxes to their host govern ments, annually. The almost overwhelming global presence of the MNC has inspired c onsiderable debate among the many players involved in international commerce. Gove rnment policymakers are interested in maximizing the benefits of hosting multinationals while simultaneously limiting their negative impacts. Multinational managers ar e interested in maximizing the profitability of their businesses given a set of available investment alternatives. In a world where MNCs are increasingly prevalent, economists ar e left with the task of analyzing their causes and impacts. The resulting body of acad emic literature is substantial and diverse in purpose. Wells, Jr. (1998) indicates that the issues that face managers public and private who are concerned with FDI and the developing countries . have not attracted sufficient attention from economists (p.101). Foreign direct investment flows are ultimately driven by multinational corporations, and hence, their managers. Thus, the interaction between government policy and th e strategic allocation of capital by MNCs needs to be addressed. Much of the thought on this matter has been aimed at modeling

PAGE 19

8 the effect of exchange rate regimes on the international allocati on of production, but the issue is much more broad. Problem Statement The study of foreign direct investment ha s attracted the atten tion of researchers from various fields of business. Economi sts have portrayed FDI as the result of production cost differentials and comparative ad vantages in resource supply. In the field of finance, FDI has been described as a tool for international port folio diversification. Finally, in the strategic management body of literature, FDI patterns are seen as being dictated by the organizational decisions of the firms management. Although these approaches may differ from one another, efforts to understand why FDI occurs have yielded some compelling result s in each line of research. Foreign direct investment can be viewed as the allocation of financial capital from an entity in one country to establish or support a business entity in another country. When defined in this manner, FDI becomes the result of a strategic decision guided by the goals of management.4 To the extent that foreign investment policy can create an environment that either fosters or impedes di rect investment, the actions of host-country governments may have a significant influen ce on FDI flows. Along these lines, the ultimate concern of this study determining the extent to which FDI flows are affected by the investment climates of developing host countries. By surveying a group of U.S.-based multinational companies, Basi (1966) identified several variables related to the investment climate that were of significant importance to the investment decision. Of the variables considered by Basi, the 4 Management is defined broadly as those individuals who have a direct influence over the international allocation of direct investors capital.

PAGE 20

9 following factors were most often cited as either crucially important or fairly important to multinational managers: (i) th e host countrys attitude toward foreign investment; (ii) political stability; (iii) limita tions on ownership; (iv) currency exchange regulations; (v) the stab ility of foreign exchange; and, (vi) tax structure. While some of these variables are difficult to quantify, Basi at least provides an illustration of the investment climate that is consistent with the how the term is used in this study. The contribution this study makes to the existing literature centers on two primary characteristics of the projects design. First, the empirical model presented here examines the relative effect of five distinct investment climate f actors on FDI. While some proxy for the investment climate is ofte n included as an expl anatory variable in existing models, fewer models have been desi gned to simultaneously examine different aspects of the investment climate. The second distinguishing characteri stic of the study is its focus on Latin America and the Caribbean. Much of the existing literature is devoted to examining FDI flows to the entire unive rse of developing count ries. However, less ambiguous conclusions may be provided to policymakers in the region by limiting the focus of the analysis to the developi ng countries of the Western Hemisphere. There are three reasons for examining the e ffect of the overall investment climate on inflows of FDI. First, early attempts to survey firms on the determinants of international production show th at variables related to the investment climate are among the most influential factors from the perspective of management.5 Second, debate on the importance of many investment climate factor s to foreign investors is ongoing. Finally, 5 See Dunning (1973) for a review of surveys conducted by various authors prior to 1973.

PAGE 21

10 the relevance of the issue to policy-makers is clear, as many investment-climate variables are under the direct cont rol of the regulatory bod ies of host countries. Examination of FDI flows to the devel oping countries of the Western Hemisphere is particularly relevant. The financial markets of these countries are increasingly accessible to foreign investors, especially t hose who reside in the United States, where nearly 20% of world FDI out flows originated during the period 1970-2002. Furthermore, the provisions of the Free Trade Agreement of the Americas (FTAA), which is scheduled for entry into force by the end of 2005, provi de for the promotion and protection of foreign investment funds, re gardless of their source. Objectives The general objectives of this study are To provide an understanding of FDI as an economic phenomenon resulting from the strategic activities of multinational corporations To examine the effect of the investment climate of developing countries in the Western Hemisphere on inflows of FDI over the last two decades. In order to achieve these rather broad goals, the study is designed with several specific objectives in mind. The specific objectives of the study are To examine the evolution of the mul tinational corporat ion (MNC) from a theoretical perspective and address the strengths and w eaknesses of several streams of theory To develop a framework for understanding FD I that integrates the elements of economic and strategic management theory To develop a policy-relevant model of FDI inflows by in cluding variables that are not only considered as elements of the inve stment climate, but are also affected by the actions of host-country policymakers.

PAGE 22

11 Scope Chapter 2 establishes the context for th e study by providing an historical account of contemporary developments in global ec onomic integration. Specific attention is given to the developing countries in the West ern Hemisphere. The chapter compares and contrasts the different types of international capital flows and presents a detailed analysis of the structure of capital flows to Latin Am erica and the Caribbean over the last thirty years. Chapter 3 presents the theo retical foundations of the study. The chapter begins with a short review of the existing empiri cal research on FDI, although the results and implications of the models are not the main point of concern. Rather, the models are used to highlight the distinction between neoc lassical economic theory and strategic management theory, as well as to illustrate the fact that research on FDI has been conducted in a number of contexts. Specifi c attention is given to the conceptual evolution of the firm in each of the two stream s of theory. Finally, a review of the most relevant theoretical literature serves to de fine the perspective from which the empirical considerations of the study derive. Chapter 4 revisits the empiri cal literature is revisited in more depth. The chapter reviews a range of models that examine issues related to the international allocation of economic activity. Some of the models pr esented explicitly address the economic determinants of FDI, while others analyze factors that affect the strategic decisionmaking processes of multinational corporat ions. Empirical research on investment location and the alternative modes of foreign ownership is also reviewed. Chapter 5 presents the empirical model us ed to analyze a set of panel data on 21 developing countries in the Western Hemisphere over a period of 18 years. The chapter

PAGE 23

12 compares the results from several altern ative estimation procedures and provides a discussion on the responsiveness of FDI to changes in the investment climates of different countries. Finally, Chapter 6 summar izes the empirical re sults and presents a set of conclusions based on the findings. Chap ter 6 also provides suggestions for future research.

PAGE 24

13 CHAPTER 2 HISTORICAL CONTEXT It is difficult to find a piece of recent literature in the fiel d of international economics within which the term globalizat ion does not appear. According to the Economic Commission for Latin Americ a and the Caribbean (ECLAC 2002), globalization refers to the de nationalization of political, le gal and cultural systems, as well as economic markets. The primary en tities driving this process are governments, private investors and financial institutions (Schmukler 2004), and its socio-cultural impacts on developing countries are far-reaching, and often difficult to measure. In terms of economic impacts, globalization has led to reductions in trade barriers between countries, increased exchange of informati on and technology, and in many cases, greater vulnerability to worldwide economic conditions It should be note d, however, that the path toward globalization has been anything but uniform across countries. ECLAC distinguishes between the thr ee traditionally recognized phases of globalization by comparing the relative le vels of capital and labor mobility, the (non)existence of free trade and internati onal institutions for economic cooperation, and the extent of standardization among national development models. The first phase of globalization has its roots in th e transportation revolution and began to take shape in the last thirty years of the 19th century. Transportation costs fell dramatically during this period, effectively reducing the distance between countries and contri buting to increased mobility of goods and labor (Philippe 2001).

PAGE 25

14 With the Bretton Woods conference in 1944 came the initial movement toward establishing international organizations for economic cooperation (specifically, the International Monetary Fund and the World Ba nk). However, during this (the second) phase of globalization, disparity remain ed among national models of economic organization, and capital and labor mobili ty were limited. Although the Bretton Woods agreement was intended to foster the flow of capital across international boarders, it wasnt until the first oil crisis in the early 1970s that international capital mobility really expanded (Phillipe 2001). Nevertheless, w ith the foundation of global institutions for economic cooperation, the stage was set for an expansion in the trade of goods and financial capital. Since 1973, the drive towards international c ooperation has affected a real expansion in internat ional trade of manufactured goods services and capital. The increased global presence of multinational corporations also contributed to this expansion. Most recently, the birth of th e information age brought about unprecedented access to information and communication techno logies, further facilitating international transactions. While this story characterizes the pa th to globalization among industrialized economies throughout the world, it does not ne cessarily account for the individual experiences of many developing economies in the Western Hemisphere. There are four major economic integration groups in Latin America and the Caribbean: the Andean Community; the Central American Co mmon Market (CACM); the Caribbean Community and Common Market (CARICOM); and the Southern Cone Common Market (MERCOSUR). Mexico, Chile and the Domi nican Republic do not hold membership in

PAGE 26

15 any of these four integration groups but do relationships with them. Given that the Dominican Republic is currently an observer to CARICOM and that a bilateral free trade agreement exists between the two, the Do minican Republic is grouped with CARICOM in the ensuing discussion. Mexico and Chile are discu ssed independently. A Taxonomy of Capital Flows According to Liberatori (2003), internati onal economic integration is per se the result of both direct and indirect mobility of resources across national borders . including migration of worker s, international trade in good s and services, capital flows and international production and investment (p.2). In addition, international trade and investment flows are the result of both equity and non-equity transacti ons. In addition to the distinction between equity and non-equi ty, capital flows are cat egorized according to the source of funds and the conditions upon wh ich they are disbursed (i.e., concessional vs. interest bearing). The Deve lopment Assistance Committee (DAC 2000)1 states that Official transactions are those undertaken by central, state or local government agencies at their own risk and responsibi lity, regardless of whether these agencies have first borrowed the necessary funds from the private sector. Private transactions are those undertaken by firms and individuals reside nt in the reporting country. (DAC 2000, p.6) Official capital consists of official development assist ance (ODA) and other official flows (OOF). According to the guidelines set up by the DAC, ODA includes flows to countries on Part I of the Development Assi stance Committee List (w hich includes all of the countries in this analysis) that satisfy th ree criteria: (i) the f unds must be provided by official agencies including st ate and local governments or th eir executive agencies; (ii) 1 The DAC is the principal body through which the Organization for Economic Cooperation and Development (OECD) deals with issues related to cooperation with developing countries. The DAC statistical reporting directives provide a basis upon which all donor countries should report disbursements of official capital to developing countries.

PAGE 27

16 the transaction must be admi nistered with the promotion of economic development and welfare of developing countries as its main objective; a nd (iii) the funds must be concessional in nature and convey a grant elem ent of at least 25%. Official development assistance flows are unique in that the debt se rvice includes no interest payment and the grant portion of these flows is free of repaym ent obligation. Other official flows, as defined by the DAC, include offici al sector transactions with aid recipients that do not satisfy the ODA criteria. Private capital flows accounted for a si gnificant share of resource flows to developing economies in the Western Hemisphe re during the last 30 years. Private capital includes commercial-b ank and trade-related lending, fo reign direct investment and portfolio investment. Bank and trade-relate d lending is comprised of commercial-bank lending and other credits extended by foreign le nders in the private se ctor. The criteria established by the World Bank dictates th at lending by commercial banks that are whollyor partly-publicly owned be ex cluded from bank and trade-related lending. Portfolio capital flows include both equity and bond investment. Portfolio equity flows are measured as the sum of country f unds, depository receipts, and direct purchases of shares by foreign investors who own less than 10% of the voting stock of a firm. Portfolio bond investment consists of a ny bond issues purchased by foreign investors who own less than 10% of the voting stock of the issuer. Foreign direct investment has historically been categorized as a private capital flow despite the fact that direct investment, as defined by the DAC may originate from private or official sources. Another characteristic of FDI is that it takes the form of both equity and non-equity investment. The World Banks comprehensive guide to FDI released in

PAGE 28

17 2003 describes some of the ambiguities that ha ve historically existe d in the measurement and classification of FDI flows and establis hes a clearer definiti on of FDI than had previously been provided. The following discussion draws heavily upon the guide. According to the World Bank, direct investors may be individuals; inco rporated or unincor porated private or public enterprises; associated groups of i ndividuals or enterpri ses; governments or government agencies; or estates, trusts, or other organizations that own direct investment enterprises in economies other th an those in which the direct investors reside. (Liberatori 2003, p.3) Furthermore, the IMF indicates that in orde r for funds to be classified as direct investment, it is necessary to establish a (or show a pre-ex isting) specific relationship between the parties involved in the transaction (Liberat ori 2003, p.2), where the term specific relationship refers to the inve stors possession or acquisition of a lasting interest or an effective voice in the mana gement of a direct investment enterprise. In order to minimize the s ubjectivity of this defin ition, a specific empirical threshold has conventionally been used to separate FDI from portfolio flows. The threshold defined by the IMF dictates that only those entities in which the foreign investor has acquired at least 10% of the ordinary shares or voting power qualify as direct investment enterprises. Foreign direct investments are made through greenfield investment (e.g., creation of new production capa city), joint ventures, or through mergers and acquisitions (e.g., the priva tization of an existing govern ment-owned enterprise). The types of capital covered by direct inve stment include: (i) equity capital; (ii) reinvested earnings; and (iii) inter-company loans. All flows of capital reported in this study are net inflows. This is important because there is a clear distin ction between net inflows a nd net flows. For instance, the term net foreign direct investment flows refers to nonresident direct investment in

PAGE 29

18 the host economy net of resident direct investment abroad. This is not what this study examines. Rather, net inflows of foreign direct investment (with which this study is concerned) refers to nonresident direct i nvestment in the host economy net of foreign direct investment funds withdrawn from th e host economy by foreign investors. When defined in this manner, a negative value for n et inflows of FDI in any given year would indicate that repatriation of FDI capital (which is an out flow) exceeded new inflows of FDI in the host economy during the period. Si milarly, a negative value for net inflows of bank and trade-related lending indicates that principal and interest (P&I) payments to foreign lenders exceeded new loan di sbursements to the borrowing country. Furthermore, a negative value for portfolio equity would indicate a certain level of portfolio capital flight, not that more portfolio capital was in vested by resident entities than received by the host economy. A Method of Characterizing Diffe rent Types of Capital Flows In addition to the definitional character istics provided by th e Organization for Economic Cooperation and Development (OECD), capital flows differ in terms of their implications for both the host country and the investor. In fact, Williamson (2000) states that there are five characteris tics relevant to distinguish ing among the various forms of capital flow: (1) cost; (2) conditionality; (3 ) risk-bearing; (4) access to intellectual property; and (5) vulnerability to capital flow reversal. Th e following discussion of these characteristics draws heavily upon the work presented by Williamson (2000). Cost Official capital, given that it ofte n comes in the form of grants or concessional lending, is tradi tionally considered to be the cheapest form of capital available to developing economies. While ther e is no clear difference between the cost of commercial-bank lending and portfolio bond inve stment, it is generally recognized that

PAGE 30

19 the cost of each is less than official capital. Portfolio equity returns tend to be highly variable, fluctuating with the performance of stock markets. However, given that markets typically demand an equity risk pr emium, it is generally accepted that the required return on portfolio equity investment is, on average, higher than the return on foreign debt. Meanwhile, foreign direct invest ment is believed to be the most expensive form of capital for developing countries. As evidence of this, Williamson points out that a conservative estimate of the annual re turn on United States FDI is about 12.4%. Conditionality Developing countries are often required by the lending multilateral development bank (MDB) to establish a clear set of macroeconomic policy objectives in order to qualify for concessi onary loan funds. In addition, development funds tend to be tied to the design and implem entation of specific development projects. Thus, disbursements of official development assistance (ODA) are considered as highly conditional. Furthermore, the negotiations that often take pla ce between multinational firms contemplating investments and host countries (e.g., performance requirements imposed by the host) result in FDI also being highly conditional, al though perhaps not to the extent of official flows. Other forms of capital are typically free of these types of conditions. Risk-bearing Risk-bearing refers to who reap s the benefits (or foots the bill) in the case of unexpectedly high (o r low) revenues and costs. Here, risk is broken down into commercial risk, interest risk and exchange rate risk. Virtually all official capital, with the exception of investments made by th e International Finance Corporation, convey commercial risk to the borrower. The same is true of bank loans and portfolio bond

PAGE 31

20 investment, except in the extreme case of le nder bankruptcy. In the case of portfolio equity investment and FDI, commercial risk is typically assumed by the investor. Interest risk is also born solely by the investor in the case of equity investment, whether portfolio or direct. For official flow s and loan capital, the placement of interest risk depends upon the terms of the loan. Fixed interest rates place the burden on the lender while floating rates convey risk to the borrower. Given that most official loans carry fixed interest rates, the lender (in most cases a multilateral development bank) usually bears the risk of changes in the mark et interest rate. Both bonds and commercial loans tend to made at floating interest rates, and therefore the burden is typically assumed by the borrower. Exchange rate risk, in the cas e of equity investment, is usually born by the investor. Given that developing countries are rarely able to borrow la rge sums of money in their local currency, exchange rate risk is typica lly conveyed to the borrower of official funds and commercial loan capital. However, this risk may be transferred to the lender if borrowed funds are denominated in the domestic currency. Intellectual property Since the pioneering work on FDI by Stephen Hymer (1976), it has been commonly asserted that dir ect investment by foreign firms brings with it the transfer of technologi cal know-how in the form of patents, trademarks and managerial expertise. As pointed out in Ch apter 1, these externaliti es associated with FDI are often referred to as sp illover effects. The conditions associated with official capital (e.g., macroeconomic policy requirements) also result in the transfer of knowledge from MDBs and other official entities to host economies. Each of the other forms of capital inflow typically doe s not facilitate the transfer of intellectual property.

PAGE 32

21 Vulnerability to reversal Although firms do shift working balances into and out of different currencies in response to change s in macroeconomic pr ospects, the sum of these transfers is likely to be small relative to the total of sunk capital invested in a foreign economy (Williamson 2000). Thus, FDI is traditionally regarded as being minimally vulnerable to reversal. Official flows also tend to be stable, and have actually been shown to be counter-cyc lical in nature. Private bank and trade-related lending, while usually grouped together, tend to be very different in terms of vulnerability to reversal. Short-term loans are more vulnera ble to reversal than long-term loans, and trade-related credits are almost always disbur sed on extremely short terms. However, the fact that trade credits are constantly rene wed as new trade transactions need to be financed makes trade-related lending more stab le forms of capital inflow. At the other extreme is bank lending, which in recent crises (e.g., the East Asian crisis and the Latin American debt crisis) has proven to be highly vulnerable to reversal. Figure 2-1. Williamsons Key Char acteristics of Capital Flows Cost of Capital Conditionality Risk Intellectual Property Vulnerability to Reversal ODA Loan Portfolio FDI Portfolio Loan FDI FDI Portfolio LoanODA ODA FDI Portfolio Loan FDI ODA Portfolio Loan Low High High High High High ODA

PAGE 33

22 Figure 2-1 summarizes Williamsons fr amework for distinguishing between different types of capital flows. The origin (or center) of the diagram corresponds to low levels of conditionality, risk bearing, etc., while the outer limits represent the highest levels. The figure shows reiterat es the fact that FDI tends to be conditional and, at least in theory, facilitates the transfer of technology. In cont rast, portfolio equity investment brings with it no conditionality or transfer of knowledge. The cost of equity investment is typically high, which reflects the investors burden of comm ercial risk, interest rate risk and exchange rate risk. Perhaps most im portantly, FDI is considered the most stable form of foreign capital inflow av ailable to developing host countries. The Structure and Evolution of Capital Flows from 1970 to 20012 Figure 2-2 shows the structure of capita l inflows to most of the developing economies of the Western Hemisphere duri ng the period 1970-2001. What is readily evident in the figure is that private investme nt and lending served as the major sources of foreign financing to these countries fo r much of the period 1970-2000. With the exception of the mid-1980s, inflows of private fu nds were generally 3to 5-times greater than official development assistance and other official flows combined. Looking specifically at the structure of private flows, bank and trade-related lending was the most abundant type of priv ate finance throughout the 1970s and early 1980s. However, this trend came to an abrupt end as the Latin American debt crisis set in. As Figure 2-3 shows, repayments of pr ivate foreign debt actually exceeded new lending in 1989 and 1993. The global economic slowdown of the late 1990s also had a detrimental effect on the level of private lending to the developing countries of the 2 All dollar-denominated figures are stated in terms of current US dollars. These figures are converted from domestic currencies using single-y ear official exchange rates.

PAGE 34

23 Western Hemisphere. Similarly, net inflows of portfolio capital were rather unstable, displaying the greatest year-toyear volatility during the 1990s. 0 30 60 90 120 1501 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0Billions of US Dollars Private Capital Flows Official Development Assistance Other Official Flows Figure 2-2. Capital Flows to Developing Economies in the Western Hemisphere (15) 0 15 30 45 60 75 901 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0Billions of US Dollars Foreign Direct Investment Bank and Trade Related Lending Portfolio Investment Figure 2-3. Private Capital Fl ows to Developing Countries in the Western Hemisphere In contrast to bank lending and portfolio investment, inflows of FDI grew during the period 1970-2001; a majority of the grow th occurred in th e period immediately following the debt crisis. Inflows of FD I stood at 0.99% of GDP in 1993, while they accounted for just under 5% in 1999. However, the most remarkable trend is that inflows

PAGE 35

24 of FDI, which once represented less than 20% of the size of priv ate lending, ended the period at over 50-times the magnitude of private lending and portfolio investment combined. Overall, the information presented in Fi gures 2-2 and 2-3 provides a reasonably good characterization of how the structure of capital inflow s to developing countries across the globe has evolved over the last th ree decades. However, some of the most interesting inferences are revealed by comp aring the cases of i ndividual countries. Accordingly, the remainder of this chapte r is devoted to examining the different experiences of developing countries in the Western Hemisphere. The Andean Community The Andean Group was established in 1969 by the Agreement of Cartagena and originally included Bolivia, Columbia, Ec uador, Peru and Chile (Andean Community General Secretariat 2003). Venezuela b ecame a party to the Agreement in 1973 and Chile subsequently withdrew in 1977. The Group was primarily established as a reaction to the poor performance of the Latin Ameri can Free Trade Agreement, but by the late 1980s, commerce orchestrated by the Agreemen t amounted to no more than 5% of the combined trade of the groups members (Hanratty and Meditz 1989). The Group's adoption of Decision 220 in 1987 loosened foreign investment regulations, allowing greater freedom for the re patriation of profits, a higher percentage of foreign ownership and investment in a wider variety of industries (Hanratty and Meditz). The name of the organization was changed to the Andean Community as pat of The Trujillo Amending Protocol of 1996. In addition to a common external tariff and rules to prevent distortions in competition, the present-day agreement includes measures to guarantee international investments.

PAGE 36

25 As Figure 2-4 shows, capital flows to the Andean Community have, for the most part, been dominated by private investment since 1970. Official development assistance flows to the group were less than 1% of GD P in every year except 1991, and even then they only reached 1.10% of a ggregate output. Similarly, other official flows (OOF) exceeded 1% of GDP only twice in 1990 and 1992 and never reached higher than 1.32% of GDP. Overall, both ODA and OOF (a s a percentage of GDP) were generally flat between 1970 and 2001 and never accounted for more than 0.5% of GDP. -2% 0% 2% 4% 6% 8% 10%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Private Capital Flows Official Development Assistance Other Official Flows Figure 2-4. Capital Flows to the Andean Community Figure 2-5 shows that ODA flows (as a perc entage of GDP) to Bolivia grew from 1970 to 1992. In fact, ODA to Bolivia peaked at just under 12% of GDP in 1992. In dollar magnitude, Peru attracted almost as much ODA as Bolivia. However, private capital flows to the Andean Community as a whole were much more substantial (as well as more volatile) than official flows. Priv ate investment declined sharply during the first half of the 1980s, but after the adoption of Decision 220 these flows expanded just as quickly as they had dropped off. Private le nding was the dominant source of capital for

PAGE 37

26 the group until 1985, when foreign direct investme nt began to grow. The shift in private capital inflows from debt to equity investment is evident in Figure 2-6. 0% 2% 4% 6% 8% 10% 12%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Bolivia Columbia Ecuador Peru Venezuela Figure 2-5. ODA Flows to Members of the Andean Community -2% 0% 2% 4% 6%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Foreign Direct Investment Bank and Trade Related Lending Portfolio Investment Figure 2-6. Private Capital Fl ows to the Andean Community Heavy lending in Venezuela during the late 1970s pushed private lending to the Andean Community to over 4.70% of the group s aggregate GDP. However, repayments of foreign debt actually exceeded new le nding as Venezuela began to service those obligations during the 1980s; in 1987, repayments of private debt by the Community as a

PAGE 38

27 whole exceeded new loan disbursements by al most $1.86 billion, leading to a net outflow of private capital equal to 1.21% of GDP. Net bank lending (i.e., new lending minus interest and principal payments on existing de bt) remained mostly negative until 1993. Then it was Columbia that attracted a majo rity of the private lending in the Andean Community. Although lending to the group in 1997 was actually higher than in 1979, increased economic activity caused bank and tr ade-related lending to account for a much smaller proportion of GDP. FDI flows to the Andean Community a lternated between ne gative and positive values from year to year before stabil izing in the 1980s. However, both FDI and portfolio investment in the Andean Comm unity expanded rapidly following Decision 220. Foreign direct investment surged to over 5% of the groups GDP in 1997 on the heels of a sharp increase in investment in Co lumbia and Venezuela. Despite a continuing surge of FDI in Bolivia, direct investment in the larger countries co ntracted over the next few years and FDI in the Community as a w hole fell from 5.07% of GDP in 1997 to just over 3% in 2001. Organized markets for equity securities ex isted in the Andean Community as early as the end of the 16th century when the emergence of joint-stock companies began in Columbia (Bolsa de Valores de Columbia, 2003) However, foreign portfolio investment in the Andean Community first occurred in Venezuela in 1989 and equity markets in Peru, Columbia and Ecuador be gan to attract foreign portfol io capital in the mid-1990s. The Andean Community is unique in that po rtfolio investment in the group during the last 32 years was split evenly betw een the equity and bond markets.

PAGE 39

28 In summary, private capital was the most significant and most volatile type of financial capital flowing in to the Andean Community over the past three decades. Private lending accounted for the majority of private capital flows to the group prior to the Latin American debt crisis of the 1980s, while FDI and portfolio investment reigned in the latter part of the 1980s and the 1990s. Disbursements of private debt contracted sharply in the 1980s and never recovered to pre-crisis levels. Although inflows of portfolio investment expanded during the 1990s, they paled in comparison to inflows of FDI and private lending. The Central American Common Market The General Treaty of Central American Economic Integratio n established the Central American Common Market (CACM) in December 1960. The agreement went into effect in June 1961 with the completion of the necessary articles of ratification by El Salvador, Guatemala and Nicaragua. Honduras and Costa Rica subsequently acceded to the agreement in 1962. The General Treaty es tablished the Central American Bank for Economic Integration, a common external duty known as the Central American Standard Import Tariff and a free trade area among CACM members. The Integration Industries Convention (Rgimen de Industrias de Inte gracinRII) was also established at that time as a means for governing foreign investme nt practices. In a ddition, the Convention granted special incentives and privileges to firms given "int egration industries" status. Over time, this component of the agreement proved to be the most difficult to implement (Merrill 1994). The CACM integration process was some what successful in the 1960s, but by the end of the decade Honduras and El Salvador had engaged in the so-called Soccer War. This dispute effectively reduced the Central American Common Market to an entity that

PAGE 40

29 existed merely on paper (Merrill 1994). Alt hough it took nearly a deca de to establish an official peace accord between the two countries Honduras was able to negotiate a set of favorable trade arrangements with other CA CM members. The movement toward the economic integration of Central America was rejuvenated in the early 1990s and Honduras officially rejoined the process in early part of 1992. Figure 2-7 shows the evolution of net capit al inflows to CACM over the last three decades. Bank and trade-related lending kept private capital flows to the group above 2% of aggregate GDP during most of the 1970s. Lending subsided and ODA flows increased three-fold throughout the 1980s and into th e early 1990s; ODA reached 7.02% of GDP in 1990. -2% 0% 2% 4% 6% 8%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Private Capital Flows Official Development Assistance Other Official Flows Figure 2-7. Capital Flows to CACM From that point, inflows of ODA decreased and private capital inflows, led by an expansion in FDI, jumped from just over 2% of GDP in 1997 to 5.56% in 1998. Inflows of private capital had subsided by 2001 a nd were once again exceeded by inflows of ODA. Other official flows neve r exceeded 1.5% of aggregate GDP.

PAGE 41

30 As shown in Figure 2-8, the sharp incr ease in ODA in the early 1990s was led by foreign aid flows to an ailing Nicaraguan economy. Foreign aid to Nicaragua reached $837 million in 1991 and accounted for over 50% of the countrys GDP. Flows of ODA fell to 18.2% of Nicaraguan GDP over the next two years, only to rebound to over 48.5% of GDP in 1996 (or roughly $933 million). 0% 10% 20% 30% 40% 50% 60%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Costa Rica El Salvador Guatamala Honduras Nicaragua Figure 2-8. ODA Flows to Members of CACM Although relatively less subs tantial than ODA flows to Nicaragua, flows of assistance to both Honduras and El Salvador were significant during the 1980s and 1990s. As Figure 2-8 shows, flows of ODA to El Salvador reached 10.64% of GDP in 1987 but gradually fell below 2% by 2001. Of ficial development flows to Honduras reached their highest level in 1999, when th ey accounted for more than 15% of GDP. As previously mentioned, private capita l flows to CACM were dominated by bank and trade-related lending during most the of the 1970s. However, as was the case with most of the countries in the Western Hemisphe re, private lending contracted significantly during the Latin American Debt Crisis. Fo reign direct investment in the group, while fairly consistent throughout the 1970s and 1980s, expanded during the 1990s. In fact,

PAGE 42

31 FDI increased from 1.75% of the groups a ggregate GDP in 1997 to over 5% of GDP in 1998. Interestingly, Nicaragua was both the pr imary recipient of ODA funds and one of the major beneficiaries of the expansion in FDI. During the 1990s, major assistance from the International Finance Corporation was directed at implementing new FDI laws and es tablishing an investment promotion agency in Nicaragua. Figure 2-9 shows that FDI flows to Nicaragua in creased from 0.0% to 13.23% of GDP during that decade. In 1998, the privatization of electricity and telecommunications providers in El Salvador contributed to substa ntial inflows of FDI (Sandrasagra 2000). Meanwhile, flows to Costa Rica grew steadily and eventually reached 4.42% of GDP. -2% 0% 2% 4% 6% 8% 10% 12% 14%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Costa Rica El Salvador Guatamala Honduras Nicaragua Figure 2-9. FDI Flows to Members of CACM Bond purchases by foreign investors were in consistent and rarely exceeded 1% of GDP in a given year among the CACM countries. The most remarkable inflow of portfolio funds was in 1993, when invest ment in Honduras rose to 4.37% GDP ($152 million). Flows of portfolio capital to Costa Rica and El Salvador in 2001 led portfolio

PAGE 43

32 investment in CACM to 0.89% of aggregate GDP, the highest level observed in the past three decades. In summary, official development agenci es were the main source of financial capital flows to CACM throughout the 1980s an d most of the 1990s. While the groups ability to obtain private credit was obviously affected by the debt cr isis of the 1980s, FDI as a percentage of GDP increased steadily fr om 1984 to 1998. It is this expansion in FDI that brought inflows of private capital to levels that exceeded inflows of ODA in the late 1990s. However, it is noted that the magnit ude of private inflows and official inflows converged during the first two years of the new millennium. The Caribbean Community and Common Market The idea of regional integrat ion began in the Caribbean with the establishment of the rather short-lived British West Indies Federation of 1958. However, at the end of the four years during which the Federation existed, the 10 members found themselves economically the same as they had been for centuries (CARICOM Secretariat 2003). Although plans for a customs union were part of the original Federation, free trade among the islands in the region was not re alized until the Caribbean Free Trade Association (CARIFTA) cam e into effect in 1968. The Caribbean Community Treaty was signed in Chaguaramas, Trinidad on July 4, 1973 and it was agreed among the four indepe ndent countries of CARIFTA Barbados, Guyana, Jamaica, and Trinidad and Tobago th at the agreement would come into effect in August of that year. The revised Treat y of Chaguaramas officially established CARICOM among these four independent signator y countries. The Treaty also set forth that eight other Caribbean nations Anti gua, Belize (British Honduras), Dominica, Grenada, St. Lucia, Montserrat, St. Vincen t and St. Kitts and Nevis would join the

PAGE 44

33 Community by May 1, 1974. Haiti became the first French speaking Caribbean State to obtain full membership in CARICOM on July 3, 2002. With the exception of Montserrat (which re mains a British territory), each of the signatory CARICOM countries is currently independent from European control. Nevertheless, nearly half of the members achieved independence subsequent to 1975. The obvious problem with analyzing capital fl ows to these countries is the lack of availability, or separability, of such da ta. As a result, seven of the 15 CARICOM countries are not included in this analys is. The remaining members do represent an anecdotal sample of CARICOM, as they acc ount for nearly 75% of the overall economic activity of the group (as measur ed by 2001 GDP). It is note d, however, that the unique experiences of the eight CARI COM countries not included in this analysis may not be characterized by the experiences of the countries that are examined. As shown in Figure 2-10, capital flows to CARICOM were dominated by those from private sources during the first half of the 1970s. However, massive repayments of private debt obligations by Jamaica in 1976 and 1978 led to net outflows of private capital in those years. Throughout the 1980s, official development assistance gradually increased in some of the largest countries in CARICOM (including Haiti, Jamaica, and the Dominican Republic) and ODA flows to the group reached 3.45% of GDP by 1990. As time passed, private investment in CARI COM began to strengthen primarily in Jamaica, Trinidad and Tobago, and the Do minican Republic. By 2001, private capital flows (as a percentage of GDP) were in exce ss of the levels seen in 1970. In fact, net inflows of private capital were almost 10-ti mes the magnitude of net ODA inflows in 2001.

PAGE 45

34 -2% 0% 2% 4% 6% 8% 10%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Private Capital Flows Official Development Assistance Other Official Flows Figure 2-10. Capital Flows to CARICOM Figure 2-11 shows the structur e of private capital flow s to CARICOM. Foreign direct investment was the primary source of private capital for CARICOM as a group during the period 1970-2001. While bank and tr ade-related lending never exceeded 1% of GDP, FDI accounted for more than 7% of GDP in the early 1970s and more than 5% -2% 0% 2% 4% 6% 8%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Foreign Direct Investment Bank and Trade Related Lending Portfolio Investment Figure 2-11. Private Capital Flows to CARICOM

PAGE 46

35 of GDP in the late 1990s. Portfolio inve stment in CARICOM expanded in both 2000 and 2001, reaching 3.15% of GDP. Most of this investment came in the form of bond purchases in by foreign investors in Jama ica, Trinidad and Tobago, and Barbados. MERCOSUR MERCOSUR is an acronym for the S outhern Cone Comm on Market and specifically refers to the South American c ountries of Argentina, Brazil, Paraguay and Uruguay. The Treaty of Asuncin was si gned on the March 26, 1991, effectively establishing MERCOSUR at the end of November that same year. The Treaty was intended to promote economic integration among countries in the region, although Chile and Bolivia were conspicuously absent fr om the agreement (Hudson 1995). Bolivia had originally intended to become the fifth me mber of MERCOSUR, although this accession has never taken place. Furthermore, Chile evaded the agreement on the contention that the other four signatory countri es would have to lower their tariffs to the Chilean level before Chile would join. The Protocol of Colonia, which was signed in January of 1994, specifically addresses the promotion and reciprocal prot ection of investments among the members of MERCOSUR. The Protocol grants nationa l treatment to investments in MERCOSUR signatory countries made by investors from other members of the group. Other protocols to the Treaty of Asuncin cover the defense of competition, the protection of intellectual property rights and dispute settlement. Figure 2-12 shows that for the last thr ee decades financial capital flows to MERCOSUR were dominated by private flows. Paraguay was the only member of MERCOSUR for which ODA represented a subs tantial proportion of GDP, ranging from 0.67% to nearly 3.5%. Although Brazil received more ODA in most years than the other

PAGE 47

36 three members of MERCOSUR combined, the immense size of the Brazilian economy kept annual receipts to under 0.5% of GDP. Other official flows, while slightly higher than ODA, never exceeded 0.75% of the gr oups aggregate GDP in a single year. -1% 0% 1% 2% 3% 4% 5% 6% 7%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Private Capital Flows Official Development Assistance Other Official Flows Figure 2-12. Capital Flows to MERCOSUR Figure 2-13 shows the evolution of privat e flows to MERCOSUR. In much the same manner as the rest of Latin Americ a, private flows to MERCOSUR consisted mostly of debt throughout the 1970s and ear ly 1980s. The boom in FDI began in 1994 and continued through the end of the decade. Brazil and Argentina attracted a vast majority of the FDI flows to the group. In fact, FDI in Argentina quadrupled in the latter half of the 1990s, and FDI in Brazil in 1999 was more than 6-times the level in 1995. Large flows of portfolio capital also poured into both Argentina and Brazil throughout the 1990s; flows to these two countries led por tfolio investment to just over 3% of MERCOSURs aggregate GDP in 1993. MERCOSUR is unique in that portfolio inve stment in the group really began to increase prior to the expansion in FDI, whereas this wasnt necessarily the case in the rest of Western Hemisphere. With the exception of Mexico, the heaviest portfolio investment

PAGE 48

37 in Latin America took place in Argentina and Br azil. Also, differences in the magnitude of private capital flows and official flows are also more obvious in MERCOSUR than in any of the other groups in the Western Hemis phere. However, the gap narrowed slightly in 2001 as private flows to the group droppe d, led mostly by contractions in both portfolio investment and FDI in Argentina. -2% -1% 0% 1% 2% 3% 4% 5% 6% 7%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Foreign Direct Investment Bank and Trade Related Lending Portfolio Investment Figure 2-13. Private Capital Flows to MERCOSUR Mexico and Chile Mexico is a party to severa l free trade agreements with countries and trade groups on both sides of the globe. The most signifi cant of these agreemen ts are the so-called Group of Three (1995) and the 1994 North Amer ican Free Trade Agreement (NAFTA). The Group of Three is a sub-regional ec onomic complementarity agreement which includes provisions for trilateral investme nt flows between Mexico, Columbia and Venezuela. Similarly, NAFTA also provide s for the national treatment of signatory countries investors and their investments.

PAGE 49

38 Figure 2-14 shows that, with the excep tion of four years in the 1980s, private funds were the most abundant type of financ ial capital flowing into Mexico during the period 1970-2001. In fact, Mexico attracted almost twice as much private capital as the Andean Community, CACM and CARICOM combined. Private flows to Mexico consisted mostly of bank and trade-relate d lending until the mid-1980s. However, credit reversals and repayments to foreign lende rs exceeded new borrowing in 1993 by more than $6 billion. -1% 0% 1% 2% 3% 4% 5% 6% 7% 8%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Private Capital Flows Official Development Assistance Other Official Flows Figure 2-14. Capital Flows to Mexico Figure 2-15 shows the structur e of private capital infl ows to Mexico during the period 1970-2001. The figure shows that portfoli o investment poured in to Mexico at an unprecedented rate in 1993, perhaps in anti cipation of NAFTA. Bond purchases by foreign investors amounted to nearly $9 billi on while portfolio equity investment soared to $14.3 billion. Total portfolio investment in 1993 reached 4.87% of GDP, whereas FDI only amounted to just over 1%. Direct inve stment expanded in the latter half of the 1990s and eventually reached 4% of GDP in 2001. Meanwhile, portfolio flows were much more volatile, ranging from -0.38% to 4.29% of GDP.

PAGE 50

39 Chile and Mexico are similar in the fact that ODA accounted for a small portion of total capital inflows over th e last three decades. In the case of Chile, ODA never surpassed $200 million in any single year. Pr ivate capital flows were substantial during the late 1970s and early 1980s, accounting fo r anywhere between 9% and 12% of GDP annually. Nearly all of the private capital flow ing into Chile at this time took the form of bank and trade-related lending. Private credit collapsed during the Latin American debt crisis and for a period of thr ee years in the late 1980s offi cial flows to Chile actually exceeded inflows from private sources. -2% -1% 0% 1% 2% 3% 4% 5% 6%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Foreign Direct Investment Bank and Trade Related Lending Portfolio Investment Figure 2-15. Private Capital Flows to Mexico Following the crisis, FDI in Chile surged to 12.3% of GDP. A slight recovery in private lending and a modest in crease in portfolio investment also contributed to private capital inflows reaching almost 18% of GD P by the end of the 1990s. As Figure 2-16 shows, the new millennium brought a sharp decl ine in FDI, and thus, net private capital inflows as a whole.

PAGE 51

40 -4% 0% 4% 8% 12% 16%1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5 2 0 0 0% of GDP Foreign Direct Investment Bank and Trade Related Lending Portfolio Investment Figure 2-16. Private Capital Flows to Chile Summary of Trends The figures in this chapter provide a graphi cal illustration of both the volatility and evolving structure of capital flows to the developing countries of the Western Hemisphere. As previously mentioned, Schm ukler (2004) argues that the evolutionary process has been shaped by three primary agen ts. First, governments have influenced the structure of capital flows in some instan ces by relaxing restrict ions on the foreign exchange transactions and allowing increased participation by forei gn investors in many sectors. Second, borrowers and investors, in choosing among different forms of financing alternatives, have also played a ro le. Finally, financial institutions have begun to offer a broader range of financing altern atives to investors and borrowers by making use of international equity and debt markets. As a result, the deve loping countries of the Western Hemisphere by and large have become more financially integrated with the developed economies of the world. However, some interesting differences have been exposed in this chapter with regard to the level of private foreign investment in the various countries within the region.

PAGE 52

41 Private capital flows to developing count ries around the globe expanded sharply during the last thirty years. The bulk of this expansion wa s accounted for by increases in foreign direct investment in many Latin American and Caribbean economies. The chapters that follow examine some of the potential reasons for such substantial increases in FDI. Chapter 3 is devoted to reviewing the theoretical considerations of researchers who have examined the issue in the past.

PAGE 53

42 CHAPTER 3 THEORETICAL FOUNDATIONS Foreign direct investment has been exam ined in a number of contexts in the economic literature. The significance of th e topic is illustrated by the variety of theoretical models and frameworks that have emerged from efforts to characterize and explain FDI. The diversity of the empirical literature all udes to the exis tence of many distinct research agendas. An area that re ceives a disproportionate amount of attention (and the area of concern in this study) is identifying the fact ors that drive flows of FDI. The economic literature offers several approaches to examining this issue. Ethier (1986) developed a general equilibri um approach to the determination of FDI flows, addressing the need to understand how the founding principles of neoclassical trade theory relate to the issue of foreign investment. Similar m odels were developed by Helpman (1984) and Markusen (1984). While th ey are theoretically elegant, efforts to test these models empirically have met with limited success. In f act, a review of the empirical literature on FDI uncovers little in the way of general eq uilibrium comparativestatic analyses. A second approach to the determination of global FDI flows is rooted in the theory of industrial organization and is aimed at explaining why individual firms make investments in foreign countries to produ ce the same goods as they produce at home (Blomstm and Kokko 1997). Earl y theoretical work in th is area, including Hymer (1976) and Vernon (1966), focused on the firm -specific characteris tics that make FDI more attractive than exporting (Buckely and Casson 1998b). Meanwhile, Dunning (1977,

PAGE 54

43 1995) expanded the set of factors to include host-country-specific variables as well as firm-specific characteristics. Empirical rese arch along these lines has shown that tariff and non-tariff barriers to trade, as well as the legal, politic al and economic conditions of host countries have a significant effect on investment decisions (Davidson 1980). In a third body of literature, researchers model FDI flows as a function of either the global supply of investment funds, the host countrys demand for investment funds, or some combination of the two. In one exampl e of such an analysis, Dasgupta and Ratha (2000) developed a two-stage approach that addresses both the supply and demand for investment funds without rely ing on general equilibrium. In the first stage, the global supply of FDI is determined by so-called push factors1 as investors in developed countries decide how much capit al to invest in the develo ping countries. In the second stage, the global supply of di rect investment funds is given, while country-specific variables2 (the so-called pull factors) dete rmine each developing countrys respective share of FDI. The remainder of this chapte r is devoted to comparing two widely-cited streams of theory that are often used to explain the gl obal allocation of FDI fl ows: neoclassical economic theory and strategic management theo ry. This will serve to illustrate the perspective from which the empirical sec tion of this study is approached. The methodology developed here is that forei gn direct investment, both vertical and 1 Variables considered in the first stage of the pro cess include global market growth (proxied by world trade as a percentage of world GDP), world GDP growth, GDP growth in developing countries as a whole, the real LIBOR rate and an index of privatization. 2 Variables examined in the second stage include th e current account balance, GNP per capita and private non-FDI flows as a percentage of GDP.

PAGE 55

44 horizontal, is the observable outcome of coordinated international economic activity3. As such, the international allocati on of FDI flows is ultimately dictated by investors as they choose among alternative investment locations based on criteria that are perceived to affect profitability. Neoclassical economic theory pays little attenti on to the actions of firms with regard to resour ce allocation, treating the firm as the proverbial black box into which resources go and out of whic h goods emerge (Demsetz 1997). Conversely, the strategic management stream of literature (which has Coases 1937 theory of internalization at its core) is concerned with explaining the st rategic allocation of resources by firms, thus providing a ro le for the decisions of management. Neoclassical Economic Theory vs Strategic Management Theory The purpose of neoclassical economic theory, with perfect competition as its core, is to understand price-guided, not manage ment-guided, resource a llocation (Demsetz 1997 p.426). The economic firm is generally em bodied in a mathematical function (i.e., a production function) defined in terms of technology. In this cons truct, the usefulness of the theory lies in its ab ility to capture fundamental economic forces and their interrelationships (Taylor and Seale 1999 p.8). The theory illustrates a state of dependency, where production takes place in the firm and consumption and resource supply take place in households (Demsetz 1997). The strategic management of resource supply and output demand conditions is not di rectly addressed by neoclassical economic theory, giving rise to the s uggestion that the economic firm is a black-box about which little is known. 3 Most foreign direct investment is horizontal in th e sense that a majority of the output of the foreign affiliate not intended for export to the parent firms home country (Markusen 1995).

PAGE 56

45 Over time, neoclassical theory has been useful in attempting to explain the domestic, and by extension in the trade literatu re, the international allocation of resources and economic welfare. In neoclassical trade theory, the notio n of the firm is generally that of a purely domestic entity competing via trade with the national champions of other countries (Markusen 1995 p.169). In addition to the assumption of perfect competition, extending the theory to the inte rnational trade of goods and services often requires the researcher to assume the existe nce of comparative advantages in resource supply. The shortcomings of this extensi on have been recognized by many and a more recent research agenda (referred to by Markusen as the new trade theory) is aimed at acknowledging that trade and gains from trade can arise independently of any pattern of comparative advantage (as trad itionally understood) as firms exploit economies of scale and pursue strategies of product differen tiation in an imperfectly competitive environment (Markusen 1995 p.169). Relaxing the assumption of perfect competition does not pose a significant problem for general-equilibrium trade analysis. Howeve r, the definition of the firm in neoclassical trade theory inherently limits its ability to provide any explanation for the widespread real-world phenomenon of th e multinational enterprise. As the boundaries of the firm expand in both geographic and product space, the need arises to address the strategic activities that occur inside th e black-box of the economic firm. A useful example is Calderon-Rossells (1985) attempt to model the effect of foreign exchange rates and production cost s on the MNCs choice between producing in foreign and/or domestic locations. In this model, the multinational firm is composed of a parent (in the home country) and a subsidiary (in a fore ign country). Monopolistic

PAGE 57

46 demand and cost functions for each are assume d. The limitations of this approach are characteristic of neoclassical trade model in that the analysis is limited to a two-country world and requires the assump tion of perfect competition, comparative advantage and zero transportation costs. The strategic management body of litera ture was inspired by Robinsons (1932) pioneering suggestion that assumptions in economic th eory should correspond to conditions in the real world (Coase 1937). Robinson (1932) takes no tice of the disparity between the firm as it is defined in economic theory and the firm as it is described by the plain man. Coase (1937) subsequently set out to provide a definition of the firm that is not only realistic in that it corresponds to what is meant by a firm in the real world (p.386) but is also tractable by th e tools of economic analysis. Coases (1937) seminal paper addressed the motives for organizing domestic assets and labor into a firm rather than ma king use of specialized market exchanges to conduct arms-length transactions. The main te net is that the cost of using the price mechanism associated with market excha nges, along with the co sts of negotiating and concluding separate contracts make it more prof itable to internalize production activities. He stated, It is true that c ontracts are not eliminated when there is a firm but they are greatly reduced (Coase 1937 pp.390-91). Coase contended that elements of th e regulatory regime in an economy also provide incentive for firms to internaliz e operations. Specifically, he contends: Another factor that should be noted is th at exchange transact ions on a market and the same transaction organised within a firm are often treated differently by Governments or other bodies with regulatory powers. If we consider the operation of a sales tax, it is clear that it is a tax on market transactions and not on the same transactions organised within the firm. Now since these are a lternative methods of organisation by the price mechanism or by the entrepreneur such a regulation

PAGE 58

47 would bring into existence firm s which otherwise would have no raison detre Of course, to the extent that firms already ex ist, such a measure as a sales tax would merely tend to make them larger than they would otherwise be. (Coase 1937 p.393) Thus, the Coasian concept of internaliz ation can be described as the firms incentive to internalize imperfect markets wh en the cost associated with transacting internally is lower. However, Coase also identified the tendency for firms to increase in size and diversify geographically as the cost s of decentralization (i.e., communication and transportation costs) decrease, albeit in a pur ely domestic sense. Nevertheless, his work ultimately led to the realization that the fi rms boundaries are set in two dimensions: product space (through vertical and horizontal integration), and geographic space (both regionally and internationally). Pointing once again to the dichotomy between the economists definition of the firm and the firm in everyday speech, Ph elan and Lewin (2000) argue that modern theorists are really attempting to explain the existence of th e corporation. For them, a corporation is broadly define d as an entity comprised of a number of people and other assets, which may have legal status as a co mpany or partnership. This rather broad definition is a reflection of the structural diversity of the real -world organizations that are referred to by modern econo mic theorists as firms. The focus of more recent research has shifted away from the coordination problems originally emphasized by Coase and towards the role of firm boundaries in providing incentives (Holmstrm and Roberts 1998 p.74). Furthermore, strategists have made significant contributions to understanding the benefits of the firm structure as well as how resources affect thei r boundaries (Phelan and Lewin 1999). An extensive body of literature is devoted to examining the rela tionship between firms and so-called hold-up problems, transactions costs, and intellectual property rights. While much of the early

PAGE 59

48 research was conducted in a domestic setti ng, Dunning (1977) provide d a framework that extends the analysis to the internatio nal allocation of ec onomic activity. Dunnings OLI Framework The Coasian theory does not specif ically address the emergence of the multinational corporation. However, a more contemporary agenda of the strategic management literature is to explain the dete rminants of foreign production, and hence, the existence of multinationals. Within this area of research, Dunnings (1977) Ownership-Location-Internaliza tion (OLI) framework receives a disproportionate amount of attention. In an effort to address why firms take ownership positions in foreign markets as opposed to exporting or conducti ng transactions at arms le ngth, Dunning contended that foreign direct investments (F DIs) are made when three c onditions are fulfilled: (1) the firm must posses ownership -specific advantages, those that are internal to the enterprise of the home country, but capable of being used with other resources in the home country or elsewhere (Dunning 1970 p.399). Ownership-specific advantages include organizational and entr epreneurial skills, patents, and firm size (which may lead to both scale econo mies and market power); (2) the host country must posses location -specific advantages, those originating only from the resources of [the home] country but available to all firms (Dunning 1970 p.399). Location-specific endowments incl ude Ricardian-type endowments, i.e., proximity to the point of sale, market size, and availability of natural resources and manpower; as well as the legal and commer cial environment in which resources are used, i.e., market structure, and govern mental legislation and policies (Dunning 1980); (3) and finally, there must be benefits to internalizing foreign production processes. Addressing the two former conditions, D unning (1980) states, The possession of ownership advantages determines which firms will supply a particular market, whereas the pattern of location endowments explains whether the firm will supply that market by exports (trade) or by local production (non-trade) [and hen ce FDI] (p.11). The third

PAGE 60

49 condition addresses the mechanism by whic h firms exploit ownership and location advantages in order to service foreign market s. In subsequent work, Itaki (1991) pointed out that Dunnings early concept of internalization differs from that of Coase in that it is interpreted as internalization of an ownership advantage rather than that of an imperfect market (p.445). This is made clear in the following statement by Dunning: The thesis is that the international co mpetitiveness of a countrys products is attributable not only to the possession of superior resour ces of its enterprises but also to the desire and ability of these enterprises to internalise the advantages resulting from this possession; and that servicing a forei gn market through foreign production confers unique benefits of this kind. (Dunning 1977 p.402) Nonetheless, Dunning subsequently reconciled hi s concept of internaliz ation with that of Coase, modifying it to include internaliza tion of both imperfect markets and ownership advantages (Itaki 1991). Dunning (1980) later revisited his original argument and a dded that the desirability of internalizing foreign production pro cesses could derive from both market imperfections and public intervention. Market impe rfections include uncertainty in future market conditions or government policies, structural imperfections (e.g., barriers to entry, high transactions costs, etc.) and c ognitive imperfections (e.g., unavailability or costly acquisition of information about the pr oduct or service being provided), while two types of public intervention were considered (at least, by Dunning) to be relevant to multinational enterprises. The first concerns the extent to which government intervenes in the production and marketing of public goods by corporations. Dunning (1977) states, the need both to generate innovations and ideas a nd to retain exclusive right to their use, has been one of the main inducements for enterprises to intern alize their activ ities (p.404). In addition, the efficient exploitation of technology often requires co mplementary resources that

PAGE 61

50 cannot be protected by patent (e.g., financial systems, organizationa l skills, marketing expertise and managerial experience). Dunning indicates that the lack of public intervention in the production and marketi ng of these complementary resources also encourages internalization by firms. The second type of public intervention th at promotes internalization is economic policy which tends to distort the internati onal allocation of resources. This includes corporate taxes and policies regarding the re mittance of dividends and other forms of profit repatriation. For example, a multinational enterprise, in an effort to record profits in the lowest tax areas, may find it desirable to control the prices at which intermediary products are exchanged by its international groups. Thus, the fi rm may internalize production across international boa rders and use transfer prici ng in order to avoid higher sales taxes resulting from the exchange of intermediary goods at external-market prices. However, Dunning recognized that as early as the 1970s, more intense governmental surveillance over transfer pricing strategies had begun to erode some of these types of benefits (Dunning 1977). In his summarization of the OLI fram ework, Kurt Pederson (2002) provides a useful figure that illustrates how the presence or absence of location and internalization advantages might affect the mode of internat ionalization preferred by firms. In Figure 21 (which is a reproduction of Pedersons fi gure), +I and +L indicate the existence of internalization and location advantages re spectively, while I and L indicate their absence. In the figure, the firms posse ssion of ownership advantages is assumed. Pederson isolates the role of locationspecific advantages, suggesting that the extent to which these types of advantages ex ist influences the firm s decision of whether

PAGE 62

51 to make a direct investment or to serve the foreign market at arms length. Loree and Guisinger (1995) classify asp ects of the policy environmen t (e.g., corporate tax rates, (O advantages assumed) -L +L -I (1) Simple export (3) Contractual agreements +I (2) Sales subsidiary (4) Foreign direct investment Reproduced from (Pederson 2002 p.6) Figure 3-1. Modes of Internationalization regulations on profit repatriation and forei gn ownership, and investment incentives) as location-specific (dis)advan tages and state that host -country governments have a normative desireto manipulate policies that are thought to affect FDI flows (p.285). In his critique of the OLI framew ork, Itaki (1991) take s issue with the inseparability of ownership and location adva ntages by pointing out that the eclectic framework is weakest when ascertaining whic h items are most decisive in attracting FDI (p.456). Using the case of a technologica l breakthrough made by a foreign affiliate as an example, Itaki contends that the ne w technology would be cl assified by Dunning as an ownership advantage of the parent MNC. However, to the extent that the cost of developing and implementing the new technology is a product of local labor costs in the foreign country, the advantag e of the new technology (in economic terms) may be both ownership-specific and location-specific. While Itakis observations are valid, the purpose of this study is not to make infere nces on which of Dunnings three conditions bears the greatest impact on the FDI deci sion. Rather, the usefulness of the OLI framework lies in its taxonomical illustration of factors that may positively or negatively affect the MNCs decision to inve st in a given foreign country.

PAGE 63

52 Foreign direct investment has been mode led by some researchers as the end-result of a multi-stage process4. Although the context varies, th e same analogy is useful in setting a backdrop for this study. In the fi rst stage, firms develop the initiative to establish some form of international pres ence. This initiative could derive from conditions specific to the home country, global economic conditions, or a mix of the two. In the second stage, the firms management makes decisions on the mode of organization and location for its international operation. Here, it is argued that managers consider location-specific attributes like the overall inve stment climate, market size and barriers to trade (Dunning 1973). The third stage of th e process involves making detailed economic profitability forecasts given a prospective lo cation for the foreign affiliate. Important factors in this stage include the level of industrial concentration and existing competition (Miller and Weigel 1972). The literature review provided in this chap ter is intended to provide insight into the first stage of the process desc ribed above, while the empirical considerations of this study pertain more directly to the second stage. Si mply put, different aspects of the investment climate are analyzed to determine their effect if any, on the extent to which foreign firms in general have found each country to be a suitable or unsuitable business location. However, one should remain cognizant of the fact that a firms decision to operate in a foreign market does not necessa rily result in foreign investment. Contractor and Kundu (1998) provide clarification of this point by stating Until a decade ago, local ad aptation by global firms was expressed by varying their business practices and methods in each country, while leavin g the ownership and organizational structures fairly invarian t across nations. Today, the modal choice issue has gone beyond the internalize or not question (Buckley and Casson, 4 See Miller and Weigel (1972) and Barre ll and Pain (1996) for two examples.

PAGE 64

53 1976), and even beyond the licensing vs. joint venture vs. merger set of alternatives (in Buckley and Casson 1996), to include other types of [noninvestment based] alliances, such as management service contracts, and franchising. The general modal choice set now includes varying levels of equity ownership, as well as several alliances of various descriptions. The manager must, today, choose from a larger set of options. (Contracto r and Kundu 1998 p.353). Thus, the literal intent of this study is to examine how firms respond to host-country investment-climate conditions, with specific a ttention given to the affect on the decision to participate via an equity -based mode of organization. In an attempt to understand how the i nvestment climate affects FDI decisions, Stobaugh, Jr. (1969) specified the following basic approaches companies take when analyzing the investment climate of potenti al host countries: (1) go-no go, (2) premium for risk, (3) range of estimates, and (4) risk analysis. The two former approaches are the least complex with regard to the required i nvestigation and calculations, while the latter two involve highly complex anal yses. Nevertheless, it is useful to illustrate how elements of the investment climate might a ffect the decisions made by firms who use the two simpler approaches (i.e., go-no go and premium for risk ). For Stobaugh, managers who use the go-no go approach sometimes use it in conjunction with the premium for risk approach and typically base the investment decision on one or two characteristics of the host country. For instance, suppose the manager is faced with a high probability that the host country curre ncy will devalue, a condition that directly affects the projected prof itability of the venture. In this case, the risk of devaluation might lead the manager to reject the investment alternative (in other words, the manager chooses the no go option) On the other hand, if the probability of devaluation is not high enough to warrant a re jection on the initial screening, the manager might require a higher return on investment give n the risk of devalua tion, thus requiring a

PAGE 65

54 premium for the risk. In either case, a host country facing a high probability of devaluation is less likely to attract FDI than a country with a low probability. This example provides an understanding of how just one element of the investment climate (fluctuations in foreign currency va lues) might affect the outcome of the FDI decision and can be extended to account for othe r investment climate factors as well as firms who use alternative inve stment approaches. Other examples of location-specific factors that could influence firms looking to invest in a specific foreign country might include the level of political (in)stability, the nature of the corporate tax regime, the extent to which foreign firms are protected fr om expropriation of assets and the level of business-related corruption. It is easy to imagine how each of these fact ors could lead a foreign firm to either reject or accept an investment project on th e basis of its perceived level of risk. However, theoretical research into the foreign investment behavior of firms also suggests that other factors often play an important role in the decision process. For instance, the existence of resources with strong global de mand in a particular country might attract foreign investment irregardless of conditi ons that would otherwise translate into a prohibitive level of risk for foreign firms. Alternatively, previous research has also shown that the level of FDI may be related to the other types of capital (e.g., official development assistance and private foreign lending) flowing into a given country. The next chapter focuses on how empirical researchers have examined the factors that affect the FDI decisions of multinational firms. Specific attention is given to the strategic management stream of literature. Later, in Chapter 5, each of countries

PAGE 66

55 examined in this analysis are analyzed for th e extent to which othe r considerations might dominate the effect of the investment climate.

