1 IMPLICATIONS OF DEVELOPED NATION STANDARDS REGIMES FOR AGRO FOOD TRADE AND RURAL DEVELOPMENT: CASE INVESTIGATIONS OF WELFARE IMPACTS IN VEGETATIVE CROPS By JASON SCOTT ENTSMINGER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012
2 2012 Jason Scott Entsminger
3 To all those who know the joys and discomforts of an agricultural life ; and to my family both near an d far, both friend and relative
4 ACKNOWLEDGMENTS First, great thanks are owed to my committee members from the University of Florida, Dr. John J van Sickle and Dr. Jeff Burkhardt, whose generous s upport, earnest encouragement and immense faith were so greatly appreciated, especially when the going got rough on this project. Thanks also go to Dr. Harald von Witzke of the Humboldt Universitt zu Berlin who served as my promoter within the I nternation al M aster of R ural D evelopment Programme Support staff and faculty at University of Florida, Ghent Universiteit, Humboldt Universitt zu Berlin, Agrocampus Ouest and Universit degli Studi di Pisa are all deserving of thanks for the individual contributi on they made during my program of study to both my general education and well being, and to this project through comments, discussions and encouraging words. Included among these are Dr. Dieter Kirschke of HUB and Dr. Guy Durand of A grocampus O uest who offered excellent commentary in their own fields of expertise. Jessica Herman and Renate Judis deserve highest recognition f or all the coordinating work they have done in support of us in ATLANTIS and me specifically Without them I truly would have been often lost. Thanks are also due to the US Department of Education and it s FIPSE program which provided financial support via a mobility grant. Finally, I owe so much to my family and to my friends who are lik e family, both old and new, in all corners of the globe. They were the ones who enabled me to grow professionally and personally and have taught me the lessons of hard work and charity. For all this and so much more I am truly humbled and immensel y grateful.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 13 Settin g the Stage ................................ ................................ ................................ .... 13 Problem Statement and Objectives ................................ ................................ ......... 19 Charting the Course ................................ ................................ ................................ 21 2 BACKGROUND ................................ ................................ ................................ ...... 31 The Role of the State and Motivations for Engaging in Development Facilitation ... 31 Trade as an Avenue for Development ................................ ................................ .... 34 Technical Barriers to Trade ................................ ................................ ..................... 39 Theoretical Considerations of SPS Measures as Trade Barriers ............................ 41 Tariff Equivalent Model ................................ ................................ ..................... 42 Supply Shift Model ................................ ................................ ........................... 44 Demand Shift Model ................................ ................................ ......................... 45 Segmented Markets Model ................................ ................................ ............... 46 Issu es Faced by Developing Nations Related to SPS Measures ............................ 46 SPS measures, Development and Structural Change ................................ ............ 51 Statement of Hypotheses ................................ ................................ ........................ 53 3 DEVELOPED NATION SPS POLICY FRAMEWORKS ................................ .......... 66 European Union ................................ ................................ ................................ ...... 68 France ................................ ................................ ................................ .............. 72 Germany ................................ ................................ ................................ ........... 74 Theorized Implications for Third countries ................................ ........................ 76 United States of America ................................ ................................ ........................ 78 US Regime ................................ ................................ ................................ ....... 78 Theorized Implications for Trading Partners ................................ ..................... 83 4 OVERVIEW OF CASE STUDY MARKETS ................................ ............................ 86 Nicaragua ................................ ................................ ................................ ............... 87 Vietnam ................................ ................................ ................................ ................... 90
6 Pineapple ................................ ................................ ................................ ................ 92 Crucifers (Broccoli and Cauliflower) ................................ ................................ ........ 93 5 ANALYSIS OF PILLAR I: MACRO LEVEL AGGREGATED WELFARE ANALYSIS ................................ ................................ ................................ ............ 101 Methodological Background ................................ ................................ .................. 101 Gravity Models ................................ ................................ ............................... 102 Partial Equilibrium Models ................................ ................................ .............. 105 General Equilibrium Models ................................ ................................ ........... 108 Selected Methodology: An Explanation ................................ ................................ 111 Basic Assumptions ................................ ................................ ......................... 113 Calculating Welfare Impacts ................................ ................................ ........... 115 Calculating Model Parameters ................................ ................................ ....... 120 Domestic demand parameters ................................ ................................ 122 Domestic supply parameters ................................ ................................ .... 123 Equilibrium points ................................ ................................ ..................... 124 Calculating Rate of Protection Equivalent ................................ ...................... 126 Data ................................ ................................ ................................ ...................... 127 Domestic Supply ................................ ................................ ............................ 128 United States ................................ ................................ ........................... 128 Europe ................................ ................................ ................................ ..... 128 Vietnam and Nicaragua ................................ ................................ ............ 130 Domest ic Demand ................................ ................................ .......................... 132 United States ................................ ................................ ........................... 132 Europe ................................ ................................ ................................ ..... 133 Vietnam and Nicaragua ................................ ................................ ............ 134 Additional Data Used to Transform, Calculate Tariff Rate and Transport Costs ................................ ................................ ................................ ........... 136 Results ................................ ................................ ................................ .................. 137 Elasticities and Model Parameters ................................ ................................ 137 Domestic supply elasticities ................................ ................................ ..... 138 Domestic demand elasticities ................................ ................................ ... 140 General discussion on elasticity estimation results ................................ .. 141 Selecting other model parameters ................................ ........................... 143 Partial Equilibrium Results ................................ ................................ .............. 144 Unintended results ................................ ................................ ................... 14 4 Tariff r ate equivalents ................................ ................................ .............. 145 Producer surpluses of developing nations ................................ ............... 146 Sensitivity Analysis on Elasticities ................................ ................................ .. 147 Discussion ................................ ................................ ................................ ............ 148 6 ANALYSIS OF PILLAR II: POTENTIAL FOR MICRO LEVEL ANALYSIS OF STRUCTURAL CHANGE IMP ACTS ON HOUSEHOLD WELFARE .................... 162 Methodological Background ................................ ................................ .................. 162 Proposing a Methodological Approach ................................ ................................ 163
7 Identifying Potential Data Sources ................................ ................................ ........ 165 Discussion ................................ ................................ ................................ ............ 166 7 SUMMARY, IMPLICATIONS AND CONCLUDING REMARKS ............................ 168 Limitations of the Adopted Methodology ................................ ............................... 169 Perfect Substitutes Assumption: Addressing Imperfect Product Substitution 169 Shift Parameters in Calculating Supply Curves ................................ .............. 171 Issues Related to the PE Market Model ................................ ......................... 172 Data Intensity and Approximation ................................ ............................ 173 Implications ................................ ................................ ................................ ........... 174 APPENDIX A KEY DEFINITIONS RELATED TO THE SPS AGREEMENT OF THE WTO ........ 175 B PRINCIPLES OF EU FOOD SAFETY LAW ................................ .......................... 176 C PROOF OF EXCESS SUPPLY/DEMAND ELASTICITY ................................ ....... 178 D META DATA SUMMARY OF ELASTICITY MEASURES ................................ ..... 179 E LISTING OF POTENTIAL VARIABLES FOR USE IN HOUSEHOLD LEVEL ANALYS IS ................................ ................................ ................................ ............ 180 LIST OF REFERENCES ................................ ................................ ............................. 182 BIOGRAPH ICAL SKETCH ................................ ................................ .......................... 196
8 LIST OF TABLES Table page 2 1 Empirical results on ODA mot ivations ................................ ................................ 55 2 2 Total value of trade in fruit and vegetable exports ................................ .............. 64 2 3 Total volume of trade in fruit and vegetable exports ................................ ........... 65 4 1 Summary of selected markets for investigation ................................ .................. 96 5 1 Results of econometric models estimating elasticities ................................ ...... 154 5 2 Imposed elasticities used for base welfare analysis and Average Year prices and quantities for calibration ................................ ................................ ............. 155 5 3 Tariffs and tariff equivalents of insurance and freight costs .............................. 156 5 4 Results of initial partial equilibrium analysis in crucifer markets ....................... 157 5 5 Results of spatial partial equilibrium analysis of SPS measures in pineapple markets. ................................ ................................ ................................ ............ 158 5 6 Results of sensitivity analysis, changing price elasticities of s upply for Vietnam and Nicaragua ................................ ................................ .................... 159
9 LIST OF FIGURES Figure page 1 1 Growth rates, exports versus G ross N ational I ncome per capita in L atin A merica and the C aribbean developing nations ................................ ................. 23 1 2 Growth rates, exports versus G ross N ational I ncome per capita in S ub S aharan A frican developing nation s ................................ ................................ ... 24 1 3 Growth rates, exports versus G ross N ational I ncome per capita in E ast A sia and the P acific developing nations ................................ ................................ ..... 25 1 4 Exports as a percent of G ross D omestic P roduct in developing nations ............ 26 1 5 Exports as a percent of G ross D omestic P roduct in select developed nations ... 27 1 6 G ross N ational I ncome per capita in developing nations over time .................... 28 1 7 G ross N ational I ncome per capita, comparing developing to developed nations ................................ ................................ ................................ ................ 29 1 8 Growth rates in G ross N ational I ncome per capita on selected groupings ......... 30 2 1 Trends in O fficial D evelopment A ssistance for select ed governments ............... 57 2 2 Trends in O fficial D evelopment A ssistance as a percent of G ross D omestic P roduct for selected government s ................................ ................................ ...... 58 2 3 Standards in the agri food sector ................................ ................................ ........ 59 2 4 S anitary and phytosanitary measures under a tariff equivalent model ............... 60 2 5 S anitary and phytosanitary measures under a supply shift model ...................... 61 2 6 S anitary and phytosanitary measures under a demand shift model ................... 62 2 7 S anitary and phytosanitary measures under a segmented market model .......... 63 3 1 Forms of government food safety regulation. ................................ ..................... 85 4 1 Political map of the Republic of Nicaragua ................................ ......................... 97 4 2 Regional map of Central America indicating Nicaragua ................................ ..... 98 4 3 Political map of the Socialist Republic of Vietn am ................................ .............. 99 4 4 Regional map of Eastern Asia indicating Vietnam ................................ ............ 100
10 5 1 Conceptual framework of partial equilibrium modeling in agricultural markets 151 5 2 Import market sh owing trade and welfare effects of removing an SPS measure ................................ ................................ ................................ ........... 152 5 3 Import market showing decomposed price wedges with an SPS measure ...... 153 5 4 S anitary and phtosanitary regime tariff equivalents at differing supply elasticities ................................ ................................ ................................ ......... 160 5 5 Percentage losses in consumer surplus at differing supply elasticities ............. 161
11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science IMPLICATIONS OF DEVELOPED NATION STANDARDS REGIMES FOR AGRO FOOD TRADE AND RURAL DEVELOPMENT: CASE INVESTIGA TIONS OF WELFARE IMPACTS IN VEGE TATIVE CROPS By Jason Scott Entsminger December 2012 Chair: John J van Sickle Major: Food and Resource Economics Trade has historically been targeted as a way out of poverty and towards economic stability for developing nations. In recent years the export of fresh fruits and vegetables has been especially targeted to achieve development in agricultural communities. Increasingly, however, these exports face more stringent sanitary and phytosanitary (SPS) requirements abroad which act as technical barriers. This thesis applies the price wedge methodology within a spatial partial equilibrium to calculate the tariff equivalent of SPS measures These are then us ed to examine changes in producer surplus of exporting nations. Analysis is within the market for fresh pineapples exported from Vietnam and Nicaragua to the United S tates and Europe. It finds, overall that European SPS regimes present higher effective rates of protection and have more negative impacts on producer surplus among nations producing for export than those of the United States. The market for broccoli and cauliflowers is also investigated, but due to model outcomes tariff equivalents of SPS regimes are not calculated. In order to implement the spatial partial equilibrium model, this work first estimates econometrica lly elasticities of supply and demand for most of the selected market s. It also pre sents an
12 avenue for more in depth analysis in the future which examines the welfare impacts on these producers as a result of structural change associated with adaptation to SPS measures in export markets abroad.
13 CHAPTER 1 INTRODUCTION Setting the Stage Trade has increasingly become the paramour targeted by economists and world policy makers alike as an avenue for the promotion of development and economic growth. Many in the realm of policy making and economic research have highlighted export led develop ment regimes as the most appropriate path 1 to follow, and it is this orientation which is still en vogue today. To this end, the liberalization of trade policies, especially removing those classical policy mechanisms which act as direct barriers to market access, has become a prominent theme within the international community and now stands for many as a central tenet of any framework for development, both in static (short and medium run) and dynamic (long run) terms. Numerous researchers have published wo rks seemingly indicating that the triptych of export trade economic growth poverty alleviation is supported both theoretically and empirically. (Bhagwati & Srinivasan, 2002) Still, there are detractors from this camp who make v alid, if not differently focused, arguments to the contrary. In fact, the diverse outcomes and implicit assumptions within empirical works have led many researchers especially during the heyday of relevant discussions to conclude that evidence is simply inconclusive. (Goldberg & Pavcnik, 2004; Greenaway, Morgan, & Wright, 2002) Anecdotally we can consider data such as t hat presented in Figures 1 1, 1 2 and 1 3 taken from the World we consider historical trends in growth rates 2 for exports as a proportion of G ross D omestic P roduct (GDP) and G ross N ational I ncome 1 Although not unanimously. For more discussion on this evolution, see Chapter 2. 2 The growth rate is defined as [(Value in period n Value in period n 1) / (Value in period n 1)]*100
14 (GNI) per capita for nations classified as developing in Lat in America and the Caribbean, Sub Saharan Africa and East Asia and the Pacific respectively. While it clearly appears that a correlated lag might exist, there is also likely to be much noise within the relationship. Most especially there are times when the two rates, even accounting for a time lag of a couple years, can change in dramatically different proportions. Consider, as an example, the case of Latin American data where, in the late 1970s and early 1980s a decrease in the exports growth rate is associated with a dramatically and proportionally larger decrease i n GNI per capita growth rates in the lag period, while, conversely, in the early 90s a very large spike in export growth rate is associated with relatively little change in GNI per capita growth rates. This would support the mixed results found within more rigorous, empirical analyses as contained within the body of literature published on the subject. In the general case, Figure 1 4 shows, that since the 1960s, the role of exports as a proportion of GDP has also grown within the developing world, but that some regions have shown dramatic increases (East Asia and the Pacific) while others moderate ones (Sub Saharan Africa). Imme diately following, in Figure 1 5 is a similar graphic for three industrialized nations the US, France and Germany as well as th e average of O rganization for E conomic C ooperation and D evelopment (OECD) members. Again, the arc of trade liberalization is evidenced in the relative increase of exports as a share of GDP in the industrialized nations as well. If we consider GNI per cap ita over time for deve loping nations, as in Figure 1 6 a relative improvement is also evidenced. Unfortunately, this fails to place the situation in the greater context. Figure 1 7 does this; here we represent GNI per capita growth in the same 4 industri alized areas compared
15 with the developing nations. The contrast is striking! Wh ile growth rates (see Figure 1 8 ) have been roughly comparable, the net effect in developed nations far outstrips that in the developing world, where growth presumably driven in large part by trade has brought contemporary GNI per capita up to the same levels that the industrialized world this data set, has widened by untenable margins. However, even if we accept the evidence that export led trade can improve development pathways in not only the long run but the short run as well, the underlying premise is often based on macro level, aggregated analysis, such as the GNI per capita meas ure presented here. This can be even more misleading, especially when one re (MDGs) 3 Some evidence points to the case where liberalization may induce overall economic growth yet is associated with simultaneous increases in distributional inequality of income effects. (Goldberg & Pavcnik, 2004; Ravallion, 2003) Literature on poverty impacts is less predomin ant mostly due to technical obstacles and lack of sufficiently robust data sources. This is a concerning aspect in terms of policy decision making, especially if arnings patterns and are (Hertel, Ivanic, Preckel, & Cranfield, 2003, p. i) Much of the literature is clear that micro level analyses, employing household or similar data, are vitally important in truly understanding the broader implications hidden or glossed over by aggregated met rics. While there still exist a 3 We recall here the MDG to reduce the incidence of poverty by 2015.
16 large amount of debate about practicalities and individual methodologies, what is becoming clea rer and clearer is that impacts are highly heterogeneous, with winners and losers amon gst the rural poor and outcomes highly differentiated across different social, political, economic and geographic attributes. Even more importantly, much of the research that was conducted on the implication of trade liberalization for growth, poverty, and inequality has focused (1) on unilateral liberalization by developing nations 4 (2) within the context of manufacturing or other, similar industrial industries, and (3) on non technical, traditional barriers to trade, such as tariffs. Recent evidence is mounting that non tariff barriers (NTBs), and most predominately technical barriers to trade (TBTs), have strikingly different impacts. Moreover, the preponderance of T BTs are imposed by developed nations and that, even when legitimate, they have a disproportionate effect on developing nations producing for export. (Disdier, Fontagn, & Mimouni, 2008) Included among these TBTs are measures ai med at protecting the health and safety of people, animals and the environment, that is, Sanitary and Phytosanitary (SPS) measures, which specifically impact agro food supply chains. These SPS measures typically apply to agro food products, both processed and un processed. They can manifest themselves as requirements on the process of production, which must be certified in the field, or as direct limits on the product itself, assessed at the border in the customs house. The literature reviewed for this proj ect makes it clear too, that SPS measures have become more stringent in the most recent decade for myriad reasons. As one World Bank report 4 While some of this was completely voluntary, a very large proportion of this unilateral liberalization was as a condition to receiving loans and other Official Development Assistance. (Greenaway & Milner, 1993)
17 into new areas, in the wake of a spate of food scares in industrialized countries, and in the face of increased scientific knowledge, official concerns over bio terrorism, and ( full citation here p. 3) Evaluating SPS impacts is of particular impor tance to the argument for trade as an avenue for development and poverty reduction; developing nations typically hold a comparative advantage in industries especially agricultural ones that demand endowments in unskilled or semi skilled labor, land and c limate, and a vast majority of source of livelihood. Conversely, many of these developing nations are in regions where the incidence of contamination may be increased due to natural factors, as well as lacking infrastructure. (Jaffee & Henson, 2004) Methods for evaluating SPS impacts are a still developing field of empirical research, where new analytical tools are increasingly entering t he fray. By and large, the focus has remained on aggregated, macro level market models like those developed to evaluate traditional policy mechanisms. However, more researchers are beginning to note the fact that, unlike broadly applicable tariff like poli cies, TBTs have differentiated impacts across actors in the supply chain, as well as regions and nations. This necessarily complicates the process for soundly informing policy decisions as one must identify associated effects and implement additional empir ical measurement tools within producing markets. This is exacerbated by the difficulty found in conducting standard cost benefit analysis of policies on SPS measures, where the full benefits to society are hard to measure concretely. There is some eviden ce, however, that the actual benefits accrued by stringent SPS measures, like those in the developing world, may have, at best, a
18 marginal impact. (Otsuki, Wilson, & Sewadeh, 2001) Finally, there is an issue of inseparability faced in most models whereby it is difficult if not impossible to differentiate empirically between impacts of public regulations by the government and private standards set by import, distribution, transformation, and retail firms. (Henson, 2006) Extending the problematic situation discussed above, Nation States specifically those with developed, post industrial economies increasingly face a dualistic set of motivations for their policies. On one hand their governments must continue to assure th e central, foundational role of the State as provider of security for their own people. Within the modern conception of this is included the protection of the domestic populace, environment, and agricultural activities from hazards. One key mechanism of this is the regulatory framework of sanitary and phytosanitary restrictions it places on products originating from outside its own realm of internal controls. On the other hand, these governments have committed themselves to activities that pr omote growth and stability in those areas where economies are transitioning or under developed. The motivations for these policies are often mixed between self interested and altruistic natures. Sometimes these policies manifest themselves in the form of O fficial Development Assistance (ODA) which, while part of a mixed growth strategy, can be costly to donor governments and have questionable levels of effectiveness in meeting targeted objectives. As highlighted above, complimentary or alternative to ODA a s a development strategy is the promotion of trade as a means for growth. Thus, the potential arises for a clearly conflicting set of policy orientations to occur whereby an over zealous or inappropriately applied attempt to ensure security on the home fro nt
19 hinders the trade in agricultural products vital to a key aspect of development in trading partner nations. Problem Statement and Objectives Therefore it is essential that we evaluate the impact of SPS regulations vis vis development objectives. Mac ro level analy tical tools which measure aggregated welfare changes are a starting point. However, they fail to fundamentally address the contemporary understanding of development needs, showing nothing of how the welfare changes might be applied within th e supply chain nor offering insight into the fundamental objective of modern development policy: poverty alleviation, employment and vulnerability reduction. Importantly, too, we highlight the fact that these policy instruments of the developed world often have consequences that may compete or conflict with the stated goals of many developed nations regarding altruistic and strategic positions on development. Therefore, the framework of this analysis attempts also to provide a view of how policy makers migh t be led to make decisions which fail to meet the full complement of state objectives if they rely solely on broad spectrum, aggregated analy tical tools. At its essence, this thesis ure, and instead focuses on a mixed or of the targeted focus on case studies instead of amalgamated cross sectional models. In an effort to initiate such a movement in the availa ble literature, this work aims to build upon a small but growing sub section which examines a key aspect of SPS compliance within developing economies: changes in supply chain structure at the production level and related impacts on welfare measured as i ncome, consumption or expenditures on households engaged in production for export. To date, central
20 empirical work has only been able to examine the first of these three measures (income), while this work attempts to identify possible ways to extend into the remaining Ferrantino (2003) in his discussion on analytic issues related to non tariff measures. Unfortunately, due to a myriad number of challenges to be addressed later, the work stops short of being able to perform a full analysis but does articulate a clear plan of action for doing so. The results and conclusions of this micro level analysis c ould then be combined with the results of a more tradition al trade model welfare analysis -which is conducted in this paper -to draw more relevant conclusions about the impacts of SPS measures in developed nations as they relate to the competing commitment of promoting development through trade. After reviewing some relevant methodological literature, the price wedge method is adopted for this macro level precursor. It is used to determine the tariff equivalents of U nited S tates and E uropean U nion SPS mea sures for selected products and trade partners. These tariff equivalents act as proxies for barrier protection rates, and, implementing standard market models, we calculate estimates of changes in aggregated welfare captured by consumers and producers, as well as net losses to society. Secondarily, much of the research in this area has focused on African nations. This thesis attempts to apply a sim ilar analysis to other nations in Latin America and Asia noting the differences in data set availability. Following the calls put forth by a number of researchers, it also specifically employs a case study structure, whereby
21 individual markets are considered instead of broad sweeping, multi market and cross sectional analyses. The objectives of this study ar e therefore as follows: To present an integrated analysis of SPS measures as a model to address the failure of current works to adequately inform divergent policy goals; To Investigate how we can model the welfare impacts of SPS/FSR measures and, once a m odel is in place, estimate gains or losses in trade terms under selected SPS/FSR regimes; To Address literature on evaluation of structural changes occurring in primary export (production) markets as a result of food safety regimes and present a clear meth odological plan for conducting further analysis. Charting the Course In Chapter 2 empirical estimations and argumentation based on the results related to meeting these objectives is provided First presented is a general background which reviews relevant l iterature on core subjects. Based on this literature, a number o f hypotheses are drawn I mmediately following the background at the end of Chapter 2 some of these hypotheses are empirically tested and others are identified for inclusion in future research In Chapter 3 an overview of relevant policy and enforcement frameworks is presented. Here, and throughout the rest of the work, the focus is placed on the developed nations employed within the empirical case studies: the United States of America and the European Union, specifically the Member States of France and Germany. Chapter 4 extends the identification of case study players as it presents the developing nations (Nicaragua and Viet n am ) which round out the case study format as well as the specific the analysis. It presents an overview of macro level investigative models within the literature, discusses the chosen model to be applied and then presents results f or the markets ide ntified in C hapters 3 and 4 structure
22 analysis is the household level investigation of welfare impacts, which is fo und in Chapter 6. Here a potential methodology and data sources are discussed Finally, Chapter 7 provides a summary, discussion of implications, and some concluding remarks.
23 Figure 1 1. Growth rates, exports v ersu s GNI per capita in LAC developing nations Source: Author. Data from World Bank databank. -15 -10 -5 0 5 10 15 20 25 30 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Growth rates, Exports (% GDP) versus GNI per capita Latin America and the Caribbean developing nations Latin America & Caribbean (developing only) Exports Latin America & Caribbean (developing only) GNI per capita
24 Figure 1 2. Growth rates, exports versus GNI per capita in SSA developing nations Source: Author. Data from World Bank databank. -20 -10 0 10 20 30 40 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Growth rates, Exports (% of GDP) versus GNI per capita Sub Saharan Africa developing nations Sub-Saharan Africa (developing only) Exports Sub-Saharan Africa (developing only) GNI per capita
25 Figure 1 3. Growth rates, exports versus GNI per capita in EAP developing nations Source: Author. Data from World Bank databank. -20 -15 -10 -5 0 5 10 15 20 25 30 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Growth Rates, Exports vs GNI per capita East Asia and Pacific developing nations East Asia & Pacific (developing only) Exports East Asia & Pacific (developing only) GNI per capita
26 Figure 1 4. Exports as a percent of GDP in developing nations Source: Author. Data from World Bank databank. 0 5 10 15 20 25 30 35 40 45 50 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Exports as a % of GDP in developing nations by region Latin America & Caribbean (developing only) LAC Sub-Saharan Africa (developing only) SSA East Asia & Pacific (developing only) EAP Middle East & North Africa (developing only) MNA Europe & Central Asia (developing only) ECA World WLD
27 Figure 1 5. Exports as a percent of GDP in select developed nations Source: Author. Data from World Bank databank. 0 5 10 15 20 25 30 35 40 45 50 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Exports as a % of GDP in select industrialized nations OECD members OED France FRA Germany DEU United States USA
28 Figure 1 6. GNI per capita in developing n ations over time Source: Author. Data from World Bank databank. 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 GNI per capita, developing nations by region East Asia & Pacific (developing only) EAP Europe & Central Asia (developing only) ECA Latin America & Caribbean (developing only) LAC Middle East & North Africa (developing only) MNA Sub-Saharan Africa (developing only) SSA World WLD
29 Figure 1 7. GNI per capita, comparing developing to developed nations Source: Author. Data from World Bank databank. 0 10000 20000 30000 40000 50000 60000 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 GNI per capita, developing and industrialized nations East Asia & Pacific (developing only) EAP Europe & Central Asia (developing only) ECA Latin America & Caribbean (developing only) LAC Middle East & North Africa (developing only) MNA Sub-Saharan Africa (developing only) SSA World WLD OECD members OED France FRA Germany DEU United States USA
30 Figure 1 8. Growth rates in GNI per capita on selected groupings Source: Author. Data from World Bank databank. -15 -10 -5 0 5 10 15 20 25 30 35 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Growth Rates, GNI per capita East Asia & Pacific (developing only) EAP Europe & Central Asia (developing only) ECA Latin America & Caribbean (developing only) LAC Middle East & North Africa (developing only) MNA Sub-Saharan Africa (developing only) SSA World WLD OECD members OED France FRA Germany DEU United States USA
31 CHAPTER 2 BACKGROUND The Role of the State and Motivations for Engaging in Development Facilitation One need only but to peer through the age old annals of political theory to discover a resounding commonality in explanations of and justifications for organized government. Classically, the common denominator is the [Nation] State 1 as the provider of collective security (a public good) to those citizens who provide it with legitimacy and resources. In engaging a Nation State, the populace is aiming to solve a classical problem of economics: the provision of a good whose opportunity costs are too great for a single individual to bear on their own. At the most basic or primal of levels, thi s public good takes most commonly the form of military defense. However, over the millennia, it has morphed and shifted into other areas of security, such as law, domestic infrastructure and healthcare. Interestingly, this increasingly includes activities which hold other attributes, such as altruism, as seen in policy objectives focusing on providing development assistance to third country partners. As the world of independent Nation States has grown more interconnected, providing what would otherwise be a purely altruistic good in many cases serves a dual purpose of ensuring domestic security for the donating Nation State as well. As Jolly and Gadbois (1989) the follo wing mandates for aid: (1) humanitarian, (2) economic and development growth of recipients, (3) fostering trade or commercial interests, and (4) serving of strategic resear chers who are skeptics of purely altruistic behavior in development assistance 1 Here the term State refers to what alternatively may be called the Polity.
32 policies, it is more likely that a mixed or dualistic set of motivations is at play. 2 (Schraeder, Hook, & Taylor, 1998; Alesina & Dollar, 2000; Grilli & Riess, 1992) In fact it is Alesina and Dollar (2000) who, at the end of the last millennium, completed an intriguing empirical analysis of strategic motivations for development assistance as measured by bilateral Official Development A ssistance flows between 1970 and 1994. 3 fficial D evelopment A ssistance (ODA) within their data set] is accounted for by four countries: U.S., Japan, (Alesina & Dollar, 2000, p. 35) Includ ed within their set of donor nation motivations are the improvement of democracy in the recipient country, the openness of its economy measured on a Sachs Warner index, the initial wealth (or political ties as measured by colonial history and an index of United Nations voting block correlations. Although they make a series of generalized conclusions based on broad spectrum models, they also investigate differences across donor nations. Selecting three of the four top providers of ODA in their study, we summ arize their findings in Table 2 1 What their results show is that while most nations have very clear geo political motivations for offering bilateral ODA, not all express via donation levels the same valu es on the reciprocal issue of 2 It is incredibly important to note that research shows that the determinants of development assistance transferred bilate rally versus multilaterally are significantly different and highly separable from one another. 3 Flows of official financing administered with the promotion of the economic development and welfare of developing countries as the main objective, and which are concessional in character with a grant element of at least 25 percent (using a fi xed 10 percent rate of discount). By convention, ODA flows comprise contributions of donor institutions. ODA receipts comprise disbursements by bilateral dono rs and multilateral institutions. Lending by export credit agencies with the pure purpose of export promotion is (OECD, 2003) (OECD, 2008)
33 trade or the more altruistic motives of poverty reduction or the promotion of democratic governance. 4 Perhaps Alesina and Dollar best summarize their findings, and hint at their implica tions, best when they wri te Factors suc h as colonial past and voting patterns in the United Nations explain more of the distribution of aid than the political institutions or economic policy of recipients. Most striking here is that a non democratic former colony gets about twice as much aid as a democratic non colony. A similar result holds for former colonies that are closed to trade versus open non colonies. (p. 55) I t is essential here to highlight a pivotal facet of this study which relates to later discussion: the analysis completed by Alesina and Dollar covered a period ending in 1994, the final year before the ascent of the World Trade Organization and the sweepin g reforms suggested by the Uruguay Round. No pertinent studies of a similar nature have been published in the available literature under the subsequent regime to applic ability under contemporary circumstances, it also does not mean one should automatically discount the earlier literature entirely. In fact, there is evidence to suggest that even prior to 1995, while still solely under the General Agreement on Tariffs and Trade, some governments member states had begun to weigh direct aid more heavily than providing trade preferences, whose impact as a special call out for development policy has diminished as nation states are increasingly placed on relatively more equal footing with one another. (Grilli & Riess, 1992) 4 Noting, of course, that these can be argued to also have clear implications for strategic position and donor nation promotion of self interests.