PAGE 67

56 CHAPTER 4 EMPIRICAL CONSIDERATIONS The literature review presented in the prev ious chapter highlighted the distinction between neoclassical economic theory and stra tegic management theory as each relates to the direct investment activities of the multin ational corporation (MNC). The purpose of this chapter is to develop an empirical mode l that can be used to analyze FDI flows to selected developing economies in the West ern Hemisphere. The chapter begins by reviewing several empirical studies specifica lly concerned with the determinants of foreign direct investment as well as the fact ors that affect the mode of organization (i.e., ownership structure) and the lo cation of FDI. The chapter en ds with an in-depth look at the variables that are analyzed in this study. The MNCs Motivations for Making Direct Investments It should be recognized that Dunnings OLI triad represents only one of many attempts to understand why MNCs establish foreign operations. While some of the alternative theories have gained impetus, mo st are derived from the same motivations and often yield results that tend to reiterate Dunnings contentions. For example, in a direct comparison of portfolio theory and the OLI framework, Morck and Yeung (1991) found that the incentives for FDI derive from the be nefits of internaliz ing foreign operations rather than benefits associated with international diversification of corporate portfolios. Morck and Yeung (1991) examined the extent to which multinationality affects a firms market capitalization (i .e., market cap), and thus th e net worth of the firms shareholders. Assuming that financial market s operate efficiently, the MNCs market cap

PAGE 68

57 ( V ) can be measured as the sum of its tangible ( T ) and intangible ( I ) asset values. Citing a host of earlier studies (including work by Dunning, Vernon, and Modigliani and Miller), the authors set out to determine the extent to which a firms degree of multinationality1 impacts its intangible asset value. Expenditures on R & D and advertising were used to contro l for the effect of intangible as sets like technical expertise and consumer goodwill. The authors also cont rol for different industries by including a series of three-digit SIC2 code dummy variables. Ordinary least squares (OLS) regression results indicated that the degree of multinationality was positively correlated w ith the MNCs share price and that the relationship was highly significant. On average, a firm with five or more subsidiaries was found to have a share price 8.41% higher than that of a firm with less than five foreign subsidiaries. More importantly, Morck and Yeung s howed that the impact of spending on intangibles like advertising or R & D also increased with the degree of multinationality. The implication of these findings, according to the authors, is that the value of multinationality is not derived from the in ternational diversification of risk, tax advantages or relative produc tion costs. Rather, the firm s possession of (ownershipspecific) intangible assets serves as a neces sary condition for FDI to positively affect the firms market value. Thus, the results call into question the notion that investors value 1 The degree of multinationality is measured by the nu mber of subsidiaries the firm has, the number of foreign nations in which the firm has subsidiaries and a series of dummy variables representing various levels of FDI (Morck and Yeung 1991). 2 Standard Industrial Classification

PAGE 69

58 the MNC as a means for international portfolio diversification, and lend credibility to the theory of internalization. Pugel (1981) tested the eff ect of four ownership-speci fic variables the possession of proprietary new technology, marketing ab ility and expertise, organizational and managerial technique, and the ability to obt ain capital at favorable rates on outward U.S. FDI intensity, which was measured as the sh are of U.S. subsidiary profits in the total after-tax industry profits in the host country. He used the proportion of scientists and engineers in the total workforce as a proxy for the extent to which new technology is generated through R & D. Advertising intens ity in the industry served as a proxy for marketing abilities and the importance of or ganizational and managerial technique was measured by the share of total employment in the industry accounted for by managers. Finally, the amount of capital n ecessary to establish a factor y of minimum efficient scale was measured by total assets (net of depr eciation) multiplied by the location-specific importance of scale economies (the meas urement of which is explained below). Using data roughly corresponding to the 3digit SIC level on U.S. manufacturing industries, Pugel found that all four ownership advantages favored outward FDI. He also included two location-specific independent va riables representing centralizing and decentralizing agents. Namely, he examin ed the importance of scale economies in production measured as the average size of the largest plants pr oducing half of the industrys output divided by total industry sh ipments and the magnitude of transport costs. The four-firm concentr ation ratio (C4) of each i ndustry was also included as a measure of oligopolistic rivalry.

PAGE 70

59 Pugel specified a double-logarithmic regres sion equation and used OLS to estimate the relationships between the dependent and expl anatory variables. As such, the resulting parameter estimates can be inte rpreted as elasticities. The importance of scale economies was shown to hinder outward FDI, as was th e magnitude of transport costs, although the effect of the latter was statistically insignificant. The C4 variable was positively related to FDI, indicating that oli gopolistic reaction leads to more FDI than would otherwise occur in an industry (Pugel 1981, p.226). Pugel takes this finding as an indication of the fact that firms operating in industries wher e production concentra tion is high tend be more likely to protect their market share in foreign countries by esta blishing a subsidiary. The findings of these studies suggest that a firms FDI decision is impacted by the need to protect proprietary information and expertise. However, the nature of this relationship has been a source of debate among theorists. Along these lines, Lee and Mansfield (1996) found that the adequacy of a developing countrys system of intellectual property protection (a location-spec ific variable observed by direct survey of a random sample of U.S. firms) affects bot h the volume and composition of FDI in a country. Least squares estimates on outward FDI data from 100 major U.S. firms in 14 developing countries suggest th at if the percentage of fi rms regarding protection in a particular country as inadequa te falls by 10 points, U.S. fore ign direct investment there might increase by about $140 million per ye ar (Lee and Mansfield 1996, p.185). Other theorists have argued that a lack of inte llectual property protection may favor FDI as opposed to licensing or contracting. In cont rast, the empirical results presented by Lee

PAGE 71

60 and Mansfield suggest that a MN Cs incentive to invest direct ly is positively related to the level of intellectual propert y protection in the host country. Decisions on the Mode of Organization The study by Lee and Mansfield provides an example of how the scope of research on FDI has broadened over time from anal yzing the export versus FDI decision to focusing on the incentives of internalization as opposed to licensing, subcontracting or franchising. According to Buckley and Ca sson (1998), the 1990s witnessed a renewed interest in why certain circumstances seem to favor some modes of market entry over others. They make the following statement wh ich, in fact, embodies the perspective from which this study is approached: Entry [into a foreign market] involves tw o interdependent de cisions on location and mode of control. Exporting is do mestically located and administratively controlled, foreign licensing is foreign located and cont ractually controlled, and FDI is foreign located and administra tively controlled. (Buckley and Casson 1998, p.541) A review of the empirical literature on alternative modes of foreign entry and ownership turns up studies on the decision betw een greenfield investment or mergers and acquisitions (see Zejan 1990, and Hennart and Park 1993), joint ventures versus whollyowned subsidiaries (see various works by Contractor and Lorange, and Beamish and Killing), and more broadly, the decision be tween foreign equity investment and nonequity contractual alliances. Researchers have found that many of the same variables play a role in each of these distinct decisions. The theoretical model of entry strategi es developed Buckley and Casson (1998) yields the generalized result that high transaction costs cau se firms to favor FDI over subcontracting and licensing if the cost is associated with arms-length technology transfer, and over franchising if the cost is a ssociated with the arms-length intermediate

PAGE 72

61 output market. However, empirical work by Contractor and Kundu ( 1998) indicates that the choice of entry mode depends upon much mo re than transactions costs, pointing out the relevance of agency costs as well as the importance of host country-specific variables and firm characteristics. The authors provi de an interesting st udy on the international hotel industry, a sector in which foreign equity ownership is at least as widespread as non-equity modes of organiza tion (Contractor and Kundu 1998). Contractor and Kundu specify their dependent variable as the m ode of organization and allow for four different modes: (i) fully owned, (ii) partially owned, (iii) management or service contract and (iv) franchise. Host country-speci fic determinants included as independent variables are Frost and Sullivans Composite Risk Index, the level of foreign business penetration (measured by FD I/GDP), the level of development (i.e., GDP per capita), and a measure of cultu ral distance between the home and host countries3. The most interesting result of the anal ysis is that foreign equity-based modes are less likely in risky countries and more likely in lower income countries. The results reached by Contractor and Kundu were supported by survey questionnaire responses reported by Kim and Hwang (1992), who surveyed 96 multinational managers on the importance of host-country environmental factors to the decision between three modes of entry (i.e., licensing, joint venture and wholly-owned subsidiary). The survey results indicated th at perceived country ri sk and the level of unfamiliarity with the host country are more important environmental factors than the uncertainty of demand and intensity of comp etition in the host country. Specifically, 3 Firm-specific structural, strategy and control factors are also consider ed by Contractor and Kundu but are less relevant to the intent of this study and are omitted for the sake of time.

PAGE 73

62 higher levels of risk and unfamiliarity were associated with lower levels of resource commitments and hence lower levels of FDI. As part of the vast liter ature inspired by Dunning seminal work, Agarwal and Ramiswami (1992) examined the interrelati onships between ownership, location and internalization (OLI) factors a nd the choice of entry mode. The authors pointed out that while there is a significant body of research devoted to examining the effect of OLI factors, the interrelati onships of the three types of variab les have largely been neglected. They concentrated on overseas leasing data from the U.S. equipment leasing industry and specified the OLI triad as Ownership advantages such as firm size, multinational experience and the ability to develop differentiated products Location advantages such as market pot ential and level of investment risk Internalization advantages such as contractual risk. Overseas leasing is carried out by foreign financiers who either export the loaned funds from the home country or establish a foreign leasing subsid iary. Furthermore, foreign leasing subsidiaries are establishe d through contractual arrangement, setting up a new foreign entity, or by direct investment into a foreign l easing company. The authors analyzed survey data consisting of respons es from Presidents and CEOs of 97 U.S. leasing firms on the effect of the OLI factors. The interacting relationships estimated by Agarwal and Ramiswami were obtained from l ogistic regression and chi-square analysis. The major findings were as follows: (1) large firms show a preference for investment modes of entry in both low and high potential markets; (2) small firms with limited multinational experience prefer the join t venture mode of entry in markets with high potential; (3) firms with greater ability to develop di fferentiated products favor

PAGE 74

63 foreign investment over exporting when contract ual risks are high; and, (4) even in high potential markets, substantial investment ri sk leads both small and large firms to export instead of investing. The results presented by Agarwal and Ra miswami highlight an important fact. That is, although the empirical methods of each study mentioned above vary, they share a common result. That is, host-country locati onal factors and their interactions with ownership-specific factors play a key role in determining the MNCs choice of entry mode (i.e., non-investment vers us equity investment-based). The investment climate, or at least investment risk, also appears to have a substant ial effect on th e decision of whether or not to establish ope rations in a foreign location. Decisions on Location While many studies (includi ng those cited thus far) have examined the effects of ownership and internalization factors on th e FDI decision, a large body of literature is specifically devoted to analyzing the locat ion decision. This is typically done by modeling the cross-country dist ribution of FDI flows from a particular country (flows from the U.S. and Japan have received a disp roportionate amount of attention), groups of countries (here, flows from the EU are the primary concern), or th e entire universe of countries. From a host-countrys perspective, it is important to understand how firms determine the geographic location of FDI. However, a randomly-selected group of managers would likely point out a diverse collection of factors that affect this decision. For instance, the manager of a labor-intensive firm may value low wage rates while the management of a capital-intens ive firm might tolerate higher wages in exchange for lower interest rates. Nevertheless, analyzi ng the distribution of out ward FDI flows does provide some insight into how firms choose foreign investment locations.

PAGE 75

64 Mody and Srinivasan (1997) analyzed th e allocation of investment funds by Japanese and U.S. investors by concentrati ng on the factors that caused investors to differentiate between countries They treat the FDI decision as a two-step process in which the foreign investor first decides how mu ch capital to invest abroad and then how that pool of investment funds will be allocate d across countries. Separate equations were estimated for each supplier country, with the de pendent variable specified as the share of FDI host-country i received in each time period. Th e authors controlled for factors like proximity in distance and methods of conducting business wh ile testing for the effects of location-specific variables like market size, the price of labor and capital, the corporate tax rate, trade propensity, country risk, infras tructure, education and the existing stock of FDI (the latter of which was incl uded as a measure of persistence). Several sets of regression coefficients were estimated. First, OLS estimators were obtained, although the authors poi nt out that these estimate s are biased when unobserved country effects are correlated with the obser ved explanatory variable s. Second, the fixedeffects model was used to eliminate the infl uence of unobserved country characteristics, thus providing coefficients that reflect [the] responsivene ss of foreign investment to changes within a country, over time (Mody and Srinivasan 1997, p.784). Third, between-estimators were provided with the inte nt to capture the vari ation in FDI shares across countries4. While the authors recognize that each of these regression techniques has the potential to result in biased estimat ors, they indicate th at the results provide different perspectives and can be used in conj unction to describe the variation in the data. Finally, generalized least squares (GLS) estim ators are obtained from a random-effects 4 Mody and Srinivasan (1997) interpret the fixed-eff ects estimators (or withinestimators) as short-run effects and the between-estimat ors as long-run effects.

PAGE 76

65 model. The authors point out that the GLS estimates represent a weighted average of withinand between-estimators, thus providing the best composite picture. Country infrastructure and primary school enrollment rates (a measure of labor quality) were found to be major influences on location only. Meanwhile, past investment in a country affected both the timing and location of FDI. Corporate ta x rates, the cost of investment, country risk and wage inflation were not signi ficant determinants of the location decision, although the la tter two did effect the timing of investment. Finally, the authors found that the determ inants of investment locat ion for Japanese and U.S. investors had converged over time. As is the case with the model presente d by Mody and Srinivasan, other studies on the policy determinants of FDI location pr imarily concentrate on the effect of hostcountry corporate tax rates. Loree and Guisinger (1995) indica te that this is primarily due to the difficulty of capturing the incidence of other policy variables. However, data compiled by the U.S. Department of Commerce and the Internal Reve nue Service in 1977 and 1982 provided information on host-country investment incentives and performance requirements, in addition to corporate tax rate s. Loree and Guisinger (1995) tested the effect of all three policy variables on the loca tion of U.S. direct investment abroad. As the authors point out, any empi rical study that attempts to model and observe the effects of policies requires the inclus ion of non-policy variables as controls of alternative explanations (p.285). Thus, they included GDP, infrastructure, and country risk as independent control variables. Loree and Guisinger made two interesti ng distinctions with regard to the measurement of variables included in their model. First, the authors indicate that total

PAGE 77

66 FDI flows include equity investment, reinvest ed earnings of foreign subsidiaries, and other longand short-term capital flows. Consequently, they specified the dependent variable as the equity compone nt of FDI, as opposed to total FDI flows. Second, they point out that models intended to m easure the response of FDI flows to changes in the absolute level of the dependent variables of ten rely on the assumption that the stock of FDI in each country is at equilibrium at the beginning of each period. Citing the difficulty of maintaining such an assumption, the authors chose to measure the variables in absolute levels. Loree and Guisinger also include a dummy variable for the host countrys status as developed or developing (as classified by the OECD). They justify the inclusion of this variable by pointing out that differe nces in the pattern of liberalization among developed and developing countries suggest that the affects of policy variables in each class may differ. So, in addition to testi ng the aforementioned relationships, the authors tested for differential effects of polic y variables among developed and developing countries. OLS coefficient estimates were repo rted for both policy and non-policy determinants of U.S. FDI location in 1977 a nd 1982. While the result s suggest that both play a significant role, policy variables may be more important since th ey are able to be altered quickly while non-policy variables take months, sometimes years to change (Loree and Guisinger 1995). Although most of the non-policy variab les had the expected signs, only the effect of infr astructure was statistically si gnificant in both years. The authors indicate that the insi gnificance of the other non-polic y variables is most likely

PAGE 78

67 due to industry specific characteristics that can only be captured by segmenting the FDI data. However, the interaction of the policy variables and the development-status dummy provide some interesting inferenc es. First, host country performance requirements had a negative influence on e quity FDI flows to both developed and developing countries. Second, investment in centives had a positive influence on equity FDI in developing host countries only5. Finally, the most significant effect was that of the corporate tax rate, which was negatively correlated with U.S. FDI in both periods irregardless of development status. Woodward and Rolfe (1993) concentrated on export-oriented FDI activity in the Caribbean Basin over a four-year period in the mid-1980s. The data available to the authors allowed them to examine the effect of location-specific characteristics on manufacturing plant openings in the region. It was assumed that investors in the region would choose one of 16 location alternatives by analyzing the level of profit their firm could achieve by locating in that country. Using conditional-logit analysis, the authors found that GNP per capita, the length of tax holidays and the likelihood of an exchange rate devaluation each had a large, positive, and statistically significant effect on locational probability. Significant negative relationships were found for wa ge rates, restrictions on the repatriation of profits, in flation and transport costs. 5 Loree and Guisinger indicate that this may be due to competitive response from other countries ultimately ending in the prisoners dilemma trap where all countries increase their [investment] incentives simultaneously but no country increases its relativ e share of foreign investment (1995, p.296).

PAGE 79

68 The Empirical Model The empirical analysis in this study draw s upon existing research into the strategic behavior of foreign direct i nvestors while attempting to shed new light on the effect of the investment climate on FDI flows to develo ping countries. The model is constructed in such a manner as to include a set of loca tion-specific factors that are considered by the entities that make FDIs. Relevance to host-country policymakers in the Western Hemisphere is considered paramount. Some researchers have suggested that the locational characteristics that influence foreign direct investors vary according to th e intended purpose of the investment. For instance, market characteristics such as size, grow th in size and income level are most relevant for investment that seeks to access the host market. Other types of investment, such as export -oriented, may be more con cerned with other locational characteristics such as wage rates or policies regarding export levels. (Loree and Guisinger 1995 p.295) In a study on FDI in developing host c ountries, Lecraw (1991) distinguishes between three types of FD I natural resource-seeking, market-seeking, and exportoriented efficiency-seeking. However, he sugge sts that despite the distinct characteristics and motives for each type investment, there exists a set of common locational factors (Lecraw 1991) that influence all infl ows of FDI irrega rdless of the type6. Lecraws view is adopted here. That is, the range of factors considered in this anal ysis are intended to represent a set of conditions f aced by all foreign direct inve stors irregardless of whether 6 Lecraw (1991) indicates th at these factors include: a change in th e corporate tax rate, the growth rate of the labor force, a change in the openness of the host countrys policy toward FDI, a change in the countrys risk rating, a change in the real exchange ra te, and the growth rate of infrastructure.

PAGE 80

69 the purpose of investment is to gain access to natural resources, the domestic market, or to establish an export platform. Many researchers have analyzed bilateral investme nt data as opposed to investment from the entire universe of s ource-countries. The most ofte n cited reasons for using this approach are that the researchers are examin ing the effect of distance between the home and host countries (i.e., the gravity appro ach), or that it is hypothesized that the relationships vary by the nati onality of the investor. Th e gravity approach is not appropriate for this analysis given that the study analyzes flows of FDI from the entire universe of sources. Furthermore, empirical results presented by Mody and Srinivasan (1997) suggest that the factor s that drive the location of FDI have converged over time for at least two of the most prominent sources of direct investment funds in the region: the U.S. and Japan. Given these findings, it is reasonable to think that the use of aggregate rather than bilateral investment da ta does not hinder the validity of the results. Foreign direct investment flows are meas ured in dollars and therefore depend on the size of the economy (Lecraw 1991, p.171). Meanwhile, the independent variables in this study are (at least theo retically) independent of ec onomic size. This makes it necessary to standardize the dependent variable across coun tries. Addison and Heshmati (2003) point out that the tradition in existi ng literature is to specify the dependent variable as FDI as a percentage of GDP. The Host-Country Investment Climate Examining the effect that the host-co untry investment climate has on FDI essentially requires the inclusi on of three types of variables on the right-hand side (RHS) of the regression equation. The first group consists of governmental policies and regulations that affect forei gn investment. Restrictions on foreign investment, foreign

PAGE 81

70 ownership, and access to foreign exchange (i.e ., barriers to profit re patriation) serve as indicators of the host governments attitude toward foreign invest ment. Factors that affect the return on capital to foreign invest ors make up the second group. Variables of this type are the corporate tax rate and cha nges in the exchange rate. The third group is comprised of measures of po litical stability and corruption. Basi (1966) surveyed 160 U.S. firms via questionnaire and asked managers to rank 15 potential determinants of FDI location as either crucially important, fairly important, or not important. In Basis survey, a majority of U.S. firms ranked political stability in the hos t country and the governments general attitude to foreign investment as crucially impor tant. It was also clear that other components of the investment climate, although not weighted as heavily as these two, received a considerable amount of attention from mu ltinational managers. Table 4-1 summarizes Basis findings on the importance of se veral investment climate factors. Table 4-1. Survey Results on the Impor tance of Investment Climate Factors Number of Firms Ranking Determinant as: Foreign Country Investment Climate Factor Crucially Important Fairly Important Not Important (1) Favorable Attitude Toward U.S. Investments 76 69 13 (2) Political Stability 101 58 1 (3) Tax Structure 30 101 27 (4) Stability of Currency Exchange 61 90 8 (5) Inflationary Trends 35 99 21 The Policy and Regulatory Environment Governmental policy towards foreign invest ment has attracted a disproportionate amount of attention over the last 10 to 15 years. In fact, Lecraw ( 1991) points to several examples where changes in host-countries at titudes toward foreign investment led to

PAGE 82

71 dramatic changes in the magnitude of inward FDI (such as in China and Korea during the 1970s and 1980s). Several measures have been proposed in the literature to proxy for the regulatory environment foreign investors face when conducting business in a given country. In response to the need for tran sparent methods of measuring and comparing regulatory regimes across countries, an in creasing number of private and public institutions have begun to pr ovide numerical ratings of the regulatory risks foreign investors face. The International Country Risk Guide (referred hereafter as ICRG ) published by the PRS Group encompasses one of the most comprehensive collections of data on the economic, financial and political risks that investors encount er in developing countries. The ICRG s composite political risk rating is comprised of twelve components measured via subjective analysis of the available political information from individual countries. Of the twelve components, the investment profile score provides the most relevant measure of the regulatory environment facing foreign direct investors. As defined by the ICRG (Sealy 2003), the investment profile is made up of three risk factors: (1) the risk for contract viabi lity/expropriation; (2) the risk for restrictions on profit repatriation; and (3) the risk for paym ent delays. Each of the three factors is assigned a score ranging from 0 to 4, with a score of 4 indicating very low risk and 0 indicating very high risk. The three compone nt scores are summed t ogether to yield an overall investment profile rating ranging from 0 to 12. A score of 4 (the highest possible) for the first component is ta ken as an indication that the host-country judicial system is likely to enforce contracts made between the foreign investor and its domes tic associates, and that there is little risk of unjustified

PAGE 83

72 expropriation of assets by the host-count ry government. A high score for the second component signifies that forei gn parent-companies are relativel y free to repatriate profits and that the level of taxation on repatriated f unds is relatively low. Finally, a high score for the third component signals the existence of legislative protection against undue delay of payment for services or products rendered. Using this methodology, the overall investment profile score is expected to be positively correlated with FDI. In order to better understand the ICRG investment profile rating, it is useful to compare the cases of two countries in which the score differs signifi cantly; Haiti had the lowest investment profile score among the c ountries in the sample (at 5.5) in 2001, while Uruguay had one of the highest (at 11.5). The inefficiency and lack of transparency in Haitis outdated legal system often hinder th e resolution of disputes between foreign investors and domestic parties. In fact there have been more than 10 cases of expropriation of private assets owned by U.S. interests in Haiti over the last 25 years. While the Haitian government publ ically indicated its desire to alleviate these types of situations, there has been little in the way of real action. In contrast, property rights in Uruguay are recognized and protected by a wellestablished and transparent legal system. In terms of dispute settlement, foreign investors are usually given a choice of arbitration or court proceedings. Bankruptcy laws dictate that creditors collect their debts first, followed by employees and then government. There have been no instances of government expropriation in Uruguay in recent history. Thus, it appears that the ICRG investment profile score does a good job of capturing the actual state of affairs at least in the two example-countries examined here.

PAGE 84

73 Political risk Political risk assessment has been one of the fastest growing areas of research in international business . [as] the discip line has flourished in the wake of the international turmoil of r ecent years (Simon 1984, p.123). Assessing differences in political risk levels across countries is a difficult proce ss that often requires highly specialized information and expertise. Thus many MNCs have again turned to outside sources for assistance in anal yzing the political risks i nherent to developing hostcountries. The academic literature is peppe red with econometric models that employ political risk indexes composed by a variet y of public and private entities. Business managers and academic researchers often choose the source that provides a measure that is most relevant to their purposes. This study examines two primary sources of political risk in developing countries: government instability and corr uption. With regard to th e first, Butler and Joaquin (1998) state, Political risk is the risk th at a sovereign host government will unexpectedly change the rules of the game under whic h businesses operate (p.599). While it is possible for an incumbent regime to have a ch ange of heart with regard to its position on foreign investment, it is more likely that pol itical risk arises from an actual regime change. Thus, government stability serves to measure the risk that the rules of the game will change. With regard to the second source, it is sometimes unclear whether corruption in local government serves as a beneficial gre ase, a minor annoyance, or a major obstacle for international investors (Wei and Shleif er 2000, p.303). However, to the extent that corruption results in so-called crony capital ism, it is possible that corrupt government practices at least encouraged the recent Latin American currency crisis by facilitating

PAGE 85

74 the misallocation of financial resources to the friends and relatives of government officials (Wei and Shleifer 2000, p.304) rath er than to their mo st productive uses. It is possible that even in the most moderate cases financial corruption in the form of demands for special payments and bribes connected with import and export licenses, exchange controls, tax assessments, police protection, or loans . may force the withdrawal or withholding of an investment (Sealy 2003, p.A-5). In the most extreme case, the uncovering of corrupt governmental activities could trigger a popular backlash [requiring a major restructuri ng,] or, at worst, a breakdown of law and order (Sealy 2003, p.A-6). For all of these reasons, corru ption is viewed here as a detrimental characteristic of the existing regulatory regime. As previously mentioned, ther e are twelve components to the ICRG composite political risk rating. Scores for government st ability and corruption ar e two of the factors included in the composite rating and are used he re as measures for the perceived level of political risk. The score for political corruption ranges from 0 to 6, with a higher score corresponding to lower risk. This score is reached through a subjective assessment of the extent to which the following conditions ex ist in the political system: excessive patronage, nepotism, job reservations, fa vor-for-favors, secret party funding, and suspiciously close ties between polit ics and business (Sealy 2003, p.A-6). The ICRG s government stability score is a meas ure of the governments ability to stay in office and carry out its declared program(s). As such, the score has three subcomponents: government unity, legislativ e strength and popular support. Each subcomponent is given a value of 0 to 4, a hi gher score indicating a higher level of unity, strength and support. The overall government stability rating is the sum of the three

PAGE 86

75 scores. The ICRG s risk ratings for corruption and government stability are each expected to be positivel y correlated with FDI. With regard to the government stability ra ting, it is useful to compare the cases of Argentina and Uruguay. The ICRG government stability score for Uruguay was 11.0 in 2001, while Argentina received a rating of 5.5. Uruguay had a stable democratic government, and there have been no instances of political violence in the recent past. On the other hand, Argentina was plagued by social tension and political unrest. Violent protests led to the death of at least 25 Argentine citizens from 2000 to 2002. Once again, the ICRG rating seems to provide a reasonably good representation of the conditions that persisted in the two countries used as examples here. Factors that affect the return on investment The paper by Stobaugh, Jr. (1969) illust rates how managers use the projected return on investment (ROI) from a foreign pr oject to discriminate between investment alternatives. As was discussed in Chapter 3, the projected ROI is sometimes used as a basis for passing on an investment opport unity altogether (referred to as the gono go decision), while in other cases managers might require a premium for risk in countries that are perceived as riskier investment locations (the exam ple used in Chapter 3 was the risk of currency devaluation). In either cas e, factors that negatively affect the projected ROI decrease the host countrys probability of attracting foreign investment. Two variables that directly affect the retu rn to foreign investor s are considered in this study. These ROI factors are intended to measure location-specifi c attributes of the host country that affect income in two ways First, the domestic (i.e., within the host country) ROI is affected by the host-country corporate tax rate. Second, the foreign

PAGE 87

76 owners ROI is subject to changes in th e foreign exchange rate. The following paragraphs elaborate on these two points. The effect of income taxa tion is straightforward; ceteris paribus a one percent increase in the corporate tax rate leads to a one percent decrease in the after-tax profit of a corporation. While the corporate tax rate may not have a significant impact on the competitiveness of an individual firm (given that all firms residing in the host country are faced with the same taxes), it may be used by governments as a tool to attract FDI that might otherwise be directed at another country. In fact, some researchers have found evidence that OECD countries do compete with each other over corporate tax rates in order to attract foreign investment. Furt hermore, Loree and Guisinger (1995) indicate that MNCs may choose to allocate a dispropor tionate amount of investment capital into countries with the lowest tax rates. The most commonly used measure for taxation in the empirical literature is the top marginal corporate tax rate, but it is often th e case that foreign firms face different levels of taxation. However, the definition of a direct investment enterprise provided by the World Bank (Liberatori 2003) requires that the firm be classified as a resident entity of the host country. As such, it is assumed here that direct investment enterprises are taxed in a manner comparable to a domestically-owned firm. The top marginal corporate tax rate is used in this analysis as the measure of taxation and is expect ed to be negatively correlated with FDI. The effect of exchange rate regimes on FDI has attracted a significant amount of attention in the empirical literature. A stable and predictable exchange rate has the effect of allowing managers of a parent-company to be confident that they will be able to

PAGE 88

77 transfer funds to and from its foreign affilia te without substantial risk of loss due to exchange. Furthermore, uncertainty caused by exchange rate volatility . makes the domestic value of foreign revenues and costs uncertain (Darby et al. 2000, p.1). That is not to say that domestic firms are not-at-all affected by the ex change rate. Rather, this study is more interested in how it affects th e attractiveness of a host country from the vantage point of a foreign investor. Goldberg (1993) points out three theoretical effects of exchange rate volatility on foreign investment: (i) sectoral profitability e ffects; (ii) location effects; and (iii) wealth effects. First, sectoral profits are subject to exchange rate-ind uced changes in product demand and cost, thus impacting the interna tional competitiveness of industries. Next, exchange rate volatility alte rs the attractiveness of do mestic and foreign production locations, and hence domestic and foreign invest ment levels. Finall y, the distribution of wealth across countries is affected by movement s in the exchange rate, which in turn also alters the demand for domestic and foreign investments. With specific attention given to the location effect, Aizenman (1992) found that aggregate investment was higher in countries wi th fixed exchange rate regimes than those with a flexible exchange rate. This finding led him to suggest that the adoption of a fixed exchange rate could encourage flows of foreign direct investment (Aizenman 1992, p.913). It is important to note that Aizenmen made this finding under the assumption that foreign investors are risk -neutral. Although, it is expected that introducing risk-adversity woul d only compound the negative effect of exchange rate volatility on foreign investment. Aizenman s finding, as well as those presented in

PAGE 89

78 previous sections, lead to the expectation th at exchange rate volat ility will be negatively correlated with net FDI inflow s in the ensuing analysis. Control Variables In addition to testing the effects of th e factors mentioned above, two additional variables are included on the right-hand-side of the equation as cont rols for alternative explanations. Although not spec ifically categorized as el ements of the investment climate, both income and the level of integra tion with foreign economies (also referred to as openness) are considered as traditional determinants of FDI flows (Addison and Heshmati 2003). Following the precedent set in the existing litera ture, productivity is measured by per capita GDP. Total trade (i.e., imports plus exports) as a percentage of GDP serves as a proxy for a countrys degree of openness or integration with the global economy. Each of the control variables is expected to have a positive effect on FDI. A linear relationship between the explanator y variables and the dependent variables is assumed. Chapter 5 addresses the valid ity of this assumption and compares two alternative models that may be used to estimate the data: the fixed-effects model and the random-effects model. A series of test statisti cs are calculated and serve as the basis for choosing the most appropriate method of estim ation. Finally, the re sponse of FDI to changes in the investment climate is exam ined and discussed for each country in the analysis.

PAGE 90

79 CHAPTER 5 MODEL SPECIFICATION, ESTIMATION, AND EMPIRICAL ANALYSIS Data and Sources The data for this analysis were obtained fr om various sources. The foreign direct investment share of GDP, total trade shar e of GDP and per-capita GDP series were obtained from the World Banks World Development Indicators 2003 CD-ROM database. Corporate tax rate data came from the World Tax Database maintained by the University of Michigan Business School1. Exchange rate data were obtained from the International Monetary Funds (IMF) International Financial Statistics website2. Finally, the investment profile, corruption and government stability scores were obtained from the PRS Groups International Country Risk Guide ( ICRG ) data wizard3. Although data on FDI were readily availabl e for many of the developing countries in the Western Hemisphere starting in 1970, a nnual risk ratings were much more difficult to obtain. In fact, the availa bility of these rati ngs limited the sample size to 21 countries observed over the 18-year period 1984-2001. Gross domestic product data for Nicaragua in 1999, 2000 and 2001 were estimated by increa sing the 1998 value by the growth rates reported by the Economist Intelligence Unit. The total number of observations in the 1 The World Tax Database compiles corporate tax rate information from a variety of sources and was accessed at http://www.wtdb.org/ 2 The IMFs International Financial Statistics database can be accessed at http://ifs.apdi.net/imf/ 3 The ICRG data wizard can be accessed at http://www.countrydata.com/wizard/ Data are available for a fee.

PAGE 91

80 Table 5-1. Descriptive Sta tistics by Country (1984-2001) FDI (% of GDP) Investment Profile Corporate Tax Rate (%) Government Stability Mean Std DevMeanSt d DevMeanStd Dev MeanStd Dev Argentina 1.962 1.9115.7501.517 30.5005.102 6.8062.023 Bolivia 3.987 4.2646.4722.841 25.8331.917 6.6112.725Brazil 1.607 1.8925.6940.987 26.1117.962 6.4442.121Chile 4.414 2.8897.4172.211 27.6119.641 7.2502.702Colombia 2.010 1.1215.7501.517 33.1113.341 6.5561.917Costa Rica 2.651 0.7996.8892.055 35.5569.218 7.4171.784Dominican Republic 2.597 1.9296.2782.886 34.77810.395 6.6942.562Ecuador 2.623 1.9235.3891.243 23.3332.425 6.9722.076El Salvador 1.033 2.1125.6392.611 28.0563.888 6.2782.372Guatemala 1.261 1.0785.9442.100 33.2785.497 6.4442.572Guyana 8.018 9.9456.0561.697 48.8896.978 6.2502.451Haiti 0.189 0.2172.7501.611 36.9446.216 4.3332.910Honduras 1.787 1.2086.1941.582 32.2229.111 6.3062.408Jamaica 2.836 2.4126.6392.127 35.2534.485 6.9172.302Mexico 1.959 1.0047.3061.808 36.0562.838 7.6111.720Nicaragua 3.430 4.3834.4721.398 33.6945.396 6.9442.363Paraguay 1.183 1.0067.1391.747 30.0000.000 6.6671.749Peru 1.899 2.3805.8612.071 34.4448.024 6.4172.451Trinidad & Tobago 5.679 4.7966.9721.803 40.1674.396 6.7781.833Uruguay 0.586 0.5087.4171.972 30.0000.000 7.1672.036Venezuela 1.852 1.8655.2781.526 40.4448.906 7.1941.758Sample 2.551 3.5356.0622.159 33.1568.342 6.6692.290 Corruption Exchange Rate Variability (%) Trade (% of GDP) Per-capita GDP (US$) Mean Std DevMeanStd De vMeanStd Dev MeanStd Dev Argentina 3.361 0.724 437.017 857.215 18.145 3.546 7,182.403 779.110 Bolivia 2.222 0.732 876.655 2,367.956 46.890 3.150 878.980 58.705 Brazil 3.500 0.618 463.218 55 8.691 18.466 3.492 4,296.506 235.791 Chile 3.333 0.485 14.669 13.037 59.003 4.790 3,959.624 1,059.139 Colombia 2.611 0.608 21.663 9.293 33.747 4.071 2,181.747 185.551 Costa Rica 4.889 0.471 12.495 4.386 78.088 10.789 3,234.244 419.880 Dominican Republic 3.222 0.548 23.659 33.641 66.303 5.991 1,575.407 249.774 Ecuador 2.972 0.436 46.817 24.846 56.924 7.165 13,519.929 4,008.081 El Salvador 2.750 0.809 8.900 16.008 53.900 9.336 1,524.806 180.581 Guatemala 2.611 0.916 13.913 17.555 40.560 6.677 1,423.258 89.060 Guyana 1.944 0.998 34.741 41.700 186.476 55.493 800.264 107.505 Haiti 1.333 0.840 9.781 14.476 39.201 8.287 433.151 69.767 Honduras 1.972 0.118 13.829 18.849 76.605 18.874 699.081 17.854 Jamaica 2.389 0.502 23.323 25.680 102.561 10.067 2,129.110 158.978 Mexico 2.889 0.471 36.308 39.946 44.711 14.202 3,315.535 234.241 Nicaragua 4.444 0.856 15,341.711 42,745.495 80.364 30.387 473.192 68.740 Paraguay 1.389 1.037 22.985 16.456 66.936 12.738 1,820.112 63.172 Peru 3.000 0.343 584.282 1,302. 757 31.144 4.517 2,225.213 219.885 Trinidad & Tobago 2.833 0.383 6.097 8.208 83.601 14.255 4,468.905 436.402 Uruguay 3.000 0.000 45.462 29.240 41.069 3.169 5,639.432 768.806 Venezuela 2.944 0.236 37.774 25.338 48.480 7.111 3,472.989 142.421 Sample 2.839 1.041 860.729 9, 661.129 60.627 38.776 3,107.328 3,057.545

PAGE 92

81 resulting balanced data set is 3024. Table 5-1 presents some descriptive statistics on the data used in the analysis. Model Specification and Estimation Panel estimates were obtained by regre ssing the dependent vari able against a set of explanatory variables for the 21 countries listed in Table 5-1. The basic regression equation takes the form it jk jit j jit j itZ X Y (5-1) where itY represents net inflows of FDI as a percentage of GDP for country i(i= 1,2,,N) in period t (t= 1,2,,T); is a constant term, and and are 1 K and 1 J vectors of unknown parameters to be estimated, respectively. The variable X represents a vector of investment climate determinants of FDI and Z is a vector of control variables, each of which varies in both country and time dimensions. The error term it in the basic panel regression model has two components and can be expressed as it i it (5-2) where i represents the variation uni que to the cros s-section and it is white noise. Greene (2000) indicates that there are two frameworks commonly used to generalize this basic model. First, the fixed-effects mode l, which is used to obtain within-estimators, takes as constant over time but sp ecific to each country. The dependent variable in the fixe d-effects model is regressed against the difference between the observed value of each of the explanatory variables and its mean, thus eliminating the effect of unobserved country (or fixed) characteristics. Thus, the fixed-effects regression equation can be expressed as

PAGE 93

82 it ki kit k k ji jit j j i i itZ Z X X Y Y ] [ ] [ ] [, (5-3) where iY is the mean of country is FDI inflows, jiXis a 1 J vector of means on jitX and kiZ is a 1 K vector of means on kitZ, taken over the T observations respectively. The intercept term (i ) in this model varies by c ountry and thus is an unknown parameter to be estimated. Furthermore, i is interpreted as the mean residual in each country i and can be expressed as k ji j ji j j i iZ X FDI (5-4) Equation 5-4 is unique to each country in that any change in the constant term represents a parametric shift. The error term it in the fixed-effects model is defined as follows: k kit k j jit j i it itZ X Y jk kit k jit j k ki k j ji j i itZ X Z X Y Y ) ( k ki kit k j ji jit j i itZ Z X X Y Y ] [ ] [ ] [ (5-5) Second, the variance component s model is used to obtain random-effects estimators. These estimates are a weighted average of betweenand within-estimators, and thus reflect both the influences across and within countries (Mody and Srinivasan 1997, p.785). The intercept term in this mode l is not allowed to vary by country, hence the estimated coefficients reflect the aver age country in the sample. Random-effects estimators are derived from the following reform ulation of the basic regression equation: it jk i jit j jit j itu Z X Y (5-6) In Equation 5-6, iu is a random disturba nce characterizing the ith observation and is constant over time (Greene 2000, p.568).

PAGE 94

83 The random-effects model involves a transfor mation similar to the one made in the fixed-effects model with one exception; wh en the mean is subtracted from each observation, it is weighted by 1, where is defined as ) (2 2 2uT (5-7) In Equation 5-7, 2 is the variance of the basic error term (it ), 2 u is the variance of the country-specific error term (iu) and T is the number of years. Some extreme cases should be explained purely for example. If the value of from the random-effects model is equal to 1, then the random-effects model is ordinaryleast-squares and the classica l regression model defined by E quation 5-1 applies. At the other extreme, a value of 0 for would indicate that all of the variation in the data is unique to the cross-section of countries. If this were th e case, then the random-effects model would be identical to the fixed-effects formulation. Ordinary-least-squares (OLS) estimates we re obtained first and serve as a basis against which to compare th e alternative methods of es timation. Following Greene (2000), the first hypothesis test was for the ex istence of individual-c ountry effects in the data. The Lagrange multiplier (LM) test statistic4 from the least-squares regression took a value of 66.3939, much larger than the critical value of 3.845. Thus, the null hypothesis of no individual-country effects was rejected. An F test comparing the fixed-effects estim ators to those from the classical OLS regression further suggest that individual-country effects ex ist in the data. The null 4 This statistic is part of the standard TSP output. A more detailed explanation of how the statistic is computed can be foun d in Greene (2000). 5 Chi-square (95%)

PAGE 95

84 hypothesis in this case is similar to the first LM heteroscedasticity test performed (i.e., there are no individual-country effects in the data, and hence there is no difference between OLS and the fixed-effects model. The computed F statistic was 6.267, which exceeds the 99% critical value of 1.88 for F(20,350). As such, there is strong statistical support for the existence of country-specific effects. Given the existence of indivi dual-country effects in the data (and consequently, the conclusion that the classical re gression equation is inappropriate), the LM test statistic based on the residuals from the fixedand ra ndom-effects models wa s used to determine the need to estimate heteroscedastic-consistent standard errors for each of the models. The test statistics took values of 133.363 a nd 103.887 for the fixedand random-effects models, respectively, leading to a rejection of the null hypothe sis of no heteroscedasticity and indicating the need to co mpute heteroscedastic-consistent6 standard errors. By estimating a set of robust standard e rror matrices for the fixedand randomeffects models, it is possible to compare the two models and choose the most appropriate. The preferred method is the Hausman test fo r correlation between the individual-country effects and the other regressors in the equa tion (Greene 2000). If the individual effects are correlated with the other regressors, the fixed-effects model is the most appropriate. Conversely, if the hypothesis th at the two are uncorrelated ca nnot be rejected, then the random-effects model is the better choice. The Hausman test is based on the Wald criterion and is asymptotically distributed ch i-squared. The test statistic of 0.63254 was considerably less than the 95% critical value of 2.167 for the chi-square distribution with 6 The ROBUST command in TSP performs this computatio n even if the nature of the heteroscedasticity is unknown.

PAGE 96

85 the necessary degrees of freedom. Thus, the null hypothesis of no correlation could not be rejected, and the random-effects model wa s deemed to be the most appropriate. The Box-Cox transformation was made to se veral explanatory va riables that were hypothesized to have a nonlinear relationship with the dependent variable. The following transformation was made to the exch ange rate, trade, corporate tax and percapita GDP variables: 1) ( x x, (5-8) where the optimum value of for each variable was estimated by scanning the range of values between and 2 (in increments of 0.1) and maximizing the log-likelihood function of the fixed-effects regression equati on. However, making the transformation to these variables did not have a significant imp act on the regression results and in each case the linear model performed at least as well as the non-linear specifica tion. Thus, the most appropriate approach was to estimate th e random-effects regression equation under the assumption of linearity in the independent variables. Results of Estimation The parameter estimates and their associated standard errors can be seen in Table 5-2. Overall, the results suggest that the model performed well. The 2 R value of .424 indicates that the right-hand-si de (RHS) variables in the mode l explained nearly half of the variation in the data. This can be considered a relatively high 2 R given the size and cross-sectional nature of the panel data set used in this analysis. Six of the seven explanatory variables were st atistically significant and only one took an unexpected sign.

PAGE 97

86 Table 5-2. Random E ffects Regression Results Variable Coefficient Error t-statistic P-value Investment Profile 0.00156060 0.00089697 1.679 0.093 Corporate Tax Rate (0.05498600) 0.02076900 (2.648) 0.008 Government Stability 0.00256093 0.00079045 3.240 0.001 Corruption (0.00019929) 0.00182247 (0.109) 0.913 Exchange Variability (0.000024 77) 0.00001345 (1.842) 0.066 Trade 0.07202500 0.00599952 12.005 0.000 Per Capita GDP 0.00000160 0.00000084 1.890 0.059 Constant (0.03032100) 0.01162300 (2.609) 0.009 The investment profile score had a positive impact on FDI and was significant at the 10% level. The parameter estimate of 0.0015 suggests that a one-unit improvement in the investment profile score would have the effect of increasing the FDI share of GDP by 0.15%. Based on the definition provided by the ICRG (International Country Risk Guide), the investment profile score takes into consideration the ri sks associated with profit repatriation, contract vi ability and expropriation, and de lay of payment. The score would increase (or improve) if a the ruling regime in a country were to ease the restrictions on profit repatriati on by allowing a greater percentage of profits to be transferred ou t of the host country or by lessening the tax (or other) penalties associated with such a transfer enact tougher legislation to en sure the enforcement of le gal contracts and limit the possibility of excessive delays in payment for goods or services rendered decrease the governments right to expropr iate funds or other property belonging to foreign investors. Thus, the positive relationship between the investment profile score and FDI is not surprising. The government stability score was the most significant of the investment climate variables examined. As defined by the ICRG, the government stability score is a measure of government unity, legislative strength and popular support. The rating is intended to

PAGE 98

87 provide an indication of the cohesiveness of the ruling regime and the extent to which opposing parties pose a threat to its ability to remain in power A one-unit increase in the government stability score was estimated to ha ve the impact of in creasing the FDI share of GDP by 0.26%. Thus, the results suggest that the potential for change in the hostcountry regulatory regime is of subs tantial concern to foreign investors. The amount of variability in the exchange rate had a negative effect on FDI, although the magnitude of this e ffect was rather small. The exchange rate variable, as specified in the model, does not reflect whethe r the change is due to an appreciation or depreciation in the host-count ry currency. Rather, it simp ly measures the level of stability in the local currencys value against the US dollar. Darby et al. (1999) showed that exchange rate volatility leads to uncerta inty in the future cash flows of a proposed foreign investment project, and in turn, to a greater likelihood that the investment project will either be postponed or rej ected altogether. However, th e results of this analysis suggest that a 1% increase in exchange rate variability decreases the FDI share of GDP by only 0.002%. It is noted that the risk of ch anges in the foreign exchange rate is a risk that can be hedged in international financ ial markets (e.g., thr ough the use of currency swaps and futures contracts), a nd that this fact may account for the lack of response in FDI to changes in the variable. The excha nge rate variable was significant at the 10% level. The parameter estimate for the corporate ta x rate suggests that higher taxation is a significant deterrent to forei gn investment. Decreasing the corporate tax rate by 1% had the effect of increasing the FDI share of GDP by as much as 5.5%. Furthermore, the tax rate variable was significant at the 1% level.

PAGE 99

88 The negative coefficient for corruption indi cates that as the risk for political corruption increases the FDI share of GDP in creases. However, the variable was not statistically significant. Nonetheless, the existing theoretical literature suggests that corruption can serve as a benefit to foreign i nvestors in the sense th at they often posses the capital necessary to buy favors from governmental agents. Although the regression results seem to offer weak support for this c ontention, the most obvious inference is that the level of political corruption does not appear to be a significant determinant of FDI in the sample countries. Finally, each of the control variables ha d a positive impact on the FDI share of GDP. The trade variable was significant at th e 1% level and the magnitude of its impact was 0.072. This result suggests that countries that trade more also attract more FDI as a share of GDP. Per-capita GDP was significan t at the 10% level, although the magnitude of its effect was much smaller. A $1,000 increase in per-capita GDP was estimated to increase the FDI share of GDP by only 0.0016%. Although the parameter estimates presented in Table 5-2 appear to be small in magnitude, their corresponding eff ect on the absolute level of FDI is considerable in many cases. This is important because it is the impact in terms of dollars with which this study is ultimately concerned. As shown below, simple algebra can be used to reveal the effect of the independent variables on the actual dollar level of FDI in each of the countries in the sample. The dependent variable is specified in th e regression equation as the FDI share of GDP for two reasons. First, as pointed out by Lecraw (1991), because the magnitude of the explanatory variables is not dependent upon economic size, there is a need to

PAGE 100

89 standardize the dependent variab le (as defined in this study) across countries. This is most often accomplished by dividing FDI by GDP. An alternative approach is to include GDP as a discrete independent variable on th e RHS of the regression equation. However, the FDI and GDP variables were correlate d with each other and GDP became the dominant variable in the regr ession equation, detracting from the significance of the other variables in the model. The regression equation was therefore estimated in shares and the effect of each variable on the dollar level of FDI was then extracted from the model by multiplying through by GDP as follows: k kit k j jit j i it itZ X GDP FDI (5-9) jk it kit k it jit j it i itGDP Z GDP X GDP FDI ) ( ) ( ) ( (5-10) Equation 5-10 shows that although the esti mated parameters are constant across countries, the magnitude of change in each countrys level of FDI is unique to that country. The remainder of this chapter is de voted to analyzing the relationship between FDI and hypothetical changes in the investme nt climate variables in each country. Equation 5-10 was used to create a benchmarking scenario where the parameter estimates were applied to the most recent data avai lable (2001). The implied impacts on FDI of ceteris paribus changes to the explanatory variables in the model are presented for each country in the analysis. For the variables that consisted of a ra ting or score, the effect of a one-unit improvement in the score is considered. Fo r the corporate tax rate, trade and per-capita GDP variables, the effect of a 10% increase is shown. Finally, the reaction of FDI to eliminating any variability in the foreign ex change rate in other words, pegging the local currency to the US dolla r is considered. In each cas e, the change in the dollar

PAGE 101

90 level of FDI is reported al ong with the percentage change from the models predicted value using the benchmark data. The percenta ge change is referred to hereafter as the responsiveness of FDI to change s in the investment climate. MERCOSUR Table 5-3 presents the benchmark values of the explanatory variables for the countries of MERCOSUR. Table 5-4 shows th e responsiveness of FD I to changes in the explanatory variables. As Ta ble 5-4 shows, FDI in Argentina was the most responsive Table 5-3. Investment Climate Variables MERCOSUR (Benchmark = 2001) Investment Profile Score Government Stability Score Corruption Score Corporate Tax Rate Exchange Rate Variability Argentina 7.5 5.5 2.5 0.35 0.00000 Brazil 8.5 6.0 2.0 0.15 0.14838 Paraguay 9.5 7.0 1.0 0.30 0.14773 Uruguay 11.5 11.0 3.0 0.30 0.08392 Table 5-4. FDI Responsiveness MERCOSUR in terms of the percentage change in FD I resulting from a change in each of the investment climate variables. The results indicate that, ceteris paribus a one-unit improvement in the investment profile scor e would increase FDI in Argentina by almost 54%. Based on the benchmark values, this repr esents a change in FDI of $404.5 million. A one-unit increase in Argentinas government stability rating affected a 92% increase in Responsiveness of FDI to a Change in: Investment Profile Score (1-unit increase) Government Stability Score (1-unit increase) Corruption Score (1-unit increase) Corporate Tax Rate (10% decrease) Exchange Rate (fix to US$) Argentina $ 404,585,286 $ 687,963,691 $ (53,537,449) $ 516,996,951 $ 0 Brazil 756,808,283 1,286,889,657 (100,145,968) 414,464,167 1,846,875 Paraguay 10,852,259 18,453,366 (1,436,044) 11,886,425 26,368 Uruguay 28,112,691 47,803,298 (3,720,061) 30,791,691 38,802 Argentina 53.91% 91.67% -7.13% 68.89% 0.00% Brazil 9.08% 15.45% -1.20% 4.97% 0.02% Paraguay 4.66% 7.92% -0.62% 5.10% 0.01% Uruguay 4.22% 7.18% -0.56% 4.63% 0.01%

PAGE 102

91 FDI, or nearly $688 million. A decline in the risk for corruption was estimated to decrease FDI by 7.13%, and cutting the corpor ate tax rate by 10% led to a 69% increase in FDI. It is noted that the value of the Argentine Peso was, in fact, pegged to the U.S. dollar in 2001. Thus, there was no change from the benchmark value and no response in FDI. At first glance, the results for Argentina might be considered suspect given their inconsistency with the results from other countries in the sample. However, there are both computational and qualitative rationales for why such extreme results were observed for Argentina. An anomaly in the benchmar k data caused the models predicted value for FDI to be unusually low. This was primar ily the result of a sharp decline in the government stability rating for Argentina in 2001. In any event, the combination of Argentinas large GDP and the low predicte d value for FDI made the impact of any adjustment to the explanatory variables (w hich is a function of both the parameter estimate and GDP) very large in terms of percentage change. However, this result is consistent with the empirical evidence. As mentioned, the ICRG government stability rating for Argentina fell from 10.0 in 2000 to 5.5 in 2001, a difference of 4.5 (the reas ons for this change are discussed below). During the same period, the investment pr ofile score increased by 2.0 points and the corruption score fell by 0.5 points. The exchan ge rate variability and corporate tax rate did not change. Working backwards from the benchmark values of the explanatory variables and FDI, the results indicate that the observed changes in the investment climate variables would cause a 66% decrease in FDI. In reality, FDI dropped by 72% from 2000 to 2001. Thus, the results of the an alysis appear to correspond well with what

PAGE 103

92 can actually be observed in the data for Argen tina. Furthermore, the observed changes in the ICRG risk rating variables can be rati onalized by analyzing the political and economic environment in Argentina during the period 2000-2001. Argentina is the third largest economy in Latin America and a member of MERCOSUR. To a large extent, foreign invest ment in Argentina during the early 1990s was in response to privatization. In the late 1990s, foreign acquisitions in a number of deregulated industries continued to account for relatively high levels of investment (EIU 2003). However, the Argentine economy was hit hard by the Brazilian currency devaluation in 1999. The global economic slow down that occurred af ter the turn of the century only aggravated the situation and the country was facing a seri ous financial crisis by late 2001. The President of Argentina (Fernando de la Ra) was forced to resign in 2001 after he failed to fulfill his promise to revive the economy (EIU 2003). The ensuing political environment was characterized by vi olent protest and disse nsion both within and between political parties. The investment climate in Argentina degr aded quickly as a re sult. Although the country remained open to foreign investment, an inconsistent regulatory environment and a weak judicial system contributed to a $20.7 billion contraction in FDI (from $23.9 billion in 2000 to just over $3 billion in 2001). The EIU (2003) states that the majority of FDI after 2001 consisted of capita l contributions from parent firms to sustain existing Argentine subsidiaries. The cont raction in FDI combined with the flight of other types of capital led to net capital outflows in 2001 and 2002. The EIU (2003) indicates that the degrad ation of the investment climate in Argentina was the primary reason for the dras tic decline in FDI from 2000 to 2001. The

PAGE 104

93 results of this analysis are consistent with th is contention. The extent to which the results appear to be overstated can be accounted for by the fact that Argentina was an economically large country in which FDI was very low relative to its size. However, this situation was not necessarily comparable to some of the other countries in MERCOSUR. Within MERCOSUR, the results indicate th at FDI was much less responsive to changes in the investment climate in the sm aller countries of Paraguay and Uruguay. For example, a one-unit improvement in the invest ment climate had the effect of increasing FDI by 4.66% and 4.22% in Paraguay and Urugua y, respectively. In reality, despite a history of treating foreign and domestic investors as equals, Paraguay and Uruguay attracted much less FDI (both in absolute le vels and as a share of GDP) than the two larger countries in MERCOSUR. The Paraguayan government continued to be the primary agent in the economy throughout the 1990s and there wa s little popular support for pr ivatizing state-controlled industries. As a result, major opportunities for direct investme nt failed to materialize and Paraguay received only a small portion of the FDI flowing in to Latin America during the 1990s investment boom (EIU 2003). Contagion effects from the problems in Brazil and Argentina suppressed economic growth late in the decade. Similar conditions existed in Uruguay at the time. During the period 1999-2002, the Uruguayan economy suffered its deepest and longest recession in 50 years (EIU 2003). Like Paraguay, the Uruguayan government resisted privatiza tion during the 1990s while otherwise remaining open to foreign i nvestment. Despite gr anting private sector access to the telecommunications utilities, transpor tation and insurance industries, the Uruguayan government maintained its monopolization of these markets through 2001.

PAGE 105

94 Although plans to further privatize these indus tries had developed, they were met with opposition among voters who were concerned about the potential impact on the labor market. Table 5-4 shows that the results for Para guay and Uruguay were very similar. In these countries, the percentage change in FDI resulting from a change in each of the investment climate variables was about half the estimated response in Brazil, with one exception. In Paraguay and Uruguay, FDI wa s more responsive to a decrease in the corporate tax rate than in Brazil. This is interesting because the Brazilian tax rate (15%) was half as high as in Paraguay and Uruguay (3 0%). Thus, it stands to reason that the marginal effect of lowering the tax rate was smaller in Brazil. Improvements in both the investment prof ile and government stability scores in Uruguay the magnitudes of which were 1.5 and 1.0, respectively from 2000 to 2001 were accompanied by very little change in FDI. The results suggest that these improvements in the investment climate, ceteris paribus would effect a $43.2 million increase in FDI. In reality, FDI incr eased from $273.5 million to $318.2 million, a difference of $44.7 million. The results for Paraguay were comparable. There are several explanations for why FD I might be less responsive to changes in the investment climate in Paraguay and Uruguay than in Argentina and Brazil. The most obvious of these reasons is th e underlying resistance to priv atization in the two former countries. The governments of Argentina and Brazil promoted the sale of many stateowned companies and opened a number of i ndustries to private investment during the 1990s. Foreign investors responded in kind by making substantial FDIs. Conversely, the empirical evidence suggests that the broad resistance to privatization in Paraguay and

PAGE 106

95 Uruguay effectively narrowed th e scope of available investment opportunities. Hence, FDI in these countries was less than 2% of GDP and showed little response to improvements in the investment climate. MERCOSUR is dominated by Brazil, the second-largest economy in Latin America. Brazil was both the primary source of imports for each of the other members of MERCOSUR and the primary destination of th eir exports. It was also the largest recipient of FDI in Latin Am erica throughout the late 1990s. However, before 1996, FDI never amounted to more than 1% of the countrys GDP. Constitutional amendments in 1995 led to the privatization of state-owned interests in th e telecommunications, energy and transportation industries in Brazil, causing the country to become a major recipient of FDI in South America. Even after the sale of state-owned assets slowed, FDI remained above 4.4% of GDP (EIU 2003). This leve l of investment contributed to positive, although sluggish economic growth despite the de valuation of the Brazilian Real in 1999. Interestingly, it has been sugge sted that other countries in the region suffered more than Brazil as a result of the Real devaluation. From 2000 to 2001, the investment profile score for Brazil increased 2.5 points, reflecting the enactment of new, more liberal investment regulations for foreign equity investors. At the same time, the political reform process stalled as President Cardoso entered his final year in office. Accordi ngly, the government stab ility score for Brazil decreased from 10.0 to 6.0. The corruption rating simultaneously fell from 3.0 to 2.0 and the corporate tax rate did not change. Ba sed on the benchmark values, the regression results indicate that FDI in Brazil would d ecrease by 38% in response to these changes.

PAGE 107

96 In reality, FDI declined from $32.8 billion to just under $22.6 billion, a change of 31%. Again, the models in-sample prediction is consistent with the actual observed value. The Andean Community Three of the countries in which FDI was the most responsive to the investment climate Columbia, Venezuela and Peru were members of the Andean Community. These are relatively large countries in term s of GDP with few restrictions to foreign investment. However, FDI remained below 3% of GDP in each during 2001. Table 5-5 presents the benchmark values of the expl anatory variables for the Andean Community and Table 5-6 shows the responsiveness of FDI to changes in the investment climate in these countries. Table 5-5. Investment Climat e Variables Andean Community (Benchmark = 2001) Investment Profile Score Government Stability Score Corruption Score Corporate Tax Rate Exchange Rate Variability Columbia 9.5 6.0 2.0 0.35 0.14513 Venezuela 7.0 7.5 2.0 0.34 0.09342 Peru 8.5 8.5 2.0 0.30 0.01818 Bolivia 9.5 9.0 2.0 0.25 0.06616 Ecuador 7.0 7.5 1.5 0.25 0.56025 Table 5-6. FDI Responsiveness Andean Community Responsiveness of FDI to a Change in: Investment Profile Score (1-unit increase) Government Stability Score (1-unit increase) Corruption Score (1-unit increase) Corporate Tax Rate (10% decrease) Exchange Rate (fix to US$) Columbia $124,115,544 $211,048,179 -$16,423,805 $158,600,325 $296,257 Venezuela 188,179,873 319,984,252 -24,901,228 233,594,139 289,144 Peru 81,398,200 138,410,882 -10,771,158 89,155,042 24,332 Bolivia 12,002,187 20,408,723 -1,588,210 10,954,946 13,060 Ecuador 27,082,634 46,051,770 -3,583,756 24,719,561 249,548 Columbia 13.66% 23.23% -1.81% 17.45% 0.03% Venezuela 10.30% 17.52% -1.36% 12.79% 0.02% Peru 10.11% 17.19% -1.34% 11.07% 0.00% Bolivia 5.97% 10.14% -0.79% 5.44% 0.01% Ecuador 2.54% 4.32% -0.34% 2.32% 0.02%

PAGE 108

97 As a result of several legi slative actions taken in the 1990s, foreign investment was permitted in nearly every industry in Columbia by the mid-point of the decade. Foreign firms subsequently invested heavily in the oil, electricity, financial services and telecommunications industries. Despite th e countrys level of openness to private enterprise, the EIU (2003) indi cates that instability in Co lumbias foreign investment regulatory regime and the increased threat of guerilla terrorism depressed FDI flows from 1999-2001. The results suggest that a one-unit change in the investment profile score would increase FDI in Columbia by $124 million (a 13.22% change from the benchmark level). The same change in the government stabil ity score increased FDI by $211 million, while lowering the corporate tax rate by 10% led to an increase of $158.6 million. In October 2000, a decree was issued by the Columbian government eliminating restrictions on the acquisition of voting stock in publicly-tra ded companies by foreign investors. Accordingly, the investment profile score increased from 3.0 in 2000 to 9.5 in 2001. However, popular support for President Se rpa declined as c oncerns over guerilla terrorism made the hard-line approach of his political ch allenger, Mr. Uribe, more attractive to many voters (EIU 2003). As the likelihood of Mr. Serpas re-election decreased, Columbias government stability scor e fell from 9.0 to 6.0. The corporate tax rate and corruption rating remained unchanged during that period, a nd the variability of the exchange rate decreased slightly. Based on the benchmark values, the cumulativ e effect of these changes would be an increase in FDI of $173 million. This is r oughly equivalent to a 20% change. Although FDI did increase from 2000 to 2001, the magn itude of the actual change was $47 million,

PAGE 109

98 much less than the models prediction. One potential explanation for this is the persistence of civil unr est in Columbia. Factors not fully accounted for in the regression equation, such as kidnappings and other interference by guerillas, paramilitary groups and criminal gangs serve as hindrances to FDI. The results suggest that although the investment climate improved substantially, these threats to foreign investors limited the potential response of FDI. Like Columbia, Venezuela had one of the wo rst investment profile ratings in Latin America throughout the 1990s. However, th e low rating in Venezuelas case was primarily due to a lack of effective systems fo r dispute settlement and tight restrictions on the repatriation of profits. Sweeping cha nges were made in late 1999 regarding the constitutional treatment of foreign invest ment in Venezuela. The new constitution allowed for 100% foreign ownership in a numbe r of industries, established legal means for the settlement of business disputes, and simplified the rules for making direct investments. However, the process of a pproving the implementing legislation continued through 2001. Although the ICRG investment profile rati ng for Venezuela reflected these changes (improving from 4.0 in 2000 to 7.0 in 2001), the data shows that a majority of investors opted to await the completion of the reform process before carrying-through with their investments. As such, FDI in Ve nezuela amounted to less than 3% of GDP in 2001. Foreign direct investment in Peru was fu eled by an extensive privatization program during the late 1990s. Although the program was ongoing through 2001, the investment climate was adversely affected by the collapse of President Fujimoris government and his subsequent resignation in mid-2000. Contra ct law also continued to be an issue for

PAGE 110

99 foreign investors in Peru. As a result, judicial reform was identified as a priority of the new Toledo administration in 2001. Table 5-6 shows that in terms of percenta ge change, the estimat ed response of FDI to changes in the investment climate was almost identical in Peru and Venezuela, although the dollar amount of these changes di ffers substantially. For example, while a one-unit increase in the investment profile score had the impact of increasing FDI by $81.3 million in Peru, the change in FDI in Venezuela was $188 million. In each case, this represents an increase of just over 10%. An improvement in the government stability score increased FDI by roughly 17% in both coun tries, while decreasing the tax rate led to an 11.1% and 12.8% increase in Peru and Venezuela, respectively. Although not reflected by the ICRG government stability rating, the political environment in Peru from late 2000 to early 2001 was volatile. The EIU (2003) indicates that many foreign investments were withheld pending a resolution to the political turmoil that followed Mr. Fujimoris resignation. In fact, FDI in 2000 was 66% lower than the level observed in 1999. Alt hough the brevity of the situation (i.e., new elections had been held and the situation was resolved by mid-2001) prevented it from being captured by annual data, its impact on FDI in Peru was significant. The estimated impacts of changes in Venezuelas investment climate were, however, consistent with what can be obser ved in the FDI data. For instance, the investment profile score for Venezuela incr eased 3.0 points from 2000 to 2001, while the government stability and corruption scor es decreased by 3.5 points and 1 point, respectively. The results of the analysis s uggest that FDI in Venezuela should have been roughly 30% higher in 2000 than in 2001. In reality, FDI increased by 29.4%.

PAGE 111

100 When compared to Columbia, Venezuela and Peru, FDI in Bolivia was less responsive to changes in the investment cl imate. However, over the entire sample, Bolivia was one of the countries in which FD I was the most responsive. The government of Bolivia actively promoted FDI in nearly every economic sector throughout the 1990s and inflows reached 12% of GDP by 1999. While the global economic slowdown in 2000 and 2001 caused global FDI flows to contra ct, FDI continued to account for more than 8% of Bolivias GDP during these year s. The energy sector was the principal destination of this investment ., although much of the economy has been liberalized by the late 1990s The results indicate that FDI in Bolivia was roughly half as responsive as in Venezuela. Also, FDI averaged 9.8% of GDP in Bolivia from 1997-2001, compared to an average of 4.2% in Venezuela. In fact throughout the entire sample, only Trinidad and Tobago had a higher average level of FDI than Bolivia. From 2000 to 2001, Bolivias investment climate, government st ability and corruption scores each declined by 0.5, 1.0 and 1.0, respectively. The regr ession results suggest that FDI in Bolivia should have been roughly 12% lower in 2001 as a result of these changes. This result is consistent with the observed 9.5% decrease in investment. Direct investment in Ecuador was much less responsive to changes in the investment climate than the rest of the A ndean Community countries. As was the case with the least responsive c ountries of MERCOSUR, the Ec uadorian government imposed heavy restrictions on private se ctor involvement in most sect ors of the economy. In fact, the Ecuadorian economy remained one of the most protected in Latin America through 2001 (EIU 2003). Another negative aspect of Ec uadors investment c limate was a lack of

PAGE 112

101 contract enforceability. Rega rdless, FDI inflows accounted fo r over 7% of the countrys GDP in 2001. Over 80% of this investment took place in the petroleum industry, which was liberalized in the early 1990s. The estimated change in FDI from a one -unit improvement in the investment profile, government stability and corruption ra tings in Ecuador were $27 million (2.54%), $46 million (4.32%) and $3.6 million (-0.34%), respectively. A 10% decrease in the corporate tax rate impacted FDI by nearly $25 million. These levels of response were among the lowest in the sample, and there are tw o potential explanations for this. First, the limitation on foreign ownership in most industries could have been sufficient to prohibit the new investment that woul d have accompanied the aforementioned improvements in the overall investment clim ate. Second, it is also possible that, by opening the petroleum industry to foreign investment, the Ecuadorian government effectively overrode the negativ e impact of the overall invest ment climate. In fact, the EIU (2003) indicates that economic growth in Ecuador has tracked the performance of the global oil sector since its oil reserves were ta pped in the 1970s. Furthermore, the EIU states that this phenomenon is the result of a lack of institutional reform, and hence investment, in other industries (EIU 2003). The results of this analysis offer support for these conclusions. CARICOM As Table 5-7 shows, with the exception of Haiti, the countries of CARICOM were generally less responsive to changes in the investment climate (in terms of percentage change) than the larger countries of MERCO SUR and the Andean Community. In fact, Guyana was the least responsive country in the sample. There are several attributes

PAGE 113

102 shared by the countries of CARICOM that may explain their re lative lack of FDI response. Table 5-7. FDI Responsiveness CARICOM Responsiveness of FDI to a Change in: Investment Profile Score (1-unit increase) Government Stability Score (1-unit increase) Corruption Score (1-unit increase)Corporate Tax Rate (10% decrease) Exchange Rate (fix to US$) Haiti $5,628,751 $9,571,224 $744,834 $7,192,667 $18,691 Dominican Republic 31,944,980 54,319,787 -4,227,175 29,157,648 14,845 Trinidad & Tobago 13,316,668 22,643,889 -1,762,151 17,016,627 1,173 Jamaica 11,723,277 19,934,459 -1,551,303 14,265,736 16,471 Guyana 1,052,049 1,788,923 -139,214 1,728,457 448 Haiti 7.42% 12.61% -0.98% 9.48% 0.02% Dominican Republic 3.80% 6.47% -0.50% 3.47% 0.00% Trinidad & Tobago 2.67% 4.55% -0.35% 3.42% 0.00% Jamaica 2.39% 4.06% -0.32% 2.90% 0.00% Guyana 1.15% 1.96% -0.15% 1.89% 0.00% In comparison to many of the other countri es in the sample, Haiti, Trinidad and Tobago, Jamaica and Guyana were relativel y small in terms of both geographic and economic size. Interestingly, the countries of Trinidad and Tobago, Jamaica and Guyana also had some of the highest average levels of FDI (as a percentage of GDP) in the sample. Foreign direct investment in Tr inidad and Tobago accounted for 11% of GDP from 1997-2001, the highest in the sample. In comparison, FDI stood at 8.0%, 7.9%, 5.7% and 0.07% of GDP in Guyana, Jamaica, the Dominican Republic and Haiti in 2001, respectively. In terms of the percentage response of FD I to a change in one of the investment climate variables, Haiti is th e obvious outlier within CARICOM. For example, while the results indicate that the ot her CARICOM countries were among the least responsive in the sample, a one-unit increase in Haitis inve stment profile score had the effect of a 7.4% increase in FDI. As mentioned above, Haiti also attracted th e least FDI of all the

PAGE 114

103 countries in the sample during 2001. Thus, the large magnitude of this impact is primarily the result of an extremely low level of FDI in the benchmark year. Table 5-8 displays the benchmark explanatory variable values for each of the CARICOM countries. The table shows that political corruption a nd economic volatility have been the norm in Haiti, and recent developments fail to provide any sign of the possibility for future improvements in thes e areas. Thus, Haitis relative inability to attract FDI could be the result of a low level of governmental stability and an unsuitable environment for foreign investment. However, the results of the analysis indicate that if conditions were to improve, FDI in Haiti would be highly responsive. Table 5-8. Investment C limate Variables CARICOM (Benchmark = 2001) Investment Profile Score Government Stability Score Corruption Score Corporate Tax Rate Exchange Rate Variability Haiti 5.5 11.0 1.0 0.35 0.20191 Dominican Republic 9.0 10.5 2.0 0.25 0.02826 Trinidad & Tobago 11.5 4.0 3.0 0.35 0.00536 Jamaica 9.5 9.5 2.0 0.33 0.08542 Guyana 8.0 9.5 3.0 0.45 0.02586 Extreme social poverty also persists as a major deterrent to foreign investment in Haiti. The adult literacy rate of just 48%, co mpared to an average of 90% in the rest of the Caribbean, reflects the l ack of an educated work force (EIU 2003). In addition, Haitis natural resource base is threatened by both an increasing popul ation and a lack of financial support for any sort of environmen tal preservation or re storation efforts. Furthermore, infrastructural development ha s been hindered by a regulatory regime that has traditionally discriminated ag ainst foreign investors (EIU 2003). In contrast to Haiti, FDI in the thr ee least responsive countries in CARICOM averaged over 5.7% of GDP from 1997-2001. For example, FDI in Guyana stood at over 8% of GDP in 2001. The results in Table 5-7 show that a one-unit improvement in

PAGE 115

104 Guyanas investment profile score led to an increase in FD I of 1.15% (roughly $1 million). A one-unit improvement in the govern ment stability score and a 10% decrease in the corporate tax rate each had the eff ect of increasing FDI by just less than 2% (approximately $1.7 million). Guyanas economy was dominated by ag riculture throughout the 1990s and sugar was the primary export commodity. As a resu lt, economic growth was subject to weather patterns and international co mmodity prices. In 2000, gold became the primary export commodity as international gold prices rose higher and Guyanas sugar production fell, the latter of which was primarily due to inclement weather (EIU 2003). Although the economy expanded rapidly during the early 1990s, growth during the late 1990s alternated between positive and negative levels. A number of industries in Guyana were successfully privatiz ed during the 1990s. While attempts to privatize the electric-pow er and airline industries failed, sales of majority stakes in the telecommunications agriculture and banking industries pushed inflows of FDI to nearly $147 million (or 40% of GDP) in 1992 (EIU 2003). This level of FDI, combined with extensive debt relie f and large inflows of foreign assistance, strengthened the financial environment in Guyana from the late 1990s through 2001. Inflows of official development assist ance (ODA) averaged more than 20% of Guyanas GDP during the period 1989-2001. Guyana saw net outflows of private bank lending during that same period. Most deve lopment projects in Guyana were funded by multilateral banks as many investment opportunities were typically too small to attract the attention of investors from the developed countries. However, a substantial amount of private investment was aimed at Guyanas timber and gold resources. These facts

PAGE 116

105 seem to suggest that foreign di rect investors, rather than increasing their investment in Guyana because of the improvements in the investment climate, may have responded to investment opportunities that presented themselv es as a result of development efforts led by World Bank and the IMF. This possibility is examined again later in the discussion on Nicaragua. In stark contrast to both Guyana and Nicaragua, effort s to pay down foreign debt caused net outflows of official capital in Trinidad and Tobago throughout most of the 1990s. Instead, the government of Trinidad actively promoted privatization, and FDI consequently poured in at a disproportionate rate. However, just as in Nicaragua and Guyana, the results indicate that changes in the investment climate would have relatively little effect on the level of FDI in Trinidad. Trinidad and Tobago had the largest FDI share of GDP in the sample in 2001 at over 9.4%. Despite being the smallest count ry geographically, the island-nations fossil fuel reserves combined with relative economic and political stability made Trinidad and Tobago an attractive target fo r foreign investors. As a result, economic growth was driven by fluctuations in oil and chemical prices, averaging 4.2% during 1997-2001 (EIU 2003). The reaction of foreign investors to recen t developments in Trinidad and Tobago support the results of the anal ysis. Dissension among the ranks of government left the Trinidadian legislature in a political deadlock in mid-2001 and three national elections were subsequently held over the ne xt two years. Accordingly, the ICRG government stability score for Trinidad fell five poin ts from 2000 to 2001, the quickest decline observed over the entire sample. However, foreign investors were undeterred even in the

PAGE 117

106 face of such political turmoil and FDI cont inued to expand through the end of 2002 (EIU 2003). In 1999, as much as 82% of U.S. FDI in Tr inidad flowed into the petrochemical, oil and natural gas industries. As previously men tioned, this can be compared to the case of Ecuador. In fact, over 80% of FDI in Ecua dor during the period 1997-2001 was in the oil and mining industry, and U.S. investors were the principal source of these funds. These data, when combined with the fact that fo reign investment was re latively unresponsive to changes in the investment climate in these c ountries, seem to sugge st that the possession of oil reserves served as the dominant consideration of foreign investors in these countries. In other words, foreign investor s tended to engage in petroleum-related investment projects regardless of the ex isting state of, or changes in, the overall investment climate. Although Jamaica is not an oil producing nati on, the country is very comparable to Trinidad and Tobago in terms of FDI response. For exampl e, a one-unit improvement in the investment profile rating ha d the effect of increasing FDI in Jamaica by 2.39% ($11.7 million), compared to 2.67% for Trinidad. Th e government of Jamaica also promoted the privatization of many industries and imposed few limits on foreign ownership during the late 1990s. One possible explanation for the small response of FDI to changes in Jamaicas investment climate is the high level of econo mic concentration. Specifically, Jamaicas tourism industry accounted for over 50% of FDI inflows and 15% of total economic activity in 2001. Jamaica is comparable to the cases of both Trinid ad and Ecuador in the sense that a single industry accounted for a major ity of FDI. This is interesting because

PAGE 118

107 the same cannot be said about most of the othe r countries in the sample. In fact, it was typically the case that FDI inflows were spr ead across a number of industries within the most responsive countries in the sample. In the cases of Jamaica, Trinidad and Ecuador, it seems plausible that FDI might have been more responsive to th e international demand for these countries primary products (tourism and mineral exports in the case of Jamaica, oil in the case of Ecuador and Trinidad) than to changes in their investment climates. The Central American Common Market Table 5-9 presents the benchmark explanat ory variable values for the countries in the Central American Common Market. Table 5-10 shows that, on average, FDI in these countries tended to be less responsive than in most of the other countries in the sample. For instance, Guatemala showed the larges t response among the CACM countries in terms of both dollars and pe rcent change. A one-unit improvement in the government stability score increased FDI by 9.26%, or $52.5 million. This can be compared to changes of 5.78%, 4.28%, 4.19% and 3.59% fo r El Salvador, Costa Rica, Honduras and Nicaragua, respectively. The investment profile score had a slightly smaller impact on FDI in the CACM countries, ranging from 5.44% (or $30.9 million) in Guatemala to 2.11% ($3.9 million) in Nicaragua. Nicaragua was the least responsive to each of the investment climate variables. A 10% decrease in the corporate tax rate had th e effect of increasing FDI in Nicaragua by only 1.93%. It is interesting to note that from 1997 to 2001 FDI averaged 9.0% of the countrys GDP, the third highest of the countries in the sample. Nicaragua is similar to the cases of Guyana and Trinid ad and Tobago in that FDI a ccounted for a large share of GDP and was relatively unres ponsive to changes in the investment climate.

PAGE 119

108 Table 5-9. Investment Climate Variables CACM (Benchmark = 2001) Investment Profile Score Government Stability Score Corruption Score Corporate Tax Rate Exchange Rate Variability Guatemala 11.0 9.0 3.0 0.31 0.03171 El Salvador 7.5 9.0 2.5 0.25 0.00029 Honduras 8.5 9.5 1.5 0.25 0.04340 Costa Rica 9.0 9.5 3.0 0.30 0.07294 Nicaragua 5.5 11.0 2.0 0.25 0.06416 Table 5-10. FDI Responsiveness CACM Responsiveness of FDI to a Change in: Investment Profile Score (1-unit increase) Government Stability Score (1-unit increase) Corruption Score (1-unit increase) Corporate Tax Rate (10% decrease) Exchange Rate (fix to US$) Guatemala $30,868,514 $52,489,345 -$4,084,730 $34,937,134 $16,097 El Salvador 20,691,608 35,184,361 -2,738,053 18,886,179 97 Honduras 9,617,406 16,353,600 -1,272,640 8,778,247 6,865 Costa Rica 24,259,981 41,252,084 -3,210,244 26,571,836 29,103 Nicaragua 3,857,020 6,558,542 -510,387 3,520,479 4,070 Guatemala 5.44% 9.26% -0.72% 6.16% 0.00% El Salvador 3.40% 5.78% -0.45% 3.10% 0.00% Costa Rica 2.52% 4.28% -0.33% 2.76% 0.00% Honduras 2.46% 4.19% -0.33% 2.25% 0.00% Nicaragua 2.11% 3.59% -0.28% 1.93% 0.00% Privatization of Nicaraguas state-controll ed industries, coupled with major reform in foreign investment legisl ation, was met with large in flows of FDI during the 1990s. However, problems with dispute settlement, corr uption and a lack of transparency within the regulatory system remained as major de terrents to FDI in Nicaragua in 2002. The ICRG investment profile score for Nicaragua reflected these conditions. In fact, Nicaragua and Haiti had score of 5.5 in 2001, the lowest score in the sample in that year. Regardless of the low investment profile score, investment in Nicaraguas telecommunications and energy industries boosted FDI to over 10% of GDP in both 1999 and 2000. Thus, both the empirical evidence and the regression results indicate that foreign direct investors in Nicaragua were attracted by something other than the investment climate. As mentioned in a prev ious section, perhaps the most obvious factor

PAGE 120

109 is the high level of official development a ssistance (ODA) flowing in to the country. As was the case with Guyana, inflows of ODA av eraged more than 32% of Nicaraguas GDP during the period 1989-2001. It is also true that Guyana a nd Nicaragua were the smallest and most trade-dependent economies in the sample. In addition, comparable results were observed for the two with regard to the percentage change in FDI from a change in the investment climate variables. These similarities are explored in further detail in Chapter 6, as are the comparable as pects of several other countries included in the analysis. Costa Rica, although geographically smalle r than Nicaragua, was nearly eighttimes larger in terms of economic size in 2001. Over time, Costa Rica evolved from an economy dominated by agriculture to one that included a diverse mi x of manufacturing (including high technology), financ ial services and tourism. At least part of the economic development observed in Costa Rica was fi nanced by foreign investment, as can be witnessed by the fact that FDI averaged near ly 3.5% of GDP from the late 1990s to 2001. The electronic equipment manufacturing se ctor accounted for 15% of all FDI in Costa Rica over the period 1985-2001. Although th is is a larger amount than the level of FDI any other industry, it is by no means an overwhelming majority, as investment was spread among a number of other sectors. Ne vertheless, the economic impact of foreign involvement (specifically in the manufactur e of computer processors) has led some experts to distinguish between two types of economic growth in Costa Rica: Intel- and non-Intel led growth (EIU 2003). The results of the analysis indicate that FDI in Costa Rica was only moderately responsive to changes in the investment climate. A one-unit improvement in the

PAGE 121

110 investment profile score had the effect of increasing FDI by 2.52%, while an equivalent change in the government stability score increased investment by 4.28%. Table 5-10 shows the effect of these changes as well as the impact of the ot her three investment climate variables on the leve l of FDI in Costa Rica. The results for Honduras were very similar to those observed for Costa Rica. The Honduran government remained committed to pr ivatizing state-cont rolled industries and liberalizing the foreign investment regi me throughout the 1990s. However, some important sales of government-owned en tities were slowed by bureaucratic considerations and had yet to be resolved by the end of 2001 (EIU 2003). Thus, FDI in Honduras accounted for just over 3% of GD P in that year. The Honduran economy was dominated by the industrial a nd service sectors; strong tex tile and tourism industries led economic growth to an annual average of 2.7% from 1997 to 2001. Substantial inflows of official develo pment assistance also contributed to Honduras economic expansion. In fact, infl ows of ODA were more than 3-times the magnitude of FDI in 2001. This can be comp ared to the cases of Nicaragua and Guyana, where official inflows were 7-times and 2-times larger than inflows of FDI, respectively. These three countries were the only in whic h receipts of offici al aid consistently exceeded inflows of FDI. Thus, it is intere sting to note that, across the entire sample, they were also the countries in which FDI was the least responsive to changes in the investment climate. From 1970 to 1997, civil war and unrest kept FDI flows into El Salvador to an average of only 0.33% of GDP. Howeve r, the political and economic environment improved rapidly once an official peace accord ending 12 years of civil war was signed in

PAGE 122

111 1992, and inflows of FDI increased from zer o in 1994 to nearly 2% of GDP in 2001. Foreign direct investment increased to near ly 9.2% of GDP in 1998 and much of this investment was attracted by major privatizations in the telecommunications and electric utilities sectors. Although that level of i nvestment ultimately proved unsustainable, the adoption of an investment pr omotion program by the government kept inflows of FDI positive from 1999 to 2001. The results of the analysis suggest that although investment conditions were generally favorable in El Salvador during the late 1990s, other factors may have kept investors from entering the country on a la rge scale. One-unit improvements in the investment profile and government stability sc ores caused FDI in El Salvador to increase by 3.4% and 5.8%, respectively. Meanwhile, a 10% decrease in the corporate tax rate increased FDI by 3.1%. Given that the US dollar was adopted by the government of El Salvador as the legal tender in January 2001, the elimination of any variability in the exchange rate had no impact on the level of FDI. The case of El Salvador is interesting for two reasons. First, despite being what the EIU (2003) called one of the most stab le economies in Latin America, the ICRG investment profile score for El Salvador fell fr om 10.0 in 1999 to 7.5 in 2001. In contrast to most of the countries in the samp le, there seems no obvi ous foundation for the observed decline in the i nvestment profile rating. The second interesting thing about El Salva dor is that, even with such an open economy, inflows of FDI showed only a m oderate response to improvements in the investment climate. In this regard, the EIU (2003) points to the countrys high rate of crime as a potential explanation for why El Salvador lagged behind other Central

PAGE 123

112 American countries like Costa Rica, Nicara gua and Honduras in it s ability to attract foreign investment. However, given an im provement in the investment climate, the results indicate that FDI in El Salvador w ould be more responsive to changes in the investment climate than in all but one of the other CACM countries. The only CACM country in which FDI was more responsive than in El Salvador was Guatemala, the largest CACM economy. Guatemalas investment profile score was also the highest among the CACM countries in 2001. Nonetheless, inflows of FDI remained low. Investment during the pe riod 1998-2001 was almost solely driven by privatization-related inflows. One explanation for the lack of widespread FDI is that the confidence of foreign investors in Guatemala has, from time to time, been shaken by major political and macroeconomic disturbances. In fact, th e EIU (2003) indicates that the political uncertainty associated with Mr. Jorge Serrano Elias attempted dissolution of Guatemalas government in 1993 was primarily reason for the lack of FDI. The EIU (2003) also states that the ne gative impact of an expansi onary monetary policy was the primary reason FDI inflows remained subdue d during the 1990s. However, monetary discipline during 2000 and 2001 restored confidence and FDI increased by nearly 1.5% of GDP. Overall, FDI in the CACM countries was not particularly responsive to changes in the investment climate. While Guatemala had the largest response in the group, it was still well below the most responsive countries in other groups. In that regard, the results for CACM are similar to those observed for the countries within CARICOM; 4 out of 5

PAGE 124

113 countries in each group had very low responses while the size of the response in the fifth country was almost double that of any other in the group. Mexico and Chile When compared to other countries in La tin America, FDI flows to Mexico and Chile were much more consis tent throughout the period 1995-2001. In fact, Chile is an interesting case in terms of both the quickness of its recovery from the Latin American debt crisis and its level of integration with the global economy since the crisis. Inflows of FDI in Chile averaged 7.6% of GDP during the period 1996-2001, which was well above other comparably-sized countries in Latin America. Chiles relative success at creating a favor able investment climate can be witnessed by examining Table 5-11. Following the debt crisis, the governmen t took a series of well-thought-out steps (the details of which are discussed below) aimed at maintaining investor confidence. As the table shows, a low tax rate and high ratings for the investment profile, government stability a nd corruption scores continued to provide incentive for foreign investors in 2001. Table 5-11. Investment C limate Variables Mexico and Chile (Benchmark = 2001) Investment Profile Score Government Stability Score Corruption Score Corporate Tax Rate Exchange Rate Variability Mexico 11.5 8.0 2.0 0.35 0.01147 Chile 10.5 10.5 4.0 0.15 0.11911 However, the results show that FDI in Chile was only moderately responsive to changes in the investment climate when compar ed to other countries in the sample. In fact, the response of FDI in Ch ile (shown in Table 5-12) was comparable to the responses measured in Nicaraguas FDI. For instance, a one-unit increase in the investment profile rating resulted in a 2.49% increase in FD I in Chile, and the same change in the

PAGE 125

114 government stability score increased FDI by 4.24%. Chile also recorded the lowest response to a decrease in the corporate tax rate. Table 5-12. FDI Responsiveness Mexico and Chile Responsiveness of FDI to a Change in: Investment Profile Score (1-unit increase) Government Stability Score (1-unit increase) Corruption Score (1-unit increase) Corporate Tax Rate (10% decrease) Exchange Rate (fix to US$) Mexico $930,473,580 $1,582,193,077 -$123,126,529 $1,189,000,245 $175,526 Chile 100,077,723 170,173,859 -13,242,958 54,807,315 196,055 Mexico 4.27% 7.26% -0.56% 5.45% 0.00% Chile 2.49% 4.24% -0.33% 1.36% 0.00% The empirical evidence seems to suggest that the results of the analysis are valid. While the investment climate in Chile was highly rated by the ICRG in 2001 (and for that matter, much of the late 1990s), the ratings were much lower in 1988. Nonetheless, inflows of FDI accounted for 4% of GDP in 1988, compared to 6.7% in 2001. Such a small change 2.7 percentage points in FDI over that 13-year period was uncharacteristic for most of the countries in th e sample. In fact, a large majority of the countries in the region attracted much le ss FDI during the late 1980s than in 2001. The fact that Chile began to recover fr om the Latin American debt crisis and attract large inflows of FDI before other countries in S outh America is largely a product of two actions taken in the 1980s. Following th e crisis, the governme nt of Chile assumed responsibility for over $6 billion worth of priv ate debt and enacted a system of debt-toequity swaps that were successful in restor ing the confidence of foreign investors (EIU 2003). As a result, FDI flows to Chile recove red quickly following the crisis (reaching 4% of GDP by 1987) and generally trende d upward during the subsequent 15-year period. Although the ICRG ratings for Chile followed the same general trend, year-to-

PAGE 126

115 year fluctuations in the inde pendent variables had a small im pact on the level of FDI. This is consistent with what the results of the analysis suggest. Among the developing countries in the Western Hemisphere, Mexico was the second largest recipient of FDI flows during the 1984-2001 period (Brazil was the largest). As the investment profile score for Mexico improved over time, FDI flows began to account for a greater percentage of GDP. The most noticeable increase in FDI came during the period immediately following the establishment of NAFTA. As might be expected, investors from the United Stat es accounted for the bulk of FDI in Mexico. However, Mexicos openness to foreign i nvestment had been established prior to NAFTA by the Foreign Investment Act of 1993. In fact, FDI jumped from $4.39 million in 1993 to $10.72 million in 1994 and never fe ll back below $9.0 million after that. Interestingly, inflows of FDI remained strong during the 1994-1995 Mexican Peso crisis despite a deterioration in both the investme nt profile and government stability ratings7. Foreign direct investment in Mexico was not particularly responsive to small changes in the investment climate. Fo r example, a one-unit improvement in the investment profile score increased FDI in Mexico by 4.27%. A one-unit change in the government stability score had the effect of a 7.26% change in FDI, and decreasing the corporate tax rate by 10% led to a 5.45% increase in FDI. As a point of reference, the level of response in Mexico was most co mparable to the countries of Uruguay and Paraguay. 7 The poor state of Mexicos political environment in 1994 was exemplified by the assassination of both Lois Donaldo Colosio and Jose Francisco Ruiz Ma ssieu, the new presidential candidate and secretary general of the ruling party at that time, respectively.

PAGE 127

116 The results of the analysis show that the level of response in Mexican FDI was also close to the median response across the sample Mexico and Brazil were comparable in terms of economic size and dollar level of FDI in 2001, although FDI in Mexico was much less responsive to changes in the investme nt climate. Within the data, there are no obvious differences in the two countries that would account for the difference in FDI response. It is with this last point that the next ch apter is concerned. That is, there are many differences and similarities between the count ries in this analysis in terms of FDI response. Some of the corollaries were exp licitly pointed out above, while others were perhaps only alluded to. The countries were segmented according to their respective regional integration groups in this chapter, but there ar e other ways in which the countries might be segmented. The most obvious segmen tation is into two groups : (i) countries in which the response of FDI to changes in the investment climate was high and (ii) countries in which the estimated response wa s low. Chapter 6 discusses some of the traits shared by the most res ponsive countries in the analys is, and likewise, the traits shared by the least responsive countries.

PAGE 128

117 CHAPTER 6 SUMMARY AND CONCLUSIONS The purpose of this study was to provide in sight into the nature of the relationship between foreign direct investment (FDI) and the investment climates of developing host countries in the Western Hemisphere. While it was expected from the outset of the study that the foreign investment pol icies adopted by these countries indeed impact the extent to which foreign investors find them to be either suitable or unsuitable locations for direct investment, the results indicate that the impor tance of investment climate often depends on the country being examined. That is to say, in some countries, changes in the investment climate resulted in large changes in the level of FDI, while in other cases, the impact of a similar change appeared to be, in effect, dampened by other factors. The responsiveness of FDI was defined as the percentage change in FDI resulting from hypothetical changes in each investment climate variable. The most recent data available was used to create a benchmarking s cenario in order to ex amine the effect of a hypothetical change in each of the right hand side (RHS) variables, independently. The results obtained from the empirical analysis make it possible to characterize the countries according to their level of FDI response. Howe ver, it is useful to revisit the cases of several specific countries before providing some broad conclusions on the nature of the observed responses in FDI across the entire sample. The Least Responsive Countries Guyana had one of the highest levels of tr ade (as a percentage of GDP), one of the highest levels of FDI (also as a percenta ge of GDP), and the lowest level of FDI

PAGE 129

118 responsiveness in the sample; the results for Nicaragua we re similar. Although there were few other likenesses between these two countries, inflows of official development assistance (ODA) to both Nicar agua and Guyana accounted for a higher percentage of GDP than in any of the other sample countri es. Honduras, another country in which FDI was minimally responsive to changes in the i nvestment climate, also received a large amount of ODA in the benchmark year. Together, the three cas es mentioned above seem to s uggest that flows of FDI and ODA to developing countries are complement ary to one another, and that this relationship may overshadow the effect of cha nges in the investment climate. In fact, there is support for this conc lusion in the existing literatur e. Specifically, Dasgupta and Ratha (2000) found that FDI re sponds positively to official capital inflows (including World Bank lending). They also found that this effect domina ted any relationship between FDI and the real and financial variables (Dasgupta and Ratha 2000, p16) included in their model. Another factor that appears to be linked with a smaller response in FDI is the concentration of foreign inte rest in a single industrial se ctor of the economy. Some countries that fit this mold are Jamaica, Tr inidad and Tobago, and Ecuador. As pointed out in Chapter 5, the tourism industry acc ounted for over 50% of FDI and roughly 15% of total economic production in Jamaica. In both Ecuador and Trinidad and Tobago, a majority of FDI was directed at the petrochemical industry. In fact, in most, if not all of the developing countries of the Western He misphere, FDI accounts for a majority of investment in the tourism and petrochemical industries. From the results of this analysis, it is again plausible to concl ude that a factor not accounted for in the regression equation

PAGE 130

119 caused FDI to be less responsive to changes in the investment climate. For example, it could be that the demand for suitable vacation resorts on the north coas t of Jamaica is so great that international reso rt developers may place less of an emphasis on the overall investment climate as a result of their inte rest in capitalizing on opportunities in the tourism industry. A similar analogy could be made for both Ecuador and Trinidad with regard to international dema nd for oil and natural gas. Finally, the least responsive countries in the sample had a history of maintaining government-owned monopolies in a number of industries. For example, while Paraguay and Uruguay had the highest average i nvestment profile ratings among both MERCOSUR and Andean Community countries, FDI accounted for less than 2% of GDP in each. Additionally, the results indicate that improvements in the investment climates of these countries would lead to only moderate increases in FDI. However, this is not surprising given the tradition of limitation on FDI that persists in these countries. The Most Responsive Countries Within both the Andean Community and MERCOSUR, countries that were more open to privatization tended to be the most responsive to changes in the investment climate. For instance, Argentina and Columb ia each initiated extensive privatization programs in the early 1990s and Brazil followed suit in the latte r years of the decade. As such, direct investment in these countries from 1997-2001 tended to be spread across a variety of economic sectors. In the case of Venezuela, FDI was also encouraged in some industries throughout the 1990s. However, economic uncertainty and extensive limitation on foreign investment was the nor m in most industries in that country throughout the 1990s. Although foreign interest in Venezuelas petrochemical reserves was high, the underlying state of the overall investment climat e served as a deterrent to

PAGE 131

120 FDI. In both Argentina and Columbia, conditions were somewhat similar in the sense that domestic instability was cited as a major problem for foreign investors. The results suggest that any improveme nt in the investment climates of Venezuela, Argentina or Brazil would lead to relatively large increases in the amount of FDI that these countries attr act. One possible conclusion is that, regardless of whether these countries may have the desired natural re sources and consumer ba se to be a suitable location for FDI, investment has been negative ly affected by the unfavorable state of their investment climates. This is perhaps also tr ue in the second largest Caribbean market in terms of population, Haiti, wh ich also showed one of the hi ghest levels of FDI response in the sample. The idea that FDI is more responsive to small improvements in the investment climate in countries where economic and politic al conditions are at their worst is very interesting. Is it that foreign investors are optimistic with respect to the possibility of improved fortunes in those countries? Altern atively, it could be that a history of instability has left the count ries with so little investme nt that a vast amount of opportunities open themselves once the invest ment climate starts to improve. Policy-Related Conclusions One of the goals of this study was to provi de a model for analyzing the investment climate that, in the end, had some relevan ce for policymakers in developing countries of the Western Hemisphere. While it is diffic ult to provide specific guidance on national foreign investment policy by examining a group of countries as a whole, the empirical analysis in Chapter 5 does yield some inte resting information for those who are charged with designing these policies. The first im portant inference is that the state of the investment climate is, in some cases, not the principal concern of fo reign direct investors

PAGE 132

121 when choosing among potential locations for inve stment. This is particularly important for policymakers who have been successful at creating a favorable investment climate in their respective country, but ha ve not reaped the benefit of increases in inward FDI. For countries in which the investment climate was favorably rated yet FDI was relatively unresponsive in compar ison to other countries in th e sample, the results suggest the need to develop a stronger industrial ba se for FDI activity Paraguay and Uruguay serve as examples of countries for which this is true. While the investment climate in these countries ranked favorably, historical restrictions on private commerce have limited the development of an economic infrastructure that is capable of absorbing large inflows of FDI. Thus, these countries attract a ve ry small amount of FDI, and improvements in the investment climate have done little to change this. Howe ver, the inference is no less relevant for countries that hist orically received a large amount of FDI. In these countries, one or two industries tend to attract a si gnificant amount of FDI, while the lack of development in other industries limits the pote ntial effect of further improvements in the investment climate. Overall, each of th e two aforementioned in stances highlight the importance of fostering industries that lend themselves to FDI activity. On the contrary, the investment climate was shown to be of acute importance to foreign direct investors in some of the most developed countries in the sample. This result supports the conclusion that while th e level of industrial development in these countries may be suitable for direct invest ment, changes in the investment climate can substantially affect the am ount of FDI received by thes e countries. Thus, the implementation of policies that are considered by foreign direct investors to be more favorable is an area worthy of attention.

PAGE 133

122 The data presented in Chapter 5 also show that the most underdeveloped countries in the Western Hemisphere attracted substant ial flows of FDI in relation their economic size. Additionally, neither impr ovements nor deteriorations in the investment climates of these countries had a large impact on their ability to attract FDI. As previously mentioned, this finding is consistent with the contentions of other researchers who purport a complementary relationship between ODA and FDI. The primary conclusion here is that the policies that must be enacted for a country to qualify for official assistance also have the potential to spur fore ign investment in the private sector. The broadest conclusion that can be drawn from the results of this study is that the responsiveness of FDI to ch anges in the investment climate varies widely across countries. For policymakers, this might sugge st that the level of priority given to reforming the foreign investment policy re gime should be determined on a country-bycountry basis. While in some countries the st ate of the investment climate appears to be a pressing issue, other countries might do well to devote their resource s to other areas of concern, as outlined above. Considerations for Future Research Panel data were used in this study to expose similarities and differences in the response of FDI across a large group of countri es. However, it would be interesting to conduct more in-depth analysis on a country-by-c ountry basis. It is likely that doing so would enable future researchers to provide insight into the relationship between the responsiveness of FDI (as defined in this st udy) and the factors not fully accounted for in this analysis, such as industrial concentration and economic stability. Testing for causality between changes in the investment climate variables and changes in FDI would also be a worthwhile exercise. Tests for causality between the

PAGE 134

123 independent variables might uncover feedback e ffects that the tests conducted in Chapter 5 did not detect. Additiona lly, testing for causality betw een FDI and the independent variables could provide furthe r evidence of the importance of foreign investment policy. Finally, one of the major problems associ ated with examining the relationship between foreign investment policy and FDI has been the difficulty of measuring the data. To a great extent, foreign investment policy consists of legal provisions and considerations that while easily elucidated with legal jargon, are nearly impossible to account for numerically. As a re sult, empirical literature on the subject is abundant with proxies for policy-related variables. Th e World Bank recognized this void in the available data and recently esta blished the Doing Business Database1 in an effort to provide more objective criteria for measuri ng business regulations and the extent to which they are enforced. The Database includes numerical measures of business regulations in 145 countries. Seven aspects of the regulator y environment are covered, including: (i) starting a busine ss; (ii) hiring and firing worker s; (iii) registering property; (iv) getting credit; (v ) protecting investors; (vi) enforc ing contract; and, (vii) closing a business. Although the lack of historical coverage preven ted the Database from being useful for this study, its potential va lue to future researchers is clear. 1 The Doing Business Database can found at http://rru.worldbank.org/DoingBusiness/

PAGE 135

124 APPENDIX DEFINITIONS OF SELECTED VARIABLES World Bank Data Foreign Direct Investment, Net Inflows (% of GDP) Foreign direct investment is defined as i nvestment to acquire a lasting management interest (10 percent or more of voting stock) in an ente rprise operating in an economy other than that of the investor. It is the su m of equity capital, re investment of earnings, other long-term capital, and short-term capital as shown in the balance of payments. This series is measured as net in flows in the reporting economy. Trade (% of GDP) Trade is the sum of exports and imports of goods and serv ices measured as a share of gross domestic product. GDP Per Capita (Constant 1995 US$) GDP is the sum of gross value added by a ll resident producers in the economy plus any product taxes and minus any subsidies not in cluded in the value of the products. It is calculated without making deductions for depr eciation of fabricated assets or for depletion and degradation of natural resources. GDP per capita is measured as gross domestic product divided by midyear population. Data are in constant U.S. dollars. International Country Risk Guide Data Investment Profile A measure of the government's attitude toward inward investment as determined by four components: the risk to operations, taxation, repatriation, and labor costs.

PAGE 136

125 Corruption A measure of corruption within the political system that is a threat to foreign investment by distorting the economic and fina ncial environment, reducing the efficiency of government and business by enabling peopl e to assume positions of power through patronage rather than ability and introducing inherent inst ability into the political process. Government Stability A measure of the government's ability to st ay in office and carry out its declared program(s), depending upon such factors as the type of governance, cohesion of the government and governing parties, approach of an election, and command of the legislature.

PAGE 137

126 LIST OF REFERENCES Addison, Tony and Heshmati, Almas. The New Global Determinants of FDI in Developing Countries: The Importance of ICT and Democratization. Discussion Paper number 2003/45 United Nations University World Institute for Development Economics Research (05/27/03). (2003): 31 pages. Agarwal, Sanjeev and Ramiswami, Sridhar N. Choice of Foreign Market Entry Mode: Impact of Ownership, Location and In ternalization Factors. Journal of International Business Studies 23.1 (1992): 1-27. Aitken, Brian J. and Harrison, Ann E. Do Do mestic Firms Benefit from Direct Foreign Investment? Evidence From Venezuela. The American Economic Review 89.3 (1999): 605-618. Andean Community General Secretariat. Who Are We?-Andean Community http://www.comunidadandina .org/ingles/who/crono.htm (06/11/03). 2003. Barrell, Ray and Pain, Nigel. An Econometr ic Analysis of U.S. Foreign Direct Investment. The Review of Economics and Statistics 78.2 (1996): 200-207. Basi, Raghbir S. Determinants of United Stat es Private Direct Investments in Foreign Countries Kent, Ohio: Kent State University Press. 1966. Birdsall, Nancy and Lozada, Carlos E. Recu rring Themes in Latin American Economic Thought: From Prebisch to the Market a nd Back. In Securing Stability and Growth in Latin America: Policy Issues and Prospects for Shock-Prone Economies eds. Ricardo Hausmann and Helmut Reisen Paris: Organisation for Economic Cooperation and Development. (1996): 11-21. Blomstrm, Magnus. Foreign Investment and Productive Efficiency: The Case of Mexico. Journal of Industrial Economics 35.1 (1986): 97-110. Blomstrm, Magnus and Kokko, Ari. Multina tional Corporations and Spillovers. CEPR Discussion Paper no. 1365 http://www.cepr.org/pubs/dps/DP1365.asp (9/12/03). (1996): 49 pages. -------- The Economics of Foreign Direct Investment Incentives. NBER Working Paper: w9489 http://www.nber.org/papers/9489 (11/16/03). February. (2003): 26 pages.

PAGE 138

127 --------. How Foreign Invest ment Affects Host Countries . Working Paper number 1745 World Bank. (1997): 52 pages. Borenzstein, Eduardo; de Gregorio, Jos and Lee, Jong-Wha. How Does Foreign Investment Affect Economic Growth? Journal of Inte rnational Economics 45.1 (1998): 115-135. Buckley, Peter J. and Casson, Mark C. The Future of the Multinational Enterprise New York: Holmes and Meier. 1976. --------. An Economic Model of International Joint Vent ure Strategy. Journal of International Business Studies 25.5 (1996): 849-876. --------. Analyzing Foreign Market Entry Strategies: Extending the Internalization Approach. Journal of In ternational Business Studies 29.3 (1998): 539-561. Butler, Kirt C. and Joaquin, Domingo C. A Note on Political Risk and Required Return on Foreign Direct Invest ment. Journal of International Business Studies 29.3 (1998): 599-607. Calderon-Rossell, Jorge R. Towards the Theory of Foreign Direct Investment. Oxford Economic Papers 37.2 (185): 292-291. Calvo, Guillermo A.; Leiderman, Leonardo, and Reinhart, Carmen M. Inflows of Capital to Developing Countries in the 1990s. Journal of Ec onomic Perspectives 10.2 (1996): 123-139. Caribbean Community Secretariat. The History of CARICOM http://www.caricom.org/archives/history.htm (06/24/03). 2003. Coase, R.H. The Nature of the Firm. Economica 4.16 (1937): 386-405. Contractor, Farok J. and Kundu, Sumit K. Modal Choice in a World of Alliances: Analyzing Organizational Forms in the In ternational Hotel Sector. Journal of International Business Studies 29.2 (1998): 327-357. Darby, Julia; Hughes Hallett, Andrew; Ireland, Jonathon and Piscitelli, Laura. The Impact of Exchange Rate Uncertainty on the Level of Investment. The Economic Journal 109.454 (1999): C55-C67. --------. Exchange Rate Uncertainty and Bu siness Sector Invest ment. Econometric Society World Congress 2000 Cont ributed Papers number 0600 http://fmwww.bc.edu/RePEc/es2000/0600.pdf (2/2/04). (2000): 34 pages. Dasgupta, Dipak and Ratha, Dilip. What Fact ors Appear to Drive Private Capital Flows to Developing Countries? And How Does Official Lending Respond? World Bank Policy Research Working Paper number 2392 http://econ.worldbank.org/docs/1147.pdf (9/14/03). (2000): 23 pages.

PAGE 139

128 Davidson, William H. The Location of Fore ign Direct Investment Activity: Country Characteristics and Experience Effects. Journal of Internatio nal Business Studies 11.2 (1980): 9-22. Demsetz, Harold. The Firm in Economic Theory: A Quiet Revolution. The American Economic Review 87.2 (1997): 426-429. Development Assistance Committee. DAC Statistical Reporting Directives. http://www1.oecd.org/dac/htm/dacdir.htm (5/15/03). 2000. Dunning, John H. The Determinants of International Production. Oxford Economic Papers 25.3 (1973): 289-336. -------- Trade, Location of Economic Activity and the MNE: A Search for an Eclectic Approach. In The International Allocation of Economic Activity eds. Bertil Ohlin, Per-Ove Hesselborn and Per Magnus Wijkman. London: Macmillan. (1977): 395-418. -------- Toward and Eclectic Theory of Intern ational Production: Some Empirical Tests. Journal of Interna tional Business Studies 11.1 (1980): 9-31. -------- Reappraising the Eclectic Paradigm in an Age of Alliance Capitalism. Journal of International Business Studies 26.3 (1995): 461-491. Easterly, William; Kremer, Michael; Pritche tt, Lant and Summers, Lawrence H. Good Policy or Good Luck? Country Growth Performance and Temporary Shocks. Journal of Monetary Economics 32.3 (1993): 459-483. Economic Commission for Latin America and the Caribbean (ECLAC). Globalization and Development. http://www.un.org/esa/usg_o campo/present/200205_globdev.pdf (7/20/03). 2002: 38 pages. Economist Intelligence Unit (EIU). C ountry Profile: [Var ious Countries]. http://www.eiu.com/site_info.asp?inf o_name=ps_country_reports&entry1=psNav (4/22/04). 2003. Ethier, Wilford J. The Multinational Fi rm. The Quarterly Journal of Economics 101.4 (1986): 805-834. Gabel, Medard and Bruner, Henry. Global Inc.: An Atlas of the Multinational Corporation New York: New Press, 2003. Goldberg, Linda S. Exchange Rates and I nvestment in United States Industry. The Review of Economic and Statistics 75.4 (1993): 575-588. Greene, William H. Econometric Analysis 4th ed. Saddle River, New Jersey: PrenticeHall, 2000.

PAGE 140

129 Hanratty, Dennis M. and Sandra W. Me ditz, eds. Columbia: A Country Study Washington DC: GPO for the Library of Congress, 1989. Hausmann, Ricardo and Gavin, Michael. Secu ring Stability and Growth in a ShockProne Region: the Policy Challenge for Latin America. In Securing Stability and Growth in Latin America: Policy Issues and Prospects for Shock-Prone Economies eds. Ricardo Hausmann and Helmut Reisen Paris: Organisation for Economic Cooperation and Development. (1996): 23-64. Helpman, Elhanan. A Simple Theory of International Trade with Multinational Corporations. The Journal of Political Economy 92.3 (1984): 451-471. Hennart, Jean-Franois and Park, Young-Ry eoul. Greenfield vs Acquisition: The Strategy of Japanese Investors in th e United States. Management Science .39.9 (1993): 1054-1070. Holmstrm, Bengt and Roberts, John. The B oundaries of the Firm Revisited. Journal of Economic Perspectives 12.4 (1998): 73-94. Hoti, Suhejla. Trends and Volatilities in International Capital Flows for Developing Economies. In Integrated Assessment a nd Decision Support, Proceedings of the First Biennial Meeting of the Internati onal Environmental Modeling and Software Society Volume 2. eds. A.E. Rizzoli and A.J. Jakeman. Manno, Switzerland: International Environmental Modeling a nd Software Society. (2002): 544-549. Hudson, Rex A., ed. Chile: A Country Study Washington DC: GPO for the Library of Congress, 1995. Hymer, Stephen. The International Operati ons of National Firms: A Study of Foreign Direct Investment Cambridge, Massachusetts: MIT Press. 1976. Itaki, Masahiko. A Critical Assessment of the Eclectic Theory of the Multinational Enterprise. Journal of In ternational Business Studies 22.3 (1991): 445-460. Kim, Yoonbai. Causes of Capital Flows in Developing Countries. Journal of International Money and Finance 19 (2000): 235-253. Kim, W. Chan and Hwang, Peter. Global Strategy and Multinationals Entry Mode Choice. Journal of Intern ational Business Studies 23.1 (1992): 29-53. Lecraw, Donald J. Factors Influencing FD I by TNCs in Host Developing Countries: A Preliminary Report. In Multinational En terprises in Less Developed Countries eds. Peter J. Buckley and Jeremy Clegg. Ne w York: St. Martins Press. (1991): 163180. Liberatori, Maura. A Database on Foreign Direct Investment. http://www1.worldbank.org/economic policy/globalization/data.html (11/01/03). 2003.

PAGE 141

130 Loree, David W. and Guisinger, Stephen E. Policy and Non-policy Determinants of U.S. Equity Foreign Direct Investment. J ournal of International Business Studies 26.2 (1995): 281-299. Lusinyan, Lusine. International Capital Flow s, Economic Growth and Financial Market Efficiency. European University Institute Working Paper : 2002/20. http://www.iue.it/PUB/ECO2002-20.pdf (7/21/03). 2002: 39 pages. Markusen, James R. Multinationals, Multi-Plant Economies and the Gains from Trade. Journal of Intern ational Economics 16.3/4 (1984): 205-226. -------- The Boundaries of the Multinational Enterprise and the Theory of International Trade. The Journal of Economic Perspectives 9.2 (1995): 169-189. Merrill, Tim, ed. Honduras: A Country Study Washington DC: GPO for the Library of Congress, 1994. Miller, Robert R. and Weigel, Dale R. The Motivation for Foreign Direct Investment. Journal of Interna tional Business Studies 3.2 (1972): 67-79. Mishkin, Frederic S. Financial Policies and the Prevention of Financial Crises in Emerging Market Economies. World Ba nk Policy Research Working Paper Series: 2683 Washington D.C.: The World Bank. (2001): 52 pages. Mody, Ashoka and Srinivasan, Krishna. Japan ese and U.S. Firms as Foreign Investors: Do They March to the Same Tune? The Canadian Journal of Economics 31.4 (1998): 778-799. Moreno, Ramon. What explains capital flows? http://www.frbsf.org/econrsr ch/wklyltr/2000/el2000-22.html (6/15/2003). 2000. Parry, Thomas G. The International Firm and National Economic Policy: A Survey of Some Issues. The Economic Journal 83.332 (1973): 1201-1221. Pedersen, Kurt. The eclectic paradigm: 25 years anniversary. Paper prepared for the First International Business and Economy Co nference, San Francisco, January 6-9, 2002. (2002): 9 pages. Phelan, Steven E. and Lewin, Peter. Arri ving at a Strategic Theory of the Firm. International Journal of Management Reviews 2.4 (2000): 305-323. Philippe, Beatrice. Sector vs. Geography: The Medium Term Development Outlook. http://www.ifecorp.com/Papers-PDFs/Philippe701.pdf (03/06/03). (2001). Pugel, Thomas A. The Determinants of Fore ign Direct Investment: An Analysis of U.S. Manufacturing Industries. Mana gerial and Decision Economics 2.4 (1981): 220228.

PAGE 142

131 Robinson, Joan. Economics is Serious Subject The Apologies of an Economist to the Mathematician, the Scientists and the Plain Man Cambridge, Massachusetts: Heffer. 1932. Rojas-Suarez, Liliana and Steven R. Weisbrod. Building Stability in Latin American Financial Markets. In Securing Stability and Growth in Latin America: Policy Issues and Prospects for Shock-Prone Economies eds. Ricardo Hausmann and Helmut Reisen. Paris: Organisation for Economic Co-operation and Development. (1996): 139-166. Sandrasagra, Mithre J. FDI to Exceed 1 Trillion Dollars. http://194.183.22.90/ips/ENG.NSF/ vwWebMainView/550BD155BE10FC0A80256 A07005D9866/ (06/17/03). 2000. Schmukler, Sergio L. Financial Globa lization: Gain and Pain for Developing Countries. Economic Review 89.2 (2004): 39-66. Sealy, Tom, ed. International Country Risk Guide East Syracuse, NY: The PRS Group, Inc., 2003. Simon, Jeffrey D. A Theoretical Perspective on Political Risk. Journal of International Business Studies 15.3 (1984): 123-143. Smarzynska, Beata K. Does Foreign Direct Investment Increase the Productivity of Domestic Firms? In Search of Spillove rs through Backward Linkages. World Bank Policy Research Working Paper: 2923 Washington D.C.: The World Bank. (2002): 30 pages. Stobaugh, Robert B., Jr. How to Analyze Foreign Investment Climates. Harvard Business Review 47.5 (1969): 100-108. Taylor, Timothy G. and Seale, James L., Jr International Trade and the Firm. In Agricultural Globalization, Trade, and the Environment eds. Charles B. Moss, Gordon C. Rausser, Andrew Schmitz, Timo thy G. Taylor and David Zilberman. Norwell, Massachuse tts: Kluwer Academic Publishers, 2001. Vernon, Raymond. International Investment and International Trade in the Product Cycle. Quarterly Journal of Economics 80.2 (1966): 190-207. Wei, Shang-Jin and Shleifer, Andrei. Lo cal Corruption and Global Capital Flows. Brookings Papers on Economic Activity 2000.2 (2000): 303-354. Wells, Louis T. (Jr.). Multinationals and the Developing Countries. Journal of International Business Studies 29.1 (1998): 101-114. Williamson, John. Issues Regarding the Composition of Capital Flows. Paper prepared for the Overseas Development Institute C onference Capital Account Liberalisation: A Developing Country Perspective, L ondon, June 21, 2000. (2000): 16 pages.

PAGE 143

132 Woodward, Douglas P. and Rolfe, Robert J. The Location of Export-Oriented Foreign Direct Investment in the Caribbean Ba sin. Journal of International Business Studies 24.1 (1993): 121-144. World Bank (WB). Global Development Finance 2001 Washington D.C.: International Bank for Reconstruction and Development. (2001). --------. CD-ROM World Development Indicators Washington D.C.: World Bank. 2003. Zejan, Mario C. New Ventures or Acquisi tions: The Choice of Swedish Multinational Enterprises. Journal of Industrial Economics 38.3 (1990): 349-355.