34 Data from the O rganization for E conomic C ooperation and D evelopment (OECD) on total ODA given may support this notion. As o ne can se e in Figure 2 1 the EU as a conglomeration has consistently provided greater amounts of ODA than the United States in terms of total value amounts. France and Germany give at roughly the same level, although consistently higher than the US. Howev er, over the last decade the trend seems to be a gradual increase in overall amounts. However, this simplistic data is not a sufficient measure of commitment to ODA. To provide a slightly more robust picture, Figure 2 2 presents the ODA totals as a perce ntage of relevant G ross D omestic P roduct (GDP) numbers, readjusted to a base year of 2005 values. Of the governments selected, France directs the largest percentage of GDP towards ODA, with Germany about on par with the EU average. The United States provi des the least. While the trend is upward, the figures are all well below the internationally agreed target of 0.7% While ODA is a more straight forward, traditionalist method of development promotion and, as applied above, a measurement of commitment to development goals the notion of trade for development, or, more specifically, trade liberalization policies as a promoter of theoretically autonomous growth have, in recent years, regained prominence within the spectrum of policy orientations. Trade a s an Avenue for D evelopment While notions of trade as a promoter of economic expansion have been about a part of the policy landscape for centuries (Findlay, 2009) targeting trade and its liberalization as a route to economic growth gained directed attention after the Second World War. in 1947, after a failed attempt to create an International Trade Organization the General Agreement on Tariffs and Trade (GATT) was drafted and signed. During its initial phase, roughly from the end of WWII until 1970, classical development
35 economists such as Nurske (1959), realized that trade could be a driver of development. The caveat to the prevailing view among academics however was that trade in and of itself coul d not suffice to induce development. For example, Prebisch (1950) argued against trade only models because of concerns about elasticities and the terms of trade. Others noted that under colonialism, in the period prior to WWII, growth in European overseas territories tended to run in cycles and that in these regions structural change was seen to occur only in those territories endowed with enough autonomy t o create import barriers, in a pre cursory form of Import Substitution Industrialization. 5 For thos e in this school of thought it boiled down to a need for government intervention in developing countries as a first step to facilitate the industry protection to facilitate le arning and build the public or collective goods of physical and social infrastructure. (Adelman, 2001) Clearly then, under this first period, the prevailing policy advice from economists was an infusion of development aid and investment first and trade liberalization later. For approximately a decade or so through the 1970s and into the early 1980s the development economics literature engaged in a debate with two schools of thought highlighting different avenues for development policy. The first school continued the perspective just beginning to transform into a more neo classica l view The other school which ultimately failed to gain dominance in the policy orientation debate s focused on issues of income distribution, advocating continued government intervention but with a new focus on choosing the right target industries 5 Which, while brought forward later in the period we discuss here by Prebisch, Singer, Furtado and the ilk, was founded in ful ly classical, 18 th century writings by Ricardo, Mill and List. (Mehmet, 1999)
36 (i.e. those that were labor intensive, especially in unskilled and semi skilled industries, such as agriculture) to readjust for the inequalities that had been created after rapid industrialization. (Adelman, 2001) By the 1980s the former would win out in the policy arena, especially among international organizations and the developed nation governments Comparative Advantage and the related Hecksher Ohlin theorem 6 Some proponents of this school advocate for it staunchly per se especially those within the so called Reagan Thatcher neoliberalism movement which ran until roughly 1996. Others accept orientations closer to the original classical models where an initial insertion It was during the end of this period, in the 1990s, amidst a series of economic crises, that the interplay of trade and economic development was solidified as a corner stone of policy action. Specifically relating to the aftermath of the Latin American Economic crises of the 1980s and 1990s economist John Williamson characterized the policy targets of leading District of Columbia W His consensus consisted of a series of ten for development, and included among them was 7 (Williamson, 2000; Williamson, 2004) Th World Trade Organization was launched Soon afterward various e conomies, notably many in Asia, shift ed from import substitution to export promotion and achiev ed some success in improving on standard development targets, including poverty, vulnerability 6 Presumably corrected to address the Leontief paradox. 7 (Williamson, 2000)
37 and inequality. (Stiglitz, 1996) Over the nearly two following decades, trade has been touted as a central pathway for more su stainable development economically, socially, and politically. By the early part of the current millennium a more holistic version of the trade for development policy orientation, at least from a multilateral standpoint, emerged. With a decade of conferen ces and summits leading up to the Millennium Summit in September 2000, the resultant Millennium Declaration laid out the Millennium Development Goals, initiating a boost of optimistic momentum for achieving development in a targeted way. 8 While trade is no t specifically mentioned, Section VIII, Paragraph 30 calls for greater policy coherence and cooperation among International Organizations, specifically including the World Trade Organization. In November 2001 the W orld T rade O rganization (WTO) began the Doha Round of negotiations, which produced the Doha Development Agenda (DDA) At this time the Ministerial Conference also adopted a Ministerial Declaration to set forth the aims of the DDA. It reaffirmed a commitment to the mixed approach poli cy orientation stated above with trade liberalization placed in the central role (see Paragraphs 2, 3 and 5). (4th WTO Ministerial Conference, 2001) Chief among the Programme of Work was reform for agricultural products, contai ned in Paragraphs 13 and 14. It also contained a call for assistance and capacity building in [expediting the movement, release and clearance of goods, including goods in tran 8 Goal Number One: Eradicate extreme poverty and hunger by halving the proportion of people who live one less than 1 USD per day over a two and a half decade period, Achieve full and productive employment and decent work for all and halving the proportion of people who suffer from hunger. (General Assembly of the United Nations, 2000)
38 has been slow, perturbed by conflicts between and among post industrial nations and those with developing and transitioning economies, especially in the realm of agricultural reforms. The Doha Rou nd has lasted some 11 plus years without final conclusion and only partial implementation of the DDA. Until such time as the Doha Round is concluded, the Uruguay Round Agreement on Agriculture, which went into force in 1995, is the prevailing accord relat ed to trade in agricultural and agro food products, with a few exceptions 9 In March 2002 several months after the start of the Doha Round, the United Nations held its own high level ministerial conference in Monterrey, Mexico. It was one of the first and most prominent ministerial level conferences to address movement on on Financing for Development was the adoption of the Monterrey Consensus which in many cases[,] the single most important external source of (United Nations, 2002, p. Par. 27) It went on to highlight the role of trade as an engine for both ec onomic growth and employment, especially noting the role of agricultural trade, and called attention to cerns over technical barriers to trade and sanitary and phytosanitary measures along with general trade liberalization in agricultural products. Six years later, a follow up conference to Monterrey was held in Doha, Qatar. It too produced a Consensus document, in which 9 O ne of which is the Agreement on the Application of Sanitary and Phytosanitary Measures, and, more tangentially, another is the Agreement on Technical Barriers to Trade. The SPS and TBT Agreements also went into force with the inception of the WTO in 1995 a long with the URAA.
39 most of the paragraphs related to trade were calling for final resolution to the DDA. Many in the media criticized the follow up as a carrying little meaning since many of the Technical Barriers to Trade As progress has been made in the liberalization of trade under GATT and now the WTO, classical impediments to market access have diminished significantly. Under th e contemporary multilateral policy orientation, market openness is the ultimate goal. there still remain deliberate barriers to access. In the broadest terms these are Non tariff Barriers (NTBs). As Hillman (1991) device or practice other than a ta riff which directly impedes the entry of imports into a country and which discriminates against imports, but does not apply with equal force on (2001) point out much of the literature on NTBs focused on the protection of domestic producers as the primary factor in the definition. In this paper, we instead underscore the distortionary aspect as the central feature of NTBs This is important specifically because w e recognize that some measures are legitimate and rooted in the fundamental purpose of the state, yet still have conflicting implications for different policy objectives. M echanisms that aim to ensure the health, safety, quality and compatibility of produ cts are generally categorized as Technical Barriers to Trade (TBT) While there is little contention that legitimate reasons can be given for TBTs, there is also concern that governments may inappropriately apply such mechanisms to artificially create d istortionary or protectionist barriers. Recognizing this challenge, two renewed multilateral agreements related to such matters came into force along with the
40 WTO. These additional agreements work to strengthen a base provision within the GATT recognizing the right of Nation States to set restrictions to guarantee health and safety. Both serve only to prohibit the misuse of regulations and standards to distort trade for non legitimate reasons. The broadest is the Agreement on Technical Barriers to Trade (th e TBT Agreement) which covers issues technical regulations (the mandatory restrictions), standards (non mandatory guidelines) and conformity assessment. 10 It is applicable to essentially all product types, with the exception of certain agricultural product s. These are covered instead by the Agreement on the Application of Sanitary and Phytosanitary Measures (the SPS Agreement). In both cases, there is a call for harmonization of national regulations by a cadre of international standards organizations. 11 Th e recommendations set within these cooperative bodies constitute the most basic level of protection and are thus assumed to be acceptable. However, protections above and beyond these can be adopted, provided they are scientifically justifiable. Crucial wit hin the se provisions are that they create a clear system of priorities: the freedom to determine what is an appropriate level of acceptable risk is a wholly national issue rooted in state sovereignty. This delicate balancing act therefore lends itself to n umerous disputes between Member States to be resolved as provided for via WTO panels. In addition to the core provisions related above, the two agreements also contain provisions related to transparency and notification as well as on the granting of Tech nical Assistance, that is, aid in various forms which assists other WTO Member 10 See Appendices for definitions. 11 For example, under the SPS Agreement, three ISOs are recognized:
41 States in meeting requirements, especially developing nations. Despite these additional provisions, developing nations still have concerns regarding the ability of their produce rs to adapt to requirements. This is especially true in relation to trade with the developed economies where restrictions are often high and different from State to State. The oretical Considerations of SPS M easures as Trade Barriers Before continuing, it is important that we take a moment to consider salient aspects of theory related to SPS measures and trade. First, one should note what we that i s, public regulations. At the most essential level, this motivation centers on the theoretical concepts of asymmetric information (on the part of consumers regarding the safety of their food), internalization of social costs not accounted for in private ma rkets (such as health care burdens and lost societal productivity from illness), and the provision of common or public goods (here, manifested in safe products via certification and testing) which may be costly or poorly incentivized in the market. (Mitchell, 2003) Second, one must recognize that SPS measures can be classified in a number of ways (Josling, Roberts, & Orden, 2004) the most basic of which is the associated locality of the regulatory impact. Does a specific regulation act on processes or merely on products? The differentiation is an important one when evaluating either a specific policy, or a set of policies. Those SPS measures that are process focused are generally seen within the l iterature as more protectionist in their outcomes because they mandate a specific system of production, removing the freedom of producing firms to choose alternatives that maximize their profit s With process standards, foreign producers tend to bear a gre ater burden in terms of certification, as well as have added costs for adaptation (whether it is meeting new standards or exiting the market). Product
42 standards are those which are control led and test ed at the border. While still holding the potential to be costly in market terms their impact has been generally seen as less impactful on foreign pro duc ers Figure 2 3 graphical ly summarized these points Third, an issue of product substitutability arises when considering international markets and food saf ety. In the models discussed below, we generally assume that products are perfect substitutes for one another ; however much theoretical and empirical work dating as far back as the 1960s suggests that consumers may differentiate among products using orig in as the signal of food safety and that th e substitution effect may not be exactly perfect. (Nimenya, Henry de Frahan, & Ndimira, 2008) In recent years, several authors have compiled overviews of the divergent theoretical conceptions of SPS measures as trade barriers. Often they have maintained a central focus on the importing country. However, a few notabl e exceptions are highlighted here The first is that of Nolte and Rau (2006) who identify four methods for modeling SPS impacts within a standard market equilibrium framework: tariff equivalencies of SPS measures, SPS measures inducing supply shifts, a s well as demand shifts, and market segmentation as a result of SPS measures. As is evidenced in the discussion by Mitchell (2003) it may be the case that combinations of these models are combined into a multi dimensio nal impact framework. Tariff Equivalent Model The tariff equivalent modeling framework is perhaps the most precipitous within the literature, being widely applied and heavily discussed. Th is model considers the fact that compliance with SPS measures in an export market places some additional costs upon the product. These added compliance costs are analogous to traditional tariff s in
43 classical trade and policy analysis. Evaluation begins at the assumption of free trade in a given international market fo llowed by the imposition of the SPS measure acting as tariff. The analysis is static, and we also assume that the standard is only felt in prices (say, a product standard). Figure 2 4 provides a graphical illustration of how this process works. We can con sider th e cost of compliance as an artificial shift in the Excess Supply curve in the international market, while assuming that the two domestic markets remain unchanged (i.e. importing and exporting nation supply and demand remain the same). Under the Ex cess Supply curve accounting for cost of compliance we determine new international market equilibrium at P A 2 and Q 2 traded Tracking down to the prevailing production curve, ES B 1 at the equivalent trade quantity, Q 2 traded provides the altered world pric e which is then used to complete the analysis. Not depicted are the standard welfare measures. In general we find an increase in Importing Producer Surplus in Country A at the expense of foreign producers in Country B as well as net welfare losses for Coun try B In their presentation, Nolte and Rau (2006) (p. 7) basing this conclusion on the fact that an SPS measure would have to apply solely to foreign firms However, they miss the point as numerous other researchers have shown that in reality; compliance with an SPS regulation may be less costly (or even have almost no cost) for domestic firms than for foreign firms, especially when framed in a developed developing nation context. While SPS measures targeting a specific nation or region would be illegitimate under WTO rules, those with a scientific basis that are applied uniformly despite having disproportionate compliance costs are not prima facie preclu ded.
44 Supply Shift Model In the previous model, a key assumption was that the policy at hand did not induce real changes in production, typically manifested in a product oriented SPS regime. However, certain nations have begun to favor SPS policies that req uire practices to improve health or environmental standards. One example of this would be the total ban of all products not certified as being free of certain agrochemicals, such as DDT. Under this framework, firms who wish to continue producing for export would face an augmented supply curve. Again, application of the model could often rest upon the costs of compliance faced by firms. In discussing this model, Nolte and Rau (2006) argu e that it may be the case that standards act as a catalyst for investme nt in more efficient production practices and or the provision of goods at a price premium. However, this fundamentally presupposes that the necessary capital and credit will be available to undertake the burden of adaptation. As we have seen, there is evi dence that in developing countries this is not always the case. Therefore we can continue with the case of strictly backward shifting implications, as they do, albeit for mainly illustrative purposes. Figure 2 5 presents the theory graphically. Again we assume a starting point of free trade in a two country model. In the scenario portrayed in Figure 2 5 the SPS measure introduced induces compliance costs on both domestic and foreign firms. This leads to shifts in the exchange market in both Excess Su pply and Excess Demand. World market prices rise and so too does demand. The key observation comes in terms of expected changes in supply quantities and is related directly to the magnitude of compliance costs within each country. In the case of a develop ing nation as exporter and developed nation as importer, one could assume that costs in the former may be
45 much larger than compliance costs in the lat t er, or could even assumed to be zero in the developed nation importer If this were the case, then the mo del would show that foreign firms decrease production while domestic firms increase production. Also in this case, one would need to recognize that Excess Demand in the international market would remain constant. In short, we would expect Producer Surplus in Country B to contract. Demand Shift Model Similar to the previous model, the demand shift model considers the augmentation of a full curve. In this case, it is assumed that consumer preferences change. While Nolte and Rau support this notion, it is a dubious matter for some. How can consumers differentiate products? Do they actually account fully for the costs, many of which are socially borne, of food safety? Evidence such as that discussed by Mitchell (2003) suggests the assumption of changing consum er preferences is not as realistic as Nolte and Rau first make it out to be. Nonetheless, the model still has validity, especially if considered as a future component of a more holistic approach to the analysis. This is in fact the graphic model they discuss, which is shown here in Figure 2 6 The analysis is at the base rather straight forward, and echoes the supply shift only model. However, the caveat here again becomes one of magnitudes; compliance costs can shift supply to a magnitude such that, even when accounting for demand increases under safer food, quantities may decrease. What is more, determinations o f welfare effects are highly dependent on the differences in these magnitudes and we cannot summarily estimate gains or losses.
46 Segmented Markets Model Up until this point, the previous three models have held one general assumption: SPS regimes considered are mandatory for all products in the market (i.e. safety levels are not differentia ted). They do not consider cases where consumers are given the option to choose among different product safety levels nor when [foreign] firms choose to produce in non compliance for sale to other markets. Such a case is presen ted in Figure 2 7 where the supply and demand curves for compliant and non compliant product are mapped on the same model In line with the notion of the supply shift model above, those products in compliance are seen as produced under a supply curve simply shifted from the base li ne of non compliance. The same logic applies for the demand side of this model. However, the insightfulness of this model for trade and welfare analysis is limited. To do so would create a complex system of segmented models across the two nations and in th e international market as a whole While feasible, the number of assumptions needed and the challenges in empirically and reliably estimating curves suggest doubts about its general applicability. Issues Faced by Developing Nations Related to SPS Measures It has become abundantly clear over the most recent decade of development research that agriculture plays a vital if not central role in the promotion of stated objectives o f poverty alleviation and increasing employment. More specifically, policies tar geting the production of high value crops for export, and moving away from commodity only schemes, ha ve gained prominence as mechanism s for attaining meaningful progress in poverty alleviation and employment This means that one of the greatest challenges to trade for development agendas which developing nations face is compliance with SPS measures in export markets. (Henson & Loader, 2001) Some
47 have suggested that SPS regimes are especially targeted as non tariff barriers fo r tropical products, and high value crops in which much of the developing world holds a comparative advantage (but a heavy reliance on for livelihoods ) (Reardon, Codron, Busch, Bingen, & Harris, 2001; ECLAC, 1998) Despite the centrality of SPS compliance to the success of this development strategy, it also challenges lesser developed nations on a number of other issues. First, data shows that participation by these governments in SPS related matters at the multilat eral level is limited, impacting what Henson and Jaffee (2008) term their activity in international bodies and standards organizations, so me researchers have been led to conclude that decisions on SPS items fail to result in considering the needs of low and middle income nations. (Gebrehiwet, Ngqangweni, & Kirsten, 2007) In many ways this is indicative of one of the cyclical traps that these nations can find themselves in. They do not have the resources to actively engage in the multilateral policy arena and therefore outcomes fail to reflect their unique viewpoints or provide strategic footing for improvement o f their situation. As Frohberg, Grote and Winter (2006) note, the structure of the international system as laid out in the SPS and TBT agreements of the WTO are partly the issue: they specifically create a listing of International Standards Organizations ( ISO ) bodies with jurisdiction to engage in consensus building in order to set what then becomes the minimum base standard 12 In terms of SPS issues, this includes at least three different organizations to which nations would need to send delegations. Thus, the continuity of a 12 See, for example Article 3, Paragraphs 1 and 2 of the SPS Agreement.
48 n ation international voice is of paramount importance as the decision making is a continuous process in which most developing economies are not involved. In contrast many of the larger industrializing nations such as Argentina, Brazil, Chile, China, India, Malaysia, Mexico, Thailand and South Africa tend to have regular participation in ISOs like the Codex Alimentarius Commission. (Henson & Jaffee, 2006) These nations often form a voting bloc in internationals bodies which while technically representing other developing nations with low participation may act without fully articulating or representing the other less er concerns and positions. Frohberg, Grote and Winter (2006 ) note that some work by the German Institute for Standardization ( Deutsches Institt fr Normung ) provides evidence that active participation in the standard setting process a llows for the realization of gains from competitiveness and cost saving and that the International Trade Center of the U nited N ation C onference on T rade and D evelopment (UNCTAD) and the WTO highlight the role that participation by governments can play in transmitting information to domestic firms in a timely manner Doing so would facilitate adaptation prior to a need for compliance, thus reducing costs. However, it is important to also note that increased participation in the ISO bodies is only a partial solution, a notion beginning to be borne out in empirical evaluations. (Jaffee S. Trade Note 25, 2005) Apart from this policy issue, developing economies also struggle directly to meet the increased levels of safety demanded by new standards and regulations in the developed world. Essentially, these challenges come to an issue of capacities 13 (Jaffee 13 It is important to note, however, that when sufficient ca pacities are available, it may be the case that more stringent SPS measures in export target markets may provide a catalyst for growth. (Jaffee &
49 & Henson, 2004) Firms engaged in production for export must either develop themselves or have access to forms of human capital, physical infrastructure institutional procedures and support networks to adapt practices that bring them into compliance with new regulations in the target market Thus there are is sues of investment for creation and maintenance of processes to facilitate compliance, under which firms will face increased fixed and variable costs. In the case of multiple regulatory standards or frameworks in different target markets, investment needs and cost changes are likely to be substantially larger than those under harmonized, international rules. Empirical work, such as that by Disdier, Fontagn and Mimouni (2008) provides evidence for this when they show, via a gravity model analysis over seve stringency levels are higher and have a greater reducing effect on imports received from developing nations as compared to other OECD members. Importantly, as Jaffee (2005) not es, the issue of SPS measures holds special value foods, including fruits and vegetables, fish, beef, was shown in Chapter 1 incomes globally have been rising, especially in the developed world. With these increases, consumers have begun to demand not only higher quality agro food products (a driver of SPS stringency) but also a greater variety of products, available during the entire year, and beyond simple commodities. (World Bank, 2005) The same report highlights the role that these high value products are playing in developing nations producing for export, with total Henson, 2005) The logic follows that if production markets can adapt and absorb t o the changes, then they may be more competitive with those who cannot and receive higher prices that include the risk reduction premium.
50 proportions having now risen above the 50% mark of their total agro food exports. In many cases, the expansion into these products was a result of both trade liberalization and targeted promotion. High value crops, such as fresh fruits and vegetables (FFV) have been held up as a pro poor move a way to alleviate rural poverty through value creation. (IFAD, 2008; AVRDC The World Vegetable Center, 2007) Table 2 2 provides a view of the growth in exports of FFV, in trade value terms, broken down b y developing and developed nations during the period of WTO implementation ( from 1992 to 2001). Similarly, Table 2 3 presents data on trade volumes. 14 The se FFV crops were most especially targeted at assisting small holders, who make up a majority of the en demically poor. Yet, coincidentally, FFV are also far more prone to face an increasing and diversified array of SPS barriers because of their non processed and highly perishable nature. Synthesizing several pieces of literature, Henson, Brouder and Mitulla h (2000) xports of high value food brought benefits to a number of developing countries at both the macro and microeconomic levels, although concerns were raised about the distribution of these benefits a (p. 1159) Although it can directly be inferred from the aggregated FFV data there is evidence to support the conclusion that small farmers in export production markets are disproportionately negatively impacted by changing SPS measures and private standards abroad. (Okello & Swinton, 2007) This has implications theoretically and empirically for how the supply chain might react to the imposition of SPS measures. 14 For both data sets, numbers exclude intra European trade.
51 SPS measure s, Development and Structural C hange Considering the unique position of agricultural products within development schemes combined with the fact that they face the most directed TBTs in the form of SPS measures, it is imperative that we look further at thi s issue of structural change that is, how the supply chain changes while firms and markets adapt and what it can mean for development targets. Intuitively the argument that SPS measures can induce of structural change logically holds. Compliance carr ies increased fixed and variable costs (investment and maintenance) which small firms may struggle to access in appropriate amounts. The nature of SPS measures, especially the increasingly prevalent process oriented regulations, means that control must be held closely to the firm to minimize risk. Thus, firms may benefit from economies of scale and from increased vertical coordination. These results have been, time and again, reported in the empirical literature (Wo rld Bank, 2005; Reardon, Barrett, Berdegu, & Swinnen, 2009; Maertens & Swinnen, 2009) So too does much of the literature make light of the fact that this movement towards higher standards has clear implications for small holders, both in terms of num bers and income. As Okello, Narrod and Roy (2007) tightly coordinated supply chain works against the smallholder in three ways: 1) information asymmetry and transaction cost s; 2) organizational constraints; and 3) (p. 19) Within the context of the United Kingdom -granted a developed nation -Yapp and Fairman (2006) examined the perceptions of small and medium sized enterprises (SMEs) within the food supply chain concerning compliance with SPS measures. They identified challenges of finance and time, noting the temporal focus of most firms was on the short run. Many of the SMEs also saw compliance as
52 something external to their operation s conducted by non firm actors such as regulatory agencies. They also noted issues with experience level, access to information, perception of support biased towards large firms, lack of interest, and low awareness levels as important factors. Their results are telling, even more so considering that these are firms in a developed economy. As discussed previously, standards which are product oriented versus those that are process orie nted are likely to have differing impacts. Previously, this discussion was centered on trade and aggregated welfare aspects, but it can be extended here too. Process standards, which dictate what practices can be implemented in order to be certified for import, are more likely to induce direct effects on producers by limiting their flexibility in profit maximization. For small farmers this is especially important as they typically do not have the same access to credit or capital needed to meet advanced s tandards. Reardon et al (2001) towards process standards has] tended to exclude small firms and farms from (p. 421) They later note tha t the strategic responses to this change by agribusiness firms and by small firms and farms was to ally with public and nonprofit sectors to form internal grades and standards as well as certification systems to access export markets and to bring instituti onal change to non tradable product markets. (Reardon, Codron, Busch, Bingen, & Harris, 2001) Thus it may be the case that process standards favor structural consolidation and increased vertical coordination to a greater extent than product oriented measur es.
53 In fact, the body of research is absolutely rife with cases examining the link between increasingly pervasive SPS measures in the developed world and declines in small farmers within developing nations, especially those heavily involved in export mar kets. In addition, throughout the development literature, the stability of rural communities is persistently tied to the availability of livelihoods via small farming operations. Mounting evidence has shown that incomes, a measure of welfare, are higher am ong small farmers producing high value crops than households not similarly engaged (Maertens & Swinnen, 2009; McCulloch & Ota, 2002) What ends up being key to the link between development and small holders may end up b eing the context in which the individual market is placed. (World Bank, 2005) However, some empirical work notably that of Maertens and Swinnen (2008) has also pointed out that the story may not be solely one of doom and gloom, and that, again, context matters. In their study, they found that while small holders may have been negatively impacted by increased standards in export markets, not only were their incomes still higher than those not involved in the industry, but that the vertical coordination in to larger agro industrial firms expanded opportunities for wage labor in high value product markets, benefitting the poorest of households. Clearly, then, the results of SPS measures on development are more complex than at first glance. Statement of Hy potheses Regarding the macro level analysis conducted within a market model framework, we proceed to test the following hypothesis: That SPS measures by developed nations have a negative impact on Producer Surplus for those engaged in export production and cause net losses in Societal Welfare.
54 That tariff equivalencies are higher in European nations than in the US for selected produc ts.
55 Table 2 1 Empirical results on ODA motivations from Alesina and Dollar 2000. Who gives foreign aid to whom and why? Journal of Economic G rowth. ODA Donor Nation Elasticity of aid Openness of Economy/Trade Liberalization Democr acy UN voting block History as former colony of the donor History as former colony of some nation other than the donor United States of America OLS 1.84 1.3 0.57 0.06 0.39 0.08 Tobit 2.08 1.46 0.62 0.06 0.41 0.09 2 SLS 1.29 0.97 0.44 0.08 0.37 0.07 Summary Openness of the recipient nation's economy increases commitment of ODA dramatically. Presence of a democratic government also holds relative weight in awards while political ties (when factoring out effect of middle eastern or religious variables) hold little. Response of aid levels to income levels is extremely high in terms of elasticity absolute values, indicating that the "richer" a recipient nation is, the faster it's ODA from the US will "decline". France OLS 0.61 1.27 0.13 0.07 1.51 0.25 Tobit 0.7 1.49 0.14 0.09 1.66 0.29 2 SLS 0.34 0.72 0.02 0.08 1.15 0.19 Summary Openness of the recipient nation's economy increases commitment of ODA dramatically. Presence of a democratic government has little impact on awards, as do political ties as measured in strategic terms of UN voting behavior. However status as a former own colony dramatically increases ODA award levels. Response of aid levels to income levels is extremely flat, indicating that the relative wealth of a recipient nation means little for their expected aid levels from the French Republic. Germany OLS 0.78 0.26 0.18 0.14 0.21 0.15 Tobit 0.81 0.26 0.18 0.14 0.21 0.16 2 SLS 0.41 0.15 0.1 0.09 0.18 0.08 Summary Unlike the previous two cases, openness of the recipient nation's economy has little impact on aid levels. Presence of a democratic government as well as colonial history also mean little in terms of increased ODA received. However, strategic alliance within UN voting blocks has, relitively, a positive effect on ODA sent to a recipient nation. In fact, of the UN coefficient is second only to that of Japan. Response of aid levels to income levels is extremely flat, indicating that the relative wealth of a recipient nation means little for their expected aid levels from the Federal Republic of Germany.
56 Summary entries by author. Numerical data taken from Alesina and Dollar March 2000. Who Gives Foreign Aid to Whom and Why? (pp. 43, 44 and 48; Tables 5, 6 and 7 ) Journal of Economic Growth
57 Figure 2 1 Trends in ODA for selected governments Source: Author. Data from OECD Statistics. 5000 15000 25000 35000 45000 55000 2002 2003 2004 2005 2006 2007 2008 2009 2010 ODA in millions of USD, constant 2010 values Trends in Official Development Assistance for selected governments France Germany United States DAC EU Members, Total
58 Figure 2 2 Trends in ODA as a percent of GDP for selected government s Source: Author. Data from OECD Statistics. 0.00% 0.05% 0.10% 0.15% 0.20% 0.25% 0.30% 0.35% 0.40% 0.45% 0.50% 2002 2003 2004 2005 2006 2007 2008 2009 2010 ODA as a percentage of GDP Trends in ODA as a % of GDP for selected governments France Germany United States DAC EU Members, Total
59 Figure 2 3 Standards in the agri food sector S ource: Author. A dapted from : Nolte and Rau May 2006 Incorporating food quality standards into equilibrium models: a review and discussion of possible approaches. (pp. 5; Figure 1) Prepared for the IATRC Summer symposi um. Bonn, Germany. Product Standards Process Standards (affecting production practices) Maximum Residue Levels o Fertilizers o Pesticides/herbicides o Preservatives/ripening agents/additives o GMO contents o Veterinary drugs Grading Classification standards Directly related to product Not directly related to product Hormones and antibiotics Food hygiene practices Certification of production practices ( such as on risk management plans) Hormones and antibiotics Food hygiene practices Certification of production practices (such as on risk management plans) Impact product quality & consumer safety Other external effects & values
60 Figure 2 4 SPS measures under a tariff equivalent m odel Source: Author. Adapted from: Nolte and Rau May 2006 Incorporating food quality standards into equilibrium models: a review and discussion of possible approaches. (pp. 8; Figure 2 ) Prepared for the IATRC Summer symposium. Bonn, Germany.
61 Figure 2 5 SPS measures under a supply shift m odel Source: Author. Adapted from: Nolte and Rau May 2006 Incorporating food quality standards into equilibrium models: a review and discussion of possible approaches. (pp. 10; Figure 4 ) Prepared for the IATRC Summer symposium. Bonn, Germany.
62 Figure 2 6 SPS measures under a d emand shift m odel Source: Author. Adapted from: Nolte and Rau May 2006 Incorporating food quality standards into equilibrium models: a review and discussion of possible approaches. (pp. 12 ; Figure 5 ) Prepared for the IATRC Summer symposium. Bonn, Germany.
63 Figure 2 7 SPS meas ures under a segmented market m odel Source: Author. Adapted from: Nolte and Rau May 2006 Incorporating food quality standards into equilibrium models: a review and discussion of possible approaches. (pp. 13 ; Figure 6 ) Prepared for the IATRC Summer symp osium. Bonn, Germany.