PAGE 144

133 BIOGRAPHICAL SKETCH Eric Bonnett was born in Bradenton, Flor ida, on December 17, 1976, and is a fifthgeneration Floridian. His wife, Rita, is also a Florida native. Eric holds both a Bachelor of Science degree and a Master of Agribusiness degree from the University of Florida. In August 2001, he was awarded the Universi ty of Florida Alumni Fellowship and subsequently received his Doctor of Philosophy degree in December 2004.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20110114_AAAAEG INGEST_TIME 2011-01-14T21:50:29Z PACKAGE UFE0008325_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 23655 DFID F20110114_AACQNJ ORIGIN DEPOSITOR PATH bonnett_e_Page_117thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
79b4c73b2fc469e6d07062b2822296bb
SHA-1
a045c7efb7f90d0684e72cab9088db9af300bfa6
25092 F20110114_AACQMV bonnett_e_Page_110thm.jpg
1b12e6b041e383574c7e3e08e81f11fe
26435a7ebf22a6546ea99d44f0af624723f94298
206197 F20110114_AACPKH bonnett_e_Page_105.jpg
3b4da9dceb80be26f03d980bb6ffbb27
66432353f3e500b39d5bce8a93803196e771f1fd
207900 F20110114_AACPJT bonnett_e_Page_090.jpg
6626e434c245f5a62df56d3eb093e177
62efe5e7d8fe7b7707480c3d65e6ff75cafbf21a
73306 F20110114_AACQNK bonnett_e_Page_117.QC.jpg
148934c99e345abba4a54dee483fb347
7e5c9b6b4e835232991b63a856cb4ec044cb3fb8
76605 F20110114_AACQMW bonnett_e_Page_110.QC.jpg
18d85a7b6d9b51c6755ea52ebf4b141a
be792a8067fe7b70117895a829e764b144fa7e80
201171 F20110114_AACPKI bonnett_e_Page_106.jpg
b4d1e5d8fa5d82c9463a3cc7d85e41ee
532aa540be2fbc87ae90f983472f669903481f18
288433 F20110114_AACPJU bonnett_e_Page_091.jpg
68e6de502fbf68f28e0bf4b827a0a524
9e7623d4895e2f2dd5a1a5345141fbc1170b625a
53244 F20110114_AACQOA bonnett_e_Page_127.QC.jpg
87735fb4495ffc3b5d26041b5d161037
770ae2970d68512f2435a7c292dffd338183c277
23770 F20110114_AACQNL bonnett_e_Page_118thm.jpg
79939f505ea814d7ed43f36d1f925ed6
8672f04b7163264b743814f25a478e2936e97fbf
24919 F20110114_AACQMX bonnett_e_Page_111thm.jpg
bae2453c30a6d4883eaefffafc6bf413
7a03e090f3808273b7642a2072eabcc3222ec7e7
193941 F20110114_AACPKJ bonnett_e_Page_107.jpg
1f415c15bf1881710b05c52f7ad78639
7b1eebb9df031728bcff68dde1ba5bf797cb97cf
174227 F20110114_AACPJV bonnett_e_Page_092.jpg
907fd3d8d1337bf1cbf05ab88a5f80a4
9d0a0d391fd0e0f530480e7758a83f4d9660b8c9
22581 F20110114_AACQOB bonnett_e_Page_128thm.jpg
5198893cfd159cec48341fbc1176cd7e
e3f379545a055e1951857424531cfd58e46ae2d2
23533 F20110114_AACQNM bonnett_e_Page_119thm.jpg
4c6a8b6d09bd8ef6cdb6cb2f7f78df0c
55e8f64411dc1ee88122b66394caa84dab252aef
78639 F20110114_AACQMY bonnett_e_Page_111.QC.jpg
46ae755a2de78cad437193e458c8935b
f7966305ca2e75219b78246f603602ab1e376f9f
206408 F20110114_AACPKK bonnett_e_Page_108.jpg
1d81b5b7336e2ddb7a3361eb473abc46
f90868cd88887cc081269cfacd5ba82d35b5bec3
160988 F20110114_AACPJW bonnett_e_Page_093.jpg
0c433e8ddf6fd7797301190b5742ead1
83392e8233ee3483b9d146f2f7c6e9b62721cbde
70065 F20110114_AACQOC bonnett_e_Page_128.QC.jpg
a686dfe7e1ff1768ad123d5f6bd299c9
2c2958cc84e97ba5cde3c9445abdd70e62fc9566
24232 F20110114_AACQNN bonnett_e_Page_120thm.jpg
2cbfe0c4c7e8da6ce3d49ad4fda4a1e8
f4671988868966b53185726d7c07a24a012431bb
22514 F20110114_AACQMZ bonnett_e_Page_112thm.jpg
2467243ab787ef772ba877162056989c
15be3b9b0a077c3ffdad24252660ef26120df614
200819 F20110114_AACPLA bonnett_e_Page_124.jpg
e56583c956af47fc318b90ec11d8f973
d0c6df4fe3a7f5971458ecc8316871eb9eb1b552
203114 F20110114_AACPKL bonnett_e_Page_109.jpg
7d07d9b2eb3acf186d938da60178c061
3744d5e5bfd8490aa34c1c6d9362e2e1a5a06006
188827 F20110114_AACPJX bonnett_e_Page_094.jpg
9e3382009020525f34029cc517e9d12f
435893be3da5bef75eed2559e361e108646f646a
24609 F20110114_AACQOD bonnett_e_Page_129thm.jpg
0e216abaac0e9db1f687e6db23029748
ff81ff38f3b2798d7555a30d94d9c5549fd2403a
24716 F20110114_AACQNO bonnett_e_Page_121thm.jpg
1ad0abb31d7d6e76c9d7329a0ba9760a
5b55fa64655339bc2a3824b6799042e525b405bf
206391 F20110114_AACPLB bonnett_e_Page_125.jpg
db274092274b6e08fb3eefa82eb3310d
29db8f5475221ad2379837aaf174d00bd3500f80
207557 F20110114_AACPKM bonnett_e_Page_110.jpg
577e7a5be01ed773430f0e36c7136979
0579d9e407f8909b1cffa8eeebe374f0b67e01e3
212013 F20110114_AACPJY bonnett_e_Page_095.jpg
9adacf1b1d9775b45fd5bf44e413b62e
6821cfb8b846c5481de752df19bbca038e5fc083
76744 F20110114_AACQOE bonnett_e_Page_129.QC.jpg
ba8c612efdf6f46c5b698986d081bcac
5cfda3d42dc5ca53d3641517d2c44f38e7d25596
77810 F20110114_AACQNP bonnett_e_Page_121.QC.jpg
241a6b09dd37bda69ae9b30445d65523
e66b379f5722f893973047dba1627c7efebf8076
200248 F20110114_AACPLC bonnett_e_Page_126.jpg
f40de899fd7ddde93cfa215533c578c1
44f88264933a8afd164fbb4203fd33297511abe0
206820 F20110114_AACPKN bonnett_e_Page_111.jpg
0ecb69e5350d8cfaf14fcc87c3683bc6
68e758194974aa341981564e2fab960f4b9c29df
174323 F20110114_AACPJZ bonnett_e_Page_096.jpg
a3d077b6438fb596f304f9c3c147127b
5a082863219c716045a3e3c9439e5a23fb69706e
24811 F20110114_AACQOF bonnett_e_Page_130thm.jpg
a87ee0f07a0cc832831a98803ce5ec50
6ad1d546bd8a95029cb6f8d84578e35faf5e799d
23738 F20110114_AACQNQ bonnett_e_Page_122thm.jpg
149216acd42c62dc3bccc9c68d1c3e28
10bcbf161e5d36e7a614286f5d81da41624ea1bd
139697 F20110114_AACPLD bonnett_e_Page_127.jpg
26c1210910dee4f992d39c5f41a0d753
152d83e9d98e8a61cf47658f97739175dec0e672
197794 F20110114_AACPKO bonnett_e_Page_112.jpg
b284d5efd737370077af5a234c495bcc
c2e6927b8ed3ca8fb088fbe5549fe7a8c6286c0c
80978 F20110114_AACQOG bonnett_e_Page_130.QC.jpg
edb65b5d8086f5d70fda30d720ae6cf9
08bbc690e2b99039c4c46141784e085b71c4e49a
74418 F20110114_AACQNR bonnett_e_Page_122.QC.jpg
5f98fe62761a17ca1d39785a3757b3dd
857c99afa23bbf9c8553c40b51a75183df9d4349
216075 F20110114_AACPLE bonnett_e_Page_130.jpg
bd9aff6ea899b83b075e87f4320b9629
04c2aa2aa39e45d54498cbfa0152f3c9c5be3a01
213120 F20110114_AACPKP bonnett_e_Page_113.jpg
80a2ad1f2a7b0f2746c2b7f6cba55573
1232d155a6914864d32a1f742578311b9fd26066
25422 F20110114_AACQOH bonnett_e_Page_131thm.jpg
71b46ed1ab51be212a91e5acca1ad989
8ea803e09aef78f0924cd4645f126860f3194d00
22960 F20110114_AACQNS bonnett_e_Page_123thm.jpg
f70bb725e838c3ffd6bf7958c4e72f72
430acadabf36e04b1be9e86c341a412e37314a84
208906 F20110114_AACPLF bonnett_e_Page_131.jpg
554b43d5ad00ea348750e86df4676e74
6bdda427f98ef174a5a6ffd83ebf2fbe8bd98c1a
215863 F20110114_AACPKQ bonnett_e_Page_114.jpg
8058b68537899548b98d2b9a07361776
89dbbeeeb0ff36ad569972302a19a9291e737c31
79968 F20110114_AACQOI bonnett_e_Page_131.QC.jpg
74418f1627b055dfc195e9356004d88c
16484fea16bccbd12424a29b88b06bef97bf1fd7
71851 F20110114_AACQNT bonnett_e_Page_123.QC.jpg
9ddd3e4990307d951453f1502a49861b
0b9e6b7a0ce0fc56e77c359f4ea4287dd066c9eb
212824 F20110114_AACPLG bonnett_e_Page_132.jpg
2e8250e25d410360e0f1b33ee6369604
a0ea46af9625dd3de8abedbd49e3a938784b885a
207128 F20110114_AACPKR bonnett_e_Page_115.jpg
c32f30620c6a688deacf2ef686cdf342
7f129dc6321ebecb8cf1c87ee05c6385bd5fd452
24297 F20110114_AACQNU bonnett_e_Page_124thm.jpg
466afb3a1d50b1beef8d90eaaf4e4518
58e0cd802045fb1b91bb55d9c3182df358e0385b
199546 F20110114_AACPKS bonnett_e_Page_116.jpg
d9e2ddc9173e33542e0152b10a8ff2c2
97d1aa9829de3f33050d65d5af6d75b05a246081
25210 F20110114_AACQOJ bonnett_e_Page_132thm.jpg
867588b9334a170ca153683516a1c7e6
16e86d720ee52a2320a2af5f031eced9ca43253f
73061 F20110114_AACQNV bonnett_e_Page_124.QC.jpg
91ad293c79d1ae47261b341ba820e08c
0c1c046d3aefe7baeb159876926507d29c230aa7
209136 F20110114_AACPLH bonnett_e_Page_133.jpg
b182898f88eebfe447670bb83ef200a8
2b7ca203e97e3f297354ed2529fef7612e207c20
200175 F20110114_AACPKT bonnett_e_Page_117.jpg
e0cbb1ebcec914080cedfe01a1e0d637
d814d4ea4b2f35fdc6fe37467e09497e9852cdf1
78779 F20110114_AACQOK bonnett_e_Page_132.QC.jpg
32465b215777710fa184eaa46e0c54de
5d8c6a3a5f011f014046b2ca7f8f27a2646ea4dd
23198 F20110114_AACQNW bonnett_e_Page_125thm.jpg
6cbf0d8c600f3cab1f193eefbf499d66
010aacefd6c44f5e9531e728ab0d91fe0a640e32
196972 F20110114_AACPLI bonnett_e_Page_134.jpg
5a646d41c8c238c7bae7cc3ae73eb51d
9bc216e7f3125fa24c21a1b9e71779e9dd194123
206837 F20110114_AACPKU bonnett_e_Page_118.jpg
0d99414253c3634d6657a529b25e3d35
a7c5b0dbace2793f827802d3ee8144f9836617d3
8200 F20110114_AACQPA bonnett_e_Page_143thm.jpg
ab3ea30b1a72ded2b075da78061aa195
d62195247d90d1ccab93f30144eac0ee918fb283
24519 F20110114_AACQOL bonnett_e_Page_133thm.jpg
adf315ce0c642c6c8c331c8be8e642f8
89fa1138ef373d932b3410632f703e91095e9562
74058 F20110114_AACQNX bonnett_e_Page_125.QC.jpg
43c846a68fff9f4d4d4e987571d5847c
7e760eed73fdf16d788d79d1bda29296ae81dcb9
80313 F20110114_AACPLJ bonnett_e_Page_136.jpg
c7fb99da637f0b941a763285140f8e27
34fa751aba087340b4359eb0e863965082ee63de
218614 F20110114_AACPKV bonnett_e_Page_119.jpg
edc7add897e9bb96647baa6501c7f396
7d723deaf7e758ff84a200a267e81560cae60337
24780 F20110114_AACQPB bonnett_e_Page_143.QC.jpg
5f40bc8e05033076a4f6ce5d8f12431f
8d27d30939baa43cc48a6d456c12951936f3f75b
45376 F20110114_AACQOM bonnett_e_Page_134thm.jpg
f6b1c790d66207509260514b9d91c48b
9b0e5bd42785ff7404a5ff900872bb2428e5860c
23199 F20110114_AACQNY bonnett_e_Page_126thm.jpg
4f9c300602f575542810d43f6d37c786
2d26249ed259289a3459f792001fb6f8dbbf236b
260368 F20110114_AACPLK bonnett_e_Page_137.jpg
0773c16f2ee324609efe3cc0ee1d46f4
762698f415a709e8cddf7ecd170b55a6779037e7
207761 F20110114_AACPKW bonnett_e_Page_120.jpg
e115241a5c5b32f98351af90967acaed
af969a8d56da69384d99b8b2af8ab20bd92939bf
23664 F20110114_AACQPC bonnett_e_Page_144.QC.jpg
6d4a239e2d0df864a52bd0a16e823a86
afec2a76522ee66124a00608be634cf0ad65a1ac
89925 F20110114_AACQON bonnett_e_Page_134.QC.jpg
7fda5d03dec2639b01e4ba9be02fc334
9a5f9fc58f4e48f91b16f8411fed586c0729dfe6
17201 F20110114_AACQNZ bonnett_e_Page_127thm.jpg
ea1c137be1146410b0220d2bbecfca31
2614585783eb795e546054523fb18fe417b45fe9
281174 F20110114_AACPLL bonnett_e_Page_138.jpg
a450e31e0dd8dda3b81e6f3b3b0d2b97
4a025e41ca45d0359a74763c70859adce05c0aae
201986 F20110114_AACPKX bonnett_e_Page_121.jpg
9383d67dc5aba681fe8b13767d591679
7d9e6f1e7c79c40b61ca1b631cacf07948d0860b
98209 F20110114_AACPMA bonnett_e_Page_012.jp2
fe11828e7cee07ed0fa3497c9d7c0196
d9815359cd191d1e1a723206111f4aa5f1cfa1a1
167061 F20110114_AACQPD UFE0008325_00001.mets FULL
53efbaffac1ac2a0c0ac234704357e02
22e7d0c23f1a89a9961ef21837b37084920555a8
59832 F20110114_AACQOO bonnett_e_Page_135.QC.jpg
5cccc3f920742bd4df7ee01dd3bc8e4f
509de7adbef4bcc12fff5300ace30e546082c209
282492 F20110114_AACPLM bonnett_e_Page_139.jpg
90a68cbf9c3767a47940e6ef9cfd4bcd
053f8fd19cf08fe73e8e6b1272613e8be453fe68
203032 F20110114_AACPKY bonnett_e_Page_122.jpg
ba51dbc23f69ecb358054d0667697082
979e6a02f5048bc2324b25056aadc88037c9bfc6
106605 F20110114_AACPMB bonnett_e_Page_013.jp2
08d4b1b03e4fd1a909fbc81c3bcdb364
986725fa292ad6e23352620f81693ea9603f822e
9272 F20110114_AACQOP bonnett_e_Page_136thm.jpg
1714958e12b241a199e3d05a9f6a4e58
038ab1c8b502c84ecf5f66e4db5e48b207543eb9
308443 F20110114_AACPLN bonnett_e_Page_141.jpg
6e43e48cb3f35321a0adcb626c8059f0
a5d09c348d0219e34c8e4efc40d4200b0068490f
193330 F20110114_AACPKZ bonnett_e_Page_123.jpg
6ad84ab7ee861dd1dc0685af97119994
9e5e931c9df350bb88f3902b7ad44f4b198dee34
105896 F20110114_AACPMC bonnett_e_Page_014.jp2
9c5590b7dbad4caab030303ea544c868
ca259f2a90ef4a24731f29e90861fa6f17be3293
29542 F20110114_AACQOQ bonnett_e_Page_136.QC.jpg
67a094af4633caf1f886cbfcb8b3486a
a439068022656be06fb99b36a7d2c666cffb6221
298071 F20110114_AACPLO bonnett_e_Page_142.jpg
786711b8b6e071d4949988cff170f46b
379a4a6b7c2651ebf06ca90038f94756187e3535
105538 F20110114_AACPMD bonnett_e_Page_015.jp2
d887cddcf96ac5cf9e4a8cb95a7068ad
f60c3fb373607a6193aa67b56163421d83f864d4
51580 F20110114_AACQOR bonnett_e_Page_138thm.jpg
4682d6cae3aaf3bf28a55631b1c1b2fd
4cf5cbc9b9a7afe94882eda831fd05fbb942d261
72623 F20110114_AACPLP bonnett_e_Page_143.jpg
d525342a8b9a04b9d726a3e1e7ca5834
f9e121ca69f2ecdea87295aaf9fa82a954cf64bb
107435 F20110114_AACPME bonnett_e_Page_017.jp2
7b6aad420fca8b56121b8a8246d1151e
0eef58424283b9383d551b2bbdacf1bed102c3c6
110672 F20110114_AACQOS bonnett_e_Page_138.QC.jpg
f05b7142111ab700622c0ddfea25939d
b1e210ef5f4dda1aaa1c63844f1c7c4864871f04
59708 F20110114_AACPLQ bonnett_e_Page_144.jpg
942506538dc6118dcb285c9820048643
915dcc1a3dda2ef2d2215e59793e216972e9454f
102282 F20110114_AACPMF bonnett_e_Page_018.jp2
a2b45dd24f482a5e21d841adc6d6cecd
c45cd746595d77976380f4aa7fcf33c7fa875ce3
51946 F20110114_AACQOT bonnett_e_Page_139thm.jpg
35e0c0b5ef6fa416e78aab9e6cfa3b56
a71c8ca48f9364a8a0238ca0ab2a6065d41be4d2
24765 F20110114_AACPLR bonnett_e_Page_001.jp2
cb230c672163dc99cf3864e47b68d91e
b7d2e56ab6be384fcd3f68d3260bcdc7809c0dd3
103797 F20110114_AACPMG bonnett_e_Page_019.jp2
9f31ef763d7c2d47f5db3cab8e1231c1
3f0659319d987138bbcfc0794781e20c4cd9be51
108554 F20110114_AACQOU bonnett_e_Page_139.QC.jpg
88a85c162b7df2f328e96857234f5f6e
279358f898e2bab474cca0fb0636cb40b9591dfe
6643 F20110114_AACPLS bonnett_e_Page_003.jp2
c34c215a66ef9c28d4331d51f3fca763
8b751482d41c1d8205f72d1fbd13a09ddd6c6af1
106651 F20110114_AACPMH bonnett_e_Page_020.jp2
21965c51f3baaa7fdc4ac2b209aae316
c7a7b936ec8294b332b12da6d7ecbd5ee77a5c9a
51673 F20110114_AACQOV bonnett_e_Page_140thm.jpg
0655f8a26fd7f8478f898b9c6872b57d
4a195978af59a8988d4f640d3b3e93e4b31abc14
88963 F20110114_AACPLT bonnett_e_Page_004.jp2
ba74ed5ff2faf30821a5ec44fbaeda19
a65ae98aad388a8139829bc934754453aba3f677
50925 F20110114_AACQOW bonnett_e_Page_141thm.jpg
6a4d2616cfb95070ddeaa2397b5c0cff
79cc7a0107b6b18c2dfe3b1505be7f097a5694a4
24181 F20110114_AACPLU bonnett_e_Page_005.jp2
8e555bc89a27571986dc8a15154d3d78
668853f56b34246dc13bc894819c4b5ab5d007a5
97992 F20110114_AACPMI bonnett_e_Page_021.jp2
eb9e429bd33d2c97f399784523be0a2d
1c3bfeadd1c9bbad7f3c3f08c853643ca68e44c8
115185 F20110114_AACQOX bonnett_e_Page_141.QC.jpg
7a4d3320a0019ec91210b004e129651e
b4ab4fb320bf7ba23b13192a0e01d8da312f3c48
1051984 F20110114_AACPLV bonnett_e_Page_006.jp2
42ae130b73af0c3cbd66fe8f612eb606
b9f65f3b63822be2bf99c7333b9cee5f950c8b89
23821 F20110114_AACPMJ bonnett_e_Page_023.jp2
9bc57261e1729f1d2ddcf2dc034582ef
db646faded968e9a574e6c46334bcb240f241e11
51098 F20110114_AACQOY bonnett_e_Page_142thm.jpg
6f3c370d76218c129dccd16e6f5ff143
17c3c01d4af39e8b9777692d347f953a44f22cef
1051962 F20110114_AACPLW bonnett_e_Page_007.jp2
f3202b42fa35139d8b3652c1ff75691f
1f3de14b0e4faa882c22c30c71bc741c2b6222c1
92209 F20110114_AACPMK bonnett_e_Page_024.jp2
bf8e79babb954b41871af42ddedf4f20
ba8ea9cbd62222afc9b61b603eac00dea6c1c5c0
112703 F20110114_AACQOZ bonnett_e_Page_142.QC.jpg
885e196b1a0bd115fd9ce2cf5c5d7348
eb28f35f21d32825c636d707029964e3d1584016
1051964 F20110114_AACPLX bonnett_e_Page_008.jp2
1b97f944a29dcef1b9320f868d370cfa
361107dd16961f61e7a72cbc4586ebbe07256918
108234 F20110114_AACPNA bonnett_e_Page_043.jp2
7a5365ad4942235add3e8c1aa6b882a5
12b1b2db04d4699eaff5fc808309d367dfd547c6
103970 F20110114_AACPML bonnett_e_Page_025.jp2
a66f2bf7ee57bcc6574beb680b49fbaa
9cf894b684a6ed13db32388a173ed42df44b7f76
1051970 F20110114_AACPLY bonnett_e_Page_009.jp2
565e8c52e537ac8b03fa54373857ab88
db60f2efa796dfeb306fef14caca3627d784b723
104984 F20110114_AACPNB bonnett_e_Page_044.jp2
c15f0df2f2ca5aea45a1b1ee84214d11
19281acd94e019646a3254dcac03c1d56e824dd1
120306 F20110114_AACPMM bonnett_e_Page_026.jp2
d210454798c4ee0244938ae0dd9e05d7
6bd2d410229c26c4327025ef83acf295d23e7bd3
66319 F20110114_AACPLZ bonnett_e_Page_011.jp2
5cd2de3faa86822e1f4a10ed8b829a12
4df66eb2d994427c17a268ef86fe97bf4e6dae17
597531 F20110114_AACPNC bonnett_e_Page_045.jp2
3c05b69c420b7ed774f5bda1e564adf9
16159ebad09b1eb7aab08787b6d2e6572b3cf076
108522 F20110114_AACPMN bonnett_e_Page_027.jp2
f8dc5917723619aa81cabaec3640a563
5f93a96bbc19205b5f725f629c294870ddf96f51
107753 F20110114_AACPND bonnett_e_Page_046.jp2
005fcb0242c2a902de470984cb5cd350
f6c9e38b56d4c109ff7a48e628a56df3b7757fd0
114651 F20110114_AACPMO bonnett_e_Page_028.jp2
98e1cf2f2ff44e35c645f7131423db35
44f6725388f95523f3c6d4828abd3ad0c7db04f3
940703 F20110114_AACPNE bonnett_e_Page_047.jp2
599151f6a04384c2db936362e42d0bf1
c7596a95e58db6163a5b13513bb5d2e4db513493
113071 F20110114_AACPMP bonnett_e_Page_029.jp2
25cad6d77e84c2057f5c072143e465ac
7e09f9f52ed466eb7bf788ecd9c8410d26a43c4e
794199 F20110114_AACPNF bonnett_e_Page_048.jp2
152ed7330d131abe390966b0aa713bce
21b92eeefb92d1c0f2237ace6431535dd4157013
103745 F20110114_AACPMQ bonnett_e_Page_030.jp2
47eb21c7fb3307b2d31f19152fc11394
2592284efc7b3828da2c78a6cbdb0d83378bf80b
918717 F20110114_AACPNG bonnett_e_Page_049.jp2
a68fecd34befe86cd4910a08d792eaae
a898fcec0c430fc99a09754a31cc6b1d89ba7a60
106470 F20110114_AACPMR bonnett_e_Page_031.jp2
e5e10177f5eb0db6fe9a857fa8ac1581
aea01da9fcd407f79f215a21d6534b28cf3edf27
829281 F20110114_AACPNH bonnett_e_Page_050.jp2
b193994286b987feebdc0220c9f9bdba
4182e2364de0bfdb5b18ab0adde818e1d443f2b1
877646 F20110114_AACPMS bonnett_e_Page_032.jp2
bca3ec455c0175f599c58ca06751f8f3
f71ae751c05ecf3cf9882250df4bee40580b39ae
886765 F20110114_AACPNI bonnett_e_Page_051.jp2
fa021dadf48d2b250a281f34d14a7f41
7980d7c003d98d4980e05a2a6edd1ee734eb2514
112189 F20110114_AACPMT bonnett_e_Page_033.jp2
28b1d3c9cea633bdfbcd91dc9f3ae7cb
0eb607ebd68a72ad20fc83d70ab0120c15950df8
702617 F20110114_AACPMU bonnett_e_Page_034.jp2
c8886d7906d11ba57357cd8284a30f9e
0e2aa3d44196cc8b9f73f27f426cad1a33341644
30191 F20110114_AACPNJ bonnett_e_Page_052.jp2
ad0b9be26d6e3b80e80bc8d4bf898f09
684a4e7e929561012497e7798da4ec307b250e8e
105313 F20110114_AACPMV bonnett_e_Page_035.jp2
3c385d45b13407c1a4fcf4ea312ba4af
6ab1e3d0fe42893348f9a715ae454e1c2f299e05
108015 F20110114_AACPNK bonnett_e_Page_054.jp2
059acd1c570d887e7a9a3c57137486b1
955d453a5abe9bc532d15d291616349681c2dafc
898396 F20110114_AACPMW bonnett_e_Page_036.jp2
7151da7c867e96a698dd184c408dfcf2
3bf58eeaace1fe9898849e5b815d17b54ce68de2
107548 F20110114_AACPNL bonnett_e_Page_055.jp2
759f014a7fc67c4fd79fdc63b0473879
d91f7f78834eb3a30f0a1aacffc8535f2b123a4f
841061 F20110114_AACPMX bonnett_e_Page_040.jp2
7b0b25c7d7f8be0c7a79fd4313e32f2d
c72175ab6aac05aae4f55f0f2d1146e781ed1249
105022 F20110114_AACPOA bonnett_e_Page_070.jp2
3b42f4a3de5b12db4a09b918d7aa8999
0a33a2c250bcd7fd8c42c14522cc11ce9c9abca5
109826 F20110114_AACPNM bonnett_e_Page_056.jp2
a8a64d99f0a08bd80dccb4b295355fe2
697262969a5e9b819f2ae2f061844eca69f46a02
947828 F20110114_AACPMY bonnett_e_Page_041.jp2
7db301ea82ecc7a9918ce33da1d5744b
bbc705e7dea1a08325e9d8c376c3feb9c52f8dfb
113156 F20110114_AACPOB bonnett_e_Page_071.jp2
5e060086b0a95f58c2fa8ef523eb7b91
00d7b885a47e112cc6fc18ff9932c14d20ee9beb
116795 F20110114_AACPNN bonnett_e_Page_057.jp2
b4b145307ae922f87642a758767b3ca5
a76bf2e602ecb0d35013ab48cd31595a0b56436d
875942 F20110114_AACPMZ bonnett_e_Page_042.jp2
5d377a5dbe857493bb387272fd5d9a3a
54e0c7bc24c7cd5160d02bed75b58876444e429e
109183 F20110114_AACPOC bonnett_e_Page_072.jp2
50e590b05b3c80c2a36572942dc3987f
e5cdb747e3a3f856f9066daf5829543eada8bcd3
116807 F20110114_AACPNO bonnett_e_Page_058.jp2
a9af61c03095de0405ea6afc09c5fb41
6d6dc587d6279eb7db8c47f50042bda564281fd0
107929 F20110114_AACPOD bonnett_e_Page_074.jp2
bbb13f659a96a09184e4d8efc56bdad8
1efaeab7f8cc1f3f27c053b2845105d00e5262e5
122714 F20110114_AACPNP bonnett_e_Page_059.jp2
64fa0bfb59b28c0d05f5daadabd275fc
2cb1857e0038deaf63fdce9933a1a6b5c4ff7c5e
115859 F20110114_AACPOE bonnett_e_Page_075.jp2
b53dc41e8e48b7c587a0b02ee8a32e26
220dc1ad0eed448dd2c11fc2f36ae54347de6112
117541 F20110114_AACPNQ bonnett_e_Page_060.jp2
40ed2ef5dc2f50fe7f70ec2d54d7f8d1
cc578959ea3ec461fd26ba7fb09b783b8d7adb14
105537 F20110114_AACPOF bonnett_e_Page_076.jp2
61e14d101905de7187876965b502fba6
21608c1a985b6dd12c3553aec22fa0a6c7684bcd
107166 F20110114_AACPNR bonnett_e_Page_061.jp2
14c510d1e2962041b4511cb274c88f88
598daa6debe2ced9941fb159904bb220ea2318b3
101273 F20110114_AACPOG bonnett_e_Page_077.jp2
fb27b2be56ca9cddbd080a0dec275268
bed2bf3f7dc9067c53a946168949d5721b380c19
100725 F20110114_AACPNS bonnett_e_Page_062.jp2
fe820c9417c8790aa93976efb5de5635
74729bdf833b8ab2e28822121a0af8edc240c4c5
98394 F20110114_AACPOH bonnett_e_Page_078.jp2
8ee2b912ce2617b14a8ed06a1ba5d29a
7464d7ce057626a8bbe5ff2662f32b2982348706
115585 F20110114_AACPNT bonnett_e_Page_063.jp2
c37a542b200a0da21cd9a588f17a04b2
9bb1bfe78747b346be06df29f070f61068559310
108741 F20110114_AACPOI bonnett_e_Page_080.jp2
2b43c3738b30937534951ce7cf115fe9
9953eed2e161b1b93be488271ba3291999b75fbe
120536 F20110114_AACPNU bonnett_e_Page_064.jp2
13a08cd523c58a03a585147e6cd7d125
b6658ba61a219d3ce7aa52bf138b1f8500c27aa9
107744 F20110114_AACPOJ bonnett_e_Page_082.jp2
8b8577a40abd9d4f8b392ecea83586c0
f94d012ce733ad0acf99449f3118695e93538c0e
103086 F20110114_AACPNV bonnett_e_Page_065.jp2
b6212e3481e4d5891883253212863588
885d7d8d3e1ded83bc8546694ffc65c380630178
11573 F20110114_AACPNW bonnett_e_Page_066.jp2
750ac7f275498a130258314cb4a0cbdb
2acf79a95c46eb3a77524f7ee639515fe4f6d5ad
105862 F20110114_AACPOK bonnett_e_Page_083.jp2
5c1cd28596ba71d79a98afc59ad3699e
845a4926a9b8178e15550ec66ba9f00d7e7f544b
97939 F20110114_AACPNX bonnett_e_Page_067.jp2
e6faf7bbb9cba13e0200ccafac7cedfb
2160da8b39606570b5e7c5819a870fa4d0f4a7dd
109018 F20110114_AACPPA bonnett_e_Page_101.jp2
75de89be67c7223339bed984bca84e90
b22b6a92d81c9fa94309449437db4cb18e082897
107010 F20110114_AACPOL bonnett_e_Page_084.jp2
ef5c74a71766d67f2a669a92e8e87f18
ae67b720cb52ae7dcc365b5dc7f6c7ab3a4e37a9
99668 F20110114_AACPNY bonnett_e_Page_068.jp2
4dc49f95a0a4bc02df2dda59b63ef8ce
b7371704444b171119825d3a1853e46d00efe18d
105717 F20110114_AACPPB bonnett_e_Page_102.jp2
09c5eeac3c79cf1086acf0b3479ba86c
20fc8a96daead988f5a14d47211d52e30584bb9f
110036 F20110114_AACPOM bonnett_e_Page_085.jp2
756e8980aac680b592785847544ead45
81d54fc40c8ce77033b4ee3ea6bc97fa9ce253e5
102701 F20110114_AACPNZ bonnett_e_Page_069.jp2
85458fad15b1d403d1e935a97075995a
cbb1b4ed33bfe6f830ea7ec614b8fdef4eb9caed
106631 F20110114_AACPPC bonnett_e_Page_103.jp2
0540a633009d99029cd0c53c46d33592
31b347394c0a1f575e060519cea509c056f7de9d
105323 F20110114_AACPON bonnett_e_Page_087.jp2
86844529778321878bd94e834f7ba79f
b482c5b61b5514e35d47a8e3cd159dd586b59887
109257 F20110114_AACPPD bonnett_e_Page_104.jp2
95340e6445de5ea284e15f2ad3540288
9917e1e9ef83076284c5214ffac333eed4f8632e
104415 F20110114_AACPOO bonnett_e_Page_088.jp2
4612ca28eed2c167aff10ea4106ac937
c20517a24ea33be84ae4abc0b809f6ce2542ae55
105573 F20110114_AACPPE bonnett_e_Page_105.jp2
82f1dc74e886d9e988522aa5fd93b74f
050a1c5a23e7c4fb444f003383b814c3c2d648ce
85979 F20110114_AACPOP bonnett_e_Page_089.jp2
b606fed30dae49d57a730120e209ee0f
7c2b0cd2203ae1ff017f219adda14efd7bb49219
102930 F20110114_AACPPF bonnett_e_Page_106.jp2
1af397d469a52808c00a3ea69c9aaf8b
f96c035051dc6b034804dff8affccfe1f563899f
1010755 F20110114_AACPOQ bonnett_e_Page_090.jp2
dc841fb98fe39c34626e378bcf5f1773
ccaccfd4a568b6832b941689bfd89d56ae6880d8
102891 F20110114_AACPPG bonnett_e_Page_107.jp2
4eb07874473371c5dda5c4f0376d55e5
c6150c355c8efea27caeafe57632cd643d4ad626
153846 F20110114_AACPOR bonnett_e_Page_091.jp2
798870f7007846289a4f5a00ca96f758
d125450562703f02510346d87809a7b6ff821d2a
106194 F20110114_AACPPH bonnett_e_Page_109.jp2
a7aba1dcf9be78f8dd99ff4a14fd9d90
19ccd5aaacbfc335671adf5ff3e10b21fd5ba7c9
90148 F20110114_AACPOS bonnett_e_Page_092.jp2
41889352dc8d45fc4293c8b6dc26b45b
1a4b89da1928673869c78f69d03366951a906b33
106067 F20110114_AACPPI bonnett_e_Page_111.jp2
62624f4eac47995d6e0733de7dc08e22
15f6db02184cd1726e723571f842e3fe64835c5b
96391 F20110114_AACPOT bonnett_e_Page_094.jp2
cb1f7e0ce27ff3bde917ad078c57e915
e6dca0c2b6ed52dca791ed52d88c0ec6bdc6dd68
102098 F20110114_AACPPJ bonnett_e_Page_112.jp2
a5b783580e3b56199f7fd3be866c980d
fa13ed29124588de5270ba3a32d2e35c0712e9bc
109066 F20110114_AACPOU bonnett_e_Page_095.jp2
3bcc34d1453dd0b9051aa9272e44fe47
837f9ed71a9055309a0dbe69d7859ffaa816e82b
106969 F20110114_AACPPK bonnett_e_Page_113.jp2
598340cb0024aaf2958a6fa3bc4cf70b
0299166c256fd977cfa4fde04cd06b5d4186b367
92550 F20110114_AACPOV bonnett_e_Page_096.jp2
b1da4d10a11d0f6df4267507f2c247ff
8ada2152b6c2f33aefc9fd9ee1a453bf5c6bc8d9
111240 F20110114_AACPOW bonnett_e_Page_097.jp2
fcf489c0b38cf9da8287f015ff0f6a12
0aca2ad78c444cae9aa4c31e091c5f21e5be4f6b
111231 F20110114_AACPQA bonnett_e_Page_130.jp2
2907171f5de60484ee5c5e52ff096ec1
dc835942cbc23332984198d6b2d647f4f475f6f0
108935 F20110114_AACPPL bonnett_e_Page_114.jp2
ad6ecd1b8727c2973c21038ca1d79903
e9cbcc8c9f935d82fdd84fe873299be6e37c2775
106759 F20110114_AACPOX bonnett_e_Page_098.jp2
4c9d26dc5bc5e0111826dd22b102bbf4
e4436a40fbd695fe439e9d7c854e2fe7b7b34535
108042 F20110114_AACPQB bonnett_e_Page_131.jp2
b83cb5068735ae030869aa2f78b1fd2c
da4dbadac942466f4c8684cc3f56172488400bc3
107973 F20110114_AACPPM bonnett_e_Page_115.jp2
018273b67a73de433eb17ca0f57e07f2
aff45e404ad407a9c3c6e79740fa1af15feaf398
104040 F20110114_AACPOY bonnett_e_Page_099.jp2
670f9bee3a3e913a901c9a6f49935976
11a198d8a7be235f6e55457a24e51e7a29fe0965
108025 F20110114_AACPQC bonnett_e_Page_132.jp2
def33c9fdd7de22d94167bb277dcadcf
279d78a42b9cdf14358f1a965519583b0f16ca73
103477 F20110114_AACPPN bonnett_e_Page_116.jp2
52daaf15a4165f67c0c0f24d5719517d
0912e085c864f169467dfc1fb6b25b5b4ab7892b
104257 F20110114_AACPOZ bonnett_e_Page_100.jp2
2d9460aea2cbdae5157af66e3c294bd8
c07e09e9395c0d6d9ba3de72da767c21d047ecc3
109003 F20110114_AACPQD bonnett_e_Page_133.jp2
cc8e11b370cf240e6c7cabb43e1dbc17
65eafb919301175bf418220f2e4566c40dbecd5a
102132 F20110114_AACPPO bonnett_e_Page_117.jp2
3a1975b6b1325bf7b373062350a8f279
4d609e0221713da096db499d77707573a41478ac
931029 F20110114_AACPQE bonnett_e_Page_134.jp2
ed4622d60b71051aaada3c68dedaff2c
a37bafbc01188995fd6126fdbebdab85b227be86
107820 F20110114_AACPPP bonnett_e_Page_118.jp2
ddbff8d80f827c06c956224a701d9119
13424c3bf335d2174836fd88da685f57fbdaad2e
79695 F20110114_AACPQF bonnett_e_Page_135.jp2
8c6028499eb44ba1178307b39f9568fa
6d554512332baccebc378bf6eb62673b72947963
113967 F20110114_AACPPQ bonnett_e_Page_119.jp2
5957f6d6124b6c97be34868daf7e7315
b2a5751564ee34bf7e5472e80e1961863b0ced06
1051985 F20110114_AACPQG bonnett_e_Page_137.jp2
23af06646f03a4086a09da776c654e90
abd51b28fedbe8d4bc66b9ef63651825457cf9b5
106681 F20110114_AACPPR bonnett_e_Page_120.jp2
d29df0ab4b17f1adef778d3c3fa9154b
bf72f90fbbe89cedbc6823219c2f10632fd60242
1051974 F20110114_AACPQH bonnett_e_Page_138.jp2
4fefd24f2a406a37a7ca43be01a78a77
5b8501b65d40fba605dc0517005e67343c7c7069
104483 F20110114_AACPPS bonnett_e_Page_121.jp2
f189c41c6b6651be1b1a7c1b2caf8866
1839ab7867a14589f4f0905968dc910d08e548dc
1051966 F20110114_AACPQI bonnett_e_Page_139.jp2
460a2bb17f09b01a8bcaa54f152cf51d
1ad8ce1f9055de1cdc0bc714e4775444281fb0ae
104141 F20110114_AACPPT bonnett_e_Page_122.jp2
3f39244388929966a7b4ff67c1298467
9c46990b68fa0c1d800e06ce64c8fd875db3d21e
1051918 F20110114_AACPQJ bonnett_e_Page_140.jp2
45630c4cbb5006280de9b4123a43d459
3bb4368751732f67d0e0a18491ff8c2f708527bd
98134 F20110114_AACPPU bonnett_e_Page_123.jp2
036e36e658be123b9129ec2366e14111
e1a9662e396055500d9744a26c0b666ff26efc77
F20110114_AACPQK bonnett_e_Page_141.jp2
12f5d4473de5a2efd4ede239b7b03940
2ffc243fa7f3f8108367a75a15bae8e438d68c6f
101797 F20110114_AACPPV bonnett_e_Page_124.jp2
4f4f572dc4531646a4a463b8e798a624
3f261b72efab0f76d2f70d784fee41a259ebe624
35008 F20110114_AACPQL bonnett_e_Page_143.jp2
caab2c47c3556c8c469f3d5b337c695d
8619319e0c658efb514ef5770436597f4698e04f
105975 F20110114_AACPPW bonnett_e_Page_125.jp2
ee08f90205d2c70d35c929dc285f1f23
63889efa26d81c4b807d5e6098f571f8c8236e09
103198 F20110114_AACPPX bonnett_e_Page_126.jp2
fe40c1d8d9caf4a2e80fd0f7fba8ac85
7b76ddc879c31a2593c78e2aeb245a06176f7a2e
1053954 F20110114_AACPRA bonnett_e_Page_016.tif
d9dafd604e6f7186a85e3abac664e6be
f3a75674e42d5b988f22e16e70fa0a4b2b1300fd
F20110114_AACPQM bonnett_e_Page_001.tif
73c948e7a64bec5e4ab1304a78f6dc85
e32793c5958d8e23a87455691afe068b9cba34b3
97200 F20110114_AACPPY bonnett_e_Page_128.jp2
4e743202b05269487fe581bc123c5ca4
3522b97ee6fdb4bbb7443be2e836ea4080d96796
F20110114_AACPRB bonnett_e_Page_017.tif
4a2decc18d56a33204adfa1e2aaa1e8f
b426b09f832e18a4c97297e7a6aaa45e43b74d06
F20110114_AACPQN bonnett_e_Page_002.tif
1a560aeb39581c662c7a067dc4c3801d
ba196487db513cd8c2dde78ecfe6a9387450b86e
106300 F20110114_AACPPZ bonnett_e_Page_129.jp2
b57d57dbd54e43f4883cff2f35a2fc7f
25804151ff02d0fa694276022fcb90d4253b1e3b
F20110114_AACPRC bonnett_e_Page_018.tif
8ab21cc98f564bd21a2be6ca9b161132
55f121511fbcab096960e007bba0f51b909b45e4
F20110114_AACPQO bonnett_e_Page_003.tif
a36cb741298b0cd11b2cebc0c385efb5
25393fc52d80591c0be760244e2fe9de89c0229f
F20110114_AACPRD bonnett_e_Page_019.tif
77674c40d65d4aba8d3b0fe6306a4457
c04a84ab1715b2649c464801761f37009173ec35
F20110114_AACPQP bonnett_e_Page_004.tif
e47e7ea00bd8fae4c782b4a7c94da6cc
1722cda031c9f0fea91af2dacc9f15a83e86667b
F20110114_AACPRE bonnett_e_Page_020.tif
8ae576ff7aee27aa7fbf33a5d2833385
aa3423fb94ff680116e75e10f3a307f95d83dd47
F20110114_AACPQQ bonnett_e_Page_005.tif
0668b660eee3b584d95a53360c388f8d
3718dfdc46601ed33de54624b12124d4452e671e
F20110114_AACPRF bonnett_e_Page_022.tif
9df46fb8f7854b45a7ec7137c17b6d0f
e04a6e8eae67fe27738d6dccfb3b475c8deebab0
25271604 F20110114_AACPQR bonnett_e_Page_007.tif
6cc4eb1e9900d81502b678b2565a569e
023827d819cb17195733246472dbbb75e29b5e37
F20110114_AACPRG bonnett_e_Page_023.tif
a8e6c920c2530ec668c6b487f67f895d
8a8cedf48f42c7e879ccf3dc3f12b23adf29a332
F20110114_AACPQS bonnett_e_Page_008.tif
3e1c5b8ccb5799a56ff35b1839f949b6
01c6aa00fc54baccaceb16f65c7281858c4fe6b6
F20110114_AACPRH bonnett_e_Page_024.tif
d8cd7cc0971f0ac2830548a461433be3
84b69271639d7515382ce1d03f83098464b072eb
F20110114_AACPQT bonnett_e_Page_009.tif
f140870847927093ed47545cf83983c8
3326d65d7c4a0db0c3a327f02c78893a49e1e3f8
F20110114_AACPRI bonnett_e_Page_025.tif
14b48b7e52665c940eebabd172d73863
1ab0b39a0b116cec680d41e4aca61ebd81679b68
F20110114_AACPQU bonnett_e_Page_010.tif
7c48158ea623cc8872866d16d9d379e2
09d186ad1f82e5a4f7ede97e9b1bd596afb99323
F20110114_AACPRJ bonnett_e_Page_027.tif
4c9f655ca50ec294bf8f69d51d1716f9
b0d2b70ebf3b2fa720191eff8f41b2f834eec9a6
F20110114_AACPQV bonnett_e_Page_011.tif
a62dc18ad156b110233e417357c881bc
fd526912f5e9dfa7319cd9849e28aec008e951b6
F20110114_AACPRK bonnett_e_Page_028.tif
2b778d427d74e0dc0e2d07989071902a
f4a048dd353598d67055462ef54bb566fc644b26
F20110114_AACPQW bonnett_e_Page_012.tif
9e2c07c87129afbb575a6138c7503191
8fff29a507ac1be178ccdfa5cae38d2243998bbe
F20110114_AACPRL bonnett_e_Page_031.tif
5b56bf65d3082f57389aff74c60efb99
7abbdd497776776907e805dc99567cbc994b7573
F20110114_AACPQX bonnett_e_Page_013.tif
c0cb38210dd466a6493715e95c6d5159
e518762a8f5a04db32f4d33c28a24df817550517
F20110114_AACPSA bonnett_e_Page_047.tif
0cd75993dc37808c93318740ce77d15e
0d8346284c8754820b6a1f04637c85dfcb776e73
8423998 F20110114_AACPRM bonnett_e_Page_032.tif
485602434db5bf9dfef59caded00a67a
7b1403c588b866c19645de1a416a1c4ecb21f0bc
F20110114_AACPSB bonnett_e_Page_048.tif
ab161b42f2beb6b10df6c2266a826288
ca31e910c16e4e3c8b2131a51667baefb72d3d6b
F20110114_AACPQY bonnett_e_Page_014.tif
7dda67d358cd7c4d80d20ef1450df44f
7bcdb34d60d4954566da5d7dcb4b55c3a8b563ec
F20110114_AACPSC bonnett_e_Page_049.tif
b9cab633e8412fb1b0e57e31fe092f16
12ccff214e31e3a6a5cc5eaa072b09fd87d03738
F20110114_AACPRN bonnett_e_Page_034.tif
6494d434be226129e9613877740851a1
7bf954a2453cd36d08b75b04da7d5ccd2f9d7f6b
F20110114_AACPQZ bonnett_e_Page_015.tif
f5eb193bf1edd8532bda3a7e10f05533
f4ccbad2eac4c0ad0063cb4e91e3f8a4c66e5543
F20110114_AACPSD bonnett_e_Page_050.tif
eae954e631d80cb025a211df6ebb74c8
451dac8d75d31cac2a4705de5c04a7da44571263
F20110114_AACPRO bonnett_e_Page_035.tif
e955392780d8a8c48650ec6f33a68f12
da5b8f0f615669ffab2214156c3c2913b961ff0e
F20110114_AACPSE bonnett_e_Page_051.tif
a91999ea94c457e202be54284f55e517
bd1a595d3acfb5a889f2794d56938f80eaa410b0
F20110114_AACPRP bonnett_e_Page_036.tif
ee1243f44a1eef3a978455611ee5863b
836e08b676633f3d1d75b3b53f59f307174c5839
F20110114_AACPSF bonnett_e_Page_052.tif
2a414f52604ce24ed71f04ff8f9ae69e
626db1c5b527336a43ad1b0105444bcaea0aaf09
F20110114_AACPRQ bonnett_e_Page_037.tif
af45c5a78027741a6743adceb32c8f4c
04a3c3d1f8a4b69b8b72945cf6746d0c96d788bc
F20110114_AACPSG bonnett_e_Page_053.tif
66dcf8a0cb3a69de092f22945433d9a1
e7794d157f9d61ae09ceaf1269cdd0b74de0aa10
F20110114_AACPRR bonnett_e_Page_038.tif
74e6d79318754bd44490270c1d540ca4
7c64a6341b39f77cbcffd6696af640492a2d8f59
F20110114_AACPSH bonnett_e_Page_054.tif
91e53f2d7d3c13506b05c60e38572483
b89036d38574610aa8b717c6ed830e02a2f87b82
F20110114_AACPRS bonnett_e_Page_039.tif
5685f3871ea2e5ef9295e29659f954f5
df62530f33ab3c67840625e3a3c2549a5677fd39
F20110114_AACPSI bonnett_e_Page_055.tif
bac82833d5d36d8f9a5bd0a26f1e4687
c94d5af22d4a323da63e692d670c865945fb06c9
F20110114_AACPRT bonnett_e_Page_040.tif
40e8438546fa3cba1afc672ab2a6ab57
90190599a3b4f62f657a478809ba0eab737ec551
F20110114_AACPSJ bonnett_e_Page_057.tif
cb38b81df99c0ce5b0f4c967d5ff1d42
68b72e53541975a20343adcb350bdb28f2f0a3f8
F20110114_AACPRU bonnett_e_Page_041.tif
97c57163d9e56cd72d009cc2d473f328
18b8f125428b1b3908587513915b21bc3efe27f8
F20110114_AACPSK bonnett_e_Page_058.tif
99763eb4c72498be4a729fd124e9abcf
26f826ea3ed4fe1fbc1c3dfd073fb1569afed508
F20110114_AACPRV bonnett_e_Page_042.tif
71734ee73194bfa468cb52f8c1033990
5c4f5166839e5ca82746e7e8f403fa7af590af28
F20110114_AACPSL bonnett_e_Page_059.tif
2e05dfebe54f6b47bf96c114f75e61ba
c72728dc2fbc8d3601053697809666d32726a96f
F20110114_AACPRW bonnett_e_Page_043.tif
927f300f97d35c4839513e831b77dbcf
c76995da3a7bdb9338b0d5f236d2735ed754bfa7
F20110114_AACPSM bonnett_e_Page_060.tif
69e276e3c5c609d19eb7dcf13aa2ed9f
11da45eeee11223dc9fdbdcfc1e8e41c33409f91
F20110114_AACPRX bonnett_e_Page_044.tif
ebf0881320a224d1140e0b7f75bd56dc
a0ad80cb9580ef9b5045bf14208c2f1506d7fc53
F20110114_AACPTA bonnett_e_Page_074.tif
a8e5358a6cf995dacf1b267096e55629
1f84451cc48c29421209338151b5d73d1434bbf4
F20110114_AACPSN bonnett_e_Page_061.tif
14cc45458811950387d9934fe6c5a72c
92ef9f7c807baad64572f5f3db7e551a553b9a70
F20110114_AACPRY bonnett_e_Page_045.tif
c3f18af5f54d1edfd51a1e28e57b39ec
41cb44a1cc91988d765354a825f13a3fd4041555
F20110114_AACPTB bonnett_e_Page_075.tif
47bdf3218ecca046a3b4daf3c6b24024
0d87f0be2d8f05e508a5b73479d2cab328950f15
F20110114_AACPRZ bonnett_e_Page_046.tif
e2bcea6d8f83f122ebb4a44d0470e4e5
3cff09c75d3cc0fe4657ce19e24316bb2ed2921a
F20110114_AACPTC bonnett_e_Page_077.tif
7ded86b2d6cf3457c07cbfe66a7db2b0
69e248a90d7544b64589245acbb31e732ada97e8
F20110114_AACPSO bonnett_e_Page_062.tif
b0854ef8c2da56e5a25628b385f80cd8
9a414d2b82f5bcc8ac1e8354033e67ced1c765ea
F20110114_AACPTD bonnett_e_Page_078.tif
18649125a3379328ef07d2ac335f64a7
3f0602887ad13b9f035b58a72d50e290800a9cbf
F20110114_AACPSP bonnett_e_Page_063.tif
b9976f6e0a3127c8c50a0377cd602e01
b4b14832b73d096ef3e710a5c45001b4781c1141
F20110114_AACPTE bonnett_e_Page_079.tif
2dbb6f18c222a62e1ac7966a6df7e6c3
8ca90b953f44246aa8991cc605bd58cc694e3867
F20110114_AACPSQ bonnett_e_Page_064.tif
770aa949ce506296ab67b5fe52650722
15e7c249a9c7130560f0f2e26deb4a8446d5a13b
F20110114_AACPTF bonnett_e_Page_080.tif
0e4f669cf97837edc8441c0f4d380b72
5c3fb0d031439d6acf0b8642ba89a22c12dde60e
F20110114_AACPSR bonnett_e_Page_065.tif
5f564e931b800c6fb04795d9a599009e
6904b2c8f831ef31bb2b271980f874b87e9d302e
F20110114_AACPTG bonnett_e_Page_081.tif
b839b20a8ecd37825fe95d58e193f082
fbeebf45289763bb9ebd0b401de98188f191fb36
F20110114_AACPSS bonnett_e_Page_066.tif
25e06e5adc471580d7a0c7255c1e7f8a
b00d7df2dd866ed06269bbbc3981a7007e269e92
F20110114_AACPTH bonnett_e_Page_082.tif
0d81ce95383306e229f46b29bbcafb2e
c7f50d1dce29ab11e4253e594986eadc08a2f925
F20110114_AACPST bonnett_e_Page_067.tif
80ef6893d70a83a532064b3a92ef43a2
b21dd4b0e6a7d3eb6085d705214820562539b418
F20110114_AACPTI bonnett_e_Page_083.tif
604ffeda7123ad23f190294975ee290a
24d4b42c3f575c97ed92d5bd26c9067c5cc59619
F20110114_AACPSU bonnett_e_Page_068.tif
0f7b12848408c4f5992f5514b9b2706c
80ceca56a5f6b44f15a1528826aaa88691d38951
F20110114_AACPTJ bonnett_e_Page_084.tif
a2c98108ab81212ac00e622111bc31cb
3f8cf794f5ce6a09c3e1a469ea76cafc675ddc32
F20110114_AACPSV bonnett_e_Page_069.tif
3896487f3fe6f46024c344efd76a6caa
f037c13bd4ebb0fffa427d1dd3157ea4f846f269
F20110114_AACPTK bonnett_e_Page_085.tif
d16fd1f7c61fd558c932e8e8ab4a352f
a331f884b472852684bb2d89fc7c3f97187c9952
F20110114_AACPSW bonnett_e_Page_070.tif
fa14c347d8ed556de118d619d97c8939
44d7721867f2ce7d6be1ad849d16aa39381667a0
F20110114_AACPTL bonnett_e_Page_086.tif
997971b9f61703db8baa02066cbaef08
08a70519742b8a9da52886ffe91873bd4915fa21
F20110114_AACPSX bonnett_e_Page_071.tif
a6f4c9bff4ce4ada58dfa3a2697f1742
4413af6bd96780415ef90944816efbde9252b804
F20110114_AACPUA bonnett_e_Page_102.tif
e1bf0db0f7c085bf55bc99ee177dcca8
b65f8633b654be34bee3c42cbf08be463ceb002d
F20110114_AACPTM bonnett_e_Page_087.tif
2e0607794605c6dc334c98e3b6ff0612
95bb79d3aa2c14bf3411d82163e4c403205474d3
F20110114_AACPSY bonnett_e_Page_072.tif
c1da501b08c7f20716ca03dc1ce2b7e8
8d09bc5fbd4c927626aa585c1102cc3e713639e8
F20110114_AACPUB bonnett_e_Page_103.tif
efc5f0afdedfed6efaae0a1fcc3cc9ee
4960b8a22a4a5095b392d90c6a4668b76dc7c1a0
F20110114_AACPTN bonnett_e_Page_088.tif
371578c2af9966eba31a43035a07909a
afdf33d8bea03aebb70904359957581bb504699b
F20110114_AACPSZ bonnett_e_Page_073.tif
e579a2284bbcb801f59598ed9edb49a0
4f62f7a9cdab50c4e507dd69740acf4423497b35
F20110114_AACPUC bonnett_e_Page_104.tif
c03b8f5d9269986d458bce8a389caebe
c8af9f4398ccbffc1e2d90ac4c6e5b011cdce2f1
F20110114_AACPTO bonnett_e_Page_089.tif
1562ca34d5d2d44ddfe9b11111c0c2e0
05610fbf47c23bac5d46cfa47a9d82e6fac1d60a
F20110114_AACPUD bonnett_e_Page_107.tif
d13eee61f181554f38e0244a15dd9b61
96dbf50183420dd41fe561bff5c7d4f0b5678df1
F20110114_AACPUE bonnett_e_Page_108.tif
0796b8205b48cd9b64a15d0cfe2bd1a5
6908e4427a0e05f663b7afb85b3d17a81a81266d
F20110114_AACPTP bonnett_e_Page_090.tif
c8b1e272d75e0d60a187278c47586249
ad27589b1060ccad743ff0af6d9a4c773369fab0
F20110114_AACPUF bonnett_e_Page_109.tif
2e53964bcb52777140a11c220e7e5567
e9fd8b5d17a2b5c9a7e48761c01405ef4a21724c
F20110114_AACPTQ bonnett_e_Page_091.tif
554129b8cada1bc9ecdd18a87319c0c2
599fd44af3087c89e0ccd66cecf2fab0d07cc18a
F20110114_AACPUG bonnett_e_Page_110.tif
eca92191fb3abba5358faedda461f3c5
6c2d3eb958e84dc25efbbee2546e41b64ed168d1
F20110114_AACPTR bonnett_e_Page_093.tif
124d65aae0ef3dbbbadb78317a8a4611
2134145032c31e901f573b1bcc8fef71957b05df
64499 F20110114_AACQAA bonnett_e_Page_140.pro
ed4989d22beeb3ee5a2cb2b2f776a456
f271216ff13eceb322c67b6db8ff96b926ee738a
F20110114_AACPUH bonnett_e_Page_111.tif
dba5c1f116bdea683451c8a2a33c0fee
49d1435a9e8bdc5e64c00e58432e4b07095601ef
F20110114_AACPTS bonnett_e_Page_094.tif
cd359ab048a552f1104cbeb3023a66dd
990410629826e483c651968d0f6f33cb661be330
60492 F20110114_AACQAB bonnett_e_Page_141.pro
aad3ef12bdd356b5ac634c332a2c3b85
d584c931aa99039c5d8a36457a4b0e0d61d02e90
F20110114_AACPUI bonnett_e_Page_112.tif
ab281f140497f382fa5cbfc3fc76c7c9
a079e29bed2342c7f9629d805c49a104fc0def7a
F20110114_AACPTT bonnett_e_Page_095.tif
b7d58a30571687e89ac098da4339588d
e2c53974096b1c3eef81991f904524b8cbf24b22
62602 F20110114_AACQAC bonnett_e_Page_142.pro
8ae1b4f5aa58efa1a863778686223052
2cea9674d0905c85e4eac37418057b99f91ceab6
F20110114_AACPUJ bonnett_e_Page_113.tif
9a8a6c9d3c8718eb96fb9a58849af678
2bfc1ca4935b4da14ae5c1a1c5f27850ad912d1c
F20110114_AACPTU bonnett_e_Page_096.tif
cd293ceb976b9e572a18569c0070a927
413d28dd4fcaf663e275c54738ae510671eb6471
14907 F20110114_AACQAD bonnett_e_Page_143.pro
5d35db0eb79e618266c4050eacc3531c
cd7018e3e03ced475e2b0a44d4d5f23a64bd9702
F20110114_AACPUK bonnett_e_Page_114.tif
3b83c2cfd085964fafd046f984fb5ca4
3ad0a957d760a6de075843395c18d5366299e1ab
F20110114_AACPTV bonnett_e_Page_097.tif
b34950b7a44d45c2b4f2ae690dcf4aea
0c0ad187a67ef514cc0b3363246c317486549ace
11543 F20110114_AACQAE bonnett_e_Page_144.pro
6119d28e4dcf5cd0098b1ba7b288c4f4
81241b6b1c9c6ca76f3fa2ee221a61f628f5d064
F20110114_AACPUL bonnett_e_Page_115.tif
475b34cc7e9280c1e94a8a72fc0eb298
30d4292c8438a964b21b75ab793274ad05a09b5b
F20110114_AACPTW bonnett_e_Page_098.tif
2b9650f3c4ee46979f374c8f24df4279
d86b0a7bea958c3d2770f690e4867a85bbeea7ba
465 F20110114_AACQAF bonnett_e_Page_001.txt
038f2c8f6023435f170fdaca3baa5ff9
b21c6ce965ad1183a92aa9f5de4aaf57b3ee3695
F20110114_AACPVA bonnett_e_Page_131.tif
3e2eb5d6cc4fd18d125b2e7ee49f0a21
e10d759468093dd1e4f866edddb0cff974b42b0b
F20110114_AACPUM bonnett_e_Page_116.tif
12e1f7c0ccb07383c8f1087844d99004
fba070e4ebc8befbabc4cd45230caea2a26fab51
F20110114_AACPTX bonnett_e_Page_099.tif
db1a5aee5a926844b89ee58e498bce42
ba05ec9527bdf87767a147bc2f7fac22676b983b
112 F20110114_AACQAG bonnett_e_Page_002.txt
d3a7ee56f8f41371cace642eebcd2ff3
2283188cf95113b54942b1bdcf2bb998002159c3
F20110114_AACPVB bonnett_e_Page_133.tif
b6055205b320619009cd716b06e1a73d
0ef6b9d0b0e687076aed066741c1aadcf37d10d5
F20110114_AACPUN bonnett_e_Page_117.tif
460197a4ed500aa0463b43532a684513
7f12aa5dfc42d8856f9e8de339747e4d27dc5382
F20110114_AACPTY bonnett_e_Page_100.tif
aea5f0ba88d173930d06d5fbced478d7
21fed394aafd01feaedd2368c5e362f224adb17a
118 F20110114_AACQAH bonnett_e_Page_003.txt
77aa92d5b751d9e1e169eb9dc6c41f2c
6db9f722094724f93b1c382ea0f85a50203c9e54
F20110114_AACPVC bonnett_e_Page_134.tif
27219c5013122f5377110ae9591cf76c
7c70eb777dbad1eb22242e26b685693b47ac0fd6
F20110114_AACPUO bonnett_e_Page_118.tif
1d50a49b24c3471df35b3ac0c005f8e2
eacde6dabfdc890a57de0d26aedc797fbe2d7100
F20110114_AACPTZ bonnett_e_Page_101.tif
b3e325c199ee686a695c1c35c6ea3314
6929f5d3880eeb0f536682aee3402d8ad1b961f8
1666 F20110114_AACQAI bonnett_e_Page_004.txt
6fa6e25817f1cfe87de3691a25661f36
c6e51cdb64dd4c37aa09f79e98ae485b194275e8
F20110114_AACPVD bonnett_e_Page_135.tif
9d93d644075e91f18a527754d5f8f055
50d266d989aaee363761d69038c3d9a813f37df0
F20110114_AACPUP bonnett_e_Page_119.tif
d73367cbe794434be4c20c464d4daf10
6f73712c1c5a408866714d07c7b07998c5ccf5f7
406 F20110114_AACQAJ bonnett_e_Page_005.txt
acbba8978b974b7a6b55a0e9fad3c2cc
82342f0d3ce92606d4487b52ba19c90a8831c0ec
F20110114_AACPVE bonnett_e_Page_136.tif
3564f224489362e407401bae8c8215e1
e386aad64bccc3011489f9a8d664bdee33b0efa4
2765 F20110114_AACQAK bonnett_e_Page_007.txt
6fefbb880f69bd31590bbea58fb005c4
8850e22f3241792f30f4f691f216329d2f733f09
F20110114_AACPVF bonnett_e_Page_137.tif
0eaa287015e76fa99b240ede847cda14
f8117ba28b193560ec2b44a0e47bdf121d13acff
F20110114_AACPUQ bonnett_e_Page_120.tif
9bb6253ee9a244381265b05c87c5afd5
4c00f13948d9d20fd60def3f549abd3546b27e13
1451 F20110114_AACQAL bonnett_e_Page_008.txt
c1780282fc696ce05754a8d303eae61f
26f00d88e03d9103de0265fc934c494f72b79de4
F20110114_AACPVG bonnett_e_Page_139.tif
dc984091aec9ed9802c807b2e9c5c76e
a089ebf838b74ccae7bd0d8e4149f26c59b8bfe0
F20110114_AACPUR bonnett_e_Page_121.tif
c651ead5d94425debe406800339961b7
c9218f4003c683f5cbfe52d1bf05acfd3fad834c
1756 F20110114_AACQBA bonnett_e_Page_024.txt
06a293956609f0e4d6f49983bc8bc59c
e70b8dac1e30a3046e470abe9113e872cac26c32
1659 F20110114_AACQAM bonnett_e_Page_010.txt
0132aa4229a09c73bce0bd58542137d9
b89ae2ca6d2df448cb5066fc7437981f19d4f60d
F20110114_AACPVH bonnett_e_Page_140.tif
c047b76732b3eb2b43b16f40185abf62
c811ce0c0cbbc1d2b85c20868c608db6453f3eb3
F20110114_AACPUS bonnett_e_Page_122.tif
00c65464c46dab6a2584e087c53c23e3
88b25e2cea0399f3045b71a47950f1e0898a6b42
1853 F20110114_AACQBB bonnett_e_Page_025.txt
36bd5485d66fcecab0a7b770e0281583
eb578929dfc2db9062e9082601607838e2098c64
1217 F20110114_AACQAN bonnett_e_Page_011.txt
c134ee07ed157f679e6186652a9210b4
fce2b4e553c2ae207a0c778df045492e52b557aa
F20110114_AACPVI bonnett_e_Page_141.tif
0be6dd126cd3291a084feb6167a216b0
76495eee8b43ef7655ea7349d8df713c2bd25eb1
F20110114_AACPUT bonnett_e_Page_123.tif
910ea14313a50df1a3a9c9bda5f1ce75
ae3e260a3f3a42231492b561aac24bfac0726e11
2333 F20110114_AACQBC bonnett_e_Page_026.txt
6073f3576f170a15ded987cad21cd6b5
05f9706a4ce09be49d9ef16816501164fde87528
1913 F20110114_AACQAO bonnett_e_Page_012.txt
140548a8eef19a6784faf74fad2acbc3
5195eb13146667bb5a4ac02db8a045022d187b10
F20110114_AACPVJ bonnett_e_Page_144.tif
369c2e6b1df7cbf08be84685fb177538
8dc82afe71209bdd05c708a464a81b120624690b
F20110114_AACPUU bonnett_e_Page_124.tif
c62ed7bca9401aa60e020ea388b860d5
25bbdf718e17f5012ce60369cfe80cd0b3bf66bd
1984 F20110114_AACQBD bonnett_e_Page_027.txt
b3dbf3ae40d2dc6e44910c50a9ede2a0
99c4a224decf1d884ffbdf8fb959a2bddd81261a
1950 F20110114_AACQAP bonnett_e_Page_013.txt
eae7478ec7e345e8fc90d66ff62da178
4d90a518569e04effdfa4e8b69eafe8a73e1be80
8669 F20110114_AACPVK bonnett_e_Page_001.pro
434ad5d0eb59fd3151644515e9f78db6
461b2d341f56f8c2683aa087d87c5053c8c052b4
F20110114_AACPUV bonnett_e_Page_125.tif
eb8701350a950e7be4ac5801366ab36d
c9cfe0fb00e09c446c8707c7c50811f998f54577
2189 F20110114_AACQBE bonnett_e_Page_028.txt
f14213857820025adc1f5252a8e62319
195a5340080405b400f13a9072773bbfaeab837f
1958 F20110114_AACQAQ bonnett_e_Page_014.txt
f4a8e61bd273fb7e522cdf82e1c5b4b1
22f856b81a8eb03df63a66e72ddd6ca3d288bf08
1176 F20110114_AACPVL bonnett_e_Page_002.pro
e3c4cb249223f8a61d25adee9599b44a
cae9ce3a76dc3610f65e6bbe891644e7d62f21bc
F20110114_AACPUW bonnett_e_Page_126.tif
4785d277a5ecaab8864591bf8873984f
d527b41ecda4062aae444d70f8e1c796a1e9dd55
2096 F20110114_AACQBF bonnett_e_Page_029.txt
f0188dcb34f841ad7ba05672c3908f1b
9879e714919f1c17c13bd4f49680acdd32d5631f
2008 F20110114_AACQAR bonnett_e_Page_015.txt
5c97274a6fc2a7c6f5ac696571981432
0529ff01ebb104f299cac24c846cf69c46af545d
1718 F20110114_AACPVM bonnett_e_Page_003.pro
de3ed758608b0f11548f7f13536122e5
f82dd788b9249361f706ae0ef72121ad8e09fad3
F20110114_AACPUX bonnett_e_Page_128.tif
13e9c7c6a977c3519e56dc398af19238
5d9c90aef8c3bd99eb2f40700a86042df63b0e66
1907 F20110114_AACQBG bonnett_e_Page_030.txt
6e6766e43ca1d717e525a62b38eee9ff
579f5ff775929f1b360967ec988586c0a4231d7b
50290 F20110114_AACPWA bonnett_e_Page_017.pro
5f8a0d8165d1072a84a66a28133ac133
c7353976dd5be7ed5793e142593c56225a11cc11
2073 F20110114_AACQAS bonnett_e_Page_016.txt
7e96d1abf94223aef470a746d4620c0a
92cea4494359a763910351bb5ca02d9a6bf77434
41049 F20110114_AACPVN bonnett_e_Page_004.pro
aedb416dd9927c228cafe689b39e9368
da0b4f48c1973138bc158292948be22826421dd0
F20110114_AACPUY bonnett_e_Page_129.tif
60751638080e5f279ba898bc23d93e25
3431bd1bd3f271d88c9a9f6c5d4d5a966c482dec
1987 F20110114_AACQBH bonnett_e_Page_031.txt
4381b41eebe9249846149071a7d5e578
1b2cbb82f446fe5b6cc05307e2498b3301ec79a3
47327 F20110114_AACPWB bonnett_e_Page_018.pro
f4c781ad990470584c148d656e1168c0
62acd2d96975dba9b8ea9c820d333d6838539cc8
2049 F20110114_AACQAT bonnett_e_Page_017.txt
e54c6ed8d0c797df5a110621ea76deff
4b04a998bc780bd53e68151bff3de0949f38a4bf
9994 F20110114_AACPVO bonnett_e_Page_005.pro
7cbc9c0e831e76af1fa08c1c860be081
d4d8323a4c765e6202051dab023a4a1b52e74410
F20110114_AACPUZ bonnett_e_Page_130.tif
252c109627e9e7c86fd8ee0874eb1b57
a842524da3fbc52abc1a491574a7fb685b00fd4c
2120 F20110114_AACQBI bonnett_e_Page_033.txt
681e6b8b1fb8b3b5659604fa09818225
376a09f76ba664977c23828075ee94f1f51d920a
50396 F20110114_AACPWC bonnett_e_Page_020.pro
3ff40a784f95faa1b373fa5e3f88cf1b
4b467e11ddf5504118051b85d17a91172d4131f6
1871 F20110114_AACQAU bonnett_e_Page_018.txt
27ab0c61691f092b6996d088dc833cc0
c2f6ba17c9a94767729f762e781bfe9f7928e859
62831 F20110114_AACPVP bonnett_e_Page_006.pro
fcea5868f7b4b148681bc95653a07640
ffdf971bdb6fa5cd3e592f6cb4f900218bd2aebf
1784 F20110114_AACQBJ bonnett_e_Page_034.txt
48bd55c219c18c42c5fd9d66f79f24a9
ebc85ea704d8064d6d7665e6eef01bacb8148c18
45584 F20110114_AACPWD bonnett_e_Page_021.pro
206002de6ee62bf0b5c43c7810e3169e
02fd1d197b30697c686c980d3dec9f0a81d9b660
1965 F20110114_AACQAV bonnett_e_Page_019.txt
9116c00392e7bffd77241c0e98e88ae8
8291a3965a3c131416d56aa76617f57152858a4e
67865 F20110114_AACPVQ bonnett_e_Page_007.pro
fc1ecff78fac7e1c6c106425aaedd9d7
81acf7a0c00fffe798096e0cf22ed7694b7a6707
1880 F20110114_AACQBK bonnett_e_Page_035.txt
1e93fdeccb12a60ce33789cd3b6e7a8d
1ab57cbf86f7cecbc26c7841ad87fbee4d95152a
48920 F20110114_AACPWE bonnett_e_Page_022.pro
98fa824ddcc84766ac67d486748c499e
0cb853f037b8550aabcf30389c8ca404881763a0
1988 F20110114_AACQAW bonnett_e_Page_020.txt
c3bcc03df911881ee578d31002a52787
10f9cf34effba604deb2f93342427123b27b321e
1867 F20110114_AACQBL bonnett_e_Page_036.txt
611852a81f37fa3c163381c97e8e453a
95e50168d7e3f516d3451e662f952da8c5d223f3
9534 F20110114_AACPWF bonnett_e_Page_023.pro
bd065e85e5f451711bb405b0750cebc1
6f29a07b166c10276b3bb5c3be2a6d781a05113c
1897 F20110114_AACQAX bonnett_e_Page_021.txt
6e2ffefcabe0bfd5dcf96965b730aad7
cc7ef32294293f34b1550ae0f15baf3c2407cae8
34769 F20110114_AACPVR bonnett_e_Page_008.pro
14d14afed47104cc4d0f50f4ee8f86a6
ec6d94a88206b58e904ff5c5a378a599995aefc0
2046 F20110114_AACQCA bonnett_e_Page_054.txt
6ef69cbc30248af2e48bf157d26eafc0
d01bc5c501d641603361d9d1fdb1e97dcee51704
1804 F20110114_AACQBM bonnett_e_Page_038.txt
d16539b2f7d523974d31eca015668288
3524d9a00f1c9e58bad2dffc7c468c4a83b32486
42549 F20110114_AACPWG bonnett_e_Page_024.pro
cd695a83c0d5934dcf616642fe9d80d4
c2a626d77657f3d52e2249fd41e85e042bdbe919
1978 F20110114_AACQAY bonnett_e_Page_022.txt
d9cfbcd0304d5d39a90df56f92577938
d9a7d72057426b9852c15e2ed6bb0a9a4a3d582a
48554 F20110114_AACPVS bonnett_e_Page_009.pro
76c5d46fa34a431642a463ded7c9e210
964aefeabaaac013e644aadad9563e6a0ff6ebfc
1999 F20110114_AACQCB bonnett_e_Page_055.txt
b9be899e003bbd94bdbcf71a2203dd9a
e6dace1314d36d59e2e2fa452913d45f5df0ca8b
1949 F20110114_AACQBN bonnett_e_Page_039.txt
38c074997a5bb36417b4000d86757c6c
f1c75194982755a6e8ac9531a5c70a68eb828631
46785 F20110114_AACPWH bonnett_e_Page_025.pro
53f39ce0e760a9404d3e904866a43854
1b4f052b9e7fb20240af6f09e13e749361837f04
422 F20110114_AACQAZ bonnett_e_Page_023.txt
960e9c5d60b38ce85fe66d340def32b7
8fac08cdb8958b2b5cf7b48dbb93190500d4f9e7
37713 F20110114_AACPVT bonnett_e_Page_010.pro
5f9b7e1390b61a567e4afa0164e5b89a
573b0fd1bc158efe71c9de141b761bb4df706521
F20110114_AACQCC bonnett_e_Page_056.txt
b7f8dd5c98c6d613e510d0a9803336a3
ddafca0af47c68a5cf04c04628d70a8b5acef5a2
1676 F20110114_AACQBO bonnett_e_Page_040.txt
ad817bcfa74b1792cfde90ca14acdd2d
6498f3f52dab9ced2f1aebb7aad655359aac740f
58264 F20110114_AACPWI bonnett_e_Page_026.pro
0a286fa986d95bb99920ca79083887a0
96fd11e6291b7feac4b9ff21a2f8bc8409b36328
30358 F20110114_AACPVU bonnett_e_Page_011.pro
a5c4640bd4c3a80ca5679305011c491d
0ccd4428488722aaff06c0f0615646b41a8f4c4e
2217 F20110114_AACQCD bonnett_e_Page_057.txt
ad44e14d9cc443e0079499dca48b0513
ffc96b34dc3d599ae899d480bfb5a115c70e880c
1633 F20110114_AACQBP bonnett_e_Page_042.txt
ba187c5d4723154275bae01dee105d90
2e57dc456c405a98b7444b3fb00caa1f508403ae
50235 F20110114_AACPWJ bonnett_e_Page_027.pro
e8f014df8c9828fc2ae1b22ef1c16dc8
b322e98b6db53d439f93a9a77d112b7075b32b42
2331 F20110114_AACQCE bonnett_e_Page_059.txt
bd475f9369ff99df3632e51e260e035b
ffdff202f75c1664ff9f5114c23802e7c63573fb
1948 F20110114_AACQBQ bonnett_e_Page_043.txt
34f0890d813c153cff705263770c6bee
2bc5c105f58f8cb7440785587a66d12b564ce684
53094 F20110114_AACPWK bonnett_e_Page_029.pro
6ed01adbce8771bc5fcc49fbbc8dee36
e7d2cd84046a87db2caf4920398c9c00e02ba342
46145 F20110114_AACPVV bonnett_e_Page_012.pro
a6518dec085a62d8d5be043f354d3942
235f59a25516533d9b313395c7abfb2d9c2a5b65
2223 F20110114_AACQCF bonnett_e_Page_060.txt
a6e75ba1c36cefef27203623cd109e1e
7d38075d035143fe1c4c380982c4ca8cabdb5fc1
1879 F20110114_AACQBR bonnett_e_Page_044.txt
22697515b74ab6ebe3d705a8c78b8aa2
1c56cb33d94b19cfee73784d99642570f38f2229
48291 F20110114_AACPWL bonnett_e_Page_030.pro
22b6af1df7eade5ab7fb2f1918112f2d
92aec355abd24f579d1728490ed9426bfc2abbdd
49053 F20110114_AACPVW bonnett_e_Page_013.pro
28d85aca666ed3579b6cbbd4fb0a412d
c14c55d34b3ef95ba28f79f32f5ec5ae45210082
1967 F20110114_AACQCG bonnett_e_Page_061.txt
94d582835561c6434ee95b998209efab
4d3b1ec0c9aab552ad2fbe62f21b0111d7e2e17d
47693 F20110114_AACPXA bonnett_e_Page_046.pro
725295b3fea7804d8d6d5097197eaf51
369378173d634e132aea9faec17979e901926b0c
1636 F20110114_AACQBS bonnett_e_Page_045.txt
38d66943715c3433763830ae1bf1c8db
6b3a6a7e8efd71a8a6a7812ff9f7609decda0ead
50320 F20110114_AACPWM bonnett_e_Page_031.pro
49e7d1ff1085664486875937e0d7e853
8c98c682cbf26eb2d7adfd51a0031becc551d1de
49054 F20110114_AACPVX bonnett_e_Page_014.pro
238b09e2db4944817f5f3e455223a0e0
3502bba94b7dfd5f47485548e3e0bbafa556a7ad
1884 F20110114_AACQCH bonnett_e_Page_062.txt
87f5a2a2c32fa390338b7a96f448201a
6605ed47fa75ecd3f17fbdaa972635997f642d51
42010 F20110114_AACPXB bonnett_e_Page_047.pro
e24dab0b71d48a10bfec492b9b0a0383
e3621195e941241d89de97bdd224d2e01a2d1d2d
F20110114_AACQBT bonnett_e_Page_046.txt
6f9ae2fd8acbc7a4ce918a75d7e574fa
e687e9ec7ecddec0502d4dd05322eb0d2c0e2d7e
36333 F20110114_AACPWN bonnett_e_Page_032.pro
9e73fd8817552e1aa7853ea38be3b4cc
bc53e16f3c22e58286cb07b8e037b6d0b8c66000
50891 F20110114_AACPVY bonnett_e_Page_015.pro
e8963eea5e5975f12501836abf2eeda7
8d51d416a010f671e156c4a0c4ef40f02c283c3f
2188 F20110114_AACQCI bonnett_e_Page_063.txt
ef1534ff9bf84c0ea13e8d51e3e574f8
4fee8a44110eef8cf02d72909f97ca7f862d3654
37813 F20110114_AACPXC bonnett_e_Page_048.pro
427332725e9e159257fa19bd79d2a219
590b9a84862f3c3143e8a2917211d677fff64f14
1986 F20110114_AACQBU bonnett_e_Page_047.txt
d8cf9a56da0f0009046b8eea26105585
fd02b1ff12e09ee444eb4cfe3691343f8997b069
53610 F20110114_AACPWO bonnett_e_Page_033.pro
6f9d98d115be748df452ed418f86a1a7
e52d1055f176bfda251bc65d4100563db2926988
52713 F20110114_AACPVZ bonnett_e_Page_016.pro
c748e9cb76639677ee0a301c6c654afd
a8c605c63c64344b7fe8cb2fcc334326eb5fbc4e
2276 F20110114_AACQCJ bonnett_e_Page_064.txt
3a3719e339e87ba1886d6b7b7ea29832
235a2761f3c5568741afd32f0d95afa4b1da2dd7
39495 F20110114_AACPXD bonnett_e_Page_049.pro
c6f287398cd8a277204da580a0d08212
c5761e70a2206407f695bdbdb613ee9e97b190d7
1830 F20110114_AACQBV bonnett_e_Page_048.txt
a4fa321110c9a8b722049c0f9eb71579
f26db3c116f40857959361494b6923c1ca57e14a
34084 F20110114_AACPWP bonnett_e_Page_034.pro
84a58aff8e6e8b483b9d01cb94c50038
f9ea5e23707d53fe3ba83840558748f4464c74b3
1926 F20110114_AACQCK bonnett_e_Page_065.txt
df048f17cf2edf2317f01f57c1911cb6
a6013c99f92eec8aef0913695523da8bd456c71f
36714 F20110114_AACPXE bonnett_e_Page_050.pro
cadb5a94c53d5f7ee3c0a117ac22926f
2cb06104cb4af1042274c389278c4639b3ec2412
1866 F20110114_AACQBW bonnett_e_Page_050.txt
54cfdedc47de017f5bea042a4e0addc7
20a99a1f3fc710efe19a74c122d6103a00a26d04
47535 F20110114_AACPWQ bonnett_e_Page_035.pro
cd161c514db8f3e530b23dad370ce656
38b82356298373957a38d1b3b78de644d2719488
199 F20110114_AACQCL bonnett_e_Page_066.txt
56278d39eec0ffe7f95394448a3135fa
2d2c533f2c47f8d177401d278e982a616d5fec01
38391 F20110114_AACPXF bonnett_e_Page_051.pro
9650c018401060ee1799770578e13718
64328f9fe7a2981c0b163f406e62919feba6b29f
1943 F20110114_AACQBX bonnett_e_Page_051.txt
a5e9a5214d9d1afba92e07475bff16ea
624b0db82e5c38da53882529a908e78a12e08f29
37667 F20110114_AACPWR bonnett_e_Page_036.pro
2fa9d7e0d62d5b082bd3beba1beb6cfc
eb0c85c7d68e71daaa36da52fbc95cdd8f291cb1
1860 F20110114_AACQCM bonnett_e_Page_067.txt
e049a8489185a37890e7353497d4b867
d9819e8516b5ae41935c8c4bc9c40e94a94ba6c9
12591 F20110114_AACPXG bonnett_e_Page_052.pro
7f258ba39400a08185900e49a89ee49f
bba831154ac6bd89ff2afbc7bc8ae650262cac19
548 F20110114_AACQBY bonnett_e_Page_052.txt
4a6751769291788eb27e8bbea9e07e49
521a554c5fc991faa40170361ade1383ec11cf7f
1962 F20110114_AACQDA bonnett_e_Page_083.txt
ff9220fa7270c931387b87fa132d7e8e
06618a9145075b20b581b406f9755726d26d3a29
1872 F20110114_AACQCN bonnett_e_Page_069.txt
4910f5e357dc6e10422386389dfdf9e3
1587e75f340ccb0287137c89f5cf461032b9e831
52135 F20110114_AACPXH bonnett_e_Page_054.pro
2c6b479d2a91c0a713048b9cdca5ddc3
de792a15a4968c7b7cf01540771eb5d990f27d1d
1818 F20110114_AACQBZ bonnett_e_Page_053.txt
f26f0336cab6a391bac2674b71ff5236
d2930683a52a79f91e630eb60e696cf8c72cce12
29271 F20110114_AACPWS bonnett_e_Page_037.pro
0f997680c68095769b1b94e3d3cf3ad3
357422bf474c34444ceffdcc60712a3b22f31ff2
1983 F20110114_AACQDB bonnett_e_Page_084.txt
6dd72464d6f03478c2b4cc394b79b1da
55f8d5f8b97ffb8be20a85fa4579e436db381931
1927 F20110114_AACQCO bonnett_e_Page_070.txt
976f093ca2fa6f47719f72e4d70e0761
e2697a7ed3a2ce4c01fffdcd1273702a7cd0a6cb
50481 F20110114_AACPXI bonnett_e_Page_055.pro
c6fde300772a8207d893cd0e988d9293
1421def791173853cd4ffcf3d98757a7da3a28e8
45595 F20110114_AACPWT bonnett_e_Page_038.pro
3519b2fb178a74a63ee47d1326927a55
e50a350e233efffd95fa4afa39ba745282293d93
2025 F20110114_AACQDC bonnett_e_Page_085.txt
9842e8b7be5b6d80ceaf178c7913e828
fd093f65db09ef0f9a75c4f892d534be4e91497b
2087 F20110114_AACQCP bonnett_e_Page_071.txt
043ae6d4751336560dd9be8790cb6107
4c8468df3b225546f3f2da1832b82551f385a094
50610 F20110114_AACPXJ bonnett_e_Page_056.pro
7ea319917c0a376334286bea0a12cffd
21d0bbeccea17ab7b005093115a984e3badb0aab
49357 F20110114_AACPWU bonnett_e_Page_039.pro
f2163a487feb6a43bb5f5df5d945d8f1
959e0f4adf8ae344c0710eb9f41a7b8364f5446f
1964 F20110114_AACQDD bonnett_e_Page_086.txt
e8b1cc7b787156a05e0c82dccb9046ec
40d1b54b095e6ef194fd874dd86bfe225f265804
2037 F20110114_AACQCQ bonnett_e_Page_072.txt
7f479b198d52982779f9251c8ad7e0e7
fce966c4261891059529ed53016bbed528d97ff1
55854 F20110114_AACPXK bonnett_e_Page_057.pro
33fc473914dc8d0fef310a05ea17e51a
a5cfb4712e7d5c68d69fe01e05009ba5251602f0
34919 F20110114_AACPWV bonnett_e_Page_040.pro
5eaa3277cadc63cbd33ac33e5bd9a3fb
79baa1c488f2f4f475f81c0c7a8831d60c028f31
1936 F20110114_AACQDE bonnett_e_Page_087.txt
f636a959acc52a72fdc9ad2806e4186e
e301a6ed20379b4673c8e5da125535356feb4d45
1981 F20110114_AACQCR bonnett_e_Page_073.txt
66bdfd2a11e70d78c2dd84f2ad43badf
31d5d7719b339802b5654c6d7c1e97ec3fe5618f
49707 F20110114_AACPXL bonnett_e_Page_061.pro
3926ec43d3d68d0b1381a65cf9e4f8d7
18bd327ef0a432f1064062e687da579dde9c06dd
39835 F20110114_AACPWW bonnett_e_Page_041.pro
92ac83186d60032a9a3c807fcb8ce87a
9afa79647cc4198b591680e3051273d30f0712cf
1919 F20110114_AACQDF bonnett_e_Page_088.txt
75beb1efb807bede73548e61c740e7b7
0da1d0ec9abd31b356e7c4dbc03dc529b548dbe5
1994 F20110114_AACQCS bonnett_e_Page_074.txt
7499f8490a631a070057162ba063e860
e647b469a025f3ee843e2e499226f325a0a64e54
46247 F20110114_AACPXM bonnett_e_Page_062.pro
00f587a380ddfa78beabe1509d5ae328
1010769a154998178b43479cbf726765fa6ffb98
37271 F20110114_AACPWX bonnett_e_Page_042.pro
c81ad7cf80ec25c903da81a0d444b972
6435ed8cf85685cdfc7cc9c7fac7b37d6016f548
1587 F20110114_AACQDG bonnett_e_Page_089.txt
762aae3ab106df7b9fd57fb8999a98a8
a8202d80e2c5d0599587e83a6ffbb0ecfda839e3
51211 F20110114_AACPYA bonnett_e_Page_080.pro
28fb16bf178f76a8561dff545a1aa6a6
dbaafdce7c14c72a4ff8e15cb75bec74224ffa89
55307 F20110114_AACPXN bonnett_e_Page_063.pro
9c65896e2786764c7c0ca7b4008795fc
3e5798c232cbf950cf9ff4d4be2a13d17d71c647
47546 F20110114_AACPWY bonnett_e_Page_044.pro
ec1742dc9e0db3684c25ff57920f87df
f282589fedb3d346d0d604bc2fa42ed61b215bbc
4767 F20110114_AACQDH bonnett_e_Page_091.txt
a3e04b6a1c80ece70e5109ace4089406
72ec8050155e956b3156ce2ea7e6d5379367d7c9
49305 F20110114_AACPYB bonnett_e_Page_081.pro
0ca97f422a878a0a168c6f3e5963cf3d
8a2ac202012d90a4edb86aadb3ead15494954823
2172 F20110114_AACQCT bonnett_e_Page_075.txt
1b4e3713a6f05bc0d3d563a805c11982
3c33237a297346d9005c20130a29ddc7dd1c9c6d
57135 F20110114_AACPXO bonnett_e_Page_064.pro
d03987b8d6d020e60fae645baeef2d09
07013905cba9fbf34c224a15b3bd108de8af8668
21594 F20110114_AACPWZ bonnett_e_Page_045.pro
6ea75dd973c6c35f8954a7a26590ad92
9d65857e0f2fed15cd7a11a6404087026ab1325b
1768 F20110114_AACQDI bonnett_e_Page_092.txt
fe34498f6775bb8a4e1791f61682ced9
576da6682ad024b7a27b583e86bf823ee1122910
49760 F20110114_AACPYC bonnett_e_Page_082.pro
749e5ab17f06a7e9ce73672c0bc229f9
49dabb798e86dde0dbea712e3ad4c48b09814de5
1960 F20110114_AACQCU bonnett_e_Page_076.txt
e231304b3143d00a3ff36e62457d17c6
342e6365bf062d4bcbc33386df7f5ce2aa76c63e
48619 F20110114_AACPXP bonnett_e_Page_065.pro
4ed609d676a488a9ec68a9944df10b90
288c1f908a377f1c30ff9dbae1462385425b898a
1697 F20110114_AACQDJ bonnett_e_Page_093.txt
d29d088b000a845b3829205006a9836b
b2f7e8dfec807d053eecf8797195de4d47902df1
50340 F20110114_AACPYD bonnett_e_Page_084.pro
7e36214b6141d509a40025e124e3df3f
1924a50e303cf7c58b8264304077ef7f757b4833
1875 F20110114_AACQCV bonnett_e_Page_078.txt
bb87e157894c5ca2206a329a20f2300d
40f285c87530d2ab469040cbb2f5a93ec16de2ed
3925 F20110114_AACPXQ bonnett_e_Page_066.pro
777c7231fadcca55794f6d272594b4df
204fcfc7afdaef1a2a8d72c7f083f18856243cc9
1934 F20110114_AACQDK bonnett_e_Page_094.txt
16d891bda09537a27d89c58b5b589833
d71d3a82fcbf367cfa08252edeaa00ca0f7ef0f4
51308 F20110114_AACPYE bonnett_e_Page_085.pro
7a10d7de8d1afe10530edfa4f6ffcd7f
74c9b1423cc3cfb1a1c08d584f000e4cb12fc38b
2088 F20110114_AACQCW bonnett_e_Page_079.txt
2f3a08b38eb1ebdaf10f518562aa6056
ab2f5188e51b7084aa36c5ecbd5bf7c87dab436e
44983 F20110114_AACPXR bonnett_e_Page_067.pro
3d22471aa521b375f0cbe0812e131b10
37a08475c9316dff68b460a0cb577b731a887b8a
2021 F20110114_AACQDL bonnett_e_Page_095.txt
af05f02f5fcc9e632549ea29e4ea79bb
66f8e720b9c46cf89f92f1ec67bd449269741b97
49673 F20110114_AACPYF bonnett_e_Page_086.pro
23931ac91d83e8d8be6526e7621577ce
25a2dc644a06d4aa91d4855b48248bcc82756c35
F20110114_AACQCX bonnett_e_Page_080.txt
db752abc3ef0dffe2192e825a62f15a7
1eb983a4542129f0c9dfb2b4e50eda0c2bb12835
47276 F20110114_AACPXS bonnett_e_Page_069.pro
bbdfecbc6fa4b19eafe9b67487719ef7
e9409649a16f6f6f872a4daa1d13e791740b7532
1909 F20110114_AACQEA bonnett_e_Page_111.txt
4ac36d26bf6d46ce38d7b888ad2c4ae1
9273c584ec6253ad7db9abe3a59c4b73d2ea0682
F20110114_AACQDM bonnett_e_Page_096.txt
85799a1888dafa48f15a6858122ef9bf
ce03015fc018347cfbfa8c09bc203f0995937491
48877 F20110114_AACPYG bonnett_e_Page_087.pro
6ee46248ca495303a75da6b76f7ba38f
2749f0a6663a21f2bf40a18be88d5e4a34a49b13
2133 F20110114_AACQCY bonnett_e_Page_081.txt
bd5e4a20c832a99ecbf74ae6a3aa7e83
0f96cb7fffccc150751b4472ca581e5ec3423b5f
1852 F20110114_AACQEB bonnett_e_Page_112.txt