64 Table 2 2 Total value of trade in fruit and vegetable exports Source: Hallam et al. 2004. The market for non traditional agricultural exports. (Section 2.2.1, Table 2.1). Food and Agriculture Organization of the United Nations w ith original source from FAO STAT 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 million US$ Fruit World Total 10673 10446 11542 12989 13793 13953 13 364 13269 13628 13826 Developing 6158 5911 6308 7267 7957 8037 7933 7963 7971 8257 Developed 4516 4535 5234 5722 5836 5916 5430 5306 5657 5569 Fruit, excluding bananas and citrus World Total 5328 5267 5848 6693 7338 7320 6989 7259 7810 7945 Developing 2387 2391 2527 3029 3459 3381 3450 3664 3950 4179 Developed 2941 2876 3321 3664 3879 3939 3539 3595 3859 3766 Vegetables World Total 3867 4545 5226 6105 5988 6303 6932 6625 7050 7563 Developing 2025 2507 2790 3469 3282 3416 3894 3765 4057 4427 Developed 1842 2038 2436 2636 2706 2887 3038 2860 2993 3136 Total fruit and vegetables World Total 14541 14991 16768 19094 19781 20256 20295 19894 20678 21389 Developing 8183 8418 9098 10736 11238 11453 11827 11728 12028 12684 Developed 6358 6573 7670 8358 8543 8803 8468 8166 8650 8705 Total fruit and vegetables excluding bananas and citrus World Total 9195 9813 11074 12798 13326 13622 13920 13884 14860 15508 Developing 4412 4898 5316 6498 6741 6797 7343 7429 8007 8606 Developed 4783 4914 5757 6300 6585 6825 6577 6454 6853 6902
65 Table 2 3 Total volume of trade in fruit and vegetable ex ports. Source: Hallam et al. 2004. The market for non traditional agricultural exports. (Section 2.2.1, Table 2.2). Food and Agriculture Organization of the United Nations with original source from FAO STAT 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 1000s of tonnes Fruit World Total 23 774 25 074 27 073 27 837 29 278 30 752 30 261 30 645 32 337 33 104 Developing 16 792 17 466 18 532 19 447 20 926 21 591 21 722 22 201 22 676 23 483 Developed 6 982 7 608 8 540 8 390 8 351 9 161 8 539 8 445 9 660 9 621 Fruit, excluding bananas and citrus World Total 8 196 8 799 9 529 9 781 10 955 11 509 11 333 12 078 13 150 13 462 Developing 4 436 4 748 5 059 5 344 6 248 6 380 6 788 7 225 7 535 7 998 Developed 3 760 4 051 4 471 4 437 4 707 5 129 4 546 4 853 5 614 5 464 Vegetables World Total 7 557 8 836 10 028 10 671 11 223 11 627 12 860 13 254 13 089 13 942 Developing 4 255 5 303 5 652 6 370 6 753 6 701 7 965 7 867 7 747 8 800 Developed 3 302 3 533 4 376 4 301 4 469 4 925 4 895 5 387 5 342 5 142 Total fruit and vegetables World Total 31 332 33 911 37 101 38 508 40 500 42 379 43 121 43 899 45 426 47 046 Developing 21 047 22 769 24 184 25 817 27 680 28 293 29 687 30 067 30 423 32 283 Developed 10 285 11 142 12 917 12 691 12 821 14 086 13 434 13 832 15 003 14 763 Total fruit and vegetables, excluding bananas and citrus World Total 15 753 17 635 19 557 20 452 22 177 23 135 24 194 25 332 26 239 27 403 Developing 8 691 10 051 10 710 11 714 13 002 13 081 14 753 15 092 15 282 16 798 Developed 7 062 7 585 8 847 8 738 9 176 10 054 9 441 10 240 10 957 10 606
66 CHAPTER 3 DEVELOPED NATION SPS POLICY FRAMEWORKS Numerous authors have highlighted in varying terms the same principle captured by Jaffee and Henson (2005) standards are changing rapidly, along with increased public awareness of food safety in hig h (p. 92) In effect, prior to the 1970s, issues of food safety assurance were of little or no import to policy makers and private distribution and retail firms and it was o nly the emerged. (Knowles, Moody, & McEachern, 2007) As reactionary measures, public and private actions have led to changes in institutions engaged in oversight and safety assurance, an increased stringency of measures, and an expansion of standards and regulations into areas not previously addressed or enf orced. Figure 3 1, taken from Henson and Caswell (1999), provides a graphical representa tion of different public regulatory mechanisms. Here, one can generally say that Performance standards are analogous to what we have previously described as Product base standards and, likewise, Specification standards are so to those which are Process bas ed. They and product individually are typically multi layered combinations of all five types conveyed within Figure 3 1. As Humphrey 1 (2011) notes, speaking in the context of the United States and the 1 d detailed examination of US and EU standards, both independently and comparatively, that was found by the author, thus it plays a central
67 (p. 2) (see also Josling, Roberts, & Orden (2004) ) Ultimately, this entails a movement away from product standards in favor of process standards, so invidious amongst developing nation production markets. Notably, these have manifested themselves in the well examined Haza rdous Critical Control Points (HACCP) and Good Agricultural Practices (GAPs) measures employed widely by regulating institutions. Also necessarily included in the shift is a need for traceability of products and the development of identity preservation (IP ) mechanisms across the supply chain. (Humphrey, 2011) It is for these reasons, as well as the evidence provided in Chapters 1 and 2 and the connection between developed nation market barriers and export led trade for development in high value horticultura l crops (fresh fruit and vegetables), that we continue with an overview of policy frameworks in selected developed nations. Here, again, we stress that, while the focus of this paper is on public regulations, private standards also play an important role a nd cannot be separated in subsequent empirical analyses. 2 However, as Humphrey argues himself choices of regulatory strategies in the face of the increasing complexity of fo od production systems and the perception that consumers face new or increased risks from (2011, p. 2) role in our discussion here. Other authors were also consulted, and are referenced when they represent significant de partures from the more current work from Humphrey. 2 We note, with regret, this point, especially considering the admonishment provided within the section 3.1 of Humphrey (2011). However, as Humphrey also notes later, the prevailing public regulation regime is also a driving influence on private firm standar ds setting.
68 European Union Before launching into an examination of modalities and practicalities, we note one specific empiri cal finding in the literature: imports into the European Union are more tariff measures than in any other O rganization for E conomic C ooperation and D evelopment (OECD) country or region. (Disdier, Fontagn, & Mimouni, 2008) In many ways, this is not surprising, considering the overwhelming predominance of the Precautionary Principle 3 as a foundational value of law within the EU and its member states. We should also note the likely potential that the result reported by Disdie r, Fontagn and Mimouni captures combined effects of public regulations and private standards. As Humphrey (2011) highlights, private standards e been developed most extensively in Europe, with GLOBALGAP (known prior to 2008 as EurepGAP) being particularly relevant because it applies to pre (pp. 7, 2) However, if we assume that both regulations and private standards are suff icient proxies for consumer risk preferences and attitudes (Echols, 1998) the specific nuances, while interesting, are insubstantial This context is important as well, because in the EU food safety regime developments have p rimarily been crafted as a response to problems faced domestically (or, internally to the customs union) and have been tailored accordingly. 3 The Precautionary Principle, especially the method in which it is applied in the EU Regulations, is perhaps one of the most contentious issues with the SPS debate. It is such a pivotal [and heated] element in fact, that the EC issued a speci al Communication (COM (2000) 1) spanning some 27 pages of where preliminary objective scientific evaluation, indicates that there are reasonable grounds for concern that th e potentially appropriate. (p. 3) (emphasis added) As Saunders (2010) notes, this is somewhat of an understatement, but it still The principle is to be applied when (a) there is scientific evidence for a threat to the environment or to health, but (b) the evidence, while sound, is not conclusive. (p. 47)
69 European food safety regimes began to undergo significant changes beginning in the late 1990s (Humphrey, 2011) an d includ ed actions of harmonization, centralization and coordination. Among such actions taken was European Council Directive 93/43/CEE, issued on 14 June, 1993, which instructed, but did not mandate, hygienic standards and the adoption of HACCP. (Borraz, Besanon, & Clergeau, 2006) Again, a driving catalyst was the recurrent and increasing incidents of agro food contamination. (Caduff & Bernauer, 2006) Notably these crises underlined an area where the creation of a single market had not been accompanied b y appropriate integration of relevant institutions and policies, with inconsistencies across Member States in terms of legislation and enforcement. (Humphrey, 2011; Vogel, 2003) The resulting policy responses and mecha nisms in the EU were: The creation of new, national level regulatory and enforcement agencies in Member States who did not previously have them, and the clarification and strengthening of responsibilities for those pre existing The adoption of Directive 29 /2000 4 dealing with protective measures against the introduction into the Community of organisms harmful to plants or plant products and against their spread within the Community The adoption of Regulation 178/2002 which promulgated the General Food Law at the European Union level The creation of the European Food Safety Authority (EFSA) The eventual replacement of several Directives with full Regulations in 2006 5 Included within the initial Regulation 178/2002 was a set of 5 guiding principles, a summary of which is taken from Humphrey (2011) and replicated in the Appendices. 4 In fact, 29/2000 was an amendment to 77/93/EEC from 1976, but the changes were so broad sweeping and under so different a European Community government, that one can consider it de novo 5 The reader should note that Directives within the EU are policy mechanisms open to interpretation and inclusion by Member States while Regulations are mandatorily applied by all Members, with no flexibility.
70 In general, the underlying principles for these actions are: a focus on the value chain (i.e. process standards); assessment, management and communication to contain risk 6 including the adoption of HACCP; codification of the Precautionary Principle; control and rectification through improved traceability; and placing the primary onus on private sector firms. In general, enforcement of policies happens at the national leve l, by the Competent Authorities of each individual Member State. The European Commission can engage in policing of these authorities through the appropriate Directorate General. Additionally, the European Food Safety Authority (EFSA) operates as a sort of clearing house for scientific information, conducting its own research in partnership with national level agencies, as well as formulating policy advice to the European Commission and other EU level legislative and executive bodies. (EFSA, 2012) This renaissance in the EU of Food Safety Regulations ( FSRs ) included a number of more stringent standards, for example specific regulations issued for animal products (EC Reg 853/2004) and non animal products (EC Reg 852/2004), inclus ive of fresh fruits and vegetables. This included the harmonization of acceptable residue levels, in most cases greatly lowering the tolerance levels (thus raising the stringency) for many residues, if not banning product traces completely. In the same yea r, EC Reg 882/2004 7 was also issued, including among its provisions specific rules controlling the import of food products into the EU and assuring the provision of technical assistance to 6 Noting that these are adopted almost verbatim from FAO and WHO documents of 1995 and 1997. (Henson & Caswell, 1999) 7 Interestingly, a list of certain products, which includes arable crops and fruits and vegetables, are partially exempt ed from the provisions of 882/2004 because they are covered by separate and independent regulations. (Preamble, Â§10) For an example, reference Regulations 2200/96 of 28 October 1996 on the common organization of the market in fruit and vegetables
71 developing third countries. The same Regulation also amended certai n portions of phytosanitary controls as originally laid out in Directive 29/2000 and also extends the potential requirement of third completed by Member States. This was culmination of a general move in the 1990s and early 2000s across the continent to tighten standards on pesticide residues in horticultural crops, to afford safer supplies of food and protect against the introduction of foreign pests and diseases. Some authors have identified this as a catalyst for the subsequent or simultaneous emergence of private process based standards such as EurepGap. (Humphrey, 2011) Today, the European food safety control systems can be summarized into three layers: (1) primary control procedures and correctiv e action taken by the firm 8 ; (2) national Competent Authorities must ensure they meet Community standards for controls on mo nitoring and verification; and (3) the Food and Veterinary Office of the EU conducts inspections to verify and monitor the first and second layers. For importers, meeting these criteria is crucial because once verification and certification is given, products are assumed safe, liability shifts and access to the market is directly granted. (Broberg, 2009) However, border testing is al so a possibil ity and, as Broberg supplies an example of, the established acceptance limits and changes to them can have a profound impact on mark et access to the Common Market. 9 One must also note that, just because base regulations are harmonized by the E U, it does not mean that all government regulations are exactly the same. As Borraz, Besanon and Clergeau explain: 8 Compa ratively, the burden for non animal products is less, in terms of both stringency and because those of animal origin must obtain prior approval and registration from European authorities. 9 See Broberg (2009) for a brief case on maximum residue limit changes in 2001 and the ir impact on Ghanian pineapple imports.
72 European treaties since Maastrict recognize the right of a member state to maintain more rigorous norms in order to uphold public health. The Amsterdam Treaty goes further and authorizes a country, once harmonization has been achieved, to adopt stricter rules if it holds new scientific evidence. In both cases, the European Commission must examine these measures and their possible generalizat ion in the face of the available scientific evidence. (2006, p. 145) Furthermore, while the base standards may be harmonized, the enforcement and control methods within the EU are also left to the Member States. Specifically this includes those agencies and organizations each government deems as Competent Authorities, meani ng that internal application may vary, even for imported goods as the various ports of entry are under national control. (FAO, 2005) As the FAO report notes, food saf ety legislation focuses on criteria and procedures rather than on detailed of relevant EU Regulations indicates that the most common practice is to incorporate impor ts into the same exact standards set. France Borraz, Besanon and Clergeau (2006) identify a food crisis as the catalyst for reforms at the national level in France, specifically the bovine spongiform encephalopathy (BSE) crisis onset in 1996. They link the BSE policy reaction finding it to be similar to what followed the blood transfusion scandal in the mid 1980s, placing French food safety policy in the greater context of health safety, especially in terms of its features and operations. In 1998 the Fr ench government promulgate d a new law on health safety, creati ng the Agence franaise de scurit sanitaire des aliments and authorizing the Institut de veille sanitaire
73 alert and investigative agency for contamination events. These were just additions to the typically which, while undergoing some reform, has r emained constant and has led to food safety in France being classified a contested policy zone. A driving force behind reforms was an attempt to disentangle policy makers from the strong agribusiness lobby in France, to create agencies for evaluation and p olicing that were independent of those tasked with economic policy, and to reinforce the application of the Precautionary Principle when science was inconclusive. (Borraz, Besanon, & Clergeau, 2006) AFSSA is charged with (1) risk assessment in foodstuff s; (2) the conducting of research and scientific support for policy advice, especially in animal derived products; and (3) veterinary medicines. It was not provided with enforcement functions, which were maintained at the ministerial level, but must be con sulted in the case of emergencies and in the formulation of national legislation and the transposition of the FSR decision making process in France, it is only able to cast its light as far as the central government will allow it to, meaning that policy creation and enforcement are still largely at the whim of politicians (instead of scientific experts) as well as bureaucrats directly linked to and with interest in domestic p roduction regulatory frameworks. (Borraz, Besanon, & Clergeau, 2006) Far and away the most important player in French sanitary and phytosanitary measures is the Ministry for Agriculture and its Directorate General of Food. (Lee, 2 008) Prior to French reforms and those at the European level already discussed, French food safety law was predominately geared towards the national government as
74 the testing and control agent implementing targeted product standards As with most European nations, France has slowly abandoned this approach moving towards placing the onus for food safety more on private firms. Notably, the French implemented safety clauses allowed by differe nt directives and the opportunities offered by (Borraz, Besanon, & Clergeau, 2006, p. 145) Germany the German system is similar to that of the French. The German Bundesinstitut fr Risikobewertunf ) has similar functions and roles to AFSSA (and EFSA, for that matter) Many of the foundational principles within the two decision making frameworks are the same and several challenges are also a ligned. (Borraz, Besanon, & Clergeau, 2006) It was even the intrusion of BSE into Germany in 2000 and the re sulting public crisis that instigated reform of the FSR process. (Steiner, 2006) Also, as Knowles Moody and McEachern (2007) Germany consumer protection [in agro food products] was made a more explicit There are notable differences, however. First, the German system of controls is relatively dec entralized with Competent Authorities often operating on a regional basis. (FAO, 2005) At the federal level, while the BfR handles most risk assessment matters, risk management, inclusive of pesticide enforcement A new governance structure which included the BfR and consumer protection and food safety agency (BVL) went into effect in 2002, and, distinguishing Germany from their westerly neighbors, the new
75 structure consolidated agencies with purview over food safety into a single Federal Ministry, reducin g fragmentation and improving coordination. Most importantly, the government. It is based on HACCP and represents a mixed regulatory scheme denoted ry industry led quality assurance scheme and a (Steiner, 2006, p. 11) The QS system is marked by a system of audits, most of which are completed at the production firm level by an independent auditor. Application to the fruit and vegetable market began in earnest during the 2004 crop year. Finally, a number of minor reforms took plac e, including one which extended the ability of consumers to engage in civil litigation to enforce firm liability for food safety. discount food retailers holding the highest percent of market share more than in any other European nation. Interestingly and importantly in light of previous di scussion points German firms were relatively late in adopting private standards, especially those a radical change in policy [by German food retail firms] following an e xpos by analyses lead us to a secondary conclusion which is no less important: if we are to fully accept the Greenpeace study cited, then there was, prior to the private refor ms, a certain percentage of fresh fruit and vegetable imports which were exceeding the maximum residue levels imposed within EU Regulations, and, previously, EU Directives.