006a1152e956296066e9597a2d2c8c07
537772e02a7fe0e6a478eb052ce592bd3bf4e25d
2268 F20110114_AACQDN bonnett_e_Page_097.txt
eebad5c43fe3effb6b2aa99cb5cdf84d
5c43930cf3c3782b0bdb00957ae91480f118fa73
48564 F20110114_AACPYH bonnett_e_Page_088.pro
e2099e36f8db9b8da5a83e00d8d97314
9f33873efd670ad61159ffa8131d2c16ec272470
1968 F20110114_AACQCZ bonnett_e_Page_082.txt
efb2d5b1c97cf1c6875a12b811997bdf
18015306237aeeb49e055402ce5c23c17bff916a
48647 F20110114_AACPXT bonnett_e_Page_070.pro
518d81352ca7c67edc9e4f33e0870848
8a3f3564bd634f1724aefbf556e9035c081b8711
2400 F20110114_AACQEC bonnett_e_Page_113.txt
8d29047d3f13a50efee0d96102ba3301
d789ca1e508d16fec2973b87d60390d119cfa5f4
1976 F20110114_AACQDO bonnett_e_Page_098.txt
3b96e743a0b28223542e065f8fd743a3
ff4182a873c3380d6460ec213396a3b5b0a879eb
42798 F20110114_AACPYI bonnett_e_Page_090.pro
19fca9783b6fbfbc57018a37f6abf02f
b43614521a5bbede113d7cef3177856f30709542
51764 F20110114_AACPXU bonnett_e_Page_072.pro
7eb1bb0ed623800c115fd98a8bcbfe5d
73665fc3c083e2e6e6c6bd41c025b0d21b964612
2298 F20110114_AACQED bonnett_e_Page_114.txt
65e37705b3b25883934b8fb394f47f1b
604d46170828c7a4d3263faa73a39d6f24c4b48c
1925 F20110114_AACQDP bonnett_e_Page_099.txt
c7ef554fcf37aa4fbefa8cc05f629243
621ac7b33a4847f3dc13af781fd31cd0aaf2ed46
94413 F20110114_AACPYJ bonnett_e_Page_091.pro
b5b7ff3bb8fcdbba5c523939da082b3e
107b2a7f60dfd6ef491d8dbbad0b17b0e442ec33
49821 F20110114_AACPXV bonnett_e_Page_073.pro
60b1566bdc74eab9b5cba0488f693599
63ceabb455f90784ee5d9e77277242daaf3d5a27
1932 F20110114_AACQEE bonnett_e_Page_115.txt
f3c4f73d309142792ad01c9184f822e4
b999aab8ddcab529a0333ae1e405895974e975a7
1947 F20110114_AACQDQ bonnett_e_Page_100.txt
af27d4f410fc9106248b232e632338e0
427bfd34aae282bbc21fc1201f5e092a821fae51
42539 F20110114_AACPYK bonnett_e_Page_092.pro
52b90b8b72464d505ca5e5ab2ebeae90
0c7ecd08e1ca02a308db70d0d08d22862695ed97
50706 F20110114_AACPXW bonnett_e_Page_074.pro
8ab2bd5d93015f16e69192bc04c89d2f
d9da3bf029a8389dd6c98e41c302dac762ec0306
1882 F20110114_AACQEF bonnett_e_Page_116.txt
3801753b867e784fef50e5f8d016701c
1e64800c738d2ac90f4137953845aa5842da7f0c
2581 F20110114_AACQDR bonnett_e_Page_101.txt
8d6ad689a15d6f86ddf89e54bfc8a8d9
45676489c9c852ef9fb44be0442d9899a5319bbe
40454 F20110114_AACPYL bonnett_e_Page_093.pro
61e8c3613bdba28954932a3ff2c2bf46
fb3e9dcfc1ad755c3bab7ffe90b475aed6053625
55348 F20110114_AACPXX bonnett_e_Page_075.pro
27afa8e8be2dd8ff64d9d80da37339f9
a05f973bb125887b84219acf05344bb7f16c410b
1863 F20110114_AACQEG bonnett_e_Page_117.txt
86f2659d29d6bc4d7f893e23b88c3ac7
c4cd32f9666cc156f4b5ea79c5d5e0062bea41f4
49570 F20110114_AACPZA bonnett_e_Page_110.pro
1118d82fa956f2f968b16fcc47f164e4
7cac7690a5b6cb3f8cba38db2140f6d5f47a7bd6
1929 F20110114_AACQDS bonnett_e_Page_102.txt
bb65d023287de46176946563bdbd60fd
77374673b77ec5b3613b70b0043823c209380140
46599 F20110114_AACPYM bonnett_e_Page_094.pro
36bee63c1eb00928253b3ad45a85dedb
6e2fd2c1197f6b306778705f6ea556e1c4595cde
49403 F20110114_AACPXY bonnett_e_Page_076.pro
5cb3860364fc638801def96a045d671e
fc584f429b8aa8ee1d08b35fffcf09602c1fc4b0
1954 F20110114_AACQEH bonnett_e_Page_118.txt
ef83a583e2f90e32fec4f40cca7c8dca
255edc7f3a3c2836154828a6ab4d4dad0b636f2b
48370 F20110114_AACPZB bonnett_e_Page_111.pro
3e06a06afc3fa4b6705f68d9cc5c3c57
e6317a41db8dd0cb46cb6fb49192d96d317a3d79
1956 F20110114_AACQDT bonnett_e_Page_103.txt
ea20988c0ab986be4bf024dcff2beb5d
3762ef8baa8e6ef228f03e072d06ef4bed35c147
51173 F20110114_AACPYN bonnett_e_Page_095.pro
26171f338c60eb94debe627865fefa30
6a1c69253586967e17e3a5b3e6191b636543d2cf
51351 F20110114_AACPXZ bonnett_e_Page_079.pro
e32858c27f1509cd9dc30c27efcef4f1
43aaa09e4013bd240e207f628f81875e5f392239
2821 F20110114_AACQEI bonnett_e_Page_119.txt
11b08088bd2372073bc036d9c34bf225
db3241b567aa4644e9b0d682c37432877bea19b6
46926 F20110114_AACPZC bonnett_e_Page_112.pro
8d3b7d22afcebefbf3b13d586edd3aa6
db46a362e8a3f4c6afa951f4bc90654df168f8fc
1974 F20110114_AACQDU bonnett_e_Page_104.txt
80d28560636cb98f096ed3a6f8d082eb
870ed8f75c995a03822174fe93b96810de6df15e
43058 F20110114_AACPYO bonnett_e_Page_096.pro
c9e8f61d9bc0e2f3f4aee2a2e6a46e33
bef184a7d6a42275fed40eed45a84b202c2d4a27
1921 F20110114_AACQEJ bonnett_e_Page_120.txt
906c97fbbb09f7e3b442f9ab094e22ce
014f92f98a1e8592cb5956c73c56e1efb9603a73
53936 F20110114_AACPZD bonnett_e_Page_113.pro
66bd00075778b60344966b6c337788e1
0f815512bd2cf94b054ffd657cda6ca5b5009a17
F20110114_AACQDV bonnett_e_Page_105.txt
4b1402e7fb4f72e6625cc2f03d7144b4
59d346870615297238f6feb5222c6077ec9d68fd
52906 F20110114_AACPYP bonnett_e_Page_097.pro
ea778650c626f0e6b93cac638533e853
803ccfe9d09c8ec0a155706500ca480bbbf17c2b
1905 F20110114_AACQEK bonnett_e_Page_121.txt
bd5a81369b094c5f35175c03e9f88bc8
7e25192a7344ce6676268a2348ac3fc8ddfa7868
54766 F20110114_AACPZE bonnett_e_Page_114.pro
cb6de7ada19b5b0605373d1750ba237e
8950cb05e5d797841c000b00393806503eb032e2
1876 F20110114_AACQDW bonnett_e_Page_106.txt
d993751ceecd438e4e40b935cd77ce54
d57384c676868a9c9e2d8cafbbee6bf0050cb2e8
50131 F20110114_AACPYQ bonnett_e_Page_098.pro
dfd2112fb77923911f1862f38c95cebb
2227841b4238e52bb694b95c0951220da069f2be
F20110114_AACQEL bonnett_e_Page_122.txt
5983c4b5f340dd4991e13333673a3b92
82b6311a247c46d894b00b1aff809834a93657e8
48891 F20110114_AACPZF bonnett_e_Page_115.pro
01b58b84299b7d21db5c906e63fffdad
c83c5bdde022ba061d32b5a127f3f81e16d51e54
F20110114_AACQDX bonnett_e_Page_108.txt
84b34d60ca975d66148106d55b7c73d7
0016911cbb17c615a799177b54b610d92b80d823
48436 F20110114_AACPYR bonnett_e_Page_099.pro
014e0b996f6f01fbc413453ffad7d4fc
c77e01842be776d3850643d2bf2b075778eb8258
1766 F20110114_AACQEM bonnett_e_Page_123.txt
5716a4f05cfebba24eddfe87f5ead649
fe1d1c8df8f0772bb7c8f3509f230e70f84acb0a
47578 F20110114_AACPZG bonnett_e_Page_116.pro
8c23c83687887a0241265273eddd0dd8
9b2b7624cf9dbe490fa0f1aa310375121a7051ed
1922 F20110114_AACQDY bonnett_e_Page_109.txt
cf43440d6a06aeca592db966aadde5a6
433748377ecd0f9e047741a1148ffc6ccc1c5749
48954 F20110114_AACPYS bonnett_e_Page_102.pro
b395baeb8e28b7041010b99dfb6809de
3baadce744fdad2ca12c183813cd6f31237313fb
2285 F20110114_AACQFA bonnett_e_Page_139.txt
3c00d5259543f126128846a75a1c9f46
6b37afbbba43a5b118253969dd2dbd16cff4a5ac
2017 F20110114_AACQEN bonnett_e_Page_124.txt
aa827b7e76638c3f5baf2cb7490e63a9
640bb2ab2f56078e3dad024f97f170e3619f4270
49581 F20110114_AACPZH bonnett_e_Page_118.pro
a7379eb13a7a9f001ebe22936c7b3f52
9878ff949cf48679319a161097f32cba1c58732e
1959 F20110114_AACQDZ bonnett_e_Page_110.txt
19a612b149cd0e257bbaf4423d06a022
da9351582d3fea5d87904034e92fa26273af059f
49431 F20110114_AACPYT bonnett_e_Page_103.pro
d3f540b989e0d3eaf4442770e5bfa869
bda31e4ff5b12d1e0fd2c7423129212c4c9c27fb
2610 F20110114_AACQFB bonnett_e_Page_140.txt
c4c01825b09437293cc8db31b33c2872
5d0585bd6512d7039c14c0fa2c15c51eb2360eb8
F20110114_AACQEO bonnett_e_Page_126.txt
46a24edb8ede116efd4d7b8af5bd2736
143ee770203e49c38cfb746362d33cea106f763a
61053 F20110114_AACPZI bonnett_e_Page_119.pro
7baa6c290660c42674b8ddb4ef0142ee
4eb5cf7d730817beacb99ca7a1dbc3bfc2131177
2445 F20110114_AACQFC bonnett_e_Page_141.txt
1878c5442b36c1dc6e8ef6e3564d9969
03c3b6deca87d5a9ccd799c8ae767e72599ae760
1327 F20110114_AACQEP bonnett_e_Page_127.txt
ff3f35ec60f1c60ac6da282d2676054f
f055b8332dae3f83ab95b25f6a30f79f8aa12735
48482 F20110114_AACPZJ bonnett_e_Page_120.pro
9a3337b5542d79b913f85a25104eaefc
c1e4ca42264106938055ac9d174a8be5933512aa
49984 F20110114_AACPYU bonnett_e_Page_104.pro
88f7622158edd46ca7717731a649da5c
7a280f077472a8d4fba4758205c6991d3649e163
2530 F20110114_AACQFD bonnett_e_Page_142.txt
1e6d3f797956d7c27dccfadaefe4f2ed
9d3efb19311ad697d0467a63e8ddae5ac5dbd2df
1864 F20110114_AACQEQ bonnett_e_Page_128.txt
d975c2cbc08d122d41682a3a4cde0e81
87b0f2d95b12df87478b365c5a25889a6f318077
48079 F20110114_AACPZK bonnett_e_Page_121.pro
03d2cb860fd0c6ea6d0d48913e498128
15c40a1d0e48e5dd6859ceec30dccb01c1052e51
48889 F20110114_AACPYV bonnett_e_Page_105.pro
cada9c9e0c21edc86e25ff6c5c4369b9
78cdb6d0d7a855461bbd04f3a9cfe7d6eed4aa04
644 F20110114_AACQFE bonnett_e_Page_143.txt
70c5547e97d41696eb3c85017c85e541
1de397c170165880aee1b229eec6e3ca79e91af2
1945 F20110114_AACQER bonnett_e_Page_129.txt
49de311ec359c4f1d382c44f62798161
5e2d58e109b421cccd75a2c898f2b381bd614d75
48260 F20110114_AACPZL bonnett_e_Page_122.pro
aa129e55519ccce8a2f7b173813218f8
ee4898048491daa499764cb2f4e3de435adbe054
47486 F20110114_AACPYW bonnett_e_Page_106.pro
6d49549b11ff1803a7ff1094b26c55a1
9c93ad0ccb338d9737c55e9245368b75c0ee1595
506 F20110114_AACQFF bonnett_e_Page_144.txt
ad4c7edfde99770ed9f51ee291f5d1d7
012eab6605d17363b56a4bf8ab0f0fc0d9c73e05
2069 F20110114_AACQES bonnett_e_Page_130.txt
b537d1e3ccae4f93aab1264161a59239
6ba0a17be069f247f887b893f7e4c40a29efa9a8
44557 F20110114_AACPZM bonnett_e_Page_123.pro
b0205bd9cc7a176f5825c0230b54b89b
814085facf3413afc53d96e216b73bec10223b93
56588 F20110114_AACPYX bonnett_e_Page_107.pro
a3adf86bbdabb9ec6a7f0a83cfa22c87
7abe855e9fe6c4c17b00aa51ae80e3e0c4b5a86c
537309 F20110114_AACQFG bonnett_e.pdf
44c745bc3e4253490e86f60e13d29037
75877688e8bb9a1739ca413a9d74ac05ff9b396a
2055 F20110114_AACQET bonnett_e_Page_131.txt
acbbf7b117722d0d05c3ea37348113c8
75b4162657a12f12a1d1770f0568c0b06568abc3
48964 F20110114_AACPZN bonnett_e_Page_124.pro
8a3c6069471a1329479407e2f6000f52
edb263d3d5367299e06340ab0ef1dd6feb6dacdf
48909 F20110114_AACPYY bonnett_e_Page_108.pro
ec68bec0f9dbeed8b5f7f09387cdee41
1e72da1b805fc583b475d0be1df7cc7891ec6a33
112520 F20110114_AACPCF bonnett_e_Page_140.QC.jpg
e067e2400277ed1630bf9f017f2948d2
ec7f69b3e1030756666bfd4afca461199a21270e
7664 F20110114_AACQFH bonnett_e_Page_001thm.jpg
2a65e20dbce7398d20209a95a35246ec
eb9816bd6b2e041219a5ab4c1ef002df98bf8e91
2026 F20110114_AACQEU bonnett_e_Page_132.txt
2ca3050133731b59f96cb23f8b829c54
6b3d8ef0ab3c6f4752173d0de4e7e99e4337cbd5
51658 F20110114_AACPZO bonnett_e_Page_125.pro
f71eabe5c1f05fdb27993cf9ab02df12
42893ba14c0379a80a119a7b9f9982e1fe9865da
48788 F20110114_AACPYZ bonnett_e_Page_109.pro
fe7728ab2a76e6004779d083ad451259
ce0bd84fb81e302606aec5cb9182a6eb7c64025e
75536 F20110114_AACPCG bonnett_e_Page_054.QC.jpg
8e698e6858852ec45f20b816845b1429
3486421d9851d00e36961069257e4ac63ec5eeb7
20745 F20110114_AACQFI bonnett_e_Page_001.QC.jpg
b12a93afec839bcdce4f2cc5b0e62635
5e514b8d700a7dfc5c18e1c84c7f87857cba0cfb
F20110114_AACQEV bonnett_e_Page_133.txt
4346c2df631229c24bf0a47d0205ac2f
bae32d385a4957eb85dda8ee2833518465ca0710
48763 F20110114_AACPZP bonnett_e_Page_126.pro
1fb7fe4ce72c58b871365dd6c98b4368
d164812d028e40b7b2d9012d31561f9aff063fbb
F20110114_AACPCH bonnett_e_Page_143.tif
94da9083eb2fa62bbd608806cb554dc0
02b7e9b9e9896f6e5c3ca4e84dbf43c1d7877810
3105 F20110114_AACQFJ bonnett_e_Page_002thm.jpg
649b3953b94067d1dfa31b01ae40f62b
5f6b88d209168298c93fda5669ca7597ffb6286f
1539 F20110114_AACQEW bonnett_e_Page_135.txt
3f7b6d23aaca3336ace39eea6b27c5f8
b16242d0de136217ecbeb8ec515a2e23d9ac2922
33092 F20110114_AACPZQ bonnett_e_Page_127.pro
70e6839136b6d506ee14f44936cacb96
c758996509477cfc6aa2f1e6b6c8fe8aea44a62c
F20110114_AACPCI bonnett_e_Page_030.tif
5796b4139ef7e141506bc26401c8ed04
a7b8f1663d038e8fe2cf3b3ebbade5c1bd11b885
5671 F20110114_AACQFK bonnett_e_Page_002.QC.jpg
948c53e2a4f91c5412f43f43b50d3c02
a03333ba5446436a7d7d01cf51a02955b9350b76
694 F20110114_AACQEX bonnett_e_Page_136.txt
1d33aeac32f4cae1b8010e7eb8798488
c760a76f44d5709c133434750b3f593f2464ee02
44712 F20110114_AACPZR bonnett_e_Page_128.pro
3aa65669d47abff5f279d443d2341492
5f7b4b7241d89b6dae357e77da0fe903948c0f0a
21707 F20110114_AACQGA bonnett_e_Page_012thm.jpg
ddb4e1457b01090d8c18e591411e1b25
80e6fc69042f5dd9da3d4a4a6bac313d52a87e8b
201047 F20110114_AACPCJ bonnett_e_Page_087.jpg
212c2ff70b13c6838700c47504da04ef
0b7be8ff3df305cea6cc88075d6f1a5b124b05ae
2999 F20110114_AACQFL bonnett_e_Page_003thm.jpg
04646f739a182b0f5b562df631096464
979519ef8188962f2764bebef1540e5f879f40c3
2128 F20110114_AACQEY bonnett_e_Page_137.txt
459d89ac5a397bbfaf594de9f13edad3
549baca90100730656f2a28a515a9a400c41d632
51425 F20110114_AACPZS bonnett_e_Page_131.pro
3285e74a1e6ddc880e88d0b1d4ee41fc
dcb404fc1ecd154eb7300609e8cd8585f201630d
54947 F20110114_AACPCK bonnett_e_Page_058.pro
677d51c5d6491787e5572c42a7b7b3a5
a1b6e762499d83e6f0f6ceb00caf5e8409d8d0c9
5681 F20110114_AACQFM bonnett_e_Page_003.QC.jpg
f7edd13be5cfa442d6c33dddfeefcd86
c0b56ebbd2a30d0948ef2fe26002d1432f1a8660
2449 F20110114_AACQEZ bonnett_e_Page_138.txt
918418da6121f9b4f14ad12d4e090186
980d4e81595d213a729fc9833ce95d6b11f812e8
51371 F20110114_AACPZT bonnett_e_Page_132.pro
5c5a25a14c8f883a908ec6b614da4f94
40d61fa6231c257eaea7a68c365460625b1f1ac1
69299 F20110114_AACQGB bonnett_e_Page_012.QC.jpg
1e6465c0106bde18f0c0ce27f7bd6be2
d8d67082d5eaa90cec3cbef14e0efd03fd51c5b4
49691 F20110114_AACPCL bonnett_e_Page_083.pro
3556194119492569211a2ee7b2c9f836
a9859df8ee1cdd8414314f2bb77210bec693bee3
20557 F20110114_AACQFN bonnett_e_Page_004thm.jpg
a124ebc063e96b1c5ae4e0cb956b787d
a1af1b51ce683f08288dda1fb1c2ac80c6e72bf9
51374 F20110114_AACPZU bonnett_e_Page_133.pro
55374e3b611949a63430b614166ed866
cdbddd4b3035c3e782d5e5707759b51c6cbcc94f
223575 F20110114_AACPDA bonnett_e_Page_075.jpg
5cdc8af645764a7dbdeac104098a0820
4ca42de52154856623b6c2899403a9c4f683c1a2
23788 F20110114_AACQGC bonnett_e_Page_013thm.jpg
4852227565e8fa72881d4d12fb5ad67f
8a9d82bdd2f3dbf691c0a45ceb391a065ae5a917
F20110114_AACPCM bonnett_e_Page_029.tif
c8d969e41fc3f1af958db656922d255e
c1ba3a186069e12a8e49b55fa4e0bfaef77d2652
6890 F20110114_AACQFO bonnett_e_Page_005thm.jpg
3ae29c90c08147db45aa9fc59cf0b8e1
9576c119ec5107951efc7e69299aca6f137490bd
F20110114_AACPDB bonnett_e_Page_092.tif
5841fd66c650610f84c6d5f0a08fc5d0
d5377483ac27eb4ae1eb67a376a5bf7031fc7c7a
75993 F20110114_AACQGD bonnett_e_Page_013.QC.jpg
8321d9eb544447090f34d29c48398cc9
5ec4ebb9ca8b2a8bffe63606672baae3ec28600f
F20110114_AACPCN bonnett_e_Page_127.tif
6ea7cd6fabe2111a41b9131817529cc5
6f3f31b0f4b48a1758cb0ebd9eb21ba1b84d9343
18764 F20110114_AACQFP bonnett_e_Page_005.QC.jpg
52151dcdb319a2d78ed4b851f6b4326a
2e2cca6731b30569b1474cd5d855ad4adea62a92
40739 F20110114_AACPZV bonnett_e_Page_134.pro
7bac3380070d2aaef27e808a189be0cf
d99a8bef9f8b7650d81e84587b8d228d8275fdf8
76319 F20110114_AACPDC bonnett_e_Page_101.QC.jpg
50835808b0754a30430204805869ff3c
9612e4e3f90843c79895d1c15eb2ddc142ad21bf
24409 F20110114_AACQGE bonnett_e_Page_014thm.jpg
ca154276f5681f38b07281584427c65d
83556a346a65d60ef9ce06a8b1fc10fece5e21a2
F20110114_AACPCO bonnett_e_Page_105.tif
e3542ea7ca26bdfe507e43e511144288
67b1f25a9362f989dd088d854f5285580842ab2e
43260 F20110114_AACQFQ bonnett_e_Page_006thm.jpg
6c8784990f524d2db43c52f86242ec11
503079d61d8051c036bbea4edaddedd934a5843f
35469 F20110114_AACPZW bonnett_e_Page_135.pro
c7a16839a9755558159d9baf781df6e1
9047469f33c8c769cfdc684cd9b7829a49fd57fd
F20110114_AACPDD bonnett_e_Page_106.tif
50b7e8ae44fa32c5a65970a46401bc6f
1d02bc6668242116d37190b006b0adb7980ab6d5
76719 F20110114_AACQGF bonnett_e_Page_014.QC.jpg
1da2ae14248a5062e5d99195945242ad
bb5954fbf26dbe854add0e7b106ad42581d8fd63
24682 F20110114_AACPCP bonnett_e_Page_016thm.jpg
bc332e14a8e16654bb9d49c3c5329473
406ef8a4572652cd6286252125779aa54923eb40
46058 F20110114_AACQFR bonnett_e_Page_007thm.jpg
2fba447c145f451486a447ef1c2bb25e
5dc82d6cfaf7365b8445119899daa2ab5a32b717
16949 F20110114_AACPZX bonnett_e_Page_136.pro
72ffb71cf0e80870a09adb17ced1ec9b
c277d9731e703e24a4ca0e300df86c69218c2e40
52230 F20110114_AACPDE bonnett_e_Page_071.pro
381f8799ec507473aa7c41ac7316e261
dbf4328020cbd60d62ef828d9f7494789c49e941
24506 F20110114_AACQGG bonnett_e_Page_015thm.jpg
6e831115578c1ccf80d72cd381f248bf
79e885e2b1482ca6a7d10838e1593b5551693256
1562 F20110114_AACPCQ bonnett_e_Page_037.txt
462d48b394fe3a673f11e066b65caf49
7ecf2e6f6216fd751bfb639b67d292a8f7a85381
95638 F20110114_AACQFS bonnett_e_Page_007.QC.jpg
19691bd837bff3744165c02038c1a27d
b43eec5bcb1f89e847c14755e67e7c682882af9e
60714 F20110114_AACPZY bonnett_e_Page_138.pro
edb5a6cec176e6a5f020c5932a3a996e
940d0a07e96c676e978b8bb82ef00f942cf70a79
73771 F20110114_AACPDF bonnett_e_Page_018.QC.jpg
15bcc23a41da297e02930307cb890ca6
438838345e72ec6f5738bbc460b90805dd389878
76483 F20110114_AACQGH bonnett_e_Page_015.QC.jpg
8daaa09522c412ac1ea824e10bd58c3e
79a923686b2155660f6143c473c0ac4f7435ff39
22093 F20110114_AACPCR bonnett_e_Page_068thm.jpg
530b3a10b95d3f896359d90170da6a67
ba7414efd944d743f1afc7054f493765293a5735
40127 F20110114_AACQFT bonnett_e_Page_008thm.jpg
ba78b4244e3d9d31fe71ca7765867382
4ed6982d93b3875115c6df79a51c97af22a08ec2
56182 F20110114_AACPZZ bonnett_e_Page_139.pro
43e64ded1c03e3db6b3e5bd9340da0a1
48f7f4a16eb8a0f2009dc504d19041c0fdbb830f
1893 F20110114_AACPDG bonnett_e_Page_068.txt
5c4a1177bbc4b193ef769c3cb806cfc0
6452608a37ff9455747c8cb449803322d02c3eea
78829 F20110114_AACQGI bonnett_e_Page_016.QC.jpg
086c3d03d98c30dceb1a8c8c0cdc7c50
8b9e59bc60765f33aa7f999d4e4cd00cceb2f8aa
49406 F20110114_AACPCS bonnett_e_Page_043.pro
af2677a6a64687bea8e0724f23b4e0c6
9816fa297f5c9d718502b7bf7104070954485910
75548 F20110114_AACQFU bonnett_e_Page_008.QC.jpg
b8dfdface4b899f3f932b0a4456124fe
62f62892fa042f09e5e4cb0a4dc4c49c79c839e3
79687 F20110114_AACPDH bonnett_e_Page_118.QC.jpg
02c69f3e1a2fffb60b4454442629e3ab
e79174b5fd632e141b79dec55317fc199a6a5f34
24311 F20110114_AACQGJ bonnett_e_Page_017thm.jpg
aa9ae934778ea5a1d74a44f14f4e7ecf
2b5cd877218faeea23d96317cc0f7edca134de65
47880 F20110114_AACPCT bonnett_e_Page_068.pro
bb48053c15155f1954b83c640bf2a85b
4f734580b1ab6a28ab26b2b454e6815a3d2bef36
44722 F20110114_AACQFV bonnett_e_Page_009thm.jpg
b55b9780c987d18a1a486f2387baf471
90a9fbc9bff0a453410e1b31e405009862138f85
F20110114_AACPDI bonnett_e_Page_033.tif
9cbeb364ae65525b430bc9df5bed1e32
d7f66685a83bde71d563076eaf273cc76492a60a
74525 F20110114_AACQGK bonnett_e_Page_017.QC.jpg
15c3b12edb9571c3acab63e537f0b1ba
269b11b166968fef2210671f9f9c0faed7be26c8
49409 F20110114_AACPCU bonnett_e_Page_129.pro
90e9f09e7a87d826b825e2ce843c5f3c
4a6a6dac3e4a397547d4d277aa1001adc210467c
19077 F20110114_AACQFW bonnett_e_Page_010thm.jpg
6587cdc70b6cf0669fe787c8fe80850e
0714df69852b1ab957c5e706bd073e1edccc9ff8
25888 F20110114_AACQHA bonnett_e_Page_026thm.jpg
f787fac87b361fcd5825fea7c7ffecc9
033175f5332d682baeffabf4106d76a98191025d
23943 F20110114_AACQGL bonnett_e_Page_018thm.jpg
00d34c16c329159691ba803212489a2a
2b928ff05528a0ed0e6dfe001233151481a5cdb9
153050 F20110114_AACPDJ bonnett_e_Page_135.jpg
11a7f7fd0f5483761c5137a2b80cb651
e459e55241eaacdb60eec0e3483167789ef45356
222385 F20110114_AACPCV bonnett_e_Page_029.jpg
7e4a1aa6c6f269940fe5d89f14bd6f54
4053b41b8f8be0acfa17cf2a53bc7eb46d01556d
59358 F20110114_AACQFX bonnett_e_Page_010.QC.jpg
f6984574eadf149cbe9ea705b75a425c
6938f407e778dd202db97f73e3aec1873d1bc121
82462 F20110114_AACQHB bonnett_e_Page_026.QC.jpg
2e54c7debf54bd53852988ddc424628f
3304612c7b9caed36b90d27ec4424b568af4c960
24124 F20110114_AACQGM bonnett_e_Page_019thm.jpg
f72f78b99fe388ba8098c1ef69116dc7
13a83bc7381ca6d6a620174dd1c0fb416f6dfcc8
41135 F20110114_AACPDK bonnett_e_Page_136.jp2
93951aa22512ad71a60d477d0abf1eac
680f21d056cc3e7c7b36c65ec5c40cfba5cde5a2
79145 F20110114_AACPCW bonnett_e_Page_056.QC.jpg
a9c8ade7f50829a4dca6462b245ff169
3077a73e8cf8dacb58351023a1b66a328be3a85d
15473 F20110114_AACQFY bonnett_e_Page_011thm.jpg
4e5a0c437c43610a4954fa0d0bebc08b
63cfa9cbfb53b8d268f39e40873dae667a13d48b
73067 F20110114_AACQGN bonnett_e_Page_019.QC.jpg
e480649eb21585c302983783dacd73e8
a899aab7cc67012045a519fef2701268f2ee36fd
75640 F20110114_AACPDL bonnett_e_Page_109.QC.jpg
afcd501f06e9a1a734706412a4b79403
2c3e4d27adae6b9487c37361d0ec723e8a5cda36
78047 F20110114_AACPCX bonnett_e_Page_082.QC.jpg
f913aa3b527ef1fd3eb917477a910206
d0308695d6ca60d5c10bd657ab947afba5a5af49
48565 F20110114_AACQFZ bonnett_e_Page_011.QC.jpg
e51dff8fbacd2992bcdc85abdbce7001
b5f5f7b701ed72a2afe1f27e2224057891d3893b
24631 F20110114_AACQHC bonnett_e_Page_027thm.jpg
167baeb007c3b09efc453c025cf9e449
42587424e21faf5f3d3e3e77dd5715f0c5ca5333
24080 F20110114_AACQGO bonnett_e_Page_020thm.jpg
323d5ca77c8fe9741054009f81196b09
28336c348335cc4003969fd276c3d19e9deae15b
55061 F20110114_AACPEA bonnett_e_Page_028.pro
59f57be28c35bc42728822e27af4878d
eb215c450553129e96afd11ad55bec0b3404ae14
72265 F20110114_AACPDM bonnett_e_Page_081.QC.jpg
46c1518cc46afa897d8d641c9c597855
14af40c75942b01f5e0eeba4a503e01b4f0feaa8
100165 F20110114_AACPCY bonnett_e_Page_038.jp2
bba1c95ad243e720853fdabae9ed7a94
e1e5f7cb150c170bbba4abaed2abec5f33cc1ba1
78640 F20110114_AACQHD bonnett_e_Page_027.QC.jpg
5b43fca53981254c642dfc569862fcd9
299b4420387f6810309ba9befb130c0eadb9ab5d
76783 F20110114_AACQGP bonnett_e_Page_020.QC.jpg
9a850f5005481e6bfb1527f3c499e87a
b6d4a9a6acd327c084dfd06cee82ced5561edabf
45609 F20110114_AACPEB bonnett_e_Page_023.jpg
f8f53626fd7e37c2e4155e670e186aba
c934b693e06e1bcf8a7f4e854bd06fd56b871bcb
1982 F20110114_AACPDN bonnett_e_Page_009.txt
491c98d2aac3d328e073341f4fcf2825
61af524db0fa6bf0dc59ab1e864e24852710eba2
2629 F20110114_AACPCZ bonnett_e_Page_006.txt
51312d57d5c08a9932cbc6843d7567dd
004831c423e829765c7003a494eb5beb78e41f33
25258 F20110114_AACQHE bonnett_e_Page_028thm.jpg
227575a93e2f84a51b43d129d9c71d22
fb04e21169b76014a908af1686cf5aa411b4d8f1
22526 F20110114_AACQGQ bonnett_e_Page_021thm.jpg
086f165f96d2fb1ac420738c3de96994
5ab1cc6690b2f3fe47c29d5ac870f1e02ed91000
219255 F20110114_AACPEC bonnett_e_Page_033.jpg
074f647e3993c0b51d913c9495931841
811a3e93ace8a517ece32e165fae120ddb8e574b
44892 F20110114_AACPDO bonnett_e_Page_048thm.jpg
4ce9d56a59ee73468740f9b2f1b000cb
445d37b0a4475a6b27d8d47b6aa57da5c6d37e5f
81241 F20110114_AACQHF bonnett_e_Page_028.QC.jpg
81f907a7e7a8124cb979b80f4d1a9c9e
60fa690b3504b247832bd30ed2231a3f66ab3381
66742 F20110114_AACQGR bonnett_e_Page_021.QC.jpg
baa7ddcb9695673b443adeacdcb59cbc
73cfee79852692e28edf4b787b51b0f6018938aa
200094 F20110114_AACPED bonnett_e_Page_088.jpg
a4985133b9d95816db9c03d60d20bd88
e633c1304c9915fcd5d74bfd40e58dfac1b737a7
89899 F20110114_AACPDP bonnett_e_Page_009.QC.jpg
09c6e34ae7955cefcedb5bffa28fdb30
c2a17c29701f38290350cca8c4ccd09c3318e0f9
24833 F20110114_AACQHG bonnett_e_Page_029thm.jpg
5d5d4edbfa379ae11c15f500d8af26b1
c447f8d913426a0781b35093c6ce4c8f95e5c294
23416 F20110114_AACQGS bonnett_e_Page_022thm.jpg
02b77671afc18dae31fcb66b2a7e9401
0f5b305edccc930a6604844177333b83574de0c1
74617 F20110114_AACPEE bonnett_e_Page_126.QC.jpg
6b1f29e1d5d06abbd857e7564137ff26
3b4579368021bcd768bb28b111075cf3e28276c7
79006 F20110114_AACPDQ bonnett_e_Page_133.QC.jpg
141aac3e5b35feae00d3900f42688d89
7463a95e59944d1ee92f68018780ba919a948ca6
83367 F20110114_AACQHH bonnett_e_Page_029.QC.jpg
0d14843370b8122e170f958dd12c5831
5d13b380332662db2b922e01b87b7cc0c733c9fe
74763 F20110114_AACQGT bonnett_e_Page_022.QC.jpg
c5ee78ee8e58617ce4cb81ebd42c1a03
3a5f3009271c559c80263f74b605e654672b86c8
F20110114_AACPDR bonnett_e_Page_132.tif
e67fbb69597fe7a6c3be574dd044a86a
2ef660726e18282815b94263dd1b7472becaee2b
1692 F20110114_AACPEF bonnett_e_Page_032.txt
3fe140cb5219f7d2b20d45b44adfad1c
0d3fd714ee9889b6d7f1fd8d47931806c02ede47
25183 F20110114_AACQHI bonnett_e_Page_031thm.jpg
292898ae87c2e88d71489efc99993886
407a42f9ff875feb61377b9b5b9d4f119faabc53
7127 F20110114_AACQGU bonnett_e_Page_023thm.jpg
f677b7496712940fcab59ff8a406c456
fb583017f71b5b11a0e5ed2f05d9e5b8d8b89171
52183 F20110114_AACPDS bonnett_e_Page_137.pro
cb0bed03ac49baacfb57135a831c6d06
cf026b192c28c68ad5ae4cecd1009ed660fd27b2
23846 F20110114_AACPEG bonnett_e_Page_086thm.jpg
ad479884146e2bd0ed408270a8ec217a
467885e3ae5cbbd5c6aee1550e77aa097603d1b4
76876 F20110114_AACQHJ bonnett_e_Page_031.QC.jpg
749c0308b221258683113264440d4dea
e21179214ac5c59e9bc6ea2629f3b9f0f8740a74
17348 F20110114_AACQGV bonnett_e_Page_023.QC.jpg
77cf436a0769de4550757387e41085b9
2db996a20de584e893e5592f182ac8f49f6608ea
1797 F20110114_AACPDT bonnett_e_Page_041.txt
d5691ba9f9bf48b3c83906a0fc362a3c
14e5f58fa7c0fd10d7c68135714606cd6b307ff1
F20110114_AACPEH bonnett_e_Page_142.tif
9606be246691597983ed27480a93838b
e4672604df4ea84a66cf76feac7679e6677940bc
28889 F20110114_AACQHK bonnett_e_Page_032thm.jpg
2303847cb7770b91124fe5dc4d5d6447
0e08874279a79d5512eb93c97a65195f796748af
21250 F20110114_AACQGW bonnett_e_Page_024thm.jpg
6a16430ddc05b111daa8c3290f5a1a52
23e71a5ebac41da3e3bef527ebc59d4d2f55b636
2655 F20110114_AACPDU bonnett_e_Page_107.txt
344d17b443625e4828fb093019cd2fd6
7018541333563691ebec5beb1f13705d6ae4d4cf
38779 F20110114_AACPEI bonnett_e_Page_089.pro
14db1eab0ef70680c5d065b73006d539
69436f2c31768f4991d27575a7026bcdfea9e566
85003 F20110114_AACQIA bonnett_e_Page_040.QC.jpg
26c941e70a88cac08c84efeb35d47f3f
5b18d739a2c3c37a645fbae9c826f8c699c6d161
71887 F20110114_AACQHL bonnett_e_Page_032.QC.jpg
6e8eb725fcfa3147c7a52ab019390123
f672f5dd672af59fc50de7df1069be721d96d1d5
68691 F20110114_AACQGX bonnett_e_Page_024.QC.jpg
148a75c806e901ee0aac295ebbf14617
6f6433d2b34e858464378b460a4db88f344533f0
1773 F20110114_AACPDV bonnett_e_Page_090.txt
2b67ac04858fa77ad41dd400be078708
b2c082df593785bf1b8d5ebcf99d9ece87a082a9
190538 F20110114_AACPEJ bonnett_e_Page_067.jpg
0a477bf6b39748acd8e166f04e72e4e6
7145767d53e352aa0f08ee058f0d28866ab78698
47458 F20110114_AACQIB bonnett_e_Page_041thm.jpg
787ec77f5df5eef108f878553424843c
545befc683deff578485a47764b456ee2a631150
26166 F20110114_AACQHM bonnett_e_Page_033thm.jpg
4bd39a05c022953b860ad551b19551c2
a5eabdd2c3f5ec4584b62c8a9f30d73688f714ae
25086 F20110114_AACQGY bonnett_e_Page_025thm.jpg
06f82fd3c5a9060638b5bea20c37adff
18b65d36dc6300eaf0b1ecb5debcb5fdcc893fef
1613 F20110114_AACPDW bonnett_e_Page_134.txt
e99fbe4ac0b75e17ae297e27d0968f59
df8fd1a60143ec167662f032d81ece8e46834176
211267 F20110114_AACPEK bonnett_e_Page_072.jpg
799d8cd8a83749ff4306915e1459b5fb
3052e95f2e986d0524265023a7cb19f13d3ffb68
93030 F20110114_AACQIC bonnett_e_Page_041.QC.jpg
a33b84fa9a6ba41c0a1ef733034f2c3c
348c0009065bc3e996e2b5b5180a09138db6d286
78981 F20110114_AACQHN bonnett_e_Page_033.QC.jpg
478172851e91eeb6f19c7508c8878167
6d7c28d039fcec4a9fea108907d7ed767af9c47e
75708 F20110114_AACQGZ bonnett_e_Page_025.QC.jpg
da73662d4d02c236b42b3e6106a7eea1
fbf83d46fd76ece9058d96af7da8a358439bf500
F20110114_AACPDX bonnett_e_Page_142.jp2
b49ec8c259d50a3d8baae9ae51099243
7c9925f4f037927fe075ea3475704a51c0675b21
105812 F20110114_AACPFA bonnett_e_Page_108.jp2
8e69b1dfc2ad6b0e0db8e8b8e4415ed9
3b2995394dbbe43b63905d077901e75b501c4fab
207555 F20110114_AACPEL bonnett_e_Page_129.jpg
a7d92d84332ab082f68f9f7960c26061
5d86eea741059fc42ecdbaea883e75500defafa7
43318 F20110114_AACQHO bonnett_e_Page_034thm.jpg
5faf04ff447b84fe5ad34a9b1ac0abb2
1b9e315feaef6b127e2e9d009026fe11398902c1
107417 F20110114_AACPDY bonnett_e_Page_086.jp2
6f7cb2c32de2c20c26124b867e39175e
d9bf3b78df0c699426443022dbe0a5f06a5a4773
24102 F20110114_AACPEM bonnett_e_Page_072thm.jpg
30ea41740898510fca68d548104bc8aa
e6f6f5fa81dc261e3f676c87465aabaa7444e480
46212 F20110114_AACQID bonnett_e_Page_042thm.jpg
9717eb30f59ee8b2578203ba2f9e2950
1eb8b488483130b0e986d0eb67848a01c9bec252
79027 F20110114_AACQHP bonnett_e_Page_034.QC.jpg
782732f6f6974b71e197aa6ac163b014
63bf6f2755fb28240cf95208e5412216d1fc6dbf
110278 F20110114_AACPDZ bonnett_e_Page_016.jp2
34ad653b95276a30d611e05f09e570d7
1e709d04cac4ad6ed8178608532b5203c4ce706f
57954 F20110114_AACPFB bonnett_e_Page_059.pro
eb3a1a573479fa7dcd8328c6723fc34e
75c0d02a6fc4495146fe2aad58a4670323ddf13c
206093 F20110114_AACPEN bonnett_e_Page_044.jpg
93f87eedb1d742c344d0b84a46a3c70c
6b0f4bf3342694a05ef97d818e1243fcadaae1db
88210 F20110114_AACQIE bonnett_e_Page_042.QC.jpg
edfb0eabfe67d01209c85de0649e02d5
61572d9bf784f89a4804439a1466477bac60305b
24341 F20110114_AACQHQ bonnett_e_Page_035thm.jpg
77e4ae74f407da29dd9b4c177aafdd42
709d36242c5bd7b7dee98c28aa65ccd3bbb5e543
48512 F20110114_AACPFC bonnett_e_Page_100.pro
5b7e041f6fce45f2ee08ae9682fcdff8
c86fa07d250f73c5461eefc26f8d39a59b1aa756
F20110114_AACPEO bonnett_e_Page_076.tif
6079a239425f56f26c60895f80783d81
7d74b074e508b118fee6e0d989512a4ae4fc83e4
77932 F20110114_AACQIF bonnett_e_Page_043.QC.jpg
b52f4f1acf94d48137ecfdeab53b9df5
cf1a48c10b0f12a095b783ef79a857d4ddcc70e8
75671 F20110114_AACQHR bonnett_e_Page_035.QC.jpg
8600bd7c2941dd0cc71a0a9475ff1ded
ca02769e1a7a2ebe8776ad5579ebe328f096b300
207237 F20110114_AACPFD bonnett_e_Page_086.jpg
a2fb802d75d705ecdf0596123816e4d6
976dfdcf59c967d38e615c2461547ab9f2a34763
188487 F20110114_AACPEP bonnett_e_Page_128.jpg
f5497f7d8f68964f45894110a727e44d
6108025a76a695e7511331e73291e80848dd24d6
24060 F20110114_AACQIG bonnett_e_Page_044thm.jpg
faad95f1a1d97a7af1f43a6d2fa9b641
260efdd5b6f87f0bd186bad0ad61645ad36ce8ef
45910 F20110114_AACQHS bonnett_e_Page_036thm.jpg
b9a85d61ea15777ba2c348958a3f9d7f
78cbee87e58d20dd07ef339726cf7b2c3d876b1a
56030 F20110114_AACPFE bonnett_e_Page_060.pro
ea88d15178bc4ef5af786dac47ab850b
6ab684dc9bdc34c042e814abb8d970ccd66d95cf
24882 F20110114_AACPEQ bonnett_e_Page_043thm.jpg
42c7d6fca36b500c22e048f8efa7b52f
5a7655e56fa3f9b3cc2c57fe8e4851c313b96d19
76473 F20110114_AACQIH bonnett_e_Page_044.QC.jpg
bdb409f7bedfca899c512b5fa1962fb9
89b353d8e1d9ec7d69f38c0b3d865d479d19b18f
44308 F20110114_AACQHT bonnett_e_Page_037thm.jpg
c8c181a98e4e1dcdca81aa558f9188f0
efbd516feb76b15000243b3bb64d605d64b40051
F20110114_AACPFF bonnett_e_Page_138.tif
55d2fbf2fe490e3cacf3e0ad5fccd9ea
0713deb46be4f9f5dd1923cd4fec62f7ecee3d23
57937 F20110114_AACPER bonnett_e_Page_001.jpg
d0de8e9d3596645b014734b608199e15
ffacfefd720c1283bd208564224401a096598588
41947 F20110114_AACQII bonnett_e_Page_045thm.jpg
a54829bc7c7bf65c7b0579768d718e36
19be25420471cb7dff83d9daac8dbe9c0338e3d9
81090 F20110114_AACQHU bonnett_e_Page_037.QC.jpg
b39251d8d2d2761e164915d79cbbc102
8893d99a8f7181d95fe0a76bbab9aa1de1b0d56e
66180 F20110114_AACPFG bonnett_e_Page_004.QC.jpg
95b21563dcc80595a1d77b25955fee8f
a6d401ac8ea041b648e3e37cdddd64bd76e34ed4
76204 F20110114_AACPES bonnett_e_Page_119.QC.jpg
e36e09977dbd465e1db9a3da7857a793
5cefe8f4780b47ebdcd1316bb8caf45ec0379a02
70953 F20110114_AACQIJ bonnett_e_Page_045.QC.jpg
68580c13562e3f1f39bbba6361afc3ba
4477035ed206b47468d22754f1f54c7290cd6220
22781 F20110114_AACQHV bonnett_e_Page_038thm.jpg
4446ae8e7955319c82b7160b34fe3feb
812be1f2832afabe1a8eff6d8246e31795351bd7
28525 F20110114_AACPFH bonnett_e_Page_144.jp2
53f2d4a278a270fc2b2cf6979d82b05c
4502736f97e0d811387080e764df22d0587897f3
2178 F20110114_AACPET bonnett_e_Page_058.txt
d5dda26a46e631330eeb4274f508ec9c
2b4619f84707b22cf54231d90a6b09dfe3436102
24115 F20110114_AACQIK bonnett_e_Page_046thm.jpg
cb920cf6f03b6c682b2edb2fb77dcf63
239582503e1dc20b157d4e90dfe78dfa6c25c29e
73530 F20110114_AACQHW bonnett_e_Page_038.QC.jpg
f4268a22975f5cff3587ebcda3edb201
4b8a286f80fb57ce22d124a8cc0ceaccc795ac0e
727426 F20110114_AACPFI bonnett_e_Page_037.jp2
2852d1c5d399b1245e2a8903ea69bab5
c0aadb08433d4c3bcfd695112089fb7b7d492f65
200202 F20110114_AACPEU bonnett_e_Page_100.jpg
1bd56ecac78c230cfb207655273d176d
bb80d694b27cb19ec0ad5ab164994f77d0f75539
23773 F20110114_AACQJA bonnett_e_Page_055thm.jpg
2395acb60ac421765515dbeb1b07b7ab
504e0df26ed64c5a5c48161ee91485d65ca028aa
79012 F20110114_AACQIL bonnett_e_Page_046.QC.jpg
645809bfc0e85106c9eca92d078a75a4
6d9645e3da2ece164d3820c8346cf0b91fd63caa
24491 F20110114_AACQHX bonnett_e_Page_039thm.jpg
33325a01e7e53b7af5aaba423fc01098
bc3db7d930340e23297223d49754776a40d5ba5d
8313 F20110114_AACPFJ bonnett_e_Page_144thm.jpg
c33439a3512941486b881e5831eb5458
571c0bd2734fecdc35c9063be35a8ee03fc722f8
F20110114_AACPEV bonnett_e_Page_006.tif
8f7b07ae03e2d8ecf63707c4ee1b3511
57c00322e7e36595727bde460be685e8e230cdfa
74643 F20110114_AACQJB bonnett_e_Page_055.QC.jpg
f14caca67f879750d971e110271bae8d
11473a18526e1d0fe42a8783df916a1579580474
47796 F20110114_AACQIM bonnett_e_Page_047thm.jpg
75e290a699e7dec6c37ca50e2394648f
81787c7666a9f80b5485f4ef27030843a3da46d5
78612 F20110114_AACQHY bonnett_e_Page_039.QC.jpg
5256b74536676a5c6ef3ef2a01eec4a7
65b199dcf85d3bf68ff96af5ddc16fdd51c757f9
47103 F20110114_AACPFK bonnett_e_Page_117.pro
304ad71ff787ca8a7c79b65a6fa29ea3
3a6cc399701f90ddcc45dbfc8ed937fb84a7bf00
73561 F20110114_AACPEW bonnett_e_Page_088.QC.jpg
514aa5e3b24dbd4db61b31902fa89ba7
cff7b4920d3230ed98b087df6e1a81b1e88a9473
24807 F20110114_AACQJC bonnett_e_Page_056thm.jpg
5d327c4bdad0c978216b5eac0064c8ae
5047e561a19271b91ab71c1cbee45ad70f8985ff
96137 F20110114_AACQIN bonnett_e_Page_047.QC.jpg
7a179fdad9ea24d26f9d6ee47fbe0234
19045d2028bc2240ba428f6a36a85cb5f231b32c
45424 F20110114_AACQHZ bonnett_e_Page_040thm.jpg
175281fbfdb274dbfb9d058de662d29e
7733e7a203431624aeb3b2c62e02af339b18d2f6
1842 F20110114_AACPGA bonnett_e_Page_077.txt
6b1a70ac3920f12d3e385106709e1cea
3e48e3456b5050eed3defa9a3d0456364d3df289
76025 F20110114_AACPFL bonnett_e_Page_030.QC.jpg
6cdfcac337b484ab742924062693883b
664c4e3bc8da78ff7d413d1b43287227ac7df79d
78377 F20110114_AACPEX bonnett_e_Page_120.QC.jpg
1b8ceb5c5cd911bb4b94ad0e1deac689
76a3f8de2cc91e505b7e081cdff09efe4cc0861d
25175 F20110114_AACQJD bonnett_e_Page_057thm.jpg
04f2f3517a83546659550c8cae06df97
379c95074908d692fe2fcaf6691e806e220543f3
80136 F20110114_AACQIO bonnett_e_Page_048.QC.jpg
b0a936cb786826c680dcdd97e8ba4df4
ed664068dfb06cffc68998dfadcd784970d92d87
86360 F20110114_AACPGB bonnett_e_Page_093.jp2
42e4a9f90973cb17e2d7f78b50127f69
62adb964c1d6d71adc09f32663ce6480cb6c09ab
47695 F20110114_AACPFM bonnett_e_Page_137thm.jpg
d690aa01bb3ea29fb2d4875cb990d612
efd5e5a228a1b7c40adf9ac26d0e05804ff62e30
23702 F20110114_AACPEY bonnett_e_Page_081thm.jpg
adc30477dd900f3d2e26313815990d52
f6ac905b00180d34cf5f06940b51120791d993c1
46944 F20110114_AACQIP bonnett_e_Page_049thm.jpg
51adbdb07faf3e15a5ba49a479e5b9b0
c66b97390b4ffccc8469f9c8a1ad1bf4d3d13f73
57007 F20110114_AACPFN bonnett_e_Page_101.pro
b45d372ae59be24454f1c016850ac9b7
7fd8a26da9e992a12c568a4366d1c4e62a6f248b
197942 F20110114_AACPEZ bonnett_e_Page_036.jpg
8f3231b11ebcea5a819a8422662c3936
da07268e2c456f2b4265b75b28d5bc3d8a3b489c
81735 F20110114_AACQJE bonnett_e_Page_057.QC.jpg
99d7c31a7ec6b64817c91c9b0bbbacf4
360d64d8eb7d49b4626017e970ebad514de6b177
94187 F20110114_AACQIQ bonnett_e_Page_049.QC.jpg
525d35d7d0edd555895d81f1b291f470
a9ca3a5ce66b28420f56a6caf12f0a180aadcb81
209808 F20110114_AACPGC bonnett_e_Page_046.jpg
1c462a414ea8ad8faf6a893a25eac033
936180ec9f668ecc2d4a1300aea7b6a970ae9bd5
179006 F20110114_AACPFO bonnett_e_Page_032.jpg
19a399aaaeaa6d929b7e74e637cbd465
41273470cc585dea83c0d8f73c7fcbea0b4a1a7c
25922 F20110114_AACQJF bonnett_e_Page_058thm.jpg
77d70f9c2ad8d2c202783c29266b9f36
7b112abee1dff9ae3052a1c2bdbb68102aa21468
45337 F20110114_AACQIR bonnett_e_Page_050thm.jpg
5183f57c639334f08b7430a74404d769
3a5b8ff83254d16fd9658d9a03ed1c3d23b00787
104677 F20110114_AACPGD bonnett_e_Page_022.jp2
c4c990c2e7ac424941089e29bb110db5
c44eda8e4d679c05985134b89bca827d4ed0aa31
77867 F20110114_AACPFP bonnett_e_Page_097.QC.jpg
19e65a6d744dbedcc3f1489067ba204f
10e0a10d40dc1c534439b1675f94937d6e92ee85
81678 F20110114_AACQJG bonnett_e_Page_058.QC.jpg
4ae94c47beeede19bce5a912b2e9e81e
6eda2241ae12351762de3d8c3a1d9c6c2d607d4e
86645 F20110114_AACQIS bonnett_e_Page_050.QC.jpg
45b890668ab351efbf84dea45e167fd4
014e98de9e7421387b2ac8451e25c9fededcf9b8
99992 F20110114_AACPGE bonnett_e_Page_081.jp2
7ca361bbfeeca24d35d7f2e11161277b
891c8a5258c33c3cad4c35ca042ebc2ecc8f12a8
44304 F20110114_AACPFQ bonnett_e_Page_053.pro
6be86fd2b9cf307073b0df9c9c54265d
7feeef883318e85d5102e0eb693814ada09aba6a
25488 F20110114_AACQJH bonnett_e_Page_059thm.jpg
68a24f35ed25a054e2a4e98527652e1c
88107ea4509b785d72e29e1b44e4fd794060aa42
45815 F20110114_AACQIT bonnett_e_Page_051thm.jpg
a76d0b0f8f9377c553dec875ded2ecbb
fcc5f6873cc9c4efcb32c08dce5cd11268383dd9
105960 F20110114_AACPGF bonnett_e_Page_079.jp2
7fc2e41062dabb45c8d049403ff68a56
d0e9aac486dee21fbc1484a34af08a7f3848859a
F20110114_AACPFR bonnett_e_Page_026.tif
f1462352714ed3871c44c29ddadf7b72
a365cea45bbeda5d55a65786e4a2a6cfb239ee6a
81153 F20110114_AACQJI bonnett_e_Page_059.QC.jpg
2a61a554297f9fe904613691890f01c3
8d7a38d1a15221290f5cb5c31b0a5ad41a3c8314
87086 F20110114_AACQIU bonnett_e_Page_051.QC.jpg
947a7587d772f2c4317cd0cb4dbf96e2
16e45f71757e2a25579f12629ffeb1420415c9b9
2210 F20110114_AACPGG bonnett_e_Page_125.txt
43ba891cdb8e4bc5b3f10d95151660df
a77aa112fa04f32e00824d4908605bfa9d6b7b38
19048 F20110114_AACPFS bonnett_e_Page_135thm.jpg
ff5f08bfa9220279310c9057b012fb18
c2fdce4972524a3adea3027e6087f846cbf69842
25065 F20110114_AACQJJ bonnett_e_Page_060thm.jpg
f64a3c1777aadacdb6c1bb94ec5b3505
ed890d9a993ff85e99e46d482934d3bef8e636be
7757 F20110114_AACQIV bonnett_e_Page_052thm.jpg
c4c86e7f15895a2f62c92cfb91323a08
1009035cf347233af57a90b9bbe48b98447ab5ec
93016 F20110114_AACPGH bonnett_e_Page_036.QC.jpg
a870fc51732121c1de125f8afc1efae5
3501774e28dd6f0149af8ec8769fb5459cd307bf
203132 F20110114_AACPFT bonnett_e_Page_061.jpg
daede4c934fa714d350903334611f8b8
5e0ee3d606a407d6519a680a1d72e2fbd5ce401f
82071 F20110114_AACQJK bonnett_e_Page_060.QC.jpg
2aef474e6a7286be923e36d3a105a182
cd8f52e31cec617b8e70b80ff43e6475541b89c5
23982 F20110114_AACQIW bonnett_e_Page_052.QC.jpg
ad29fcc0bda175ef10fb08c60ce62cf1
d7bc3d063fbb8fe8bada0ae73e3b09052c17b098
1810 F20110114_AACPGI bonnett_e_Page_049.txt
0c9d0dae286b34acf19e7b288df17f8d
7ad68353f37fc3c6573f8517bb8daae22a75a697
48962 F20110114_AACPFU bonnett_e_Page_019.pro
fce6db33b909493235329113e1b3ae3f
1214e64826683241943e649304e05f7b4f407b29
22918 F20110114_AACQKA bonnett_e_Page_069thm.jpg
fb4bb033e67b7d56082d56df89ae413a
1261fcfc9b8941cbbbb59a9626aeee03c6aa99dc
25196 F20110114_AACQJL bonnett_e_Page_061thm.jpg
ce8f970850a1d00cf0e5d293830548a6
3e9e7bc3ab9d0b5f355f54a6e3bf5c9e8c174945
F20110114_AACQIX bonnett_e_Page_053thm.jpg
e6c0d7095d7a0e7d083e7f840c4f0cc8
bf6d0637243a2573fc459d9a8b95c7ad4383c121
205384 F20110114_AACPGJ bonnett_e_Page_020.jpg
1a2e8d00c3984cfbdcfd7385da4f0f66
e0236557862a3ec7cd1d50ab7dc2c5f0035e9d74
F20110114_AACPFV bonnett_e_Page_056.tif
88e5cab1290ce5af3533bbeb2bc8b267
f1a2f6a36400dd20b77ffdc9641a932284d9ec16
73357 F20110114_AACQKB bonnett_e_Page_069.QC.jpg
fb7a4e0354a4f2857f1bacebf0e07805
d0f72cc8efc3c136c3db22ded1ccad57cf481512
75954 F20110114_AACQJM bonnett_e_Page_061.QC.jpg
b999282b5d1753041ba33a611c587860
56ee790533a7cfb6fd5ae7fdd754bd1677ccedbd
69110 F20110114_AACQIY bonnett_e_Page_053.QC.jpg
14fd2f171de50b280afc93778bb3f1ff
f9784eab4431ed29dffc8221e817e50a85177366
52026 F20110114_AACPGK bonnett_e_Page_130.pro
690ee43521f78846d4dfc1620fa3747b
9387fbe5835037be2eaa87c107c8fd9b4edeb196
46557 F20110114_AACPFW bonnett_e_Page_077.pro
8a4094657fdce5fdde1047d85edaa4cc
3e41778a5d7bf1e234d3f954c4a9366bb0a3c7f5
24535 F20110114_AACQKC bonnett_e_Page_070thm.jpg
2941bb9fd0545e067594451bc549b0f3
c88d94f787863f76787ba05bb8a09374589c4227
22884 F20110114_AACQJN bonnett_e_Page_062thm.jpg
f810a4689c30bb68a2f9e3318bbf9e6f
6739592fc3c9384169e1704de93103a0e4f18927
23691 F20110114_AACQIZ bonnett_e_Page_054thm.jpg
591b099ca2ecaa087f62b21efa3668e2
bc5ce7e462ba266ff0fe72d95c8cb3d804a4d754
96146 F20110114_AACPGL bonnett_e_Page_053.jp2
1fb8ee1630125fbef2ce4727ead98e93
67f25eb07b1f3f81dbc6d39c4aaf3fa184adcecd
82079 F20110114_AACPFX bonnett_e_Page_010.jp2
249ddb90b4f7785cf927eb704e143899
781a619931963227100e9bbf718b6c9574e3eb0e
173930 F20110114_AACPHA bonnett_e_Page_004.jpg
68a08f0ab9fdcbbfd674c43ead1e06e9
cb8fefab2e86bf2f5e85e482a16fbcc054519769
76335 F20110114_AACQKD bonnett_e_Page_070.QC.jpg
7d68c175e679068d713a507cda5c32b8
6bd857ba9caef883c669e898fb74047437307669
74363 F20110114_AACQJO bonnett_e_Page_062.QC.jpg
04bb199df5b28af90f60e0ff9afb4d91
4c16858ca1fa6ed3024489f16b34d16a0318e11a
5397 F20110114_AACPGM bonnett_e_Page_002.jp2
2b0d6ec4abde55f17332ebf5ec66276b
c3b2bca19ed4ed54bdd2ce6f477c79da4f1b0415
108443 F20110114_AACPFY bonnett_e_Page_039.jp2
a44a35030ff207330265b73166e4051a
1f725241dcfd26ab78b7b511db717cd405d31cb9
48261 F20110114_AACPHB bonnett_e_Page_005.jpg
c6c1c604880c1608276e7ca0102173e7
3768d85bef8f4108d6161ff94935d5544c0474cd
24902 F20110114_AACQKE bonnett_e_Page_071thm.jpg
3dbd12a311d2ecc4e94e9c220ab26a5f
ba90d6e4c2a67d1573f7118d55e79f91db874014
24741 F20110114_AACQJP bonnett_e_Page_063thm.jpg
75daa198885b667f95f0ce44bb878dcf
2e43c20bc0ac7f387e74434017f4a4a74b3a3917
47307 F20110114_AACPGN bonnett_e_Page_078.pro
f4273aba9bebaa80b810d2d2e57d555e
359b0f1c8595d177c3c396d7337dc9c805dbe2dc
106250 F20110114_AACPFZ bonnett_e_Page_110.jp2
f1fbbbfc08a3ef5ca78caceea63cf6ab
e596c26a165e11c55bed24968a2bf63b543de931
225466 F20110114_AACPHC bonnett_e_Page_006.jpg
582fd98bf9c11414ebe0c5aa5e9a128b
0d60586b56aebbceb10da3cb8b367a979cdb2f02
83421 F20110114_AACQJQ bonnett_e_Page_063.QC.jpg
2fbff643059413403f352f50fda0dce4
5d7aae64fbd990ce4d33bba44504844fd1202820
24604 F20110114_AACPGO bonnett_e_Page_030thm.jpg
89465cfa64975b2f822f373af09618e8
e5c88ed423a633f1c4ad15bf184a9db436ceb9ed
80602 F20110114_AACQKF bonnett_e_Page_071.QC.jpg
becbf11a9c6024b7a0b12a2f9ec9bba4
0ecb4ddc1d580e1168555ea4e912734ed41c2ed1
25875 F20110114_AACQJR bonnett_e_Page_064thm.jpg
859896df81c859d69618c67c9ba3f2fe
0c7bf636d56e7c026c037db4e2aca18c1180a612
107775 F20110114_AACPGP bonnett_e_Page_073.jp2
aaf0fbf4ecab079401eb2e7bb9cd6e4f
2d58424fd04ba6d752f349429d61d0e48c419903
254893 F20110114_AACPHD bonnett_e_Page_007.jpg
9bf11aea42a641565fe18d2b56cdbb7d
38bb11fa892eaf4ad97a7e65ad9e874405c2309f
77901 F20110114_AACQKG bonnett_e_Page_072.QC.jpg
a394239a69530ff61da1d301466627dd
7f5e1515df41dda5344a85ccc2781bb7c6ee43e3
83629 F20110114_AACQJS bonnett_e_Page_064.QC.jpg
9b76257483fa1dd5fbb63024f8767d0b
7678f46bf20023dda38b8384c0972d66d62f22b7
84555 F20110114_AACPGQ bonnett_e_Page_006.QC.jpg
371ec444ea8f149064742ba34bbf58a4
0131368369ee644b8bb976cdeb862812546ed105
173679 F20110114_AACPHE bonnett_e_Page_008.jpg
23e342ce92f3da144fb1a70d42e2a87f
52a549dbba00b891156a53d4a7e97be069998855
24885 F20110114_AACQKH bonnett_e_Page_073thm.jpg
4e4f776e2460bc479171bd35bc92b7fe
a076d244161e332586dafebcae9317d777b54a9f
23985 F20110114_AACQJT bonnett_e_Page_065thm.jpg
99589088ff3757caf38e27deb43a25e9
b58c7f5d0eed8097b9ae67f12ce7f48e8d71e797
101542 F20110114_AACPGR bonnett_e_Page_137.QC.jpg
a404ee98f264d5b81cda1ba31125b2d0
0a17cad914bcdd45bac463e30e74c612245bc177
211006 F20110114_AACPHF bonnett_e_Page_009.jpg
06e06fb2f819377b28dcf946839b3a86
d026171978c9785ce51739401c38c9e9bb3742c1
76929 F20110114_AACQKI bonnett_e_Page_073.QC.jpg
7c6b6da773c0fdbe5561813b63e9c607
4b157cd7728d6ae6bed83946cf5bd714d1bb726e
73286 F20110114_AACQJU bonnett_e_Page_065.QC.jpg
b1538a3f5fc222dcdda0dd6a5d3766d6
8c071960a304b3333d34b4cee51351f6d45b2e68
F20110114_AACPGS bonnett_e_Page_021.tif
78648c84d274a53c8e8e459ebd06aab4
18dd61b51dce6fb3a676f8d87b5a7434cbb92862
167741 F20110114_AACPHG bonnett_e_Page_010.jpg
1cb4c21c3e5ed465f74a4fd6e53d8c5f
8182282c011f987facb325438facbaf81272ae18
24994 F20110114_AACQKJ bonnett_e_Page_074thm.jpg
1de03d95e46bf25bd42ca9d087e4e8fd
6691a4f3b01c335ffca60a0f69528a000139cfe6
4449 F20110114_AACQJV bonnett_e_Page_066thm.jpg
e18c4d822c58173a964ff5f99b0a3e65
c220d4bd0273786b19c99cb747d089480f35bfa8
72933 F20110114_AACPGT bonnett_e_Page_127.jp2
8c04d1960df44033e09652ff0163b865
d0840eaff298a2b0cdf028c5e9c397393e6d7775
126746 F20110114_AACPHH bonnett_e_Page_011.jpg
8e3bc93a94eb0ae9847e9663aa2f8a64
52be8bffa16ba8658536bbea1d0f52e480d722c0
78513 F20110114_AACQKK bonnett_e_Page_074.QC.jpg
8205ed93d7e200f8fddc2530a3039fd7
9c253ed12d9970e2416b7a3c19d6832466ec1d6a
10114 F20110114_AACQJW bonnett_e_Page_066.QC.jpg
67abc8c0445e3a08da116b6da22519ab
fd8501666c763b13c1683218bbbfd2e051316a37
297366 F20110114_AACPGU bonnett_e_Page_140.jpg
036f8c874e319b56abe7bd9ece1c2555
dcaf5c152e67a90f556fb9c563d7b011b051694e
191872 F20110114_AACPHI bonnett_e_Page_012.jpg
9293918484a1dcc78ae221e5c7cc9b99
e7c57574bc60c11a02c2f2113ca31e7f09fa5fd2
23778 F20110114_AACQLA bonnett_e_Page_084thm.jpg
1de2b748c910dabaab123a0ce560c219
1dc945f1b11e6a40ccee341e9380c3b6dbeadc74
26146 F20110114_AACQKL bonnett_e_Page_075thm.jpg
485f5528f06b50f7ba7aaa2464579638
3af7f9a8a20288df17c8c4113663c79a4db68ea1
22463 F20110114_AACQJX bonnett_e_Page_067thm.jpg
510ba93c68f203bb05f62f577bcf71b2
842a5a216abfb4ed6f60556b50c1c5942abff1de
216716 F20110114_AACPGV UFE0008325_00001.xml
c3f9d9c842c575f864cffcf29e25b048
6ff8f0f20ceec4fa835ea2acbcafb4a53febe790
202043 F20110114_AACPHJ bonnett_e_Page_013.jpg
9b10e3c472dee37d6dc4da8840d85aab
80b40862d99a243197ee34da67161e9392335e3e
74574 F20110114_AACQLB bonnett_e_Page_084.QC.jpg
d036d93c3d4bbfc58f21a162f9c98d6b
274f753406f79a17841b549586781041eb44edce
82161 F20110114_AACQKM bonnett_e_Page_075.QC.jpg
14d94c263cc6390cc3faa6b725f0892f
5bea8a967acfa88669212b022b639673bf62aa3e
70730 F20110114_AACQJY bonnett_e_Page_067.QC.jpg
354673a077ddefa230adf58c12de1106
42178470683784eb1d6ba44e73f8883dbedc5701
204449 F20110114_AACPHK bonnett_e_Page_014.jpg
7e0d9ba65bbbcf130f785f1ed36d2829
36220273adfa97d85095de90601d6437c565b7e7
25372 F20110114_AACQLC bonnett_e_Page_085thm.jpg
8b496486fcf24af53a2815d750a60000
2760765723af0951498daa6b8d18956ddad60324
24349 F20110114_AACQKN bonnett_e_Page_076thm.jpg
d0d665bf1b4dc16525cf752a7333d2f2
279aa4983db722e70130d75b0e4ccf27ae6e0647
71189 F20110114_AACQJZ bonnett_e_Page_068.QC.jpg
1de3357754c89fdc020aa25445ae75ac
a4c841f4a33842ae8ef25199f9fe26f77e9250b3
203386 F20110114_AACPIA bonnett_e_Page_035.jpg
96abdea36ac5eb03763f4950dceca2f1
4ee16ff48fa419b5322c3477ca883c094b260c8b
204210 F20110114_AACPHL bonnett_e_Page_015.jpg
d916aa3d659090f4fea20e90222bd541
614e536204027099494115c159905a2e0f8b8b9f
78305 F20110114_AACQLD bonnett_e_Page_085.QC.jpg
4dfc9469f8ec54467c4cb3a4b0138bed
2a98fb5e4ffa2a0e51b581c0b7700ba842a9487d
78535 F20110114_AACQKO bonnett_e_Page_076.QC.jpg
6d8287fe9e8646d52d17541a4a139125
bb61223ad6cbf95738ac5f54cdbff0c5ebc75f60
14576 F20110114_AACPGY bonnett_e_Page_002.jpg
89f84cb4551ec53560f4c52e84d025d2
6e1bfb1b2926bf98a27fa2ed77a3c3aa0cd49a47
171016 F20110114_AACPIB bonnett_e_Page_037.jpg
85dc04588fcc36514f5374c4a75e236d
a8f031db8c74cd8d7a5ab188ef2c3ac9307a7bc2
214883 F20110114_AACPHM bonnett_e_Page_016.jpg
ac3a4d5591e0857ccada7c6130dfd5bc
ecf9b1acdc9f8719969dbe890f1d757c9a185f43
77251 F20110114_AACQLE bonnett_e_Page_086.QC.jpg
2c1b59eba50a4a5d006bc0f4ad6aa24d
db3daae200e6c850a220381ece94497c913b7fe3
22702 F20110114_AACQKP bonnett_e_Page_077thm.jpg
e0654074a949fce9bb93dda7f2273d6d
6e099d662e06dd15a0c43f222ad10182dd4ee308
16530 F20110114_AACPGZ bonnett_e_Page_003.jpg
f4025ae8b49392141b9be47b8582b7f8
af5fbb853d5034ee3a0a521c52aef4746b4ec98e
198454 F20110114_AACPIC bonnett_e_Page_038.jpg
33fb3e0abef27fe0365dca231f5c9550
daa7fff5bce33fba775c973bc23ede3072232d87
207050 F20110114_AACPHN bonnett_e_Page_017.jpg
734d3db99c45563538124bda40761924
0c198941dffeb521dcdcd0763312deed8691cea9
23333 F20110114_AACQLF bonnett_e_Page_087thm.jpg
a3830a40e860a85b570f3f40493cbd0c
30689b322e3161aef9bea1d635fc1e6d328c5482
72175 F20110114_AACQKQ bonnett_e_Page_077.QC.jpg
abe92007f4d64346ac9c83e28b2600ee
5f33c61884001118d14e8ff95b07a9aa82122684
209481 F20110114_AACPID bonnett_e_Page_039.jpg
e01aab56dd52e46de4e5795e89469df9
80a2799e6de118c5c18f0ca7472f70d8f7780ba2
200455 F20110114_AACPHO bonnett_e_Page_018.jpg
97467117bf276bae5a587b190da80f19
4ea236dd98078ed01773bc121a2acd5d6c4466fd
21934 F20110114_AACQKR bonnett_e_Page_078thm.jpg
131f264b88c1ef03fbb68745bc22e940
67e7598ea030f7a6e12a9596b9d4e9379ca43a37
200675 F20110114_AACPHP bonnett_e_Page_019.jpg
dcb83fe1d279cc8d7bc6cbd167b618a3
d0319ac5a80a5a7dc805d5b86468f357d2071961
75445 F20110114_AACQLG bonnett_e_Page_087.QC.jpg
6376b5dd6e08b9cbf8cdef6f0bd9050c
ce0dfbfc9ed251d637d8fc252746aeb9aeeb8b8a
69216 F20110114_AACQKS bonnett_e_Page_078.QC.jpg
6431833801659ba4ff6ea9a6e37667d4
7459c987035d7c294c69b753a9689f5cd83de9b3
190029 F20110114_AACPIE bonnett_e_Page_040.jpg
ed6d94a36abf958a5718708497c21961
614372c804a07688af265191a19ca2f0a4e0bba0
187307 F20110114_AACPHQ bonnett_e_Page_021.jpg
a67bab673dd826933202f460bd14567f
627a2bdb3cde6bb76d07dae48b23dce613fb3f3b
23538 F20110114_AACQLH bonnett_e_Page_088thm.jpg
1e4a1134a08b637202873b456f6f9127
e37e602964087faa9cfa800c8e6e64a7c614f78d
23226 F20110114_AACQKT bonnett_e_Page_079thm.jpg
6949482a68d2ce754ece7650607bf4ff
a0d983f0410beeffd15576d488e4e44a17647c9b
207948 F20110114_AACPIF bonnett_e_Page_041.jpg
d679e962887a522d728590449d5bba1c
842c47017bd5435ea0e3392125d8a79bbf59d68c
205537 F20110114_AACPHR bonnett_e_Page_022.jpg
cf3db4471b0ca8c53a52585d2838c392
4105ed8072866a206adbfc95507f0bb266ba1f97
19807 F20110114_AACQLI bonnett_e_Page_089thm.jpg
97a7d15b4364ea93aa4f5f430f1d9f47
ad4036052f752b94986439327161435461fbf243
71131 F20110114_AACQKU bonnett_e_Page_079.QC.jpg
251b801f76076e883f4f61e447b00bb4
977c73e06da2d5515a4d6ff1f726ba527f1add20
193884 F20110114_AACPIG bonnett_e_Page_042.jpg
a69c0a062534ea336fdbc58acf59d1ec
575f732584b12e7a19a5432d133867d1eca25143
179505 F20110114_AACPHS bonnett_e_Page_024.jpg
383812215f69ce982cc01037971b3db1
17a89fa30879e469a7d1c58158d328f84b2cb931
61255 F20110114_AACQLJ bonnett_e_Page_089.QC.jpg
a037bb6461c8f7b7288f2e164b2844b9
2dc6b2ee1e0f4e011dd4da29b77c238e8c7cc081
23763 F20110114_AACQKV bonnett_e_Page_080thm.jpg
152073738f5cacb5fcf326067c2c26c0
7dd23af93fbd1f2212df931275bb8ac1309556bc
209976 F20110114_AACPIH bonnett_e_Page_043.jpg
2697b6a0c8e84c5eb0873958ae3ef981
a972f858bf768aa6d52902f02fdf2abcad941c32
202482 F20110114_AACPHT bonnett_e_Page_025.jpg
7481fd21ef8ef377a032df2db4f470ca
87fb230df2c98801490fc202e5ca06fac32ffd7d
46932 F20110114_AACQLK bonnett_e_Page_090thm.jpg
8e07d9cb512f93813f38ee89fd88c17a
2fec860943e734328b7ddeaf7d2ae32df5b0a2c5
78594 F20110114_AACQKW bonnett_e_Page_080.QC.jpg
7cb61abf43cec2feb8e10ef7d06488a3
61536202ef24f81ee66a3fdac36173f27ec97cdb
146284 F20110114_AACPII bonnett_e_Page_045.jpg
57d27e2b7e9d080380be1c182568e3f3
806b34f36e65146346e43280a4bd86754a65a7ff
235511 F20110114_AACPHU bonnett_e_Page_026.jpg
e0d3ecae40be614f1cffdfcaad3b9db0
8342294d518954bc5ab5e23ef9b022f39d934ec0
75456 F20110114_AACQMA bonnett_e_Page_098.QC.jpg
36ce9b2efd854b033b6ae09e1b80258e
dabe61bf49f8ab569d697b7802ea5d6d938bfdc6
90354 F20110114_AACQLL bonnett_e_Page_090.QC.jpg
669eb777494c2397dd1661a70e6de066
4409d53b42af873c0e19bc90f6d88830cc6a72cf
24147 F20110114_AACQKX bonnett_e_Page_082thm.jpg
813a7a8de1ac0f6dd177be77208eedde
77c85ab670aa5d0729517dd66f3cc354fd3f668d
206933 F20110114_AACPIJ bonnett_e_Page_047.jpg
a1e70a390bfd239c6132444f3c062062
560cb7d0814595a2580fd015a2275b621e16538d
209964 F20110114_AACPHV bonnett_e_Page_027.jpg
326e892b76ce8e1cf931f1bb745578d7
43fe43b7c7e055e734e7507f73c0287fef545003
23556 F20110114_AACQMB bonnett_e_Page_099thm.jpg
fc5242df7ef3ea68afc2382e6fc5bf7d
2f35cb90c52bc1c38ebef3cd5dbb3861ad70f4ea
25672 F20110114_AACQLM bonnett_e_Page_091thm.jpg
6e00c24dee637a624f7f37fc776a1da5
18ed35ba218e5ce7e3867b7411dd4618272282e8
F20110114_AACQKY bonnett_e_Page_083thm.jpg
cc9cf06ee1b2509a6887c7412ca99bcc
f36a7e11f1c1258fa8ea81ec3ab6f8acd8687113
176586 F20110114_AACPIK bonnett_e_Page_048.jpg
19de29695604f780fb1d9860f5d08935
5e8bbeb104799a1200d158197f84f4ef197bee60
223534 F20110114_AACPHW bonnett_e_Page_028.jpg
c047875071c923744e68ac0839edc1fc
669e75f5f3e18ee62f49bd3bf416bba313adaa49
76303 F20110114_AACQMC bonnett_e_Page_099.QC.jpg
f100fdf85730918a816980c0ba8c3ef2
32440e83ae4b165121fa8a44b7d39ef0c738a844
89273 F20110114_AACQLN bonnett_e_Page_091.QC.jpg
1579b70c479deee5cfb810c90452a23e
fe5fe4e380200925ebe016fa1164b8877d6f67ea
75557 F20110114_AACQKZ bonnett_e_Page_083.QC.jpg
986f6b76b6a391ef557be9640d3c4de7
538e5866c7415f453d3c0d15095ae9775ac00bbb
200745 F20110114_AACPJA bonnett_e_Page_065.jpg
c4004d4d1e7daeb3cf1e6997a1e1c6be
5c1a4c03d28f1fdc605450bba460312d0cc175d0
203532 F20110114_AACPIL bonnett_e_Page_049.jpg
f6056ca45064e89f75b5dba05f30ca63
9106959425c52f5dfb10b5ed4fc060e6aef31dd0
201483 F20110114_AACPHX bonnett_e_Page_030.jpg
1f9998f66a32d0b355d0c7299279803c
51727b0e8a6994c38687867ab8b827e17a569932
24205 F20110114_AACQMD bonnett_e_Page_100thm.jpg
92acc765ba4e5ee39177c4dd949ffc83
9112c105665464094b33c0f307d1dba6a1d0332f
21793 F20110114_AACQLO bonnett_e_Page_092thm.jpg
82c98ab1845ade95623ac6cdd85a72f3
724d46bf696b6bdb33e54cd0ae0275f81ac5bdd9
23722 F20110114_AACPJB bonnett_e_Page_066.jpg
b745de7414a4df1e22dbce1267f44a6e
bf815a316b5747f24fa693c770f11ae56ad38b42
185699 F20110114_AACPIM bonnett_e_Page_050.jpg
70f9b34c8ffb9419e8eca4846a6ca80b
ba24e413246f4a872e2034514c576aacd9b3768e
207677 F20110114_AACPHY bonnett_e_Page_031.jpg
d66ed8100c683215668b71b135fd9bb8
f1910df9e8a7031eff206d4285272af30f2e9296
73441 F20110114_AACQME bonnett_e_Page_100.QC.jpg
d2741ebf66cc1bc07c211b332746f3ef
021c8049aec86acbdeec6a888ce64e16f8f30605
64970 F20110114_AACQLP bonnett_e_Page_092.QC.jpg
a818b5f5812b2cd94cd924b904b3495a
d46b2647e321255cb9fc559ebd2256992bfa8f41
193088 F20110114_AACPJC bonnett_e_Page_068.jpg
719c79ac1fe0d3b0c40aceacd5198eda
8f02d505d4c8b768326b5a5ca43c31acf844e2c4
196026 F20110114_AACPIN bonnett_e_Page_051.jpg
bfd98eacd4b1559a266e5785eab66391
5270c7e2ae098413bc9793c3f541f2fa4fac584a
169248 F20110114_AACPHZ bonnett_e_Page_034.jpg
68d288c5f2e61b32827f38aee5520da0
5b1c44eb532e895fadb208eb621c6f8d4583da18
22712 F20110114_AACQMF bonnett_e_Page_101thm.jpg
fec1dd1fcad7952b540ae2454d735d49
134bc22dc99972cb4f84e9adde17cbcb44c8690c
21864 F20110114_AACQLQ bonnett_e_Page_093thm.jpg
da8c7fc8ee2fe4208336e45ee130e36b
54a34b5f1152040da515a0df7290a42f56862c22
58826 F20110114_AACPIO bonnett_e_Page_052.jpg
d1edb94aa251c4745a09f33d30898e60
2e860f860d47e534b41eef0eb18ba7b17aae89a8
197650 F20110114_AACPJD bonnett_e_Page_069.jpg
e5cbb9d18c26cd8f55f11f7530882967
28dd3b223a51d7a77d0ff55e49bde6e2d57f67c7
24892 F20110114_AACQMG bonnett_e_Page_102thm.jpg
b3fc85b5711c3103dc5c4f9009d1e1f2
29c268e5528db665a2cf1ddb6d365128b412597e
56790 F20110114_AACQLR bonnett_e_Page_093.QC.jpg
7fe349f6042f8165a9675879d6dac7bf
1b04b62a0e56de1f2962a6e0cf2134ac8b862d0b
186127 F20110114_AACPIP bonnett_e_Page_053.jpg
f6f9fbc00a9c9c1199dac1c101f41953
e9dc6dc02d30f1567fcc05db564132b34545bab4
206090 F20110114_AACPJE bonnett_e_Page_070.jpg
cd222d437f78bc2c79bd90c43b55411a
8ffa3c7202919fec3fc4b4cd0b6f44e8b546972e
23101 F20110114_AACQLS bonnett_e_Page_094thm.jpg
6303288b434b8d41b80a0ed41ac04144
a41dbd4505ddee0a034278af2273c6884442036b
207684 F20110114_AACPIQ bonnett_e_Page_054.jpg
58e5ecfd53f8f46fb5067749ca06221c
556edfe79b6fb7273102be31a74895220dac13b1
75312 F20110114_AACQMH bonnett_e_Page_102.QC.jpg
c58deb6b389a049e7e28f189ae9adec5
e800438671160658464d25e8e6467e5c08ac244d
72287 F20110114_AACQLT bonnett_e_Page_094.QC.jpg
bd9f75705e75db3b0e18b2bd0abf39d2
7319801873aadbc2a009d88c45d2a9c2041a04ce
205240 F20110114_AACPIR bonnett_e_Page_055.jpg
30e3e9c579f79e4a85edbf80f4f8a43c
cbda5078facd3ed9f11f7de6b0eaea5c36b054a3
217271 F20110114_AACPJF bonnett_e_Page_071.jpg
00cdac6c52bfb3dcaa473c3a6002662f
70c0ff97a137704fe7c31fe9cf41a382c98fe04d
24995 F20110114_AACQMI bonnett_e_Page_103thm.jpg
acfd98a33c8fa948d90486fd248f892f
6d2c1071e5acfa03b3511b575b43ec653820c9a1
24680 F20110114_AACQLU bonnett_e_Page_095thm.jpg
7c378ee4a9e5ad4313cecdc4634efb21
6a7ad5da45f250b6d88276b1117eed8eba37a885
212190 F20110114_AACPIS bonnett_e_Page_056.jpg
23b522ffceb65f6a997253917a68283f
0c513d9dac3af13a53bf1eb710d014ce99fe3ff3
206685 F20110114_AACPJG bonnett_e_Page_073.jpg
102d2c29acf812e6e5dceb7f01d94559
6c23286e09fc363d5272f683e9f86a9e8068cd80
78435 F20110114_AACQMJ bonnett_e_Page_103.QC.jpg
f119f934dbeb4e746655707b79a77d9a
7846c0c2f176162863b6d64431cd6041d457db0f
77581 F20110114_AACQLV bonnett_e_Page_095.QC.jpg
6a1757c025f140b3dfd9692ea1a55a55
d577ee3093d04482d85bf8d7ff160e1fd8585ddc
223736 F20110114_AACPIT bonnett_e_Page_057.jpg
944933532d63dcf17cbb970c0076b804
0191b2fc08d3d9de2df6bffeac2b351b4164ad41
212112 F20110114_AACPJH bonnett_e_Page_074.jpg
ae9a0741c8ee3febb002bb005431a5bf
6131fa9c01dbf2219072f316f24372ae30ad1f67
25624 F20110114_AACQMK bonnett_e_Page_104thm.jpg
8bc18747f76a65270330f01c1a7feae2
75c150c67dba7514bd28ba0b16bff0c54259a6e0
22596 F20110114_AACQLW bonnett_e_Page_096thm.jpg
27faaa1f413608cf97672a7f0cade08a
ae6fe0f91c26f3686e355ab3eea1f9ec6d95a394
228585 F20110114_AACPIU bonnett_e_Page_058.jpg
d8dddc79b154a06d8e3ccb6899a4f033
aedd8a8bec49df0cc4f427cb3b3ef2eed207e598
205386 F20110114_AACPJI bonnett_e_Page_076.jpg
cb2dcbc57b47431295a7eccbf132f816
7376d5c790022442f6cdebbd12cab74bceea3636
75110 F20110114_AACQNA bonnett_e_Page_112.QC.jpg
cbb82786119a10dfa38d7f5cfcb5dbf8
0fa92189a8b022bd21a7c496a6f2ffff00c2a34a
79245 F20110114_AACQML bonnett_e_Page_104.QC.jpg
7f83601db0c55959681897135917368b
a70cd2647303eda065b589b46bff2925cac8d587
65446 F20110114_AACQLX bonnett_e_Page_096.QC.jpg
1944276ecb68ca9a228ca973eff6fc5e
20e93ad14c89877d4574feaeef51be352fd68335
237003 F20110114_AACPIV bonnett_e_Page_059.jpg
f63d8d9f668591359eacd91c352648fc
7536af0d64de2be6535aa49d86b625f02d3ef198
195730 F20110114_AACPJJ bonnett_e_Page_077.jpg
d99b8a4373a8a67783b1e8521066abb7
dd807d246ecb9f8d131bea6d3aff5c9887df46f2
23237 F20110114_AACQNB bonnett_e_Page_113thm.jpg
fd518a1cd4858aa4f4d6c6ec696d83f0
9217ef4e69d0079ca9ee01b555172308c0b3ac00
25103 F20110114_AACQMM bonnett_e_Page_105thm.jpg
d41b3c0f8894908c5b178f30555721b7
90e0fa9b0e3be67b6e63b50e7b998bd5115543d7
24261 F20110114_AACQLY bonnett_e_Page_097thm.jpg
9808bfe0da08f969cf335fd5d4b78c07
32e43226068b0a1f4f7eb565844dac853e152298
226727 F20110114_AACPIW bonnett_e_Page_060.jpg
3f554d6500f5926ae140fb46fe4e67d8
d6678a55b9380d1d6abdba5927cf0b667b063617
186321 F20110114_AACPJK bonnett_e_Page_078.jpg
7d117a38f83d7db8519c00dbca2da110
1aae82bfeb0107a6357cc46687cbfde4765f7d8e
74240 F20110114_AACQNC bonnett_e_Page_113.QC.jpg
b3c1fb14e4e61363fcc4dc5646bb3f5b
2aa5e04330462ba690610924ca29d1506405fcc5
77787 F20110114_AACQMN bonnett_e_Page_105.QC.jpg
eabaa2a993e423819479e66b1919d475
525556fab8c18b19aa84ba97971bb54ba5fc6144
23894 F20110114_AACQLZ bonnett_e_Page_098thm.jpg
a4b63e534f0e193f149d81fc7e6ff79f
fef82896f5f79827375da72b5765705fe9d6ded0
199555 F20110114_AACPIX bonnett_e_Page_062.jpg
f202dc97b9326378338d047e77e14002
58a3aa33c6c69cc441d757046c715a4f3d8fb2bf
212692 F20110114_AACPKA bonnett_e_Page_097.jpg
a2fdb5e2a6b9ae5ce874cd09188136a9
eb4aac8d2fe9e19d5ce2fb35c29bb77f4c9aff12
204350 F20110114_AACPJL bonnett_e_Page_079.jpg
1bd7668e4b9fe1281897bb4688e1770f
2f8409f5485d0556d6df9525822cb37536f77116
24549 F20110114_AACQND bonnett_e_Page_114thm.jpg
d0431fcbba52688d2efd8c9017d9c00f
6567a0b92839227cc8df4bd187cd04e8e5e238bd
23692 F20110114_AACQMO bonnett_e_Page_106thm.jpg
64880d73b545934ebeb317ebe5535017
647cb6195f82e24605a1413ed54b6401ef7fba6e
224712 F20110114_AACPIY bonnett_e_Page_063.jpg
f66906e89f65ca0f1776140f1340bf2b
4ae6a56d65791efb1ef256718f9fc2893fe9e1c1
205809 F20110114_AACPKB bonnett_e_Page_098.jpg
7e0d0dcbcfa9efaa1496d32e889a2262
1b8fd5ec69d0a56d9efef90cb60131952b576ff3
210187 F20110114_AACPJM bonnett_e_Page_080.jpg
ed2978f039f87fec713db33069d7a479
6c5e51a5233df6ec375888b4345d110e9ab12335
79675 F20110114_AACQNE bonnett_e_Page_114.QC.jpg
25a6dc060e668bfe89c08f1bed5652ed
28508c0f5dfeef34d4b95b06b58f4e0f54016345
73010 F20110114_AACQMP bonnett_e_Page_106.QC.jpg
a01eebe1840d811e556e221752d5be66
8672f75cf76bdcb414ee4e3715b3ca054787fd1b
233682 F20110114_AACPIZ bonnett_e_Page_064.jpg
87d2337b72a7b836918ea464026199dd
1a32834924dd0c8e280fc5740cf803e801c5046e
202336 F20110114_AACPKC bonnett_e_Page_099.jpg
39d4d6ef2ec5549cc404dbf0a10fc6c2
de0189e18707e3958110220af70e3bf97e60d8f3
194974 F20110114_AACPJN bonnett_e_Page_081.jpg
6acac8e62ec9300d7110aeaee1424dbb
0fc9859e6257986417577865667463b40c208a15
25267 F20110114_AACQNF bonnett_e_Page_115thm.jpg
ad65ac3b48ea3a387b6cf189d58d4a31
939b3a5818e53ea03435669851602a4bdad25c93
22124 F20110114_AACQMQ bonnett_e_Page_107thm.jpg
e602e5939818a183a3d0c5cee52a1736
53e7db4a50da49b336a6b6efb8b0f2fdf0928a92
207240 F20110114_AACPKD bonnett_e_Page_101.jpg
6040a7f1c07f5013aa4f37c5d3fe4625
0d78656869cd48eda2441b2bf3c9869e1a01fede
208255 F20110114_AACPJO bonnett_e_Page_082.jpg
90be3a7184c9b928c0c1ddeee6087aec
0c8ed83fa9228594c07d0b8f66ac444daf5ea290
78309 F20110114_AACQNG bonnett_e_Page_115.QC.jpg
36c182d3a5a56bed206f33f966fdfa6e
78beaa172cbfccd35825f5a2c7b4030b162781e2
69135 F20110114_AACQMR bonnett_e_Page_107.QC.jpg
f94a8699e08eb6bf1467139ae694104b
2a277f489e164bb5b5ff7be04f1a653384d6727f
206951 F20110114_AACPKE bonnett_e_Page_102.jpg
e50413793555f61519b22a526ed996c5
b9ca170654cc34bd34d57f368e3bc47e2bbdeb76
206874 F20110114_AACPJP bonnett_e_Page_083.jpg
1523383c63dbf9c63a835264a4da6a71
e61bf48ad9011ca046a02c097db0ab3157c5be02
25351 F20110114_AACQNH bonnett_e_Page_116thm.jpg
e100d6319cd046769dbd018d3f278751
734b9d902679b0bc4d29d93b7364fa84ac2df486
24247 F20110114_AACQMS bonnett_e_Page_108thm.jpg
a364304d8adf6eeff389f7a3be06b05b
df00bd9e5fc98dcf24138c38aedc051de72ee121
208763 F20110114_AACPKF bonnett_e_Page_103.jpg
79f849d9672aecf82db9a3838831d15e
6d279d31fb455e15d445077c04f2cc054ba2c239
207185 F20110114_AACPJQ bonnett_e_Page_084.jpg
989ce9956d04940840ee75a318b10d51
fc90e3a874013f5c51d09cc48750e2f45b7d1d28
78648 F20110114_AACQMT bonnett_e_Page_108.QC.jpg
397e8b0b4051a1f1b950fe7074418d13
c80144bc583d82fa8b8a45d6e9379e81898a51ce
209939 F20110114_AACPJR bonnett_e_Page_085.jpg
3f0704067aff80c4d483296062f336a4
5b97ac24d338e2ffdb74d225474dec15c031cef1
74399 F20110114_AACQNI bonnett_e_Page_116.QC.jpg
e40c5be43e1e946e6a90b1ff91ecef71
8f15babbfac1bf7fdb1ea23090a6cf3a22b6a34e
24004 F20110114_AACQMU bonnett_e_Page_109thm.jpg
2cdb74e47a8287fc01e2cd6936d3b798
f30c9e43d5271cac3562af0344bc4708d0451232
210386 F20110114_AACPKG bonnett_e_Page_104.jpg
92d992ff24cf49bd455582be24e9e020
322793a031f7f2db7794e6eb4b09f67d0699e19b
164457 F20110114_AACPJS bonnett_e_Page_089.jpg
c9940337a5f2289e663836f8ddd9f201
8e2c9c645617cf0883651f89b1f99410f6fffb5e