76 Theorized Implications for T hird countries First, one should note that throughout European SPS measures are created within a very specific contextual view which is myopic with regard to European produc ers. Broberg (2009) highlights three consequences of this myopia for agro food firms in the developing world: Maximum allowable level s of naturally occurring toxins are normally calibrated to standards levels experienced by firms operating in a European climate Certification requirements are made with the assumption that firms have easy access to accredited laboratories The assumption t hat the level of administrative competence of firms and public authorities is fairly sophisticated T here are some interesting caveats to the challenges developing nations face vis vis these consequences and in general with regard to the application of EU SPS measures to third country firms. One caveat comes from Humphrey (2011) who highlights the outcome of a case before the European Court of Justice Challenging the view of Graffham (2006) food safety regulations might suggest that importers have a responsibility to ensure that food of non animal origin imported from third countries has been produced in conditions that following analysis: In practice, it is hard to believe that this is the way import controls operate. There are clearly numerous examples of importers acting on a consignment basis, buying and selling produce about which they have limited knowledge. There is still a considerable amount of "arms length" trading in food of non animal origin. When the fresh produce importers stress the need for stringent controls, they generally referred to the customers' requirements, not those of EU legislation. (Humphrey, 2011, p. 24)
77 It is important not to forget that European laws both at the community and national levels are mixed policy instruments with both process and product standards still at play. This combined with the Euro centered myopia Broberg alludes to bec omes effects that are present in European decision making processes on limit values. Here, the rule is to adopt the lowest possible level present amongst the set o f measures. As the measure chosen to set the limit could be well below the asses sed level at which there is risk to human health. This also leaves one to wonder what happens in products which cannot be produced widely within the EU. This is not where the complexity ends. For a number of products, there are no investigations to determi ne measurement values and the limits then revert to the limit of determination (LOD)which is often equivalent to zero tolerance. Ironically, a number of the products which face this dilemma are the same generic, older generation products so common within t he developing world and small farmers. Broberg concludes that this has led to an effective zero tolerance level for a number of tropical crops. Thus, in terms of European markets, producers in third countries, particularly those in developing nations a nd/or tropical climates, face uncertainty about the application of procedures and fragmented and incoherent requirements and regulations, in addition to barriers of capital and climate. These necessarily increase the transactions costs for firms, especiall y those who would seek to supply either directly or indirectly multiple markets. Further, we have previously highlighted the literature which
78 shows that process standards, such as HACCP have a particularly profound cost to small and medium enterprises, a nd that most of these conclusions were realized in research completed in a developed nation context. United States of America US Regime If one had to concentrate into two words the differences in the principles behind European and US food safety regimes (FSRs) those words would be, without doubt, American policy makers reject the application of this principle as a basis for decision making (Henson & Caswell, 1999) and a rev iew of international negotiations and dispute settlement case briefings extends this into the multi lateral arena. At a deeper level, and one focused more on practicalities instead of principles, the American approach to FSRs differs also in the approach t o regulation and the assignment of responsibility, both in terms of enforcement and in terms of liability. (Humphrey, 2011) In general, American regulation and enforcement, especially related to imports, is significantly more centralized than in the EU. (Leibovitch, 2008) The history of federal food products regulation finds its most solid roots in The Pure Food and Drugs Act of 1906. (Leibovitch, 2008) Since that time, the system of FSRs has been one of mixed levels: federal an d state, congressional and executive. While one must be cognizant of the fact that the several States have the ability to create their own laws relating to food safety and enforcement, for our discussion these are of minimal importance 10 as international tr ade regulation is predominately a Federal 10 U.S. State and Local Governments play an important role by adding their muscle to federal efforts, state and local governments put considerable energy into food
79 prerogative. Generally, at this level, many FSRs are created via Executive Regulations. Since the Administrative Procedure Act of 1946, any decision process on new or changed regulations must involve public partici pation, be published in the Federal Register, not become effective less than 30 days after publication and need be accompanied by the publication of pertinent statements on organization and procedural rules. (FAS of USDA, 1999) Primary authority over food products is held by the Food and Drug Administration (FDA) which is part of the Department of Health and Human Services (HHS), and the Food Safety Inspection Service, which is promulgated within the US Department of Agriculture (USDA). (FAS of USDA, 1999; Humphrey, 2011) The prescription and assurance of phytosanitary measures is handled within the Animal and Plant Health Inspection Service (APHIS), also part of the USDA. Additionally a number of other agencies and bureaus have roles in regulation and, esp ecially, enforcement, including the Agricultural Marketing Service (AMS, part of the USDA), the US Bureau of Customs and Border Protection (part of the Department of Homeland Security but formerly the US Customs Service of the Department of the Treasury), the Centers for Disease Control and Prevention (CDC, part of HHS) and the Environmental Protection Agency (EPA). For the fresh fruit and vegetable (FFV) market, actions and procedures by both APHIS and AMS are crucial, as the major concern within the US fo r these products is not so much consumer protection (health risk) but the introduction of pests inspection. Some of this is done cooperatively with federal agencies, both to maximize staff effectiveness and to ensure that state standards meet federal rules. There are cooperative federal and state p rograms for fish, dairy and other food product inspection. Roughly half the U.S. states have their own meat and poultry inspection programs. State inspected meat and poultry products may only be sold within that state. Local governments inspect restaurants fast food spots and similar outlets. They can close establishments for sanitary violations. (FAS of USDA, 1999)
80 and diseases that might harm American agriculture (environmental and economic risk). (Karova, Roberts, Granta, & Peterson, 2009) In terestingly, the US has one striking similarity with the EU : animal origin products especially those which are imported face stringent controls including process and products standards in place since the mid 1990s (Humphrey, 2011 ) Restrictions for other items are less stringent and HACCP only applies to low acid canned foods, fruit juices and meat and seafood products. In fresh produce the focus is on microbial contamination, but even here the relevant authority, issued beginning i n 1998, is only a set of guidelines which are non mandatory. (Humphrey, 2011) In many respects, this is typical of the American approach to FSRs, where the focus is on the most harmful risks but there is an acknowledgement that the broad spectrum of food crops face different vectors and levels of contamination. In many ways this also goes back to the rejection of the Precautionary Principle by the federal government, where agencies like the FDA only act to intervene when there is sound, indisputable scienc e supporting such action, leaving the precaution up to private firms and individuals. (Humphrey, 2011) This has left the role of government regulation mainly to that of product centered standards for imports, with relatively higher tolerance rates compared to the EU. Here then it is also important to note the difference in vision s regarding private liability. As Henson and Caswell (1999) state incentive for quality assurance Following a may or sometimes may not provide a defense in consumer product liability cases. Such cases, for example for damages to people who suffered a food borne illness, often generate large settlements. However, the degre e of care taken may provide some protection in settlement negotiations and in suits between co mpanies in the supply chain. I USA ex post liability plays a secondary role as an incentive to the
81 immediate market impacts of being responsible for, or associated with, a food (p. 594) Again then, we see an American viewpoint where the need to intervene and correct for limited This has had an important effect on the development of private standards by US importing, distribution, processing and retail firms. Starting in the 1990s, amid the food safety scares, US importing, distribution, processing and retail firms began implementing systems which varied from firm to firm, leading to a lack of uniformity at the private level. The laissez faire attitude of the American government led companies to implement their own systems, although the government often require s HACCP and/or GAP compliance and, to a lesser extent, verification and t hird party certification activities. (Humphrey, 2011) In very recent years the American FSR landscape has been changing. A number of contamination events occurred in the late 1990s and early 2000s culminating in 2005 and 2006 with a widely publicized set of E. coli outbreaks from leafy greens. One episode in particular, the case of contamination in California spinach, led to a large private public initiative for FSR reform within that State -which happens to be the largest agricu ltural economy in the US and, incidentally one of the largest total economies in the world. (Humphrey, 2011; The Associated Press, 2007) At roughly the same time, although not necessarily linked, FSR reform began to gain traction at the federal level as well. Much of this concern was driven by new awareness of the increasing role imported foods played in the American diet; as recent testimony from the Government Accounting Office before congress made clear approximat ely brought to light that some 60% of FFVs are imported into the US. (Shames, 2010) Traceability also became a buzz word amongst policy circles, although largely due to post 9/11
82 concerns over bioterrorism. A report issued by t he FDA in 2007 called for a number of reforms, among them the implementation greater use of third party certification schemes by the private sector to allow government agencies to retarget their inspection resources to the greatest risk points. (FDA, 2007) These movements have culminated in the passage of the Food Safety and Modernization Act (FSMA) of 2011 which introduce d dramatic shifts in American food safety regimes and makes the most sweeping legislative changes for fresh produce since the 1938 Food, Drug and Cosmetics Act. (Bignebat, et al., 2011) Prior to the FSMA, the FDA had notif ied Congress that it wanted to expand its (FDA, 2009) With the FSMA, it received its wish in the form of direct authoriza tion. In addition, the new act provides for the issuance of mandatory recalls by the agency and for significant increases in the inspection and testing of facilities, including foreign firms. More generally, the FSMA represents a paradigm shift in American FSR towards preventative measures and a level of stringency not before seen within its context, and a large part of the shift is likely to be in relation to fresh produce imports. Because of the newness of FSMA, it is impossible to speculate on the detail s of its enforcement However, it is interesting to note, one caveat: in the US system there are laws which enact authorization which are often separate from those which enact spending The FSMA is the former, and, although it was passed, recent politica l developments are likely to impede its immediate implementation due to the lack of funding being allocated via separate legislative mechanisms. (Bottemiller, 2011; Bittman, 2011)
83 Theorized Implications for Trading Partn ers Act (FSMA) by President Obama in January 2011 introduces a significant new element into the discussion of food safety standards and how they impact upon producers in developin g countries (p. 2) This critically sets the stage and provides the impetus for the development of the analytical framework that wil l be developed and applied in Chapter 5 as well as the exploratory discussion in Chapter 6 However, in the current sectio n, our concern is on how the past food safety regime in the US can have influenced third countries since this is the location in which our empirical analysis will take place. In many respects, this past is similar in terms of the structure of American FSRs to those of European Union: mixed instruments of both product and process procedures. Private standards play ed a greater role in history and this is also where much of the policy incoherence has come from for exporters to the US, instead of from public regulatory and enforcement incoherence and disharmony as in the EU. Further, until the FSMA, there was no public regulation related to process standards, and product s tandards have typically centered highly on the prevention of phytosanitary risks, many of which are beyond the direct control of farmers and typically face mitigation efforts higher within the food supply chain. In general, it is a safe assumption to inco rporate as principles the same consequences discussed by Broberg (2009) in the European context. The issue, however, then becomes one of magnitudes. Most analyses, including empirical investigations, suggest that while American FSRs [of the past] had similar impacts on the supply chain, trade and foreign producers, these were much smaller in terms of quantity for the majority of products. Those products which have
84 faced increased intervention notably those where pre e xport certification of farm practices have been made mandatory have experienced impacts of a far greater magnitude than those who simply face more stringent product standards. (Alvarez, 2006; Deodhar, Rastogi, & Krissoff, 2007) Whether this is true in the future is impossible to predict at this point.
85 Figure 3 1 Forms of gov ernment food safety regulation. Source: Author. Adapted from : Henson and Caswell. 1999 Food safety regulation: an overview of contemporary issues. (pp. 5 95; Figure 3 ) Food Policy Examples: (1) a maximum level; (4) a requirement that a product is heated to a certain temperature (process standard) or has a particular composition (product standard); (5) a positive list system as generally operates for food additives. Low degree of intervention High degree of intervention Information provision requirements (labeling) 1 Targets 2 Performance 3 Specification 4 Prior approval for use 5 Standards
86 C HAPTER 4 OVERVIEW OF CASE STUDY MARKETS In Chapter 3 we discussed the sanitary and phytosanitary (SPS) policy frameworks within the three developed nations to be examined within this project. In Chapter 4 we will turn our attention towards the selected developing nations to be included in the analy sis as well as the selected fresh fruit and vegetable (FFV) product markets for analysis. Here we make a special explanatory note on selection; unlike the choice about which developed nations to examine which was based on their role as FFV importers and O fficial D evelopment A ssistance contributors Here the developing nations choice was necessarily more arbitrary. Due to the nature of this project and its limit ations, selection was based on the potential for data set availability for future expansions in to the second analysis pillar method, given the objective of this project to serve as a model for empirical analysis it is sufficient for our purposes. Moreover, as we will see, the selected developing nation s offe r important case studies specifically due to their background s and features. Additionally, selection criteria for the product market s were more rigorous and akin to those employed in the decision on the developed nation selection. Markets investigated ar e pineapples and an aggregation of cauliflower and broccoli Trade directionalities include Nicaragua to US and selected EU nations as well as Vietnam to US and selected EU markets. Table 4 1 presents a summary of the markets to be analyzed.