Permanent Link: http://ufdc.ufl.edu/UFE0008325/00001

Material Information

Title: Foreign Direct Investment and the Investment Climate of Developing Countries in the Western Hemisphere
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0008325:00001

Permanent Link: http://ufdc.ufl.edu/UFE0008325/00001

Material Information

Title: Foreign Direct Investment and the Investment Climate of Developing Countries in the Western Hemisphere
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0008325:00001


This item has the following downloads:


Full Text












FOREIGN DIRECT INVESTMENT AND THE INVESTMENT CLIMATE
OF DEVELOPING COUNTRIES IN THE WESTERN HEMISPHERE















By

ERIC T. BONNETT


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2004


































Copyright 2004

by

Eric T. Bonnett
































This dissertation is dedicated to my lovely wife, Rita.















ACKNOWLEDGMENTS

I would first like to thank the members of my supervisory committee for making

my time as a graduate student at the University of Florida as enjoyable as it has been.

During my years in the Food and Resource Economics Department I became distinctly

aware of how well respected Drs. Timothy Taylor, Gary Fairchild, P.J. van Blokland and

Roy Crum are among those in the academic world. More recently, I have also come to

realize the level of respect those outside of academia possess for these men. Their

contributions to this project are greatly appreciated and will never be forgotten. Also I

would be remiss if I did not mention my appreciation for the hours Dr. Ronald Ward

spent assisting me with this research project.

My parents deserve special recognition for instilling in me the confidence to pursue

such an ambitious goal as writing a doctoral dissertation. While they have always

encouraged me to aggressively pursue my goals, they also taught me the importance of

being able to maintain a sound conscience. I will forever be in debt to my mother and

father for passing these most admirable personal traits along to me.

As I find myself at the terminal point of this project, I am reminded of a very

special individual who was taken from this world in a most untimely manner. If it were

not for Dr. Patrick Byrne, I most likely would have never attended graduate school. He

was a source of inspiration for many students during his tenure at the University of

Florida, and his legacy lives on with each of them.









Finally, I would like to thank my wife, Rita, to whom this dissertation is dedicated.

Just as she is essential to my happiness, her unremitting encouragement and support were

essential to the successful completion of this project. I shall always treasure her

commitment to our relationship, and I will forever carry the knowledge that I am a better

person because of her.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES .................................................... ....... .. .............. viii

LIST OF FIGURES ......... ......................... ...... ........ ............ ix

A B STR A C T ................................................. ..................................... .. x

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

Impacts of FDI on Developing Host Countries..........................................................3
P problem atic Situation ............ .............................................................. ........ .. .. ....
Problem Statem ent .................. ......................................... ................ .8
O b j e c tiv e s ........................................................................................................1 0
S c o p e ..............................................................................1 1

2 H ISTORICAL CON TEX T ................................................... ........................ 13

A Taxonom y of Capital Flow s .......................... ..................... ................... 15
A Method of Characterizing Different Types of Capital Flows.............................18
The Structure and Evolution of Capital Flows from 1970 to 2001 ............................22
The Andean Community ...................... ............................... 24
The Central American Common Market..........................................................28
The Caribbean Community and Common Market............................................32
M ER CO SU R .......... .. ........................... ............ .. ........ .... 35
M exico and Chile ........................................ .. .. .... ........ ..... .... 37
Sum m ary of Trends .......................... .............. ................. .... ....... 40

3 THEORETICAL FOUNDATIONS ........................................ ....................... 42

Neoclassical Economic Theory vs. Strategic Management Theory .........................44
D running's O LI Fram ew ork ............................................... ............................. 48









4 EM PIRICAL CON SIDERATIONS ........................................ ....................... 56

The MNC's Motivations for Making Direct Investments .......................................56
Decisions on the M ode of Organization............... .............................................60
Decisions on Location ......................................... ..........................63
The Em pirical M odel .................. ..................................... ...... .. .............. 68
The Host-Country Investm ent Climate .................................... ............... 69
The Policy and Regulatory Environm ent .................................... .................70
P political risk......................... ......... .................................. 73
Factors that affect the return on investment.............................................. 75
C control V ariab les.......... ...................................................................... ........ .. ....... .. 7 8

5 MODEL SPECIFICATION, ESTIMATION, AND EMPIRICAL ANALYSIS .......79

D ata and Sources ...................................................... 79
M odel Specification and Estim ation....................................... ......... ............... 81
R results of E stim action .......... .... .. ........................ .......... ...... .. ....... .... 85
M ERCOSUR ............ ................................ ....................90
The Andean Community .................................. .....................................96
C A R IC O M ......................... ...................... ................... ......... 101
The Central American Common Market........... .................................107
M exico and Chile ...... ........... .................... ......... ................ 113

6 SUMMARY AND CONCLUSIONS..... ........... ......... ...............117

The Least Responsive Countries ................................ .....................117
The Most Responsive Countries................................ ....................119
Policy-R elated C onclusions............................................. ............................. 120
Considerations for Future Research...... ................. ...............122

APPENDIX DEFINITIONS OF SELECTED VARIABLES .......................................124

L IST O F R E FE R E N C E S ......... .. ............. .............................................................. 126

BIOGRAPHICAL SKETCH .............. ............ ............... 133
















LIST OF TABLES

Table pge

4-1 Survey Results on the Importance of Investment Climate Factors ........................70

5-1 Descriptive Statistics by Country (1984-2001)............... .................... ..................80

5-2 Random Effects Regression Results ............................................. ............... 86

5-3 Investment Climate Variables MERCOSUR ............. ..... ..................90

5-4 FDI Responsiveness MERCOSUR ...... ................ ...............90

5-5 Investment Climate Variables Andean Community......................................96

5-6 FDI Responsiveness Andean Community............... ..........................................96

5-7 FDI Responsiveness CARICOM ................................................ 102

5-8 Investment Climate Variables CARICOM ............................... ............... .103

5-9 Investment Climate Variables CACM ..................................... .................108

5-10 FD I Responsiveness CA CM ........................................ .......................... 108

5-11 Investment Climate Variables Mexico and Chile ......................... ............113

5-12 FDI Responsiveness Mexico and Chile ....... ....................................114
















LIST OF FIGURES

Figure p

2-1 Williamson's Key Characteristics of Capital Flows .............................................21

2-2 Capital Flows to Developing Economies in the Western Hemisphere ....................23

2-3 Private Capital Flows to Developing Countries in the Western Hemisphere ..........23

2-4 Capital Flows to the Andean Community ......................................... ...........25

2-5 ODA Flows to Members of the Andean Community .............................................26

2-6 Private Capital Flows to the Andean Community................. .....................26

2-7 Capital Flow s to CA CM ................................................. .............................. 29

2-8 ODA Flows to M embers of CACM ........................................ ...... ............... 30

2-9 FD I Flow s to M em bers of CA CM ..................... ................................................ 31

2-10 Capital Flows to CARICOM ............ ................................. ................. 34

2-11 Private Capital Flows to CARICOM ................................................. 34

2-12 Capital Flow s to M ERCO SU R ........................................ .......................... 36

2-13 Private Capital Flows to MERCOSUR ......................................... ...............37

2-14 C capital F low s to M exico ............................................................... .....................38

2-15 Private Capital Flow s to M exico.................................... ........................... ......... 39

2-16 Private Capital Flow s to Chile ........................................... .......................... 40

3-1 M odes of Internationalization ............................................ .... .......... 51















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

FOREIGN DIRECT INVESTMENT AND THE INVESTMENT CLIMATE OF
DEVELOPING COUNTRIES IN THE WESTERN HEMISPHERE

By

Eric T. Bonnett

December 2004

Chair: Timothy G. Taylor
Cochair: Gary F. Fairchild
Major Department: Food and Resource Economics

The structure of capital flows to developing countries around the globe has changed

considerably since the 1970s. In the Western Hemisphere specifically, foreign direct

investment (FDI) accounted for a much larger share of total inflows than any other type

of capital by the end of the 1990s. This trend has not escaped recognition by economic

theorists, and a large literature has emerged as researchers attempt to understand why

some countries attract more FDI than others.

This study examines the impact of changes in the investment climate on a country's

ability to attract FDI flows relative to other countries. Specific attention is given to the

developing countries of the Western Hemisphere. The investment climate is broadly

defined as (i) governmental policies and regulations that affect the relative "openness" of

the country to FDI, (ii) factors that impact the potential return on capital to foreign

investors, and (iii) the level of political risk and corruption in the host country.









Analyzing the impact of changes in these types of factors over time revealed some

interesting results. First, there is evidence to suggest that the state of the investment

climate is an important consideration for foreign investors. Second, FDI in large (in

terms of GDP), relatively unstable economies tends to be the most responsive to small

changes in the investment climate. Conversely, in small economies that either receive

substantial amounts of official development assistance or are dominated by a single

industry (e.g., the production of oil), FDI tends to be less responsive to changes in the

investment climate.

Finally, the results make clear that, across the sample of countries, there is no

single model that can explain all of the differences in the level of response to changes in

the investment climate. In other words, the relative impact of investment climate

variables on the level of FDI differs according to the economic, political and social

conditions inherent to each country. Thus, it is likely that future research into the issue

will yield more interesting results if analysis is conducted on a country-by-country basis.














CHAPTER 1
INTRODUCTION

The volatility of capital flows to developing countries in Latin America and the

Caribbean is well established. Investors in the region often hold short-term assets and

withdraw their funds at the first sign of economic distress (Rojas-Suarez and Weisbrod

1996). This, in turn, contributes to the vulnerability of Latin American and Caribbean

economies to external economic shocks such as changes in commodity prices and

international interest rates (Birdsall and Lozada 1996).

The financial crises that struck many developing countries after the surge in capital

flows during the 1990s led to increased skepticism about the benefits of attracting foreign

capital (World Bank 2001). However, the literature makes clear that just as there are

many different ways to define a financial crisis, there are equally as many theories of

what causes them. Nevertheless, it is generally agreed that the sharp rise in incidence of

financial crises in the last 20 years is not independent of the observed increase in the

magnitude and frequency of international capital flows.

From 1970 to 1992, Latin American and Caribbean economies were between two

and three times as volatile as industrialized economies (Hausmann and Gavin 1996).

While much of the observed volatility is a product of inconsistent macroeconomic policy,

other factors also had an influence. According to Hausmann and Gavin (1996):

another reason for the volatility of Latin American [and indeed, Caribbean]
macroeconomic outcomes is the large external disturbances that routinely buffet the
region. The most important of these are sudden changes in the terms of trade and
in international capital flows. (Hausmann and Gavin 1996, p. 27)









Large capital inflows cause rapid monetary expansion, inflationary pressures, real

exchange rate appreciation and widening current account deficits (Calvo et al. 1996).

Consequently, when flows are interrupted, the current account and exchange rate

experience reverse adjustments (Hoti 2002). Furthermore, these terms-of-trade shocks

have large and statistically significant effects on the variance of economic growth rates

over time (Easterly et al. 1993).

In many of the developing economies in the Western Hemisphere, FDI has become

the dominant component of capital inflows, accounting for a much higher percentage of

gross domestic product (GDP) in the 1990s than in the previous two decades. However,

there are exceptions to this rule, and it is often those countries that are the least

economically and politically stable that are the least successful at attracting FDI. Some

have argued that economic instability leads to uncertainty for the firms and investors who

make direct investments. As a result, investors choose to locate in less "risky" countries.

The cross-country variation in FDI since the early 1990s (discussed further in

Chapter 2) serves as the inspiration for this study. While a large body of empirical

literature addresses the global factors that cause expansions and contractions in global

FDI flows, less attention has been given to explaining the differences in FDI flows across

countries. More specifically, there is a need for additional insight into the factors that

affect the distribution of FDI flows across the developing countries of the Western

Hemisphere.

Empirical research suggests that the package of assets that accompanies FDI brings

many benefits to the host country including economic growth, development of domestic

industries, increased employment and a higher standard of living (Parry 1973).









Theoretical support for each of these contentions has developed over time while there

have been fewer attempts to argue against the benefits of FDI. What is made clear by

attempts to test the theory is that the benefits realized by the hosts of FDI funds are

conditional upon many factors.

Impacts of FDI on Developing Host Countries

The increase in foreign equity flows to developing countries in the Western

Hemisphere reflects the fact that countries have increased the number of investment

opportunities by promoting the privatization of government enterprises and facilitating

the development of deeper and more liquid financial markets (Moreno 2000). However,

Wells, Jr. (1998) points out that although tensions between foreign investors and

developing host countries had clearly weakened by the mid 1990s, there were few

definite conclusions as to the net impact of FDI in the lesser developed countries (LDCs).

Since then, a vast literature on the subject has emerged.

Much of the research into the economic impacts of FDI is aimed at examining the

indirect or "spillover" effects of hosting direct investment. It has been argued that

indirect effects primarily take the form of technological spillovers and competition

effects. In fact, Blomstrom and Kokko (2003) contend that the expectation that

technological know-how will spillover from multinational corporations (MNCs) to

indigenous firms has been strong enough for host countries to lower barriers to entry,

open up new sectors to foreign investment and, in some cases, provide investment

incentives to foreign firms. Nonetheless, empirical research into the matter is fraught

with conflicting results.

For instance, in a study on manufacturing industries in Mexico, Blomstrom (1986)

found that the presence of foreign MNCs in an industry was positively correlated with









structural efficiency. However, the results did not suggest that the increase in efficiency

was due to the transfer of technology from the MNCs to domestic firms (whether through

imitation or labor migration). Rather, increases in competitive pressure appeared to have

a more important effect.

Aitken and Harrison (1999) found that "foreign equity participation [in Venezuelan

plants] is positively correlated with plant productivity [but,] foreign investment

negatively affects the productivity of domestically owned plants" (p.605). The authors

indicate that in the presence of these offsetting effects, the net impact of FDI on the

productive efficiency of the host economy as a whole is negligible. More recent research

suggests that the extent to which foreign participation impacts the productivity of

indigenous firms appears to depend upon the type of linkages that are examined.

While Aitken and Harrison (1999) examine the effect of FDI on the productivity of

all firms in the economy, Smarzynska (2002) limits her analysis to the effect on upstream

firms (i.e., the backward linkage between foreign affiliates and their local suppliers).

Pointing out that spillovers "are more likely to be vertical rather than horizontal in

nature1" (Smarzynska 2002, p.2), she states, "a rise of ten percent in the foreign presence

in downstream industries is associated with a 0.38 percent increase in output of each

domestic firm in the upstream sector" (Smarzynska 2002, p.16). Furthermore,

Smarzynska found no difference between the effect of wholly-owned foreign subsidiaries

and joint ventures with both foreign and domestic investors.





1 This is due to the multinational firm's incentive to transfer knowledge to its suppliers in order to increase
the quality or decrease the price of its inputs, and prevent the horizontal leakage of information that might
enhance the performance of its competitors.









It has been shown by some researchers that controlling for certain domestic

conditions leads to more conclusive findings on the net impact ofFDI. One such case is

that of the relationship between FDI and economic growth in developing host countries.

Although empirical findings on the nature of this relationship do vary, the World Bank

(2001) indicates that differences in host countries' absorptive capacity2 accounts for a

majority of the variance. In fact, where absorptive capacity is high, it is generally agreed

that FDI has a positive impact on productivity and hence economic growth (World Bank

2001).

The study by Borenzstein et al. (1998) serves to illustrate this point more clearly.

Analyzing data on FDI flows to 69 developing countries over a period of 20 years, the

authors found that FDI had a positive effect on economic growth. Furthermore, the

magnitude of this effect was shown to depend strongly upon the available stock of human

capital, a variable commonly used as a measure of absorptive capacity. That is, higher

levels of human capital caused FDI to have a larger positive impact on growth.

Furthermore, with sufficient levels of human capital, FDI appeared to be more productive

than, and complementary to domestic investment. The authors also note that the positive

impact of FDI persisted even after controlling for initial income, human capital,

government consumption and the parallel market premium for foreign exchange.

Aside from the results of a handful of empirical studies, it is generally agreed that

FDI has the potential to facilitate the transfer of ideas from industrialized to developing

countries, thus increasing productivity in the latter (World Bank 2001). In addition to the

stability of FDI relative to other forms of foreign capital, the potential for increased

2 Proxies for absorptive capacity include but are not limited to openness to trade, the amount and quality of
infrastructure and human capital, and inflation.









productivity provides some rationale for why many developing countries have shown an

appetite for FDI.

Problematic Situation

It has been suggested by the Economic Commission for Latin America and the

Caribbean (ECLAC 2002) that the high volatility of net capital inflows (other than FDI)

during the last ten years has undermined the stability of economic growth in the region.

When net capital inflows are high, domestic credit and liquidity grow too quickly, while

the opposite is true in times of recession (ECLAC 2002). This phenomenon tends to

amplify boom-bust cycles, thus destabilizing economic growth. As an example, Mishkin

(2001) states

The financial crises that struck Mexico in 1994 and the East Asian countries in
1997 led to a fall in the growth rate of GDP on the order often percentage points.
The financial crises in Russia in 1998 and Ecuador in 1999 have had similar
negative effects on real output. Not only did these crises lead to sharp increases in
poverty, but to political instability as well. (Mishkin 2001, p.1)

The link between capital flow volatility and economic growth has led many

researchers to gain interest in the factors that drive capital flows. There is considerable

debate over whether capital flows to developing countries are driven by forces external to

the countries themselves, or by domestic factors.3 The existing empirical literature

suggests that the answer depends upon how the issue is addressed and that analyzing the

sum of all capital flows may be misleading. After all, each form of capital has its own

characteristics and set of circumstances (Lusinyan 2002).

Foreign direct investment accounts for a much larger share of GDP than any other

form of foreign capital inflow for many of the developing countries in the Western


3 Moreno (2000) provides one of the more concise reviews of the empirical literature addressing this issue.









Hemisphere, and much of this investment activity has been driven, or at least encouraged,

by the actions of host-country governments. In addition to privatizing government

enterprises, developing countries have adopted policies intended to foster direct

investment by MNCs. It is therefore surprising that the empirical literature does not

contain more models that attempt to measure the effect of these policies.

Gabel and Bruner (2003) report that multinational corporations and their

subsidiaries employ nearly 200 million people, generate $1.5 trillion in wages and pay

over $1.2 trillion in taxes to their host governments, annually. The almost overwhelming

global presence of the MNC has inspired considerable debate among the many players

involved in international commerce. Government policymakers are interested in

maximizing the benefits of hosting multinationals while simultaneously limiting their

negative impacts. Multinational managers are interested in maximizing the profitability

of their businesses given a set of available investment alternatives. In a world where

MNCs are increasingly prevalent, economists are left with the task of analyzing their

causes and impacts. The resulting body of academic literature is substantial and diverse

in purpose.

Wells, Jr. (1998) indicates that the "issues that face managers public and private -

who are concerned with FDI and the developing countries .. have not attracted

sufficient attention from economists" (p.101). Foreign direct investment flows are

ultimately driven by multinational corporations, and hence, their managers. Thus, the

interaction between government policy and the strategic allocation of capital by MNCs

needs to be addressed. Much of the thought on this matter has been aimed at modeling









the effect of exchange rate regimes on the international allocation of production, but the

issue is much more broad.

Problem Statement

The study of foreign direct investment has attracted the attention of researchers

from various fields of business. Economists have portrayed FDI as the result of

production cost differentials and comparative advantages in resource supply. In the field

of finance, FDI has been described as a tool for international portfolio diversification.

Finally, in the strategic management body of literature, FDI patterns are seen as being

dictated by the organizational decisions of the firm's management. Although these

approaches may differ from one another, efforts to understand why FDI occurs have

yielded some compelling results in each line of research.

Foreign direct investment can be viewed as the allocation of financial capital from

an entity in one country to establish or support a business entity in another country.

When defined in this manner, FDI becomes the result of a strategic decision guided by

the goals of management.4 To the extent that foreign investment policy can create an

environment that either fosters or impedes direct investment, the actions of host-country

governments may have a significant influence on FDI flows. Along these lines, the

ultimate concern of this study determining the extent to which FDI flows are affected by

the investment climates of developing host countries.

By surveying a group of U.S.-based multinational companies, Basi (1966)

identified several variables related to the investment climate that were of significant

importance to the investment decision. Of the variables considered by Basi, the

4 "Management" is defined broadly as those individuals who have a direct influence over the international
allocation of direct investors' capital.









following factors were most often cited as either "crucially important" or "fairly

important" to multinational managers: (i) the host country's attitude toward foreign

investment; (ii) political stability; (iii) limitations on ownership; (iv) currency exchange

regulations; (v) the stability of foreign exchange; and, (vi) tax structure. While some of

these variables are difficult to quantify, Basi at least provides an illustration of the

"investment climate" that is consistent with the how the term is used in this study.

The contribution this study makes to the existing literature centers on two primary

characteristics of the project's design. First, the empirical model presented here

examines the relative effect of five distinct investment climate factors on FDI. While

some proxy for the investment climate is often included as an explanatory variable in

existing models, fewer models have been designed to simultaneously examine different

aspects of the investment climate. The second distinguishing characteristic of the study is

its focus on Latin America and the Caribbean. Much of the existing literature is devoted

to examining FDI flows to the entire universe of developing countries. However, less

ambiguous conclusions may be provided to policymakers in the region by limiting the

focus of the analysis to the developing countries of the Western Hemisphere.

There are three reasons for examining the effect of the overall investment climate

on inflows of FDI. First, early attempts to survey firms on the determinants of

international production show that variables related to the investment climate are among

the most influential factors from the perspective of management.5 Second, debate on the

importance of many investment climate factors to foreign investors is ongoing. Finally,


SSee Dunning (1973) for a review of surveys conducted by various authors prior to 1973.









the relevance of the issue to policy-makers is clear, as many investment-climate variables

are under the direct control of the regulatory bodies of host countries.

Examination of FDI flows to the developing countries of the Western Hemisphere

is particularly relevant. The financial markets of these countries are increasingly

accessible to foreign investors, especially those who reside in the United States, where

nearly 20% of world FDI outflows originated during the period 1970-2002. Furthermore,

the provisions of the Free Trade Agreement of the Americas (FTAA), which is scheduled

for entry into force by the end of 2005, provide for the promotion and protection of

foreign investment funds, regardless of their source.

Objectives

The general objectives of this study are

* To provide an understanding of FDI as an economic phenomenon resulting from
the strategic activities of multinational corporations

* To examine the effect of the investment climate of developing countries in the
Western Hemisphere on inflows of FDI over the last two decades.

In order to achieve these rather broad goals, the study is designed with several

specific objectives in mind. The specific objectives of the study are

* To examine the evolution of the multinational corporation (MNC) from a
theoretical perspective and address the strengths and weaknesses of several streams
of theory

* To develop a framework for understanding FDI that integrates the elements of
economic and strategic management theory

* To develop a policy-relevant model of FDI inflows by including variables that are
not only considered as elements of the investment climate, but are also affected by
the actions of host-country policymakers.









Scope

Chapter 2 establishes the context for the study by providing an historical account

of contemporary developments in global economic integration. Specific attention is

given to the developing countries in the Western Hemisphere. The chapter compares and

contrasts the different types of international capital flows and presents a detailed analysis

of the structure of capital flows to Latin America and the Caribbean over the last thirty

years.

Chapter 3 presents the theoretical foundations of the study. The chapter begins

with a short review of the existing empirical research on FDI, although the results and

implications of the models are not the main point of concern. Rather, the models are used

to highlight the distinction between neoclassical economic theory and strategic

management theory, as well as to illustrate the fact that research on FDI has been

conducted in a number of contexts. Specific attention is given to the conceptual

evolution of the firm in each of the two streams of theory. Finally, a review of the most

relevant theoretical literature serves to define the perspective from which the empirical

considerations of the study derive.

Chapter 4 revisits the empirical literature is revisited in more depth. The chapter

reviews a range of models that examine issues related to the international allocation of

economic activity. Some of the models presented explicitly address the economic

determinants of FDI, while others analyze factors that affect the strategic decision-

making processes of multinational corporations. Empirical research on investment

location and the alternative modes of foreign ownership is also reviewed.

Chapter 5 presents the empirical model used to analyze a set of panel data on 21

developing countries in the Western Hemisphere over a period of 18 years. The chapter






12


compares the results from several alternative estimation procedures and provides a

discussion on the responsiveness of FDI to changes in the investment climates of

different countries. Finally, Chapter 6 summarizes the empirical results and presents a

set of conclusions based on the findings. Chapter 6 also provides suggestions for future

research.














CHAPTER 2
HISTORICAL CONTEXT

It is difficult to find a piece of recent literature in the field of international

economics within which the term "globalization" does not appear. According to the

Economic Commission for Latin America and the Caribbean (ECLAC 2002),

globalization refers to the denationalization of political, legal and cultural systems, as

well as economic markets. The primary entities driving this process are governments,

private investors and financial institutions (Schmukler 2004), and its socio-cultural

impacts on developing countries are far-reaching, and often difficult to measure. In terms

of economic impacts, globalization has led to reductions in trade barriers between

countries, increased exchange of information and technology, and in many cases, greater

vulnerability to worldwide economic conditions. It should be noted, however, that the

path toward globalization has been anything but uniform across countries.

ECLAC distinguishes between the three traditionally recognized phases of

globalization by comparing the relative levels of capital and labor mobility, the

(non)existence of free trade and international institutions for economic cooperation, and

the extent of standardization among national development models. The first phase of

globalization has its roots in the transportation revolution and began to take shape in the

last thirty years of the 19th century. Transportation costs fell dramatically during this

period, effectively reducing the distance between countries and contributing to increased

mobility of goods and labor (Philippe 2001).









With the Bretton Woods conference in 1944 came the initial movement toward

establishing international organizations for economic cooperation (specifically, the

International Monetary Fund and the World Bank). However, during this (the second)

phase of globalization, disparity remained among national models of economic

organization, and capital and labor mobility were limited. Although the Bretton Woods

agreement was intended to foster the flow of capital across international boarders, it

wasn't until the first oil crisis in the early 1970s that international capital mobility really

expanded (Phillipe 2001). Nevertheless, with the foundation of global institutions for

economic cooperation, the stage was set for an expansion in the trade of goods and

financial capital.

Since 1973, the drive towards international cooperation has affected a real

expansion in international trade of manufactured goods, services and capital. The

increased global presence of multinational corporations also contributed to this

expansion. Most recently, the birth of the information age brought about unprecedented

access to information and communication technologies, further facilitating international

transactions.

While this story characterizes the path to globalization among industrialized

economies throughout the world, it does not necessarily account for the individual

experiences of many developing economies in the Western Hemisphere. There are four

major economic integration groups in Latin America and the Caribbean: the Andean

Community; the Central American Common Market (CACM); the Caribbean

Community and Common Market (CARICOM); and the Southern Cone Common Market

(MERCOSUR). Mexico, Chile and the Dominican Republic do not hold membership in









any of these four integration groups but do relationships with them. Given that the

Dominican Republic is currently an observer to CARICOM and that a bilateral free trade

agreement exists between the two, the Dominican Republic is grouped with CARICOM

in the ensuing discussion. Mexico and Chile are discussed independently.

A Taxonomy of Capital Flows

According to Liberatori (2003), "international economic integration is per se the

result of both direct and indirect mobility of resources across national borders .

including migration of workers, international trade in goods and services, capital flows

and international production and investment" (p.2). In addition, international trade and

investment flows are the result of both equity and non-equity transactions. In addition to

the distinction between equity and non-equity, capital flows are categorized according to

the source of funds and the conditions upon which they are disbursed (i.e., concessional

vs. interest bearing). The Development Assistance Committee (DAC 2000)1 states that

Official transactions are those undertaken by central, state or local government
agencies at their own risk and responsibility, regardless of whether these agencies
have first borrowed the necessary funds from the private sector. Private
transactions are those undertaken by firms and individuals resident in the reporting
country. (DAC 2000, p.6)

Official capital consists of official development assistance (ODA) and other official

flows (OOF). According to the guidelines set up by the DAC, ODA includes flows to

countries on Part I of the Development Assistance Committee List (which includes all of

the countries in this analysis) that satisfy three criteria: (i) the funds must be provided by

official agencies including state and local governments or their executive agencies; (ii)


1 The DAC is the principal body through which the Organization for Economic Cooperation and
Development (OECD) deals with issues related to cooperation with developing countries. The DAC
statistical reporting directives provide a basis upon which all donor countries should report disbursements
of official capital to developing countries.









the transaction must be administered with the promotion of economic development and

welfare of developing countries as its main objective; and (iii) the funds must be

concessional in nature and convey a grant element of at least 25%. Official development

assistance flows are unique in that the debt service includes no interest payment and the

grant portion of these flows is free of repayment obligation. Other official flows, as

defined by the DAC, include official sector transactions with aid recipients that do not

satisfy the ODA criteria.

Private capital flows accounted for a significant share of resource flows to

developing economies in the Western Hemisphere during the last 30 years. Private

capital includes commercial-bank and trade-related lending, foreign direct investment and

portfolio investment. Bank and trade-related lending is comprised of commercial-bank

lending and other credits extended by foreign lenders in the private sector. The criteria

established by the World Bank dictates that lending by commercial banks that are

wholly- or partly-publicly owned be excluded from bank and trade-related lending.

Portfolio capital flows include both equity and bond investment. Portfolio equity

flows are measured as the sum of country funds, depository receipts, and direct purchases

of shares by foreign investors who own less than 10% of the voting stock of a firm.

Portfolio bond investment consists of any bond issues purchased by foreign investors

who own less than 10% of the voting stock of the issuer.

Foreign direct investment has historically been categorized as a private capital flow

despite the fact that direct investment, as defined by the DAC, may originate from private

or official sources. Another characteristic of FDI is that it takes the form of both equity

and non-equity investment. The World Bank's comprehensive guide to FDI released in









2003 describes some of the ambiguities that have historically existed in the measurement

and classification of FDI flows and establishes a clearer definition of FDI than had

previously been provided. The following discussion draws heavily upon the guide.

According to the World Bank,

direct investors may be individuals; incorporated or unincorporated private or
public enterprises; associated groups of individuals or enterprises; governments or
government agencies; or estates, trusts, or other organizations that own direct
investment enterprises in economies other than those in which the direct investors
reside. (Liberatori 2003, p.3)

Furthermore, the IMF indicates that in order for funds to be classified as direct

investment, "it is necessary to establish a (or show a pre-existing) specific relationship

between the parties involved in the transaction" (Liberatori 2003, p.2), where the term

"specific relationship" refers to the investor's possession or acquisition of a lasting

interest or an effective voice in the management of a direct investment enterprise.

In order to minimize the subjectivity of this definition, a specific empirical

threshold has conventionally been used to separate FDI from portfolio flows. The

threshold defined by the IMF dictates that only those entities in which the foreign

investor has acquired at least 10% of the ordinary shares or voting power qualify as direct

investment enterprises. Foreign direct investments are made through greenfield

investment (e.g., creation of new production capacity), joint ventures, or through mergers

and acquisitions (e.g., the privatization of an existing government-owned enterprise).

The types of capital covered by direct investment include: (i) equity capital; (ii)

reinvested earnings; and (iii) inter-company loans.

All flows of capital reported in this study are "net inflows." This is important

because there is a clear distinction between "net inflows" and "net flows." For instance,

the term "net foreign direct investment flows" refers to nonresident direct investment in









the host economy net of resident direct investment abroad. This is not what this study

examines. Rather, "net inflows of foreign direct investment" (with which this study is

concerned) refers to nonresident direct investment in the host economy net of foreign

direct investment funds withdrawn from the host economy by foreign investors. When

defined in this manner, a negative value for "net inflows of FDI" in any given year would

indicate that repatriation of FDI capital (which is an outflow) exceeded new inflows of

FDI in the host economy during the period. Similarly, a negative value for "net inflows

of bank and trade-related lending" indicates that principal and interest (P&I) payments to

foreign lenders exceeded new loan disbursements to the borrowing country.

Furthermore, a negative value for "portfolio equity" would indicate a certain level of

portfolio capital flight, not that more portfolio capital was invested by resident entities

than received by the host economy.

A Method of Characterizing Different Types of Capital Flows

In addition to the definitional characteristics provided by the Organization for

Economic Cooperation and Development (OECD), capital flows differ in terms of their

implications for both the host country and the investor. In fact, Williamson (2000) states

that there are five characteristics relevant to distinguishing among the various forms of

capital flow: (1) cost; (2) conditionality; (3) risk-bearing; (4) access to intellectual

property; and (5) vulnerability to capital flow reversal. The following discussion of these

characteristics draws heavily upon the work presented by Williamson (2000).

Cost. Official capital, given that it often comes in the form of grants or

concessional lending, is traditionally considered to be the cheapest form of capital

available to developing economies. While there is no clear difference between the cost of

commercial-bank lending and portfolio bond investment, it is generally recognized that









the cost of each is less than official capital. Portfolio equity returns tend to be highly

variable, fluctuating with the performance of stock markets. However, given that

markets typically demand an equity risk premium, it is generally accepted that the

required return on portfolio equity investment is, on average, higher than the return on

foreign debt. Meanwhile, foreign direct investment is believed to be the most expensive

form of capital for developing countries. As evidence of this, Williamson points out that

a conservative estimate of the annual return on United States FDI is about 12.4%.

Conditionality. Developing countries are often required by the lending

multilateral development bank (MDB) to establish a clear set of macroeconomic policy

objectives in order to qualify for concessionary loan funds. In addition, development

funds tend to be tied to the design and implementation of specific development projects.

Thus, disbursements of official development assistance (ODA) are considered as highly

conditional. Furthermore, the negotiations that often take place between multinational

firms contemplating investments and host countries (e.g., performance requirements

imposed by the host) result in FDI also being highly conditional, although perhaps not to

the extent of official flows. Other forms of capital are typically free of these types of

conditions.

Risk-bearing. "Risk-bearing" refers to who reaps the benefits (or foots the bill) in

the case of unexpectedly high (or low) revenues and costs. Here, risk is broken down

into commercial risk, interest risk and exchange rate risk. Virtually all official capital,

with the exception of investments made by the International Finance Corporation, convey

commercial risk to the borrower. The same is true of bank loans and portfolio bond









investment, except in the extreme case of lender bankruptcy. In the case of portfolio

equity investment and FDI, commercial risk is typically assumed by the investor.

Interest risk is also born solely by the investor in the case of equity investment,

whether portfolio or direct. For official flows and loan capital, the placement of interest

risk depends upon the terms of the loan. Fixed interest rates place the burden on the

lender while floating rates convey risk to the borrower. Given that most official loans

carry fixed interest rates, the lender (in most cases a multilateral development bank)

usually bears the risk of changes in the market interest rate. Both bonds and commercial

loans tend to made at floating interest rates, and therefore the burden is typically assumed

by the borrower.

Exchange rate risk, in the case of equity investment, is usually born by the investor.

Given that developing countries are rarely able to borrow large sums of money in their

local currency, exchange rate risk is typically conveyed to the borrower of official funds

and commercial loan capital. However, this risk may be transferred to the lender if

borrowed funds are denominated in the domestic currency.

Intellectual property. Since the pioneering work on FDI by Stephen Hymer

(1976), it has been commonly asserted that direct investment by foreign firms brings with

it the transfer of technological know-how in the form of patents, trademarks and

managerial expertise. As pointed out in Chapter 1, these externalities associated with

FDI are often referred to as spillover effects. The conditions associated with official

capital (e.g., macroeconomic policy requirements) also result in the transfer of knowledge

from MDBs and other official entities to host economies. Each of the other forms of

capital inflow typically does not facilitate the transfer of intellectual property.










Vulnerability to reversal. Although firms do shift working balances into and out

of different currencies in response to changes in macroeconomic prospects, the sum of

these transfers is likely to be small relative to the total of sunk capital invested in a

foreign economy (Williamson 2000). Thus, FDI is traditionally regarded as being

minimally vulnerable to reversal. Official flows also tend to be stable, and have actually

been shown to be counter-cyclical in nature. Private bank and trade-related lending,

while usually grouped together, tend to be very different in terms of vulnerability to

reversal. Short-term loans are more vulnerable to reversal than long-term loans, and

trade-related credits are almost always disbursed on extremely short terms. However, the

fact that trade credits are constantly renewed as new trade transactions need to be

financed makes trade-related lending more stable forms of capital inflow. At the other

extreme is bank lending, which in recent crises (e.g., the East Asian crisis and the Latin

American debt crisis) has proven to be highly vulnerable to reversal.
Cost of Capital

High *FDI


SPortfolio
Loan

Vulnerability Portfolio High additionall
*I Loran onditionality


Intellectual Property Risk


Figure 2-1. Williamson's Key Characteristics of Capital Flows









Figure 2-1 summarizes Williamson's framework for distinguishing between

different types of capital flows. The origin (or center) of the diagram corresponds to low

levels of conditionality, risk bearing, etc., while the outer limits represent the highest

levels. The figure shows reiterates the fact that FDI tends to be conditional and, at least

in theory, facilitates the transfer of technology. In contrast, portfolio equity investment

brings with it no conditionality or transfer of knowledge. The cost of equity investment

is typically high, which reflects the investor's burden of commercial risk, interest rate

risk and exchange rate risk. Perhaps most importantly, FDI is considered the most stable

form of foreign capital inflow available to developing host countries.

The Structure and Evolution of Capital Flows from 1970 to 20012

Figure 2-2 shows the structure of capital inflows to most of the developing

economies of the Western Hemisphere during the period 1970-2001. What is readily

evident in the figure is that private investment and lending served as the major sources of

foreign financing to these countries for much of the period 1970-2000. With the

exception of the mid-1980s, inflows of private funds were generally 3- to 5-times greater

than official development assistance and other official flows combined.

Looking specifically at the structure of private flows, bank and trade-related

lending was the most abundant type of private finance throughout the 1970s and early

1980s. However, this trend came to an abrupt end as the Latin American debt crisis set

in. As Figure 2-3 shows, repayments of private foreign debt actually exceeded new

lending in 1989 and 1993. The global economic slowdown of the late 1990s also had a

detrimental effect on the level of private lending to the developing countries of the

2 All dollar-denominated figures are stated in terms of current US dollars. These figures are converted from
domestic currencies using single-year official exchange rates.











Western Hemisphere. Similarly, net inflows of portfolio capital were rather unstable,


displaying the greatest year-to-year volatility during the 1990s.


150


-- Private
20 Capital Flows


S90
SOfficial
SDevelopment
Assistance
o 60
o Other Official
= Flow s
30
30


0 [: -- -- -



Figure 2-2. Capital Flows to Developing Economies in the Western Hemisphere


90

75
5/ ---- Foreign
60 Direct
SInvestment
S45
I \ Bank and
30 Trade
I Related
S \V Lending
o 15 -
- Portfolio
m _. Investment


(15)



Figure 2-3. Private Capital Flows to Developing Countries in the Western Hemisphere

In contrast to bank lending and portfolio investment, inflows of FDI grew during


the period 1970-2001; a majority of the growth occurred in the period immediately


following the debt crisis. Inflows of FDI stood at 0.99% of GDP in 1993, while they


accounted for just under 5% in 1999. However, the most remarkable trend is that inflows









of FDI, which once represented less than 20% of the size of private lending, ended the

period at over 50-times the magnitude of private lending and portfolio investment

combined.

Overall, the information presented in Figures 2-2 and 2-3 provides a reasonably

good characterization of how the structure of capital inflows to developing countries

across the globe has evolved over the last three decades. However, some of the most

interesting inferences are revealed by comparing the cases of individual countries.

Accordingly, the remainder of this chapter is devoted to examining the different

experiences of developing countries in the Western Hemisphere.

The Andean Community

The Andean Group was established in 1969 by the Agreement of Cartagena and

originally included Bolivia, Columbia, Ecuador, Peru and Chile (Andean Community

General Secretariat 2003). Venezuela became a party to the Agreement in 1973 and

Chile subsequently withdrew in 1977. The Group was primarily established as a reaction

to the poor performance of the Latin American Free Trade Agreement, but by the late

1980s, commerce orchestrated by the Agreement amounted to no more than 5% of the

combined trade of the group's members (Hanratty and Meditz 1989).

The Group's adoption of Decision 220 in 1987 loosened foreign investment

regulations, allowing greater freedom for the repatriation of profits, a higher percentage

of foreign ownership and investment in a wider variety of industries (Hanratty and

Meditz). The name of the organization was changed to the Andean Community as pat of

The Trujillo Amending Protocol of 1996. In addition to a common external tariff and

rules to prevent distortions in competition, the present-day agreement includes measures

to guarantee international investments.










As Figure 2-4 shows, capital flows to the Andean Community have, for the most

part, been dominated by private investment since 1970. Official development assistance

flows to the group were less than 1% of GDP in every year except 1991, and even then

they only reached 1.10% of aggregate output. Similarly, other official flows (OOF)

exceeded 1% of GDP only twice in 1990 and 1992 and never reached higher than

1.32% of GDP. Overall, both ODA and OOF (as a percentage of GDP) were generally

flat between 1970 and 2001 and never accounted for more than 0.5% of GDP.

10%

8% -
S--- Private
Capital Flow s
6%

0 Official
Development
SAssistance
2%
S.- Other Official
0% 7' Flow s

-2%



Figure 2-4. Capital Flows to the Andean Community

Figure 2-5 shows that ODA flows (as a percentage of GDP) to Bolivia grew from

1970 to 1992. In fact, ODA to Bolivia peaked at just under 12% of GDP in 1992. In

dollar magnitude, Peru attracted almost as much ODA as Bolivia. However, private

capital flows to the Andean Community as a whole were much more substantial (as well

as more volatile) than official flows. Private investment declined sharply during the first

half of the 1980s, but after the adoption of Decision 220 these flows expanded just as

quickly as they had dropped off. Private lending was the dominant source of capital for












the group until 1985, when foreign direct investment began to grow. The shift in private


capital inflows from debt to equity investment is evident in Figure 2-6.


12%


10% f

I \
8% /\ I
8 I \'\

Q p\ \I
(-

S6%


4% -


2% .
.... ... I . ....


0%



Figure 2-5. ODA Flows to Members of the Andean Community


6% I


4%


a.
2%



0%
0%


- Bolivia
S- Columbia
- Ecuador
Peru
-- Venezuela



















-- Foreign
Direct
Investment


Bank and
Trade
Related
Lending

- Portfolio
Investment


-2% !

0) 0) 0)0 0) 0

Figure 2-6. Private Capital Flows to the Andean Community


Heavy lending in Venezuela during the late 1970s pushed private lending to the


Andean Community to over 4.70% of the group's aggregate GDP. However, repayments


of foreign debt actually exceeded new lending as Venezuela began to service those


obligations during the 1980s; in 1987, repayments of private debt by the Community as a









whole exceeded new loan disbursements by almost $1.86 billion, leading to a net outflow

of private capital equal to 1.21% of GDP. Net bank lending (i.e., new lending minus

interest and principal payments on existing debt) remained mostly negative until 1993.

Then it was Columbia that attracted a majority of the private lending in the Andean

Community. Although lending to the group in 1997 was actually higher than in 1979,

increased economic activity caused bank and trade-related lending to account for a much

smaller proportion of GDP.

FDI flows to the Andean Community alternated between negative and positive

values from year to year before stabilizing in the 1980s. However, both FDI and

portfolio investment in the Andean Community expanded rapidly following Decision

220. Foreign direct investment surged to over 5% of the group's GDP in 1997 on the

heels of a sharp increase in investment in Columbia and Venezuela. Despite a continuing

surge of FDI in Bolivia, direct investment in the larger countries contracted over the next

few years and FDI in the Community as a whole fell from 5.07% of GDP in 1997 to just

over 3% in 2001.

Organized markets for equity securities existed in the Andean Community as early

as the end of the 16th century when the emergence of joint-stock companies began in

Columbia (Bolsa de Valores de Columbia, 2003). However, foreign portfolio investment

in the Andean Community first occurred in Venezuela in 1989 and equity markets in

Peru, Columbia and Ecuador began to attract foreign portfolio capital in the mid-1990s.

The Andean Community is unique in that portfolio investment in the group during the

last 32 years was split evenly between the equity and bond markets.









In summary, private capital was the most significant and most volatile type of

financial capital flowing into the Andean Community over the past three decades.

Private lending accounted for the majority of private capital flows to the group prior to

the Latin American debt crisis of the 1980s, while FDI and portfolio investment reigned

in the latter part of the 1980s and the 1990s. Disbursements of private debt contracted

sharply in the 1980s and never recovered to pre-crisis levels. Although inflows of

portfolio investment expanded during the 1990s, they paled in comparison to inflows of

FDI and private lending.

The Central American Common Market

The General Treaty of Central American Economic Integration established the

Central American Common Market (CACM) in December 1960. The agreement went

into effect in June 1961 with the completion of the necessary articles of ratification by El

Salvador, Guatemala and Nicaragua. Honduras and Costa Rica subsequently acceded to

the agreement in 1962. The General Treaty established the Central American Bank for

Economic Integration, a common external duty known as the Central American Standard

Import Tariff and a free trade area among CACM members. The Integration Industries

Convention (Regimen de Industrias de Integraci6n-RII) was also established at that

time as a means for governing foreign investment practices. In addition, the Convention

granted special incentives and privileges to firms given "integration industries" status.

Over time, this component of the agreement proved to be the most difficult to implement

(Merrill 1994).

The CACM integration process was somewhat successful in the 1960s, but by the

end of the decade Honduras and El Salvador had engaged in the so-called "Soccer War."

This dispute effectively reduced the Central American Common Market to an entity that










existed merely on paper (Merrill 1994). Although it took nearly a decade to establish an

official peace accord between the two countries, Honduras was able to negotiate a set of

favorable trade arrangements with other CACM members. The movement toward the

economic integration of Central America was rejuvenated in the early 1990s and

Honduras officially rejoined the process in early part of 1992.

Figure 2-7 shows the evolution of net capital inflows to CACM over the last three

decades. Bank and trade-related lending kept private capital flows to the group above 2%

of aggregate GDP during most of the 1970s. Lending subsided and ODA flows increased

three-fold throughout the 1980s and into the early 1990s; ODA reached 7.02% of GDP in

1990.

8%


6%

6 -- Rivate Capital
a 4% --Flow s

o0 Official
S2% -'Development
-Assistance

0% l~ /- Other Official
'Flow s

-2%



Figure 2-7. Capital Flows to CACM

From that point, inflows of ODA decreased and private capital inflows, led by an

expansion in FDI, jumped from just over 2% of GDP in 1997 to 5.56% in 1998. Inflows

of private capital had subsided by 2001 and were once again exceeded by inflows of

ODA. Other official flows never exceeded 1.5% of aggregate GDP.










As shown in Figure 2-8, the sharp increase in ODA in the early 1990s was led by

foreign aid flows to an ailing Nicaraguan economy. Foreign aid to Nicaragua reached

$837 million in 1991 and accounted for over 50% of the country's GDP. Flows of ODA

fell to 18.2% of Nicaraguan GDP over the next two years, only to rebound to over 48.5%

of GDP in 1996 (or roughly $933 million).

60%

50%
0-- Costa Rica

S....... El Salvador
.. 30%
30 -.-. -Guatamala

20% Honduras

10% -- Nicaragua

0% -


Figure 2-8. ODA Flows to Members of CACM

Although relatively less substantial than ODA flows to Nicaragua, flows of

assistance to both Honduras and El Salvador were significant during the 1980s and

1990s. As Figure 2-8 shows, flows of ODA to El Salvador reached 10.64% of GDP in

1987 but gradually fell below 2% by 2001. Official development flows to Honduras

reached their highest level in 1999, when they accounted for more than 15% of GDP.

As previously mentioned, private capital flows to CACM were dominated by bank

and trade-related lending during most the of the 1970s. However, as was the case with

most of the countries in the Western Hemisphere, private lending contracted significantly

during the Latin American Debt Crisis. Foreign direct investment in the group, while

fairly consistent throughout the 1970s and 1980s, expanded during the 1990s. In fact,










FDI increased from 1.75% of the group's aggregate GDP in 1997 to over 5% of GDP in

1998. Interestingly, Nicaragua was both the primary recipient of ODA funds and one of

the major beneficiaries of the expansion in FDI.

During the 1990s, major assistance from the International Finance Corporation was

directed at implementing new FDI laws and establishing an investment promotion agency

in Nicaragua. Figure 2-9 shows that FDI flows to Nicaragua increased from 0.0% to

13.23% of GDP during that decade. In 1998, the privatization of electricity and

telecommunications providers in El Salvador contributed to substantial inflows of FDI

(Sandrasagra 2000). Meanwhile, flows to Costa Rica grew steadily and eventually

reached 4.42% of GDP.

14%

12%

10% Costa Rica


S% -- Guatamala
0
s? 4% ---------------- "---- ^ 'r^ i
4 \ / \ Honduras
2%
2% i00/ ..- ..- .- ., .. ---Nicaragua
0%
-2%


Figure 2-9. FDI Flows to Members of CACM

Bond purchases by foreign investors were inconsistent and rarely exceeded 1% of

GDP in a given year among the CACM countries. The most remarkable inflow of

portfolio funds was in 1993, when investment in Honduras rose to 4.37% GDP ($152

million). Flows of portfolio capital to Costa Rica and El Salvador in 2001 led portfolio









investment in CACM to 0.89% of aggregate GDP, the highest level observed in the past

three decades.

In summary, official development agencies were the main source of financial

capital flows to CACM throughout the 1980s and most of the 1990s. While the group's

ability to obtain private credit was obviously affected by the debt crisis of the 1980s, FDI

as a percentage of GDP increased steadily from 1984 to 1998. It is this expansion in FDI

that brought inflows of private capital to levels that exceeded inflows of ODA in the late

1990s. However, it is noted that the magnitude of private inflows and official inflows

converged during the first two years of the new millennium.

The Caribbean Community and Common Market

The idea of regional integration began in the Caribbean with the establishment of

the rather short-lived British West Indies Federation of 1958. However, at the end of the

four years during which the Federation existed, the 10 members found themselves

economically the same as they had been for centuries (CARICOM Secretariat 2003).

Although plans for a customs union were part of the original Federation, free trade

among the islands in the region was not realized until the Caribbean Free Trade

Association (CARIFTA) came into effect in 1968.

The Caribbean Community Treaty was signed in Chaguaramas, Trinidad on July 4,

1973 and it was agreed among the four independent countries of CARIFTA Barbados,

Guyana, Jamaica, and Trinidad and Tobago that the agreement would come into effect

in August of that year. The revised Treaty of Chaguaramas officially established

CARICOM among these four independent signatory countries. The Treaty also set forth

that eight other Caribbean nations Antigua, Belize (British Honduras), Dominica,

Grenada, St. Lucia, Montserrat, St. Vincent and St. Kitts and Nevis would join the









Community by May 1, 1974. Haiti became the first French speaking Caribbean State to

obtain full membership in CARICOM on July 3, 2002.

With the exception of Montserrat (which remains a British territory), each of the

signatory CARICOM countries is currently independent from European control.

Nevertheless, nearly half of the members achieved independence subsequent to 1975.

The obvious problem with analyzing capital flows to these countries is the lack of

availability, or separability, of such data. As a result, seven of the 15 CARICOM

countries are not included in this analysis. The remaining members do represent an

anecdotal sample of CARICOM, as they account for nearly 75% of the overall economic

activity of the group (as measured by 2001 GDP). It is noted, however, that the unique

experiences of the eight CARICOM countries not included in this analysis may not be

characterized by the experiences of the countries that are examined.

As shown in Figure 2-10, capital flows to CARICOM were dominated by those

from private sources during the first half of the 1970s. However, massive repayments of

private debt obligations by Jamaica in 1976 and 1978 led to net outflows of private

capital in those years. Throughout the 1980s, official development assistance gradually

increased in some of the largest countries in CARICOM (including Haiti, Jamaica, and

the Dominican Republic) and ODA flows to the group reached 3.45% of GDP by 1990.

As time passed, private investment in CARICOM began to strengthen primarily in

Jamaica, Trinidad and Tobago, and the Dominican Republic. By 2001, private capital

flows (as a percentage of GDP) were in excess of the levels seen in 1970. In fact, net

inflows of private capital were almost 10-times the magnitude of net ODA inflows in

2001.














10%


8%
---- Private
6% Capital
Flow s

4% Official
Development
SAssistance

S----Other
\ / L Official
0%~ Flow s


-2%



Figure 2-10. Capital Flows to CARICOM

Figure 2-11 shows the structure of private capital flows to CARICOM. Foreign


direct investment was the primary source of private capital for CARICOM as a group


during the period 1970-2001. While bank and trade-related lending never exceeded 1%


of GDP, FDI accounted for more than 7% of GDP in the early 1970s and more than 5%


-2% !



Figure 2-11. Private Capital Flows to CARICOM


-- Foreign
Direct
Investment


Bank and
Trade
Related
Lending

- Portfolio
Investment


0
O? <0
5' ,0


d

f
I
,I
r
/ ~LC~ ~s
J


t









of GDP in the late 1990s. Portfolio investment in CARICOM expanded in both 2000 and

2001, reaching 3.15% of GDP. Most of this investment came in the form of bond

purchases in by foreign investors in Jamaica, Trinidad and Tobago, and Barbados.

MERCOSUR

MERCOSUR is an acronym for the Southern Cone Common Market and

specifically refers to the South American countries of Argentina, Brazil, Paraguay and

Uruguay. The Treaty of Asunci6n was signed on the March 26, 1991, effectively

establishing MERCOSUR at the end of November that same year. The Treaty was

intended to promote economic integration among countries in the region, although Chile

and Bolivia were conspicuously absent from the agreement (Hudson 1995). Bolivia had

originally intended to become the fifth member of MERCOSUR, although this accession

has never taken place. Furthermore, Chile evaded the agreement on the contention that

the other four signatory countries would have to lower their tariffs to the Chilean level

before Chile would join.

The Protocol of Colonia, which was signed in January of 1994, specifically

addresses the promotion and reciprocal protection of investments among the members of

MERCOSUR. The Protocol grants national treatment to investments in MERCOSUR

signatory countries made by investors from other members of the group. Other protocols

to the Treaty of Asuncion cover the defense of competition, the protection of intellectual

property rights and dispute settlement.

Figure 2-12 shows that for the last three decades financial capital flows to

MERCOSUR were dominated by private flows. Paraguay was the only member of

MERCOSUR for which ODA represented a substantial proportion of GDP, ranging from

0.67% to nearly 3.5%. Although Brazil received more ODA in most years than the other










three members of MERCOSUR combined, the immense size of the Brazilian economy

kept annual receipts to under 0.5% of GDP. Other official flows, while slightly higher

than ODA, never exceeded 0.75% of the group's aggregate GDP in a single year.

7%

6%
-- Private
5% Capital Flows

4%
(. Official
3% Development
o Assistance
^ 2%
1 % Other Official
S% Flows
0% V

-1%


Figure 2-12. Capital Flows to MERCOSUR

Figure 2-13 shows the evolution of private flows to MERCOSUR. In much the

same manner as the rest of Latin America, private flows to MERCOSUR consisted

mostly of debt throughout the 1970s and early 1980s. The boom in FDI began in 1994

and continued through the end of the decade. Brazil and Argentina attracted a vast

majority of the FDI flows to the group. In fact, FDI in Argentina quadrupled in the latter

half of the 1990s, and FDI in Brazil in 1999 was more than 6-times the level in 1995.

Large flows of portfolio capital also poured into both Argentina and Brazil throughout

the 1990s; flows to these two countries led portfolio investment to just over 3% of

MERCOSUR's aggregate GDP in 1993.

MERCOSUR is unique in that portfolio investment in the group really began to

increase prior to the expansion in FDI, whereas this wasn't necessarily the case in the rest

of Western Hemisphere. With the exception of Mexico, the heaviest portfolio investment










in Latin America took place in Argentina and Brazil. Also, differences in the magnitude

of private capital flows and official flows are also more obvious in MERCOSUR than in

any of the other groups in the Western Hemisphere. However, the gap narrowed slightly

in 2001 as private flows to the group dropped, led mostly by contractions in both

portfolio investment and FDI in Argentina.

7%

6% A

5% -- Foreign Direct
Investment
4%

e 3% Bank and
0( \ Trade Related
"S 2% --------------------- --------^---- Ledn
7 2/% \---- \ Lending

1% 4 \ \ Portfolio
\ Investment

-1%

-2%



Figure 2-13. Private Capital Flows to MERCOSUR

Mexico and Chile

Mexico is a party to several free trade agreements with countries and trade groups

on both sides of the globe. The most significant of these agreements are the so-called

Group of Three (1995) and the 1994 North American Free Trade Agreement (NAFTA).

The Group of Three is a sub-regional economic complementarity agreement which

includes provisions for trilateral investment flows between Mexico, Columbia and

Venezuela. Similarly, NAFTA also provides for the national treatment of signatory

countries' investors and their investments.










Figure 2-14 shows that, with the exception of four years in the 1980s, private

funds were the most abundant type of financial capital flowing into Mexico during the

period 1970-2001. In fact, Mexico attracted almost twice as much private capital as the

Andean Community, CACM and CARICOM combined. Private flows to Mexico

consisted mostly of bank and trade-related lending until the mid-1980s. However, credit

reversals and repayments to foreign lenders exceeded new borrowing in 1993 by more

than $6 billion.

8%

7%
S-- Private
6% Capital Flows

5%
a. 4% Official
SDevelopment
4 3% IAssistance

2%
2 Other Official
1% \Flows

0%
0% ------ -' ----^--^^^
-1%



Figure 2-14. Capital Flows to Mexico

Figure 2-15 shows the structure of private capital inflows to Mexico during the

period 1970-2001. The figure shows that portfolio investment poured into Mexico at an

unprecedented rate in 1993, perhaps in anticipation of NAFTA. Bond purchases by

foreign investors amounted to nearly $9 billion while portfolio equity investment soared

to $14.3 billion. Total portfolio investment in 1993 reached 4.87% of GDP, whereas FDI

only amounted to just over 1%. Direct investment expanded in the latter half of the

1990s and eventually reached 4% of GDP in 2001. Meanwhile, portfolio flows were

much more volatile, ranging from -0.38% to 4.29% of GDP.










Chile and Mexico are similar in the fact that ODA accounted for a small portion of

total capital inflows over the last three decades. In the case of Chile, ODA never

surpassed $200 million in any single year. Private capital flows were substantial during

the late 1970s and early 1980s, accounting for anywhere between 9% and 12% of GDP

annually. Nearly all of the private capital flowing into Chile at this time took the form of

bank and trade-related lending. Private credit collapsed during the Latin American debt

crisis and for a period of three years in the late 1980s official flows to Chile actually

exceeded inflows from private sources.

6%


Direct
4% Investment

3%
Bank and
2% A I \ Trade
0 2% Related
0 / I Lending
^ 1%
'""1 / ./ --- Portfolio
0% .--\ \ Investment

-1%

-2%



Figure 2-15. Private Capital Flows to Mexico

Following the crisis, FDI in Chile surged to 12.3% of GDP. A slight recovery in

private lending and a modest increase in portfolio investment also contributed to private

capital inflows reaching almost 18% of GDP by the end of the 1990s. As Figure 2-16

shows, the new millennium brought a sharp decline in FDI, and thus, net private capital

inflows as a whole.










16%

%-- Foreign Direct
% Investment

S8%
Q /Bank and
Trade Related
4% Lending

0% ----Portfolio
Investment

-4%



Figure 2-16. Private Capital Flows to Chile

Summary of Trends

The figures in this chapter provide a graphical illustration of both the volatility and

evolving structure of capital flows to the developing countries of the Western

Hemisphere. As previously mentioned, Schmukler (2004) argues that the evolutionary

process has been shaped by three primary agents. First, governments have influenced the

structure of capital flows in some instances by relaxing restrictions on the foreign

exchange transactions and allowing increased participation by foreign investors in many

sectors. Second, borrowers and investors, in choosing among different forms of

financing alternatives, have also played a role. Finally, financial institutions have begun

to offer a broader range of financing alternatives to investors and borrowers by making

use of international equity and debt markets. As a result, the developing countries of the

Western Hemisphere by and large have become more financially integrated with the

developed economies of the world. However, some interesting differences have been

exposed in this chapter with regard to the level of private foreign investment in the

various countries within the region.






41


Private capital flows to developing countries around the globe expanded sharply

during the last thirty years. The bulk of this expansion was accounted for by increases in

foreign direct investment in many Latin American and Caribbean economies. The

chapters that follow examine some of the potential reasons for such substantial increases

in FDI. Chapter 3 is devoted to reviewing the theoretical considerations of researchers

who have examined the issue in the past.














CHAPTER 3
THEORETICAL FOUNDATIONS

Foreign direct investment has been examined in a number of contexts in the

economic literature. The significance of the topic is illustrated by the variety of

theoretical models and frameworks that have emerged from efforts to characterize and

explain FDI. The diversity of the empirical literature alludes to the existence of many

distinct research agendas. An area that receives a disproportionate amount of attention

(and the area of concern in this study) is identifying the factors that drive flows of FDI.

The economic literature offers several approaches to examining this issue.

Ethier (1986) developed a general equilibrium approach to the determination of

FDI flows, addressing the need to understand how the founding principles of neoclassical

trade theory relate to the issue of foreign investment. Similar models were developed by

Helpman (1984) and Markusen (1984). While they are theoretically elegant, efforts to

test these models empirically have met with limited success. In fact, a review of the

empirical literature on FDI uncovers little in the way of general equilibrium comparative-

static analyses.

A second approach to the determination of global FDI flows is rooted in the theory

of industrial organization and is aimed at explaining why individual firms make

investments in foreign countries to produce the same goods as they produce at home

(Blomstom and Kokko 1997). Early theoretical work in this area, including Hymer

(1976) and Vernon (1966), focused on the firm-specific characteristics that make FDI

more attractive than exporting (Buckely and Casson 1998b). Meanwhile, Dunning (1977,









1995) expanded the set of factors to include host-country-specific variables as well as

firm-specific characteristics. Empirical research along these lines has shown that tariff

and non-tariff barriers to trade, as well as the legal, political and economic conditions of

host countries have a significant effect on investment decisions (Davidson 1980).

In a third body of literature, researchers model FDI flows as a function of either the

global supply of investment funds, the host country's demand for investment funds, or

some combination of the two. In one example of such an analysis, Dasgupta and Ratha

(2000) developed a two-stage approach that addresses both the supply and demand for

investment funds without relying on general equilibrium. In the first stage, the global

supply of FDI is determined by so-called "push" factors1 as investors in developed

countries decide how much capital to invest in the developing countries. In the second

stage, the global supply of direct investment funds is given, while country-specific

variables2 (the so-called "pull" factors) determine each developing country's respective

share of FDI.

The remainder of this chapter is devoted to comparing two widely-cited streams of

theory that are often used to explain the global allocation of FDI flows: neoclassical

economic theory and strategic management theory. This will serve to illustrate the

perspective from which the empirical section of this study is approached. The

methodology developed here is that foreign direct investment, both vertical and




1 Variables considered in the first stage of the process include global market growth (proxied by world
trade as a percentage of world GDP), world GDP growth, GDP growth in developing countries as a whole,
the real LIBOR rate and an index of privatization.
2 Variables examined in the second stage include the current account balance, GNP per capital and private
non-FDI flows as a percentage of GDP.









horizontal, is the observable outcome of coordinated international economic activity3. As

such, the international allocation of FDI flows is ultimately dictated by investors as they

choose among alternative investment locations based on criteria that are perceived to

affect profitability. Neoclassical economic theory pays little attention to the actions of

firms with regard to resource allocation, treating the firm as the proverbial "black box"

into which resources go and out of which goods emerge (Demsetz 1997). Conversely,

the strategic management stream of literature (which has Coase's 1937 theory of

internalization at its core) is concerned with explaining the strategic allocation of

resources by firms, thus providing a role for the decisions of management.

Neoclassical Economic Theory vs. Strategic Management Theory

The purpose of neoclassical economic theory, with perfect competition as its core,

"is to understand price-guided, not management-guided, resource allocation" (Demsetz

1997 p.426). The economic firm is generally embodied in a mathematical function (i.e.,

a production function) defined in terms of technology. In this construct, the usefulness

of the theory lies in its ability to "capture fundamental economic forces and their

interrelationships" (Taylor and Seale 1999 p.8). The theory illustrates a state of

dependency, where production takes place in the firm and consumption and resource

supply take place in households (Demsetz 1997). The strategic management of resource

supply and output demand conditions is not directly addressed by neoclassical economic

theory, giving rise to the suggestion that the economic firm is a "black-box" about which

little is known.




3 Most foreign direct investment is "horizontal" in the sense that a majority of the output of the foreign
affiliate not intended for export to the parent firm's home country (Markusen 1995).









Over time, neoclassical theory has been useful in attempting to explain the

domestic, and by extension in the trade literature, the international allocation of resources

and economic welfare. In neoclassical trade theory, the notion of the firm is generally

that of a purely domestic entity "competing via trade with the national champions of

other countries" (Markusen 1995 p. 169). In addition to the assumption of perfect

competition, extending the theory to the international trade of goods and services often

requires the researcher to assume the existence of comparative advantages in resource

supply. The shortcomings of this extension have been recognized by many and a more

recent research agenda (referred to by Markusen as the "new trade theory") is aimed at

acknowledging that "trade and gains from trade can arise independently of any pattern of

comparative advantage (as traditionally understood) as firms exploit economies of scale

and pursue strategies of product differentiation in an imperfectly competitive

environment" (Markusen 1995 p.169).

Relaxing the assumption of perfect competition does not pose a significant problem

for general-equilibrium trade analysis. However, the definition of the firm in neoclassical

trade theory inherently limits its ability to provide any explanation for the widespread

"real-world" phenomenon of the multinational enterprise. As the boundaries of the firm

expand in both geographic and product space, the need arises to address the strategic

activities that occur inside the black-box of the economic firm.

A useful example is Calderon-Rossell's (1985) attempt to model the effect of

foreign exchange rates and production costs on the MNC's choice between producing in

foreign and/or domestic locations. In this model, the multinational firm is composed of a

parent (in the home country) and a subsidiary (in a foreign country). Monopolistic









demand and cost functions for each are assumed. The limitations of this approach are

characteristic of neoclassical trade model in that the analysis is limited to a two-country

world and requires the assumption of perfect competition, comparative advantage and

zero transportation costs.

The strategic management body of literature was inspired by Robinson's (1932)

pioneering suggestion that assumptions in economic theory should correspond to

conditions in the real world (Coase 1937). Robinson (1932) takes notice of the disparity

between the firm as it is defined in economic theory and the firm as it is described by the

"plain man." Coase (1937) subsequently set out to provide a definition of the firm that is

"not only realistic in that it corresponds to what is meant by a firm in the real world"

(p.386) but is also tractable by the tools of economic analysis.

Coase's (1937) seminal paper addressed the motives for organizing domestic

assets and labor into a firm rather than making use of specialized market exchanges to

conduct arm's-length transactions. The main tenet is that the cost of using the "price

mechanism" associated with market exchanges, along with the costs of negotiating and

concluding separate contracts make it more profitable to internalize production activities.