87 Nicaragua The Republic of Nicaragua is the first of two case study mark ets Located in Central America, the nation is bordered by Honduras to the north, Costa Rica to the south, and the Pacific Ocean and Caribbean Sea to the east and west, respectively. Its terrain is characterized by extensive coastal plains rising from the eastern seaboard into a set of interior mountains, and, to a much smaller extent, the same proceeding from the western seaboard which also contains two large inland bodies of water. (See Figures 4 1 and 4 climatologically, with mountainous areas being cooler. Of the almost 120 thousand square kilometers of land held by the nation, 14.81% is arable and 1.82% is dedicated to permanent c rops. (CIA, 2012) It is the largest country by area in Central America and also has a relatively low population density. (CIA, 2012; Merrill, 1993) Importantly, the country has also struggled greatly with development. Numerous sources place it amongst the poorest countries in the Americas. Figures published by the United Nations Food and Agriculture Organization estimate that around 17 percent of its population living in extreme poverty while other estimates in the yea r 2005 place some 46% of the population below the established poverty line. (FAO; CIA, 2012) Income inequality has also been historically high, with a 1998 GINI coefficient of 60.3, although this has greatly improved ac cording to 2010 calculations 1 (CIA, 2012) The proportions of GDP coming from remittances and usually around one third of exports centered on the US market. (Seelke, 2008; The World Bank, 2012) Educational 1 Ironically, the most recent GINI index calculation (40.5 in 2010 with a rank of 58 th inco me inequality at a better position than that of the US (45. In 2007 with a rank of 42). (CIA, 2012)
88 attainment is also low, with only roughly 28% of children who begin first grade ever finishing their sixth grade year and little or no enforcement of compulsory educatio n. (US Department of State, 2012) Much of the impoverished some 76 percent live in rural areas where they struggle to make a living from agriculture and fishing. (FAO, 2005; IFAD, 2012) I n fact, nearly half (43%) of the population in Nicaragua lives in rural areas and seasonal migration is not uncommon as many land less laborers go to the Pacific region to harvest export crops there during summer season (November through January). (IFAD, 2012; FAS, 2012) These rural populations have, by and large, been small holders reliant on subsistence agriculture, although there has been growth in production of agricultural products for export. (Seelke, 2008) Histo rically, much of these exports were centered on a very select number of commodity crops usually coffee, cotton and/or bananas which led to a cyclical, boom and bust system. (Merrill, 1993) For the approximately two decades following the Second World War, agriculture in the country saw some diversification, a trend which would eventually be stifled for a period, only to very recently be renewed, especially with non traditional crops. (Merrill, 1993; Seelke, 2008) In 1991, agriculture accounted for approxim ately 45% of the national workforce, while today that proportion has dropped to 28%. (Merrill, 1993; CIA, 2012) Likewise the share of total GDP from agriculture has dropped although not by the same magnitude from an estimated 24% in 1991 to an estimated 17.4% in 2011 2 (Merrill, 1993; CIA, 2012) One cannot disengage the development struggles the nation has faced from its history of crises, both political and climatological. From 1936 to 1979, an anti Marxist, 2 Alternative figures place these numbers at 29% of the workforce and 19.6% of GDP. (FAS, 2012)
89 familial dictatorship headed by Anastasio So moza and, later, his two sons held control of the country. (Seelke, 2008) As we have noted previously, during this regime agriculture had seen great improvement in terms of diversification and an increasingly export oriented nature. After the Sandinista N ational Liberation Front (FSNL) led a conflict and turmoil This (Merrill, 1993) Giving into mounting pressures the leader of the FSNL and de facto President, Daniel O r tega, conceded to free and fair elections held in February 1990 during which Violeta Barrios de Chamorro was elected to the Presidency. During her 7 years in office, her administration promoted a r e diversification of agriculture as well as the role of exports for development. (US Department of State, 2012; FAS, 2012; Seelke, 2008) Her time in office was perhaps the most effective of recent governments at promoting this movement, although, like her successors, she also faced political discord and had a difficult time implementing reforms. With only two additional administrations after that of Chamorro, in 2007 Mr. Ortega was re elected by plurality to the presidency and installed into office in Janua ry 2007. Under his regime export promotion is still one of the pillars of the most recent National Human Development Plan (NHDP) submitted to the World Bank. A year election, the National Assembly ratified the Central American Free Trade Agreement with the United States and neighboring countries, which went into effect in April of 2006. Apart from its political crises, the nation has faced a number of natural disasters which have impacted agriculture. Prone to hurricanes, it has faced a number of major storms causing great damage, including
90 Hurricane Mitch in 1998. In other years the nation has faced severe droughts as well as volcanic eruptions wh ich have had significant impacts on crops and rural livelihoods. Despite a rough history, recent trading agreements with the US, CAFTA DR, and with the EU, the forthcoming European Central American Association Agreement taking effect in 2013, may offer newly liberalized avenues for exports. Even with these agreements, however non traditional goods, such as specialty fruits produced in the nation will still face the SPS measures of The US and EU markets, and policy analy ses related to these agreements (Central America Data, 2012) Vietnam Unlike our first market for examination, the second the Socialist Republic of Vietnam faces a relatively better development situation. The nation lies at the very center of South East Asia, bordered to the north by China and by Laos and Cambodia to the west. (See Figures 4 3 and 4 4) Along the east lie s almost 3,500 k ilometers of coastline which stretches from the Gulf of Thailand and the South China Sea up into the Gulf of Tonkin. The south ern part of the nation, especially in the infamous Mekong Delta is classified climatologically as tropical, while the north is monsoonal. It is characterized mostly by low lying and flat deltas, with central highlands and the occasional mountainous zone in the northern extremities. Approximately 20% of its land is ar able, and about 7% is dedicated to permanent crops. (CIA, 2012) Its roughly 331,000 square kilometers are densely populated and agriculture accounted for less than a quarter of its total economic output in 2011, at 22% share of employment are contradictory, although the sector held at least half of its employment, at 48% according to the CIA, and possibly as high as 70% according to
91 the Foreign Agriculture Service of the USDA and the World Bank (CIA, 2012; FAS, 2012; The World Bank Group in Vietnam, 2005) Poverty is highly concentrated in rural areas as well, with 90% of those in poverty living in these areas. (The World B ank Group in Vietnam, 2005) The people are governed under a centrally planned, communist state and government owned enterprises account for approximately 40% of all GDP. (CIA, 2012) However, in recent years many reforms ha ve been undertaken, both in terms of domestic and foreign policy which stemmed from the doi moi policy implemented in 1986 Included within these reforms was movement on land tenure which has helped spur growth in agricultural products and their exports. (The World Bank Group in Vietnam, 2005) Notably, in January 2007 the nation acceded to the WTO and has increasingly engaged in international trading agreements such as the Trans Pacific Partnership which is under development. Recent years have seen relatively high growth rates around 7 % and it has been reported that Vietnam has greatly reduced the this to the growth of its exports 3 (FAS, 2009) The United States is its chief export market, at roughly 18%, and Germany is the fourth, at 3.7%. (CIA, 2012) Vietnam is a former French colony, having been annexed in the lat t er part of the 19 th century. After the Second World War the nation declared independence but for roughly a decade the French maintained control until a Communist movement in the north led to the division of the nation. Eventually, this led to the Conflict in Vietnam (colloq uially, the Vietnam War) until a cease fire in 1973 created two states, followed 3 Importantly for this project, the same report notes that the nation is struggling to gain clearance for the export of many of its tropical agricultural products to the US.
92 two years later by the forceful unification of the North and South under a Communist regime. During the 1980s, the nation was plagued by military conflicts with China and Camb odia. Since that time, conflict has subsided and there has been a reinvestment in national development programs. Today, the country is held up in many respects as a model of transition and positive development practice. (C anadian International Development Agency, 2012; The World Bank, 2012) Pineapple The first product market to be examined is pineapple scientific name Ananas comosus In terms of importance in the tropical fruit market, this perennial plant often is the sec ond largest only after bananas, contributing to over 20 % of the world production of tropical fruits (UNCTAD, 2012) When bananas are excluded, pineapples are the clear majority leader in th is market; at t he turn of the current millennium they accounted for 51% of the total volume of the global tropical fruit market. (Scott & Riggs, 2011) According to FAO data, t welve countries account for 90 % of the worl d demand for fresh pineapple; the US leads the demand followed by Germany Italy, France, Japan, Belgium, Canada, Spain, the United Kingdom Korea, the Netherlands and Singap ore The world market is dominated by two companies, Fresh Del Monte (Del Monte) and the Dole Food Company, Inc. (Dole) Behind these companies are Fyffes and Chiquita. All four companies are highly integrated on the vertical supply chain axis, directly holding control over production, packaging, export, shipping, import, and ripening of the product (Scott & Riggs, 2011; USAID, 2012) Since 1960, pineapple production worldwide has risen by 400% and since 1998 growth was around 50% following the introduction of the "Gold" variety, developed and patented by Fresh Del Monte In some places like Costa Rica, half of the pineapple crop
93 is now grown for sale on the export market. (Cruz Medina & Garca, 2005) Historically, Hawai i was for many years the world's largest pineapple produce r and nearly only source for US pineapples but in recent decades this has changed drastically as the two major producers moved operations into Latin America and South East Asia (Scott & Riggs, 2011) Conversely, African production of the crop, in general, has been in decline in recent years. (CBI, 2008) Geographically, production in these regions is ideal as optimum production is achieved between 23 and 24 degrees Centigrade and growth retardat ion at temperatures below 15 and over 30 degrees Centigrade. The time to maturation can be anywhere from 12 to 36 months, depending on specific cultivar and climatic conditions. Typical production practices in many regions see the application of fertilizer four times during an assumed 18 month growing cycle. (UNCTAD, 2012) The primary SPS regulations facing fresh pineapple markets are the imposition of required export permits and inspects at port for pests and maximum agro chem ical residue allowances. One issue of note recently for pineapples destined for the European Union has been uncertainty about the applicable maximum residue level that would be in effect for Ethepon, an agent used on fresh fruit to assist with ripening a nd coloration. Initially set at 2 miligrams per kilogram, the standard was eventually changed in 2009 to 0.5 ppm. (Central America Data, 2008; Central America Data, 2009) In addition European consumers are the largest market for organic pineapple products and have continuously increased their demand for products which meet this certification. (CBI, 2008) Crucifers (Broccoli and Cauliflower) Unlike pineapples, the second product market examined here has a much wider range in terms of both production and consumption. Because of reporting patterns, we
94 combine broccoli ( Brassica oleracea L. Italica ) and cauliflower ( Brassica oleracea Brassicaceae ) into a single market, termed crucifers. This aggregation is not expected to pose any issues, since the two are near perfect substitutes for one another and are consistently classified together 4 In fact, broccoli and cauliflower are only two of the wide r family of crucifers which also include vegetables such as Brussels sprouts, cabbage, collards, kale, kohlrabi, radishes, rutabaga and cress. (UF/IFAS Extension, 2012) However for ease of reference, in this work when we discu mean only broccoli and cauliflower. Both crops are considered cool season vegetables, which require full sun and mildly acidic soils. These features mean that for warmer climates it is best planted in fall as a winter crop. (The Old Farmer's Almanac, 2012) There are a number of tropical cultivars which prefer slightly warmer temperatures; typically they require 20 to 30 days of night time temperatures between 18 and 22 degrees Centigrade for optimum growth. (Tjeertes, 2004) Typical planting to harvest periods are between 60 and 150 days, although these can be drastically shorter in high temperature and topical cultivars reaching as low as 40 to 55 days in these regions. (Tjeertes, 2004) These features make the crops a possible candidate for production by tropical zones during the off season in the main consumption markets of North America and Europe. Combined, the United Kingdom and Germany account for nearly a quarter of world trade in the two crops, with Germany and France being the third and fourth largest consumption markets, respectively, within both the European Union and in the world (CCITD, 2012) Importantly for this analysis, Spain, France and the US were also the top 4
95 exporting nations in 2009, accounting for 58.1% of total world exports. Despite this, imports in the US, France and Germany have still shown increases i n recent years. (ERS, 2011; United Nations Statistics Division, 2012) Values in the market have been driven by value addition processes such as preparations of fresh broccoli and cauliflower into prepackaged, pre cut s ervings. (Lucier, 1999) No data could be found about the propensity for such activities to take place in the country of actual production, but this aspect would hold promise as a potential income generation activity. Primary SP S measures affecting the crop are administrative fees for import permits, as well as the maximum residue limit for fungicides such as ametoctradin and other soil fumigants and treatments, which are common to the production process since the se crops are pr edisposed to nematode s and soil borne diseases. (EFSA, 2012; UF/IFAS Extension, 2012; APHIS, 2012)
96 Table 4 1 Summary of selected markets for investigation Product Exporter Importer Crucifers Vietnam France Germany United States Pineapples Nicaragua European Unio n USA Vietnam European Union USA
97 Figure 4 1 Political map of the Republic of Nicaragua. Source: CIA World Fact Book online.
98 Figure 4 2 Regional map of Central America indicating Nicaragua Source: CIA World Fact Book online.
99 Figure 4 3 Political map of the Socialist Republic of Vietnam Source: CIA World Fact Book online.
100 Figure 4 4 Regional map of Eastern Asia indicating Vietnam Source: CIA World Fact Book online.
101 CHAPTER 5 ANALYSIS OF PILLAR I : MACRO LEVEL AGGREGATED WEL FARE ANALYSIS Methodological Background The overwhelming majority of macro level analyses related to sanitary and phytosanitary (SPS) measures are completed within the trade context. These studies focus on trade effects such as changes in trade volumes or values as well as welfare derived from trade in products. In the early 2000s, at the height of th e SPS debate within the academic community, a number of methodological overviews were published which summarize d the divergent approaches to empirical analysis common within the body of literature. Among these are Beghin and Bureau (2001) and Rau and Schlueter (2009) as well as a comprehensive report from the OECD (2003) 1 and work from Maskus, Otsuki and WIlson (2001) each of which investigate d and describe d several general categorizations of macro level approaches to analysis. Generally these include d the price wedge or tariff equivalent method, inventory based approach, survey based approach, count or stringency m easures, gravity model analysis, risk assessment for cost benefit analyses, varied micro economic approaches, trade flow data analysis, and sectoral or market model (Partial and General Equilibrium models) approaches. Importantly, only a number of these f ocus on and have application to analysis of welfare impacts. The following sub sections present summaries the three most prevalent: g ravity models, partial equilibrium models and general equilibrium models. It should also be noted that much of the literatu re especially that of the foundational period following the establishment of the World T rade O rganization (WTO) largely focuses on 1 Much of the OECD re port appears to be a verbatim contribution from the Beghin and Bureau article published two years prior, with notable additions and updates as well as considerable expansion of representative case studies.