He stated, "It is true that contracts are not eliminated when there is a firm but they are

greatly reduced" (Coase 1937 pp.390-91).

Coase contended that elements of the regulatory regime in an economy also

provide incentive for firms to internalize operations. Specifically, he contends:

Another factor that should be noted is that exchange transactions on a market and
the same transaction organised within a firm are often treated differently by
Governments or other bodies with regulatory powers. If we consider the operation
of a sales tax, it is clear that it is a tax on market transactions and not on the same
transactions organised within the firm. Now since these are alternative methods of
"organisation" by the price mechanism or by the entrepreneur such a regulation









would bring into existence firms which otherwise would have no raison d'etre...Of
course, to the extent that firms already exist, such a measure as a sales tax would
merely tend to make them larger than they would otherwise be. (Coase 1937 p.393)

Thus, the "Coasian" concept of internalization can be described as the firm's

incentive to internalize imperfect markets when the cost associated with transacting

internally is lower. However, Coase also identified the tendency for firms to increase in

size and diversify geographically as the costs of decentralization (i.e., communication and

transportation costs) decrease, albeit in a purely domestic sense. Nevertheless, his work

ultimately led to the realization that the firm's boundaries are set in two dimensions:

product space (through vertical and horizontal integration), and geographic space (both

regionally and internationally).

Pointing once again to the dichotomy between the economist's definition of the

firm and the firm in "everyday speech," Phelan and Lewin (2000) argue that modern

theorists are really attempting to explain the existence of the "corporation." For them, a

corporation is broadly defined as an entity comprised of a number of people and other

assets, which may have legal status as a company or partnership. This rather broad

definition is a reflection of the structural diversity of the real-world organizations that are

referred to by modern economic theorists as firms.

The focus of more recent research "has shifted away from the coordination

problems originally emphasized by Coase and towards the role of firm boundaries in

providing incentives" (Holmstrom and Roberts 1998 p.74). Furthermore, strategists have

made significant contributions to understanding the benefits of the firm structure as well

as how resources affect their boundaries (Phelan and Lewin 1999). An extensive body of

literature is devoted to examining the relationship between firms and so-called hold-up

problems, transactions costs, and intellectual property rights. While much of the early









research was conducted in a domestic setting, Dunning (1977) provided a framework that

extends the analysis to the international allocation of economic activity.

Dunning's OLI Framework

The "Coasian" theory does not specifically address the emergence of the

multinational corporation. However, a more contemporary agenda of the strategic

management literature is to explain the determinants of foreign production, and hence,

the existence of multinationals. Within this area of research, Dunning's (1977)

Ownership-Location-Internalization (OLI) framework receives a disproportionate amount

of attention.

In an effort to address why firms take ownership positions in foreign markets as

opposed to exporting or conducting transactions at arm's length, Dunning contended that

foreign direct investments (FDIs) are made when three conditions are fulfilled:

(1) the firm must posses ownership-specific advantages, those that are "internal to the
enterprise of the home country, but capable of being used with other resources in the
home country or elsewhere" (Dunning 1970 p.399). Ownership-specific advantages
include organizational and entrepreneurial skills, patents, and firm size (which may
lead to both scale economies and market power);

(2) the host country must posses location-specific advantages, those "originating only
from the resources of [the home] country but available to all firms" (Dunning 1970
p.399). Location-specific endowments include Ricardian-type endowments, i.e.,
proximity to the point of sale, market size, and availability of natural resources and
manpower; as well as the legal and commercial environment in which resources are
used, i.e., market structure, and governmental legislation and policies (Dunning
1980);

(3) and finally, there must be benefits to internalizing foreign production processes.

Addressing the two former conditions, Dunning (1980) states, "The possession of

ownership advantages determines which firms will supply a particular market, whereas

the pattern of location endowments explains whether the firm will supply that market by

exports (trade) or by local production (non-trade) [and hence FDI]" (p. 11). The third









condition addresses the mechanism by which firms exploit ownership and location

advantages in order to service foreign markets. In subsequent work, Itaki (1991) pointed

out that Dunning's early concept of internalization differs from that of Coase in that it "is

interpreted as internalization of an "ownership advantage" rather than that of an imperfect

market" (p.445). This is made clear in the following statement by Dunning:

The thesis is that the international competitiveness of a country's products is
attributable not only to the possession of superior resources of its enterprises but
also to the desire and ability of these enterprises to internalise the advantages
resulting from this possession; and that servicing a foreign market through foreign
production confers unique benefits of this kind. (Dunning 1977 p.402)

Nonetheless, Dunning subsequently reconciled his concept of internalization with that of

Coase, modifying it to include internalization of both imperfect markets and ownership

advantages (Itaki 1991).

Dunning (1980) later revisited his original argument and added that the desirability

of internalizing foreign production processes could derive from both market

imperfections and public intervention. Market imperfections include uncertainty in

future market conditions or government policies, structural imperfections (e.g., barriers to

entry, high transactions costs, etc.) and cognitive imperfections (e.g., unavailability or

costly acquisition of information about the product or service being provided), while two

types of public intervention were considered (at least, by Dunning) to be relevant to

multinational enterprises.

The first concerns the extent to which government intervenes in the production and

marketing of public goods by corporations. Dunning (1977) states, "the need both to

generate innovations and ideas and to retain exclusive right to their use, has been one of

the main inducements for enterprises to internalize their activities" (p.404). In addition,

the efficient exploitation of technology often requires complementary resources that









cannot be protected by patent (e.g., financial systems, organizational skills, marketing

expertise and managerial experience). Dunning indicates that the lack of public

intervention in the production and marketing of these complementary resources also

encourages internalization by firms.

The second type of public intervention that promotes internalization is economic

policy which tends to distort the international allocation of resources. This includes

corporate taxes and policies regarding the remittance of dividends and other forms of

profit repatriation. For example, a multinational enterprise, in an effort to record profits

in the lowest tax areas, may find it desirable to control the prices at which intermediary

products are exchanged by its international groups. Thus, the firm may internalize

production across international boarders and use transfer pricing in order to avoid higher

sales taxes resulting from the exchange of intermediary goods at external-market prices.

However, Dunning recognized that as early as the 1970s, more intense governmental

surveillance over transfer pricing strategies had begun to erode some of these types of

benefits (Dunning 1977).

In his summarization of the OLI framework, Kurt Pederson (2002) provides a

useful figure that illustrates how the presence or absence of location and internalization

advantages might affect the mode of internationalization preferred by firms. In Figure 2-

1 (which is a reproduction of Pederson's figure), +I and +L indicate the existence of

internalization and location advantages respectively, while -I and -L indicate their

absence. In the figure, the firm's possession of ownership advantages is assumed.

Pederson isolates the role of location-specific advantages, suggesting that the

extent to which these types of advantages exist influences the firm's decision of whether









to make a direct investment or to serve the foreign market at arm's length. Loree and

Guisinger (1995) classify aspects of the policy environment (e.g., corporate tax rates,

(O advantages assumed) -L +L
-I (1) Simple export (3) Contractual agreements
+I (2) Sales subsidiary (4) Foreign direct investment
Reproduced from (Pederson 2002 p.6)
Figure 3-1. Modes of Internationalization

regulations on profit repatriation and foreign ownership, and investment incentives) as

location-specific disadvantagess and state that host-country governments have a

"normative desire...to manipulate policies that are thought to affect FDI flows" (p.285).

In his critique of the OLI framework, Itaki (1991) takes issue with the

inseparability of ownership and location advantages by pointing out that the eclectic

framework is "weakest when ascertaining which items are most decisive in attracting

FDI" (p.456). Using the case of a technological breakthrough made by a foreign affiliate

as an example, Itaki contends that the new technology would be classified by Dunning as

an ownership advantage of the parent MNC. However, to the extent that the cost of

developing and implementing the new technology is a product of local labor costs in the

foreign country, the advantage of the new technology (in economic terms) may be both

ownership-specific and location-specific. While Itaki's observations are valid, the

purpose of this study is not to make inferences on which of Dunning's three conditions

bears the greatest impact on the FDI decision. Rather, the usefulness of the OLI

framework lies in its taxonomical illustration of factors that may positively or negatively

affect the MNC's decision to invest in a given foreign country.









Foreign direct investment has been modeled by some researchers as the end-result

of a multi-stage process4. Although the context varies, the same analogy is useful in

setting a backdrop for this study. In the first stage, firms develop the initiative to

establish some form of international presence. This initiative could derive from

conditions specific to the home country, global economic conditions, or a mix of the two.

In the second stage, the firm's management makes decisions on the mode of organization

and location for its international operation. Here, it is argued that managers consider

location-specific attributes like the overall investment climate, market size and barriers to

trade (Dunning 1973). The third stage of the process involves making detailed economic

profitability forecasts given a prospective location for the foreign affiliate. Important

factors in this stage include the level of industrial concentration and existing competition

(Miller and Weigel 1972).

The literature review provided in this chapter is intended to provide insight into the

first stage of the process described above, while the empirical considerations of this study

pertain more directly to the second stage. Simply put, different aspects of the investment

climate are analyzed to determine their effect, if any, on the extent to which foreign firms

in general have found each country to be a suitable or unsuitable business location.

However, one should remain cognizant of the fact that a firm's decision to operate in a

foreign market does not necessarily result in foreign investment. Contractor and Kundu

(1998) provide clarification of this point by stating

Until a decade ago, local adaptation by global firms was expressed by varying their
business practices and methods in each country, while leaving the ownership and
organizational structures fairly invariant across nations. Today, the modal choice
issue has gone beyond the "internalize or not" question (Buckley and Casson,

4 See Miller and Weigel (1972) and Barrell and Pain (1996) for two examples.









1976), and even beyond the "licensing vs. joint venture vs. merger" set of
alternatives (in Buckley and Casson 1996), to include other types of [non-
investment based] alliances, such as management service contracts, and
franchising. The general modal choice set now includes varying levels of equity
ownership, as well as several alliances of various descriptions. The manager must,
today, choose from a larger set of options. (Contractor and Kundu 1998 p.353).

Thus, the literal intent of this study is to examine how firms respond to host-country

investment-climate conditions, with specific attention given to the affect on the decision

to participate via an equity-based mode of organization.

In an attempt to understand how the investment climate affects FDI decisions,

Stobaugh, Jr. (1969) specified the following basic approaches companies take when

analyzing the investment climate of potential host countries: (1) go-no go, (2) premium

for risk, (3) range of estimates, and (4) risk analysis. The two former approaches are the

least complex with regard to the required investigation and calculations, while the latter

two involve highly complex analyses. Nevertheless, it is useful to illustrate how

elements of the investment climate might affect the decisions made by firms who use the

two simpler approaches (i.e., go-no go and premium for risk).

For Stobaugh, managers who use the go-no go approach sometimes use it in

conjunction with the premium for risk approach and typically base the investment

decision on one or two characteristics of the host country. For instance, suppose the

manager is faced with a high probability that the host country currency will devalue, a

condition that directly affects the projected profitability of the venture. In this case, the

risk of devaluation might lead the manager to reject the investment alternative (in other

words, the manager chooses the "no go" option). On the other hand, if the probability of

devaluation is not high enough to warrant a rejection on the initial screening, the manager

might require a higher return on investment given the risk of devaluation, thus requiring a









premium for the risk. In either case, a host country facing a high probability of

devaluation is less likely to attract FDI than a country with a low probability.

This example provides an understanding of how just one element of the investment

climate (fluctuations in foreign currency values) might affect the outcome of the FDI

decision and can be extended to account for other investment climate factors as well as

firms who use alternative investment approaches. Other examples of location-specific

factors that could influence firms looking to invest in a specific foreign country might

include the level of political instabilityt, the nature of the corporate tax regime, the

extent to which foreign firms are protected from expropriation of assets and the level of

business-related corruption.

It is easy to imagine how each of these factors could lead a foreign firm to either

reject or accept an investment project on the basis of its perceived level of risk.

However, theoretical research into the foreign investment behavior of firms also suggests

that other factors often play an important role in the decision process. For instance, the

existence of resources with strong global demand in a particular country might attract

foreign investment irregardless of conditions that would otherwise translate into a

prohibitive level of risk for foreign firms. Alternatively, previous research has also

shown that the level of FDI may be related to the other types of capital (e.g., official

development assistance and private foreign lending) flowing into a given country.

The next chapter focuses on how empirical researchers have examined the factors

that affect the FDI decisions of multinational firms. Specific attention is given to the

strategic management stream of literature. Later, in Chapter 5, each of countries






55


examined in this analysis are analyzed for the extent to which other considerations might

dominate the effect of the investment climate.














CHAPTER 4
EMPIRICAL CONSIDERATIONS

The literature review presented in the previous chapter highlighted the distinction

between neoclassical economic theory and strategic management theory as each relates to

the direct investment activities of the multinational corporation (MNC). The purpose of

this chapter is to develop an empirical model that can be used to analyze FDI flows to

selected developing economies in the Western Hemisphere. The chapter begins by

reviewing several empirical studies specifically concerned with the determinants of

foreign direct investment as well as the factors that affect the mode of organization (i.e.,

ownership structure) and the location of FDI. The chapter ends with an in-depth look at

the variables that are analyzed in this study.

The MNC's Motivations for Making Direct Investments

It should be recognized that Dunning's OLI triad represents only one of many

attempts to understand why MNCs establish foreign operations. While some of the

alternative theories have gained impetus, most are derived from the same motivations and

often yield results that tend to reiterate Dunning's contentions. For example, in a direct

comparison of portfolio theory and the OLI framework, Morck and Yeung (1991) found

that the incentives for FDI derive from the benefits of internalizing foreign operations

rather than benefits associated with international diversification of corporate portfolios.

Morck and Yeung (1991) examined the extent to which multinationality affects a

firm's market capitalization (i.e., "market cap"), and thus the net worth of the firm's

shareholders. Assuming that financial markets operate efficiently, the MNC's market cap









(V) can be measured as the sum of its tangible (T) and intangible (1) asset values. Citing

a host of earlier studies (including work by Dunning, Vernon, and Modigliani and

Miller), the authors set out to determine the extent to which a firm's degree of

multinationality1 impacts its intangible asset value. Expenditures on R & D and

advertising were used to control for the effect of intangible assets like technical expertise

and consumer goodwill. The authors also control for different industries by including a

series of three-digit SIC2 code dummy variables.

Ordinary least squares (OLS) regression results indicated that the degree of

multinationality was positively correlated with the MNC's share price and that the

relationship was highly significant. On average, a firm with five or more subsidiaries

was found to have a share price 8.41% higher than that of a firm with less than five

foreign subsidiaries. More importantly, Morck and Yeung showed that the impact of

spending on intangibles like advertising or R & D also increased with the degree of

multinationality.

The implication of these findings, according to the authors, is that the value of

multinationality is not derived from the international diversification of risk, tax

advantages or relative production costs. Rather, the firm's possession of (ownership-

specific) intangible assets serves as a necessary condition for FDI to positively affect the

firm's market value. Thus, the results call into question the notion that investors value





1 The degree of multinationality is measured by the number of subsidiaries the firm has, the number of
foreign nations in which the firm has subsidiaries and a series of dummy variables representing various
levels of FDI (Morck and Yeung 1991).
2 Standard Industrial Classification









the MNC as a means for international portfolio diversification, and lend credibility to the

theory of internalization.

Pugel (1981) tested the effect of four ownership-specific variables the possession

of proprietary new technology, marketing ability and expertise, organizational and

managerial technique, and the ability to obtain capital at favorable rates on outward

U.S. FDI intensity, which was measured as the share of U.S. subsidiary profits in the total

after-tax industry profits in the host country. He used the proportion of scientists and

engineers in the total workforce as a proxy for the extent to which new technology is

generated through R & D. Advertising intensity in the industry served as a proxy for

marketing abilities and the importance of organizational and managerial technique was

measured by the share of total employment in the industry accounted for by managers.

Finally, the amount of capital necessary to establish a factory of minimum efficient scale

was measured by total assets (net of depreciation) multiplied by the location-specific

importance of scale economies (the measurement of which is explained below).

Using data roughly corresponding to the 3-digit SIC level on U.S. manufacturing

industries, Pugel found that all four ownership advantages favored outward FDI. He also

included two location-specific independent variables representing "centralizing" and

"decentralizing" agents. Namely, he examined the importance of scale economies in

production measured as the average size of the largest plants producing half of the

industry's output divided by total industry shipments and the magnitude of transport

costs. The four-firm concentration ratio (C4) of each industry was also included as a

measure of oligopolistic rivalry.









Pugel specified a double-logarithmic regression equation and used OLS to estimate

the relationships between the dependent and explanatory variables. As such, the resulting

parameter estimates can be interpreted as elasticities. The importance of scale economies

was shown to hinder outward FDI, as was the magnitude of transport costs, although the

effect of the latter was statistically insignificant. The C4 variable was positively related

to FDI, indicating that "oligopolistic reaction leads to more FDI than would otherwise

occur in an industry" (Pugel 1981, p.226). Pugel takes this finding as an indication of the

fact that firms operating in industries where production concentration is high tend be

more likely to protect their market share in foreign countries by establishing a subsidiary.

The findings of these studies suggest that a firm's FDI decision is impacted by the

need to protect proprietary information and expertise. However, the nature of this

relationship has been a source of debate among theorists. Along these lines, Lee and

Mansfield (1996) found that the adequacy of a developing country's system of

intellectual property protection (a location-specific variable observed by direct survey of

a random sample of U.S. firms) affects both the volume and composition of FDI in a

country.

Least squares estimates on outward FDI data from 100 major U.S. firms in 14

developing countries suggest that "if the percentage of firms regarding protection in a

particular country as inadequate falls by 10 points, U.S. foreign direct investment there

might increase by about $140 million per year" (Lee and Mansfield 1996, p.185). Other

theorists have argued that a lack of intellectual property protection may favor FDI as

opposed to licensing or contracting. In contrast, the empirical results presented by Lee









and Mansfield suggest that a MNC's incentive to invest directly is positively related to

the level of intellectual property protection in the host country.

Decisions on the Mode of Organization

The study by Lee and Mansfield provides an example of how the scope of research

on FDI has broadened over time from analyzing the export versus FDI decision to

focusing on the incentives of internalization as opposed to licensing, subcontracting or

franchising. According to Buckley and Casson (1998), the 1990s witnessed a renewed

interest in why certain circumstances seem to favor some modes of market entry over

others. They make the following statement which, in fact, embodies the perspective from

which this study is approached:

Entry [into a foreign market] involves two interdependent decisions on location
and mode of control. Exporting is domestically located and administratively
controlled, foreign licensing is foreign located and contractually controlled, and
FDI is foreign located and administratively controlled. (Buckley and Casson 1998,
p.541)

A review of the empirical literature on alternative modes of foreign entry and

ownership turns up studies on the decision between greenfield investment or mergers and

acquisitions (see Zejan 1990, and Hennart and Park 1993), joint ventures versus wholly-

owned subsidiaries (see various works by Contractor and Lorange, and Beamish and

Killing), and more broadly, the decision between foreign equity investment and non-

equity contractual alliances. Researchers have found that many of the same variables

play a role in each of these distinct decisions.

The theoretical model of entry strategies developed Buckley and Casson (1998)

yields the generalized result that high transaction costs cause firms to favor FDI over

subcontracting and licensing if the cost is associated with arm's-length technology

transfer, and over franchising if the cost is associated with the arm's-length intermediate









output market. However, empirical work by Contractor and Kundu (1998) indicates that

the choice of entry mode depends upon much more than transactions costs, pointing out

the relevance of agency costs as well as the importance of host country-specific variables

and firm characteristics. The authors provide an interesting study on the international

hotel industry, a sector in which foreign equity ownership is at least as widespread as

non-equity modes of organization (Contractor and Kundu 1998).

Contractor and Kundu specify their dependent variable as the mode of organization

and allow for four different modes: (i) fully owned, (ii) partially owned, (iii) management

or service contract and (iv) franchise. Host country-specific determinants included as

independent variables are Frost and Sullivan's "Composite Risk Index", the level of

foreign business penetration (measured by FDI/GDP), the level of development (i.e.,

GDP per capital and a measure of cultural distance between the home and host

countries3. The most interesting result of the analysis is that foreign equity-based modes

are less likely in risky countries and more likely in lower income countries.

The results reached by Contractor and Kundu were supported by survey

questionnaire responses reported by Kim and Hwang (1992), who surveyed 96

multinational managers on the importance of host-country environmental factors to the

decision between three modes of entry (i.e., licensing, joint venture and wholly-owned

subsidiary). The survey results indicated that perceived country risk and the level of

unfamiliarity with the host country are more important environmental factors than the

uncertainty of demand and intensity of competition in the host country. Specifically,



3 Firm-specific structural, strategy and control factors are also considered by Contractor and Kundu but are
less relevant to the intent of this study and are omitted for the sake of time.









higher levels of risk and unfamiliarity were associated with lower levels of resource

commitments and hence lower levels of FDI.

As part of the vast literature inspired by Dunning' seminal work, Agarwal and

Ramiswami (1992) examined the interrelationships between ownership, location and

internalization (OLI) factors and the choice of entry mode. The authors pointed out that

while there is a significant body of research devoted to examining the effect of OLI

factors, the interrelationships of the three types of variables have largely been neglected.

They concentrated on overseas leasing data from the U.S. equipment leasing industry and

specified the OLI triad as

* Ownership advantages such as firm size, multinational experience and the ability to
develop differentiated products

* Location advantages such as market potential and level of investment risk

* Internalization advantages such as contractual risk.

Overseas leasing is carried out by foreign financiers who either "export" the loaned

funds from the home country or establish a foreign leasing subsidiary. Furthermore,

foreign leasing subsidiaries are established through contractual arrangement, setting up a

new foreign entity, or by direct investment into a foreign leasing company. The authors

analyzed survey data consisting of responses from Presidents and CEOs of 97 U.S.

leasing firms on the effect of the OLI factors. The interacting relationships estimated by

Agarwal and Ramiswami were obtained from logistic regression and chi-square analysis.

The major findings were as follows: (1) large firms show a preference for

investment modes of entry in both low and high potential markets; (2) small firms with

limited multinational experience prefer the joint venture mode of entry in markets with

high potential; (3) firms with greater ability to develop differentiated products favor









foreign investment over exporting when contractual risks are high; and, (4) even in high

potential markets, substantial investment risk leads both small and large firms to export

instead of investing.

The results presented by Agarwal and Ramiswami highlight an important fact.

That is, although the empirical methods of each study mentioned above vary, they share a

common result. That is, host-country locational factors and their interactions with

ownership-specific factors play a key role in determining the MNC's choice of entry

mode (i.e., non-investment versus equity investment-based). The investment climate, or

at least investment risk, also appears to have a substantial effect on the decision of

whether or not to establish operations in a foreign location.

Decisions on Location

While many studies (including those cited thus far) have examined the effects of

ownership and internalization factors on the FDI decision, a large body of literature is

specifically devoted to analyzing the location decision. This is typically done by

modeling the cross-country distribution of FDI flows from a particular country (flows

from the U.S. and Japan have received a disproportionate amount of attention), groups of

countries (here, flows from the EU are the primary concern), or the entire universe of

countries. From a host-country's perspective, it is important to understand how firms

determine the geographic location of FDI. However, a randomly-selected group of

managers would likely point out a diverse collection of factors that affect this decision.

For instance, the manager of a labor-intensive firm may value low wage rates while the

management of a capital-intensive firm might tolerate higher wages in exchange for

lower interest rates. Nevertheless, analyzing the distribution of outward FDI flows does

provide some insight into how firms choose foreign investment locations.









Mody and Srinivasan (1997) analyzed the allocation of investment funds by

Japanese and U.S. investors by concentrating on the factors that caused investors to

differentiate between countries. They treat the FDI decision as a two-step process in

which the foreign investor first decides how much capital to invest abroad and then how

that pool of investment funds will be allocated across countries. Separate equations were

estimated for each supplier country, with the dependent variable specified as the share of

FDI host-country i received in each time period. The authors controlled for factors like

proximity in distance and methods of conducting business while testing for the effects of

location-specific variables like market size, the price of labor and capital, the corporate

tax rate, trade propensity, country risk, infrastructure, education and the existing stock of

FDI (the latter of which was included as a measure of persistence).

Several sets of regression coefficients were estimated. First, OLS estimators were

obtained, although the authors point out that these estimates are biased when unobserved

country effects are correlated with the observed explanatory variables. Second, the fixed-

effects model was used to eliminate the influence of unobserved country characteristics,

thus providing coefficients that "reflect [the] responsiveness of foreign investment to

changes within a country, over time" (Mody and Srinivasan 1997, p.784). Third,

between-estimators were provided with the intent to capture the variation in FDI shares

across countries4. While the authors recognize that each of these regression techniques

has the potential to result in biased estimators, they indicate that the results provide

different perspectives and can be used in conjunction to describe the variation in the data.

Finally, generalized least squares (GLS) estimators are obtained from a random-effects


4 Mody and Srinivasan (1997) interpret the fixed-effects estimators (or "within-estimators") as short-run
effects and the between-estimators as long-run effects.









model. The authors point out that the GLS estimates represent a weighted average of

within- and between-estimators, thus providing the best composite picture.

Country infrastructure and primary school enrollment rates (a measure of labor

quality) were found to be major influences on location only. Meanwhile, past investment

in a country affected both the timing and location of FDI. Corporate tax rates, the cost of

investment, country risk and wage inflation were not significant determinants of the

location decision, although the latter two did effect the timing of investment. Finally, the

authors found that the determinants of investment location for Japanese and U.S.

investors had converged over time.

As is the case with the model presented by Mody and Srinivasan, other studies on

the policy determinants of FDI location primarily concentrate on the effect of host-

country corporate tax rates. Loree and Guisinger (1995) indicate that this is primarily

due to the difficulty of capturing the incidence of other policy variables. However, data

compiled by the U.S. Department of Commerce and the Internal Revenue Service in 1977

and 1982 provided information on host-country investment incentives and performance

requirements, in addition to corporate tax rates. Loree and Guisinger (1995) tested the

effect of all three policy variables on the location of U.S. direct investment abroad. As

the authors point out, any empirical study that "attempts to model and observe the effects

of policies requires the inclusion of non-policy variables as controls of alternative

explanations" (p.285). Thus, they included GDP, infrastructure, and country risk as

independent control variables.

Loree and Guisinger made two interesting distinctions with regard to the

measurement of variables included in their model. First, the authors indicate that total









FDI flows include equity investment, reinvested earnings of foreign subsidiaries, and

other long- and short-term capital flows. Consequently, they specified the dependent

variable as the equity component of FDI, as opposed to total FDI flows. Second, they

point out that models intended to measure the response of FDI flows to changes in the

absolute level of the dependent variables often rely on the assumption that the stock of

FDI in each country is at equilibrium at the beginning of each period. Citing the

difficulty of maintaining such an assumption, the authors chose to measure the variables

in absolute levels.

Loree and Guisinger also include a dummy variable for the host country's status

as "developed" or "developing" (as classified by the OECD). They justify the inclusion

of this variable by pointing out that differences in the pattern of liberalization among

developed and developing countries suggest that the affects of policy variables in each

class may differ. So, in addition to testing the aforementioned relationships, the authors

tested for differential effects of policy variables among developed and developing

countries.

OLS coefficient estimates were reported for both policy and non-policy

determinants of U.S. FDI location in 1977 and 1982. While the results suggest that both

play a significant role, policy variables may be more important since they are able to be

altered quickly while non-policy variables take months, sometimes years to change

(Loree and Guisinger 1995). Although most of the non-policy variables had the expected

signs, only the effect of infrastructure was statistically significant in both years. The

authors indicate that the insignificance of the other non-policy variables is most likely









due to industry specific characteristics that can only be captured by segmenting the FDI

data.

However, the interaction of the policy variables and the development-status

dummy provide some interesting inferences. First, host country performance

requirements had a negative influence on equity FDI flows to both developed and

developing countries. Second, investment incentives had a positive influence on equity

FDI in developing host countries only5. Finally, the most significant effect was that of

the corporate tax rate, which was negatively correlated with U.S. FDI in both periods

irregardless of development status.

Woodward and Rolfe (1993) concentrated on export-oriented FDI activity in the

Caribbean Basin over a four-year period in the mid-1980s. The data available to the

authors allowed them to examine the effect of location-specific characteristics on

manufacturing plant openings in the region. It was assumed that investors in the region

would choose one of 16 location alternatives by analyzing the level of profit their firm

could achieve by locating in that country. Using conditional-logit analysis, the authors

found that GNP per capital, the length of tax holidays and the likelihood of an exchange

rate devaluation each had a large, positive, and statistically significant effect on locational

probability. Significant negative relationships were found for wage rates, restrictions on

the repatriation of profits, inflation and transport costs.







5 Loree and Guisinger indicate that this may be due to competitive response from other countries ultimately
"ending in the prisoner's dilemma trap where all countries increase their [investment] incentives
simultaneously but no country increases its relative share of foreign investment" (1995, p.296).









The Empirical Model

The empirical analysis in this study draws upon existing research into the strategic

behavior of foreign direct investors while attempting to shed new light on the effect of

the investment climate on FDI flows to developing countries. The model is constructed

in such a manner as to include a set of location-specific factors that are considered by the

entities that make FDIs. Relevance to host-country policymakers in the Western

Hemisphere is considered paramount.

Some researchers have suggested that the locational characteristics that influence

foreign direct investors vary according to the intended purpose of the investment. For

instance,

market characteristics such as size, growth in size and income level are most
relevant for investment that seeks to access the host market. Other types of
investment, such as export-oriented, may be more concerned with other locational
characteristics such as wage rates or policies regarding export levels. (Loree and
Guisinger 1995 p.295)

In a study on FDI in developing host countries, Lecraw (1991) distinguishes

between three types of FDI natural resource-seeking, market-seeking, and export-

oriented efficiency-seeking. However, he suggests that despite the distinct characteristics

and motives for each type investment, there exists a set of common locational factors

(Lecraw 1991) that influence all inflows of FDI irregardless of the type6. Lecraw's view

is adopted here. That is, the range of factors considered in this analysis are intended to

represent a set of conditions faced by all foreign direct investors irregardless of whether





6 Lecraw (1991) indicates that these factors include: a change in the corporate tax rate, the growth rate of
the labor force, a change in the openness of the host country's policy toward FDI, a change in the country's
risk rating, a change in the real exchange rate, and the growth rate of infrastructure.









the purpose of investment is to gain access to natural resources, the domestic market, or

to establish an export platform.

Many researchers have analyzed bilateral investment data as opposed to investment

from the entire universe of source-countries. The most often cited reasons for using this

approach are that the researchers are examining the effect of distance between the home

and host countries (i.e., the "gravity" approach), or that it is hypothesized that the

relationships vary by the nationality of the investor. The gravity approach is not

appropriate for this analysis given that the study analyzes flows of FDI from the entire

universe of sources. Furthermore, empirical results presented by Mody and Srinivasan

(1997) suggest that the factors that drive the location of FDI have converged over time

for at least two of the most prominent sources of direct investment funds in the region:

the U.S. and Japan. Given these findings, it is reasonable to think that the use of

aggregate rather than bilateral investment data does not hinder the validity of the results.

Foreign direct investment flows are measured in dollars and "therefore depend on

the size of the economy" (Lecraw 1991, p.171). Meanwhile, the independent variables in

this study are (at least theoretically) independent of economic size. This makes it

necessary to standardize the dependent variable across countries. Addison and Heshmati

(2003) point out that the tradition in existing literature is to specify the dependent

variable as FDI as a percentage of GDP.

The Host-Country Investment Climate

Examining the effect that the host-country investment climate has on FDI

essentially requires the inclusion of three types of variables on the right-hand side (RHS)

of the regression equation. The first group consists of governmental policies and

regulations that affect foreign investment. Restrictions on foreign investment, foreign









ownership, and access to foreign exchange (i.e., barriers to profit repatriation) serve as

indicators of the host government's attitude toward foreign investment. Factors that

affect the return on capital to foreign investors make up the second group. Variables of

this type are the corporate tax rate and changes in the exchange rate. The third group is

comprised of measures of political stability and corruption.

Basi (1966) surveyed 160 U.S. firms via questionnaire and asked managers to rank

15 potential determinants of FDI location as either "crucially important," "fairly

important," or "not important." In Basi's survey, a majority of U.S. firms ranked

political stability in the host country and the government's general attitude to foreign

investment as "crucially important." It was also clear that other components of the

investment climate, although not weighted as heavily as these two, received a

considerable amount of attention from multinational managers. Table 4-1 summarizes

Basi's findings on the importance of several investment climate factors.

Table 4-1. Survey Results on the Importance of Investment Climate Factors
Number of Firms Ranking Determinant as:
Crucially Fairly Not
Foreign Country Investment Climate Factor Important Important Important
(1) Favorable Attitude Toward U.S. Investments 76 69 13

(2) Political Stability 101 58 1
(3) Tax Structure 30 101 27
(4) Stability of Currency Exchange 61 90 8
(5) Inflationary Trends 35 99 21

The Policy and Regulatory Environment

Governmental policy towards foreign investment has attracted a disproportionate

amount of attention over the last 10 to 15 years. In fact, Lecraw (1991) points to several

examples where changes in host-countries' attitudes toward foreign investment led to









dramatic changes in the magnitude of inward FDI (such as in China and Korea during the

1970s and 1980s). Several measures have been proposed in the literature to proxy for the

regulatory environment foreign investors face when conducting business in a given

country. In response to the need for transparent methods of measuring and comparing

regulatory regimes across countries, an increasing number of private and public

institutions have begun to provide numerical ratings of the regulatory risks foreign

investors face.

The International Country Risk Guide (referred hereafter as ICRG) published by

the PRS Group encompasses one of the most comprehensive collections of data on the

economic, financial and political risks that investors encounter in developing countries.

The ICRG's composite political risk rating is comprised of twelve components measured

via subjective analysis of the available political information from individual countries.

Of the twelve components, the "investment profile" score provides the most relevant

measure of the regulatory environment facing foreign direct investors.

As defined by the ICRG (Sealy 2003), the investment profile is made up of three

risk factors: (1) the risk for contract viability/expropriation; (2) the risk for restrictions

on profit repatriation; and (3) the risk for payment delays. Each of the three factors is

assigned a score ranging from 0 to 4, with a score of 4 indicating "very low risk" and 0

indicating "very high risk." The three component scores are summed together to yield an

overall investment profile rating ranging from 0 to 12.

A score of 4 (the highest possible) for the first component is taken as an indication

that the host-country judicial system is likely to enforce contracts made between the

foreign investor and its domestic associates, and that there is little risk of unjustified









expropriation of assets by the host-country government. A high score for the second

component signifies that foreign parent-companies are relatively free to repatriate profits

and that the level of taxation on repatriated funds is relatively low. Finally, a high score

for the third component signals the existence of legislative protection against undue delay

of payment for services or products rendered. Using this methodology, the overall

investment profile score is expected to be positively correlated with FDI.

In order to better understand the ICRG investment profile rating, it is useful to

compare the cases of two countries in which the score differs significantly; Haiti had the

lowest investment profile score among the countries in the sample (at 5.5) in 2001, while

Uruguay had one of the highest (at 11.5). The inefficiency and lack of transparency in

Haiti's outdated legal system often hinder the resolution of disputes between foreign

investors and domestic parties. In fact, there have been more than 10 cases of

expropriation of private assets owned by U.S. interests in Haiti over the last 25 years.

While the Haitian government publically indicated its desire to alleviate these types of

situations, there has been little in the way of real action.

In contrast, property rights in Uruguay are recognized and protected by a well-

established and transparent legal system. In terms of dispute settlement, foreign investors

are usually given a choice of arbitration or court proceedings. Bankruptcy laws dictate

that creditors collect their debts first, followed by employees and then government.

There have been no instances of government expropriation in Uruguay in recent history.

Thus, it appears that the ICRG investment profile score does a good job of capturing the

actual state of affairs at least in the two example-countries examined here.









Political risk

Political risk assessment has been "one of the fastest growing areas of research in

international business ... [as] the discipline has flourished in the wake of the

international turmoil of recent years" (Simon 1984, p.123). Assessing differences in

political risk levels across countries is a difficult process that often requires highly

specialized information and expertise. Thus, many MNCs have again turned to outside

sources for assistance in analyzing the political risks inherent to developing host-

countries. The academic literature is peppered with econometric models that employ

political risk indexes composed by a variety of public and private entities. Business

managers and academic researchers often choose the source that provides a measure that

is most relevant to their purposes.

This study examines two primary sources of political risk in developing countries:

government instability and corruption. With regard to the first, Butler and Joaquin

(1998) state, "Political risk is the risk that a sovereign host government will unexpectedly

change the 'rules of the game' under which businesses operate" (p.599). While it is

possible for an incumbent regime to have a "change of heart" with regard to its position

on foreign investment, it is more likely that political risk arises from an actual regime

change. Thus, government stability serves to measure the risk that the "rules of the

game" will change.

With regard to the second source, it is sometimes unclear whether corruption in

local government serves as "a beneficial 'grease,' a minor annoyance, or a major obstacle

for international investors" (Wei and Shleifer 2000, p.303). However, to the extent that

corruption results in so-called "crony capitalism," it is possible that corrupt government

practices at least encouraged the recent Latin American currency crisis by facilitating









"the misallocation of financial resources to the friends and relatives of government

officials" (Wei and Shleifer 2000, p.304) rather than to their most productive uses.

It is possible that even in the most moderate cases "financial corruption in the form

of demands for special payments and bribes connected with import and export licenses,

exchange controls, tax assessments, police protection, or loans may force the

withdrawal or withholding of an investment" (Sealy 2003, p.A-5). In the most extreme

case, the uncovering of corrupt governmental activities could trigger a "popular backlash

[requiring a major restructuring,] or, at worst, a breakdown of law and order" (Sealy

2003, p.A-6). For all of these reasons, corruption is viewed here as a detrimental

characteristic of the existing regulatory regime.

As previously mentioned, there are twelve components to the ICRG composite

political risk rating. Scores for government stability and corruption are two of the factors

included in the composite rating and are used here as measures for the perceived level of

political risk. The score for political corruption ranges from 0 to 6, with a higher score

corresponding to lower risk. This score is reached through a subjective assessment of the

extent to which the following conditions exist in the political system: "excessive

patronage, nepotism, job reservations, 'favor-for-favors', secret party funding, and

suspiciously close ties between politics and business" (Sealy 2003, p.A-6).

The ICRG's government stability score is a measure of the government's ability to

stay in office and carry out its declared programss. As such, the score has three

subcomponents: government unity, legislative strength and popular support. Each

subcomponent is given a value of 0 to 4, a higher score indicating a higher level of unity,

strength and support. The overall government stability rating is the sum of the three









scores. The ICRG's risk ratings for corruption and government stability are each

expected to be positively correlated with FDI.

With regard to the government stability rating, it is useful to compare the cases of

Argentina and Uruguay. The ICRG government stability score for Uruguay was 11.0 in

2001, while Argentina received a rating of 5.5. Uruguay had a stable democratic

government, and there have been no instances of political violence in the recent past. On

the other hand, Argentina was plagued by social tension and political unrest. Violent

protests led to the death of at least 25 Argentine citizens from 2000 to 2002. Once again,

the ICRG rating seems to provide a reasonably good representation of the conditions that

persisted in the two countries used as examples here.

Factors that affect the return on investment

The paper by Stobaugh, Jr. (1969) illustrates how managers use the projected

return on investment (ROI) from a foreign project to discriminate between investment

alternatives. As was discussed in Chapter 3, the projected ROI is sometimes used as a

basis for passing on an investment opportunity altogether (referred to as the go- no go

decision), while in other cases managers might require a premium for risk in countries

that are perceived as riskier investment locations (the example used in Chapter 3 was the

risk of currency devaluation). In either case, factors that negatively affect the projected

ROI decrease the host country's probability of attracting foreign investment.

Two variables that directly affect the return to foreign investors are considered in

this study. These ROI factors are intended to measure location-specific attributes of the

host country that affect income in two ways. First, the domestic (i.e., within the host

country) ROI is affected by the host-country corporate tax rate. Second, the foreign









owner's ROI is subject to changes in the foreign exchange rate. The following

paragraphs elaborate on these two points.

The effect of income taxation is straightforward; ceterisparibus, a one percent

increase in the corporate tax rate leads to a one percent decrease in the after-tax profit of

a corporation. While the corporate tax rate may not have a significant impact on the

competitiveness of an individual firm (given that all firms residing in the host country

are faced with the same taxes), it may be used by governments as a tool to attract FDI

that might otherwise be directed at another country. In fact, some researchers have found

evidence that OECD countries do compete with each other over corporate tax rates in

order to attract foreign investment. Furthermore, Loree and Guisinger (1995) indicate

that MNCs may choose to allocate a disproportionate amount of investment capital into

countries with the lowest tax rates.

The most commonly used measure for taxation in the empirical literature is the top

marginal corporate tax rate, but it is often the case that foreign firms face different levels

of taxation. However, the definition of a direct investment enterprise provided by the

World Bank (Liberatori 2003) requires that the firm be classified as a resident entity of

the host country. As such, it is assumed here that direct investment enterprises are taxed

in a manner comparable to a domestically-owned firm. The top marginal corporate tax

rate is used in this analysis as the measure of taxation and is expected to be negatively

correlated with FDI.

The effect of exchange rate regimes on FDI has attracted a significant amount of

attention in the empirical literature. A stable and predictable exchange rate has the effect

of allowing managers of a parent-company to be confident that they will be able to









transfer funds to and from its foreign affiliate without substantial risk of loss due to

exchange. Furthermore, "uncertainty caused by exchange rate volatility ... makes the

domestic value of foreign revenues and costs uncertain" (Darby et al. 2000, p.1). That is

not to say that domestic firms are not-at-all affected by the exchange rate. Rather, this

study is more interested in how it affects the attractiveness of a host country from the

vantage point of a foreign investor.

Goldberg (1993) points out three theoretical effects of exchange rate volatility on

foreign investment: (i) sectoral profitability effects; (ii) location effects; and (iii) wealth

effects. First, sectoral profits are subject to exchange rate-induced changes in product

demand and cost, thus impacting the international competitiveness of industries. Next,

exchange rate volatility alters the attractiveness of domestic and foreign production

locations, and hence domestic and foreign investment levels. Finally, the distribution of

wealth across countries is affected by movements in the exchange rate, which in turn also

alters the demand for domestic and foreign investments.

With specific attention given to the location effect, Aizenman (1992) found that

aggregate investment was higher in countries with fixed exchange rate regimes than those

with a flexible exchange rate. This finding led him to suggest that "the adoption of a

fixed exchange rate could encourage flows of foreign direct investment" (Aizenman

1992, p.913). It is important to note that Aizenmen made this finding under the

assumption that foreign investors are risk-neutral. Although, it is expected that

introducing risk-adversity would only compound the negative effect of exchange rate

volatility on foreign investment. Aizenman's finding, as well as those presented in









previous sections, lead to the expectation that exchange rate volatility will be negatively

correlated with net FDI inflows in the ensuing analysis.

Control Variables

In addition to testing the effects of the factors mentioned above, two additional

variables are included on the right-hand-side of the equation as controls for alternative

explanations. Although not specifically categorized as elements of the investment

climate, both income and the level of integration with foreign economies (also referred to

as "openness") are considered as traditional determinants of FDI flows (Addison and

Heshmati 2003). Following the precedent set in the existing literature, productivity is

measured by per capital GDP. Total trade (i.e., imports plus exports) as a percentage of

GDP serves as a proxy for a country's degree of openness or integration with the global

economy. Each of the control variables is expected to have a positive effect on FDI.

A linear relationship between the explanatory variables and the dependent variables

is assumed. Chapter 5 addresses the validity of this assumption and compares two

alternative models that may be used to estimate the data: the fixed-effects model and the

random-effects model. A series of test statistics are calculated and serve as the basis for

choosing the most appropriate method of estimation. Finally, the response of FDI to

changes in the investment climate is examined and discussed for each country in the

analysis.















CHAPTER 5
MODEL SPECIFICATION, ESTIMATION, AND EMPIRICAL ANALYSIS

Data and Sources

The data for this analysis were obtained from various sources. The foreign direct

investment share of GDP, total trade share of GDP and per-capita GDP series were

obtained from the World Bank's World Development Indicators 2003 CD-ROM database.

Corporate tax rate data came from the World Tax Database maintained by the University

of Michigan Business School1. Exchange rate data were obtained from the International

Monetary Fund's (IMF) International Financial Statistics website2. Finally, the

investment profile, corruption and government stability scores were obtained from the

PRS Group's International Country Risk Guide (ICRG) data wizard3.

Although data on FDI were readily available for many of the developing countries

in the Western Hemisphere starting in 1970, annual risk ratings were much more difficult

to obtain. In fact, the availability of these ratings limited the sample size to 21 countries

observed over the 18-year period 1984-2001. Gross domestic product data for Nicaragua

in 1999, 2000 and 2001 were estimated by increasing the 1998 value by the growth rates

reported by the Economist Intelligence Unit. The total number of observations in the




1 The World Tax Database compiles corporate tax rate information from a variety of sources and was
accessed at http://www.wtdb.org/.
2 The IMF's International Financial Statistics database can be accessed at http://ifs.apdi.net/imf/.

3 The ICRG data wizard can be accessed at http \ \ \ .countrydata.com/wizard/. Data are available for a
fee.











Table 5-1. Descriptive Statistics by Country (1984-2001)
FDI Investment Corporate Tax Rate Government
(% of GDP) Profile (%0) Stability
Mean Std Dev Mean Std Dev Mean Std Dev Mean Std Dev
Argentina 1.962 1.911 5.750 1.517 30.500 5.102 6.806 2.023
Bolivia 3.987 4.264 6.472 2.841 25.833 1.917 6.611 2.725
Brazil 1.607 1.892 5.694 0.987 26.111 7.962 6.444 2.121
Chile 4.414 2.889 7.417 2.211 27.611 9.641 7.250 2.702
Colombia 2.010 1.121 5.750 1.517 33.111 3.341 6.556 1.917
Costa Rica 2.651 0.799 6.889 2.055 35.556 9.218 7.417 1.784
Dominican Republic 2.597 1.929 6.278 2.886 34.778 10.395 6.694 2.562
Ecuador 2.623 1.923 5.389 1.243 23.333 2.425 6.972 2.076
El Salvador 1.033 2.112 5.639 2.611 28.056 3.888 6.278 2.372
Guatemala 1.261 1.078 5.944 2.100 33.278 5.497 6.444 2.572
Guyana 8.018 9.945 6.056 1.697 48.889 6.978 6.250 2.451
Haiti 0.189 0.217 2.750 1.611 36.944 6.216 4.333 2.910
Honduras 1.787 1.208 6.194 1.582 32.222 9.111 6.306 2.408
Jamaica 2.836 2.412 6.639 2.127 35.253 4.485 6.917 2.302
Mexico 1.959 1.004 7.306 1.808 36.056 2.838 7.611 1.720
Nicaragua 3.430 4.383 4.472 1.398 33.694 5.396 6.944 2.363
Paraguay 1.183 1.006 7.139 1.747 30.000 0.000 6.667 1.749
Peru 1.899 2.380 5.861 2.071 34.444 8.024 6.417 2.451
Trinidad & Tobago 5.679 4.796 6.972 1.803 40.167 4.396 6.778 1.833
Uruguay 0.586 0.508 7.417 1.972 30.000 0.000 7.167 2.036
Venezuela 1.852 1.865 5.278 1.526 40.444 8.906 7.194 1.758
Sample 2.551 3.535 6.062 2.159 33.156 8.342 6.669 2.290
Exchange Rate Trade Per-capita GDP
Corruption Variability (%) (% of GDP) (US$)

Mean Std Dev Mean Std Dev Mean Std Dev Mean Std Dev
Argentina 3.361 0.724 437.017 857.215 18.145 3.546 7,182.403 779.110
Bolivia 2.222 0.732 876.655 2,367.956 46.890 3.150 878.980 58.705
Brazil 3.500 0.618 463.218 558.691 18.466 3.492 4,296.506 235.791
Chile 3.333 0.485 14.669 13.037 59.003 4.790 3,959.624 1,059.139
Colombia 2.611 0.608 21.663 9.293 33.747 4.071 2,181.747 185.551
Costa Rica 4.889 0.471 12.495 4.386 78.088 10.789 3,234.244 419.880
Dominican Republic 3.222 0.548 23.659 33.641 66.303 5.991 1,575.407 249.774
Ecuador 2.972 0.436 46.817 24.846 56.924 7.165 13,519.929 4,008.081
El Salvador 2.750 0.809 8.900 16.008 53.900 9.336 1,524.806 180.581
Guatemala 2.611 0.916 13.913 17.555 40.560 6.677 1,423.258 89.060
Guyana 1.944 0.998 34.741 41.700 186.476 55.493 800.264 107.505
Haiti 1.333 0.840 9.781 14.476 39.201 8.287 433.151 69.767
Honduras 1.972 0.118 13.829 18.849 76.605 18.874 699.081 17.854
Jamaica 2.389 0.502 23.323 25.680 102.561 10.067 2,129.110 158.978
Mexico 2.889 0.471 36.308 39.946 44.711 14.202 3,315.535 234.241
Nicaragua 4.444 0.856 15,341.711 42,745.495 80.364 30.387 473.192 68.740
Paraguay 1.389 1.037 22.985 16.456 66.936 12.738 1,820.112 63.172
Peru 3.000 0.343 584.282 1,302.757 31.144 4.517 2,225.213 219.885
Trinidad & Tobago 2.833 0.383 6.097 8.208 83.601 14.255 4,468.905 436.402
Uruguay 3.000 0.000 45.462 29.240 41.069 3.169 5,639.432 768.806
Venezuela 2.944 0.236 37.774 25.338 48.480 7.111 3,472.989 142.421
Sample 2.839 1.041 860.729 9,661.129 60.627 38.776 3,107.328 3,057.545









resulting balanced data set is 3024. Table 5-1 presents some descriptive statistics on the

data used in the analysis.

Model Specification and Estimation

Panel estimates were obtained by regressing the dependent variable against a set

of explanatory variables for the 21 countries listed in Table 5-1. The basic regression

equation takes the form

Y = a + X,,, + E k Z,,, + E, (5-1)

where Y, represents "net inflows of FDI as a percentage of GDP" for country i (i=

1,2,...,N) in period t (t= 1,2,...,7); a is a constant term, and / and 0 are Kxl and

J x 1 vectors of unknown parameters to be estimated, respectively. The variable X

represents a vector of investment climate determinants of FDI and Z is a vector of

control variables, each of which varies in both country and time dimensions. The error

term ,, in the basic panel regression model has two components and can be expressed as

E,t = A, + V,t, (5-2)

where u, represents the variation unique to the cross-section and u,, is white noise.

Greene (2000) indicates that there are two frameworks commonly used to

generalize this basic model. First, the fixed-effects model, which is used to obtain

ii ithin-estimators, takes a as constant over time but specific to each country. The

dependent variable in the fixed-effects model is regressed against the difference between

the observed value of each of the explanatory variables and its mean, thus eliminating the

effect of unobserved country (or "fixed") characteristics. Thus, the fixed-effects

regression equation can be expressed as









[Y,, -,.]= a, + ,j ,[Xj,, x,.] + IOk[Zk, -Zk,.] + I, (5-3)

where ,. is the mean of country i 's FDI inflows, X,,. is a J x 1 vector of means on X,,

and Z,. is a K x 1 vector of means on Z,,, taken over the T observations respectively.

The intercept term (a, ) in this model varies by country and thus is an unknown

parameter to be estimated. Furthermore, a, is interpreted as the mean residual in each

country i and can be expressed as

a, = FDI.- J J XJ,. kq Z,. (5-4)

Equation 5-4 is unique to each country in that any change in the constant term represents

a parametric shift. The error term c,, in the fixed-effects model is defined as follows:

s, =Y,, -a, -Y,8,X X,, k Zk,,

=Y, -(,. -j,/jX, --k AZk.*) J8JXj,, -kAZkit

= [L, Y,]- I Y,,[X ,, X,.]- IY [Zk,, -Zk- ]. (5-5)

Second, the variance components model is used to obtain random-effects

estimators. These estimates are a weighted average of between- and ii ithin-estimators,

and thus reflect "both the influences across and within countries" (Mody and Srinivasan

1997, p.785). The intercept term in this model is not allowed to vary by country, hence

the estimated coefficients reflect the average country in the sample. Random-effects

estimators are derived from the following reformulation of the basic regression equation:

Y, = a + P,X,, + Zk Z,, +u, + (5-6)

In Equation 5-6, u, "is a random disturbance characterizing the ith observation and is

constant over time" (Greene 2000, p.568).









The random-effects model involves a transformation similar to the one made in the

fixed-effects model with one exception; when the mean is subtracted from each

observation, it is weighted by 1 A, where A is defined as

OE
2
A C- (5-7)
(o2 + T-)

In Equation 5-7, ao is the variance of the basic error term (sE,), a2 is the variance of the

country-specific error term (u ) and Tis the number of years.

Some extreme cases should be explained purely for example. If the value of A

from the random-effects model is equal to 1, then the random-effects model is ordinary-

least-squares and the classical regression model defined by Equation 5-1 applies. At the

other extreme, a value of 0 for A would indicate that all of the variation in the data is

unique to the cross-section of countries. If this were the case, then the random-effects

model would be identical to the fixed-effects formulation.

Ordinary-least-squares (OLS) estimates were obtained first and serve as a basis

against which to compare the alternative methods of estimation. Following Greene

(2000), the first hypothesis test was for the existence of individual-country effects in the

data. The Lagrange multiplier (LM) test statistic4 from the least-squares regression took

a value of 66.3939, much larger than the critical value of 3.845. Thus, the null hypothesis

of no individual-country effects was rejected.

An F test comparing the fixed-effects estimators to those from the classical OLS

regression further suggest that individual-country effects exist in the data. The null

4 This statistic is part of the standard TSP output. A more detailed explanation of how the statistic is
computed can be found in Greene (2000).

5 Chi-square (95%)









hypothesis in this case is similar to the first LM heteroscedasticity test performed (i.e.,

there are no individual-country effects in the data, and hence there is no difference

between OLS and the fixed-effects model. The computed F statistic was 6.267, which

exceeds the 99% critical value of 1.88 for F(20,350). As such, there is strong statistical

support for the existence of country-specific effects.

Given the existence of individual-country effects in the data (and consequently, the

conclusion that the classical regression equation is inappropriate), the LM test statistic

based on the residuals from the fixed- and random-effects models was used to determine

the need to estimate heteroscedastic-consistent standard errors for each of the models.

The test statistics took values of 133.363 and 103.887 for the fixed- and random-effects

models, respectively, leading to a rejection of the null hypothesis of no heteroscedasticity

and indicating the need to compute heteroscedastic-consistent6 standard errors.

By estimating a set of robust standard error matrices for the fixed- and random-

effects models, it is possible to compare the two models and choose the most appropriate.

The preferred method is the Hausman test for correlation between the individual-country

effects and the other regressors in the equation (Greene 2000). If the individual effects

are correlated with the other regressors, the fixed-effects model is the most appropriate.

Conversely, if the hypothesis that the two are uncorrelated cannot be rejected, then the

random-effects model is the better choice. The Hausman test is based on the Wald

criterion and is asymptotically distributed chi-squared. The test statistic of 0.63254 was

considerably less than the 95% critical value of 2.167 for the chi-square distribution with



6 The ROBUST command in TSP performs this computation even if the nature of the heteroscedasticity is
unknown.









the necessary degrees of freedom. Thus, the null hypothesis of no correlation could not

be rejected, and the random-effects model was deemed to be the most appropriate.

The Box-Cox transformation was made to several explanatory variables that were

hypothesized to have a non-linear relationship with the dependent variable. The

following transformation was made to the exchange rate, trade, corporate tax and per-

capita GDP variables:

x3-
x =- (5-8)


where the optimum value of 3 for each variable was estimated by scanning the range of

values between -2 and 2 (in increments of 0.1) and maximizing the log-likelihood

function of the fixed-effects regression equation. However, making the transformation to

these variables did not have a significant impact on the regression results and in each case

the linear model performed at least as well as the non-linear specification. Thus, the most

appropriate approach was to estimate the random-effects regression equation under the

assumption of linearity in the independent variables.

Results of Estimation

The parameter estimates and their associated standard errors can be seen in Table

5-2. Overall, the results suggest that the model performed well. The R2 value of .424

indicates that the right-hand-side (RHS) variables in the model explained nearly half of

the variation in the data. This can be considered a relatively high R2 given the size and

cross-sectional nature of the panel data set used in this analysis. Six of the seven

explanatory variables were statistically significant and only one took an unexpected sign.









Table 5-2. Random Effects Regression Results
Variable Coefficient Error t-statistic P-value
Investment Profile 0.00156060 0.00089697 1.679 0.093
Corporate Tax Rate (0.05498600) 0.02076900 (2.648) 0.008
Government Stability 0.00256093 0.00079045 3.240 0.001
Corruption (0.00019929) 0.00182247 (0.109) 0.913
Exchange Variability (0.00002477) 0.00001345 (1.842) 0.066
Trade 0.07202500 0.00599952 12.005 0.000
Per Capita GDP 0.00000160 0.00000084 1.890 0.059
Constant (0.03032100) 0.01162300 (2.609) 0.009

The investment profile score had a positive impact on FDI and was significant at

the 10% level. The parameter estimate of 0.0015 suggests that a one-unit improvement in

the investment profile score would have the effect of increasing the FDI share of GDP by

0.15%. Based on the definition provided by the ICRG (International Country Risk

Guide), the investment profile score takes into consideration the risks associated with

profit repatriation, contract viability and expropriation, and delay of payment. The score

would increase (or improve) if a the ruling regime in a country were to

* ease the restrictions on profit repatriation by allowing a greater percentage of
profits to be transferred out of the host country or by lessening the tax (or other)
penalties associated with such a transfer

* enact tougher legislation to ensure the enforcement of legal contracts and limit the
possibility of excessive delays in payment for goods or services rendered

* decrease the government's right to expropriate funds or other property belonging to
foreign investors.

Thus, the positive relationship between the investment profile score and FDI is not

surprising.

The government stability score was the most significant of the investment climate

variables examined. As defined by the ICRG, the government stability score is a measure

of government unity, legislative strength and popular support. The rating is intended to









provide an indication of the cohesiveness of the ruling regime and the extent to which

opposing parties pose a threat to its ability to remain in power. A one-unit increase in the

government stability score was estimated to have the impact of increasing the FDI share

of GDP by 0.26%. Thus, the results suggest that the potential for change in the host-

country regulatory regime is of substantial concern to foreign investors.

The amount of variability in the exchange rate had a negative effect on FDI,

although the magnitude of this effect was rather small. The exchange rate variable, as

specified in the model, does not reflect whether the change is due to an appreciation or

depreciation in the host-country currency. Rather, it simply measures the level of

stability in the local currency's value against the US dollar. Darby et al. (1999) showed

that exchange rate volatility leads to uncertainty in the future cash flows of a proposed

foreign investment project, and in turn, to a greater likelihood that the investment project

will either be postponed or rejected altogether. However, the results of this analysis

suggest that a 1% increase in exchange rate variability decreases the FDI share of GDP

by only 0.002%. It is noted that the risk of changes in the foreign exchange rate is a risk

that can be hedged in international financial markets (e.g., through the use of currency

swaps and futures contracts), and that this fact may account for the lack of response in

FDI to changes in the variable. The exchange rate variable was significant at the 10%

level.

The parameter estimate for the corporate tax rate suggests that higher taxation is a

significant deterrent to foreign investment. Decreasing the corporate tax rate by 1% had

the effect of increasing the FDI share of GDP by as much as 5.5%. Furthermore, the tax

rate variable was significant at the 1% level.









The negative coefficient for corruption indicates that as the risk for political

corruption increases the FDI share of GDP increases. However, the variable was not

statistically significant. Nonetheless, the existing theoretical literature suggests that

corruption can serve as a benefit to foreign investors in the sense that they often posses

the capital necessary to "buy favors" from governmental agents. Although the regression

results seem to offer weak support for this contention, the most obvious inference is that

the level of political corruption does not appear to be a significant determinant of FDI in

the sample countries.

Finally, each of the control variables had a positive impact on the FDI share of

GDP. The trade variable was significant at the 1% level and the magnitude of its impact

was 0.072. This result suggests that countries that trade more also attract more FDI as a

share of GDP. Per-capita GDP was significant at the 10% level, although the magnitude

of its effect was much smaller. A $1,000 increase in per-capita GDP was estimated to

increase the FDI share of GDP by only 0.0016%.

Although the parameter estimates presented in Table 5-2 appear to be small in

magnitude, their corresponding effect on the absolute level of FDI is considerable in

many cases. This is important because it is the impact in terms of dollars with which this

study is ultimately concerned. As shown below, simple algebra can be used to reveal the

effect of the independent variables on the actual dollar level of FDI in each of the

countries in the sample.

The dependent variable is specified in the regression equation as the FDI share of

GDP for two reasons. First, as pointed out by Lecraw (1991), because the magnitude of

the explanatory variables is not dependent upon economic size, there is a need to









standardize the dependent variable (as defined in this study) across countries. This is

most often accomplished by dividing FDI by GDP. An alternative approach is to include

GDP as a discrete independent variable on the RHS of the regression equation. However,

the FDI and GDP variables were correlated with each other and GDP became the

dominant variable in the regression equation, detracting from the significance of the other

variables in the model. The regression equation was therefore estimated in shares and the

effect of each variable on the dollar level of FDI was then extracted from the model by

multiplying through by GDP as follows:

FDI,J/GDP, = a, + f/ X,,, + Yk Zkt (5-9)

FDI, =a, (GDP,t) + Xj,t (GDP,) + k k Zkt (GDP,t) (5-10)

Equation 5-10 shows that although the estimated parameters are constant across

countries, the magnitude of change in each country's level of FDI is unique to that

country. The remainder of this chapter is devoted to analyzing the relationship between

FDI and hypothetical changes in the investment climate variables in each country.

Equation 5-10 was used to create a benchmarking scenario where the parameter estimates

were applied to the most recent data available (2001). The implied impacts on FDI of

ceterisparibus changes to the explanatory variables in the model are presented for each

country in the analysis.

For the variables that consisted of a rating or score, the effect of a one-unit

improvement in the score is considered. For the corporate tax rate, trade and per-capita

GDP variables, the effect of a 10% increase is shown. Finally, the reaction of FDI to

eliminating any variability in the foreign exchange rate in other words, pegging the

local currency to the US dollar is considered. In each case, the change in the dollar