102 evaluating developed nation market impacts. Only more recently has increased attention been paid to applications used to ana lyze less developed and transition economies. Gravity Models Gravity models offer one, well known method of examining the residuals in regressions on trade flows and trade determinants, providing an opportunity to investigate the amount of trade foregone which cannot be attributed to standard tariffized barriers. In essence, the application of gravity models represents an intriguing cross over of disciplines; economists notably Tinbergen (1962) Law of Universal Gravitation to explain and examine trading relationships between regions, such as Nation States. the functional form is expressed as ( 5 1 ) where F ij is the flow in this case, of traded goods or services between regions i to j G is some constant of unitary measure, M GD P as presented in work by Deardorff and Stern (1998) ) and D and/or other barriers, or vector set thereof, between the two regions. (Head, 2000) The economic intuition, so states Head (2000) is that M i effectively acts as the amount producing country i is willing to supply the market and M j the amount that consuming nation j is willing to demand. D Perhaps Rau and Schlueter (2009) summarize the application of this methodology to the N on T ariff B arrier (NTB) case best when they write that Gravity models are quantity based econometric models. Contrary to simulation models which utilize price terms directly, gr avity models include price terms only implicitly which present a function of observable and
103 unobservable variables. As such gravity models do not provide welfare economics, but the estimated trade flow impact can be transformed into price effects via elast icities to obtain tariff equivalents. (p. 11) Later they go on examining Technical Trade Barrier ( TBT ) effects, such as those stemming from S PS cross sectional richness of trade data Completing the most recent literature review of gravity model applications to TBT cases available to this author, they further note that the following esti mation methodologies have been applied within the field: least squares non linear least squares fixed effects and random effects estimation binary choice estimation Hausman Taylor estimation two stage estimation procedures pseudo maximum likelihood estimation Several years prior to the review by Rau and Schlueter, Beghin and Bureau (2001) had also noted hypothetically that applications of variance or principle component analysis may also prove to be robust improvements in methodology if applied to they also proposed several ways to integrate SPS measures into gravity models via this same said term using several of the analytical tools mention ed in the previous section, but not expanded upon in this work specifically. As they state, introduce as explanatory variables, information on regulations, estimates using the methods described [earlier in their work] (number, frequency of regulations, survey based impacts), or in certain cases the level of standards themselves, provided that
104 there is some variability across countries or over time (e.g. the level of chemical (Beghin & Bureau, 2001, p. 118) However, as with any methodology, gravity models especially applied to SPS measures are far from flawless. As noted previously, their application to welfare measures is highly indirect and such applicat ions have not been well explored within the literature at least not enough to merit their application in a work such as this one. Beghin and Bureau also point out that the method ignores the correcting effect that SPS regulations can have in terms of mark et failures, thus capturing only the trade restriction impact but none of the off setting benefits to society 2 They also note that the technique is highly sensitive to model assumptions, making it less than ideal as an application to specific and detail p roduct markets. Additionally, Rau and Schlueter (2009) make note of several challenges in estimation which we summarize as : Many potential trade relationships on a product specific level become effectively non existent, being dropped to correct for selection bias in the log linearization of the gravity equation. Estimates can be greatly biased by variables which may easily be omitted, namely : o Potential unobserved f irm level heterogeneity caused by an omitted variable which measures the impact of the number of exporting fir ms o Omitting unobserved country pair heterogeneity such as multilateral resistance (relative trade barriers) Heteroscedasticit y may be present in trade data 2 Although they do note that, because the sign of the coefficient in the regressions is not constrained, (Beg hin & Bureau, 2001, p. 118)
105 Partial Equilibrium Models Partial equilibrium (PE) models represent the most simplistic, transparent, flexible and rapid method for investigating the impacts of TBTs. In the general case a PE methodology relies upon standard market equilibriu m models mapping supply and demand curves. In the case of SPS measures, these are typically two nation markets Previous sections of this work have presented graphical rep resentations of such markets. Established theory allows us to calculate welfare (consumer and producer surplus, government revenues and deadweight losses) based on these presentations of supply and demand schedules, found within the bounded definite integ rals of the respective curves. Recall here that microeconomic theory tells us that the derivative of a utility or profit curve, in the case of consumers and pro ducers respectively, results in a [Marshallian] demand or supply schedule, again respectively. T hus it follows intuitively that measuring the bounded area under the inverse demand schedule and/or above the supply schedule is operationally identical to the definite integral of these functions. (Perali, 2003; Jechlitschkaj, Kirschke, & Schwarz, 2007; Just, Hueth, & Schmitz, 2005; Nicholson, 2004) In order to apply the calculations, the analyst must have access to key information points (parameters) about the mapped curves, including relevant elasti cities. In some circumstances this information is already available within the literature. When it is not, the researcher must estimate it econometrically based on applicable market data. Figure 5 1 offers a graphical representation with which we can conceptualize a potential structure for such analysis in an agricultural context. Here we note the
106 centrality of price along with production and supply quantities. In the most rudimentary form, econome tric models can be estimated based upon the relationship between these two items (price and quantity) wit h the extended branches of the model in F igure 5 1 acting as shifting variables 3 This is a distinguishing factor of PE models; as Parapparathu (2007) makes note this methodology is characterized typically as being closer to or approaching a strict ceteris paribus assumption, usually with only price and quantity being endogenous variables with most others being exogeno us. Econometric techniques applied within the literature to determine parameters are greatly varied, but generally take one of two forms: linearized regression models or n on linear, polynomial models. Analyses of SPS measures under this approach genera lly apply a tariff rate equivale nt methodology using a price wedge although other notable theoretical models can also be used under a PE framework (refer to previous discussions in this work for he main int erest, compared to gravity models, is that they make it possible to assess not only the impact of regulations on trade flows, but also o ations of welfare changes at relatively disaggregated levels; i t is well know that consumer and producer welfare changes can be mapped, as can government revenues and societal welfare losses arriving from consumption effects and production inefficiencies. (Jechlitschkaj, Kirschke, & S chwarz, 2007; Schmitz, Moss, Schmitz, & Furtan, 2010) Notably for future extensions of this project, Rau and Schlueter (2009) highlight that PE 3 For demand side estimations we would notably add factors such as population, income, and other socio cultural variables.
107 insight on the impact of 4 (Rau & Schlueter, 2009, p. 13) Published works such as that by Overton, Beghin and Foster (1995) from the 1990s began the solid application of PE models, albeit simplistic, to NTB problems and since this period marked improvements in econometrics and greater integration of core theoretical concepts have led to increasing refinement. Also, unlike m ost gravity models, PE applications allow for analysis at a high level of specification (e.g. targeted sector or product markets). This targeting can reach very detailed levels, such as the 6 digit level of harmonized standards, as just one example. This has led to a large application of the method, especially in literature examining concerns and/ or disputes between trading partner countries (Rau & Schlueter, 2009, p. 13) meaning their application to policy analysis and as a method for informing policy makers is well documented (Lang, 2005; Tvora, 2008; Iacovone, 2005) Granted, the method does have a number of downfalls, incl uding one which is important to researchers concerned with analyzing policy from a de velopment perspective. First, are p roblem s associated with calculating model parameters chiefly price elasticities While there is a myri ad number of econometric approaches stemming from a plethora of examinations, identifying which method is best in a given case can still pose a challenge. What is more, these differing approaches have often produced results which are wildly different. (Se e as an example, some of the literature on own price elasticities of demand for tomatoes in the American market (Johnecheck, Wilde, & Caswell, 2010; Thompson, 2003) ) 4 See as an example work by Lusk and Anderson (2004) on US COOL impacts for meat producers.
108 This leads to the second downfall namely, that r esults are highly dependent upon the assumptions made This is within the econometric estimation of needed parameters, but also with regards to the functioning of policy instruments for which the SPS case is especially noteworthy as well as the underlyi ng economic theory applied. The result is that for welfare outcomes analytical focus necessarily be place d on proportions and changes, with the actual resulting values having minimal economic meaning As always, these models are abstractions and their resu lts are calculated under an ideal world far from real life. Notably too, the theory go only go so far, and the quality and refinement of data sets employed will also carry some impact. Finally, PE models d o not capture the gre ater effects within the econ omy because they have a limited scope of focus. This may be crucial for development policy decision making in economies where markets are highly connected; links in across different sectors say the fertilizer and seed input industry and the value addition chain may mean in our case that SPS impacts on a raw product market have a ripple effect into other areas which PE models fail to provide insight on. General Equilibrium Models General equilibrium (GE) models usually provide an avenue for expanding beyond the limited scope of PE models. While they include features similar to those in a PE framework, GE methods allow the researcher to examine a fuller compliment of the changes which occur within the economy due to a sh ock in the marketplace. Thus they and supply that is impacts in markets which are related or linked to those directly facing the policy change, an aspect not available within PE meth odologies. (Iacovone, 2005; Maskus, Otsuki, & Wilson, 2001) These spillover effects give insight as to cross sectoral effects
109 account for intra sectoral linkages within a consistent microeconomic framework, and add the possibility of including the government as an agent (and hence budgetary costs) in the overall evaluation of the welfare impact of such or such measure (Beghin & Bureau, 2001, p. 123) Computable GE (CGE) models also generally incorporate macroeconomic factors as endogenous variables, enabling them to capture effects of stabilization and structural adjustments. (Wobst, 2000) This generally leads are of note, and summarized as : foreign trade specification following the Armington assumption where imports and exports are imperfect substitutes for domestic produce, (b) high import dependency due to fixed rel ative input output ratios, (c) segmented factor markets, which restrict migration between agricultural and non agricultural sect ors, (d) fixed sectoral capital, which captures the rigid investment structure of the economy, and (e) minimum quantities of marketed and non marketed household demand in order to guarantee minimum levels of food consumption. (Wobst, 2000, p. 3) Because of their complexity, we do not delve further into the specifics of CGE models in this work, a nd readers are referred to more appropriate sources for examinations of the myriad GE systems which have been developed. Generally, however, within an IS LM BP framework, the impacts of the SPS measure are said to be excised directly on the B alance of P ayments (BP) curve whose movement will cause disequilibrium in the external trade balanc e. Here the effects are changes in the level of income and then the domestic interest rate. (Oyejide, Ogunkola, & Bankole, 2000) The implem entation of this framework could be tested empirically using the computable general equilibrium models.
110 As has been discussed, often in trade cases we must accept that certain market imperfections exist, notably that we must occasionally relax the assump tion of perfect substitutes. This is a difficult task to achieve in GE models. Further, in order to account for the fixed compliance costs, a modeling framework reflecting imperfect competition is necessary and this poses challenges in GE models as a rout inely applicable and robust approach is missing. Imperfect competition can be rather easily introduced in PE models. The explicit differentiation between variable and fixed compliance costs allows for determining the economic impact of regulatory measures, in particular the effect of market structure. In this regard PE models have the obvious advantage of capturing sectors (and policy measures) at a more disaggregated level than GE models Also, GE models are typically close to or approaching the mutatis m utandis assumption with very few exogenous variables. This means that they are, necessarily, data intensive endeavors. Generally, t he weaknesses of GE approaches include: the level of aggregation the quality of data typically employed crude specifications of standards (such as SPS measures) highly complex theoretical specifications a general lack of sufficient data in transition and less developed economies Accounting for these short coming s and the general aspect of the approach has led t o well known dubiousness of GE models. So much so that Maskus, Otsuki and Wilson c ompetitive prospects and rationalize industry. (p. 29)
111 Selected Methodology : An Explanation Considering the discussion above, we select a spatial partial equilibrium framework within which to evaluate the welf are impacts of increasing standards in fresh fruit and vegetable markets. While the model clearly has shortcomings important to our less developed nation focus, it has still been widely applied to such a context, most notably as far back as Golub and Finge r (1979) who examine d the classic case of export tariffs in these types of economies The method is selected because it (1) offers a direct, straight forward and transparent method for evaluating welfare changes to produ cers and consumers in different nations (2) relaxes demands for data which is not often available for developing nations and (3) offers the greatest promise for increased refinement in future work to be conducted. Tomek and Robinson (1977) provide a discussion in favor of the adoption of such a method for the general analysis of societal welfare. As many review, the literature is rife with applications of the price wedge method to form a tariff equivalent. Calvin and Krissoff (1998) provide an example of a simple form model in their examination of US apple trade with Japan, as do a number of othe r authors consulted for this project albeit with varying degrees of complexity and diverse project aims. Thus, we will apply the same methodology here, using a partial equilibrium model to evaluate changes in welfare. The general notion of the method was specifically addressed in previous section s and the graphica l model was presented in Figure 2 4 on the idea that NTBs can be gauged in terms of their impact on the domestic price in comparison to a reference p (Beghin & Bureau, 2001, p. 111) This is especially applicable to the case of SPS measures and
112 other standards regime changes which have been shown to asymmetrically impact producer groups having ffect via additional costs which raise the equilibrium price, placing downward pressure on total demand, in turn reducing imports and boosting domestic production. 5 (Hooker & Caswell, 1999; Roberts, Orden, & Josling, 1999; Iacovone, 2005) The tariff rate is equal to the price residue after correcting for factors such as real tariffs, transport and handling. Granted, the method, like any, presents a number of concerns to the researcher which must be addre ssed. Beghin and Bureau (2001) note, first off, that very rarely is observable data available on we have noted previously, it is diffi cult if not impossible to differentiate effects or the specific measures causing the tariff equivalent level of protection. Third, there are numerous practical challenges to applications, most especially if attempting to apply to large scale models such as aggregated industries. 6 This point is reinforced by claims is most appropriate for analyzing homogeneous products (Calvin & Krissoff, 2005, p. 13) In this study we begin by attempting to c alculate our own elasticity measures econometrically partly because comparable elasticities are not already present within the literature and partly to ensure that underlying econometric and data issues are uniform across the markets to be examined. The f ollowing sections will deal with methodological po ints related to this estimation. The formulation of our econometric 5 Most authors state that no tariff revenue is accrued and thus when calculating welfare, the government revenue component is ignored. While this author thinks there may be a compelling logical argument to counter this that the tariff revenue amount can be trea ted as a proxy for expected savings on contamination that is for another paper and therefore we will assume the same for this work. 6 Beghin and Bureau (2001) note that analyses should be greater than the 2 digit level classification level of either the H armonized System or Standard Global Trade Analysis Project.
113 models used to estimate elast icities relies heavily upon a training manual published in cooperation with the United Nations Food and Agric ulture Organization by Perali (with an accompanying module by Conforti on welfare analysis applications), as well as additional sources within the literature. It is crucial to note that the focus of this paper is not on parameter estimation; such a topic is rife with opportunity for a more in depth examination worthy of an entirely distinct study Thus any self calculations are simply to provide a basis upon which to explore applications of the PE model to conduct a welfare analysis. Using selected parameters and applying additional methodologies to be discussed, we then c alculate tariff equivalents. T his data can then be input into adapted iterations of partial equilibrium models in Microsoft Excel based on those from Jechlitschkaj, Kirschke and S chwarz (2007) to calculate changes in welfare measures. It is also important to stress one point: when analyzing welfare results, the actual values are relatively arbitrary. What is to be of greater importance is the rel ation of welfare changes, especially magnitudes and proportions of the tariff equivalent losses to the free trade scenario (i.e. the scenario in which there is no SPS measure distorting trade). Included within the examination is a sensitivity analysis in which we estimate welfare changes at one standard deviation less and one standard deviation more than wn price elasticity of supply. Basic Assumptions With this model, we first must assume that all markets clear, reaching some equilibrium quantity and price. Thus our study will be confined to the situation of a perfectly functioning market that satisfies both of the Fundamental Theorems of Welfare Econom ics. In general, the presence of equilibrium implies competitive (Walrasian)
114 conditions and Pareto efficiency Of course, we note that t he working assumption of Pareto efficiency is far from the reality of imperfect markets. As Perali (2003) notes, [i] n m arket economies characterized by a strong government intervention or in economies in transition, the data are not generated by an efficient market mechanism. In such situations, neoclassical economic theory has a very weak explanatory power. Prices are not expression of a market equilibrium signaling consumers desire for a specific good and its relative scarcity in the market, but is the realization of government decisions. Government rather than market behavior should be more properly modeled in such occas ions. (p. 6) However, our methodology is not sophisticated enough to incorporate these issues and, in line with standard economics procedure, we shall continue with the assumption. Further, we shall assume that producers are price takers in the marketplace. Not only is this necessary for the above assumptions to hold, but it is also expressed within the econometric expressions of supply presented later in which production decisions are based on lagged prices. Th ere is sufficient reason to believe this assumption holds well, especially in the case of less developed and/or FFV markets. In addition, traditional relationships between buyers and sellers are ignored, and the assumption that the decision to export the co mmodity is based solely on a chosen optimization rule is made. The large country assumption is applied to the import markets. We also assume that the supply and demand functions are derived from Cobb Douglass style utility and profit functions. This enabl es us to apply simple log log estimation techniques, but it also assumes that imported products are perfect substitutes for domestic ones. Econometrically we impose short run models, which is most important for production estimates because it necessarily m eans that land and labor supply are treated as fixed. It is further held that there are no complementary or substitute commodity goods for those analyzed here, a point which is most relevant to
115 the econometric estimations of the model parameters. In many r espects, concern about this assumption should be mitigated by the fact that there are likely no close substitutes to the products examined Nonetheless, one cannot completely discount the impacts of the assumption on model outcomes. Finally, in line with B aldwin (1991) a price wedge methodology requires the strong assumption that no other significant factors contribute to the price wedge between two countries. By extension w e will assume that the only non tariff measures within the marketplace are SPS regimes. Calculating Welfare Impacts Again, recall the graphi cal model presented in Figure 2 4 which presents a three paneled Partial Equilibrium market for a single good. This model the one which we employ is a spatial PE framework. In this model there is a world represented which has an international market (the center panel) and is comprised of only two nations, one an importer who presents an Excess (Import) Demand sch edule, ED A and one an exporter who presents an initial Excess (Export) Supply schedule, ES B 1 Each nation faces its own internal supply and demand schedules such that quantity supplied in country i is defined as ( 5 2 ) and such that qua ntity demanded in country i is defined as ( 5 3 ) This allows us to define some quantity traded by a country as a difference ( 5 4 )
116 under which a positive value for the quantity tr ades indicates exports and thus excess supply to the international market while a negative value indicates imports and thus excess demand from the international market when the economy is open. Invoking the assumptio n of market clearing, we impose an equi librium condition such that: ( 5 5 ) This equilibrium point determines some quantity and price, Q T 1 and P w 1 respectively. The price, P w 1 is the World Market Price under free trade conditions; in our case, in which there are no standard protection measures and no SPS measures which restrict trade. This is the starting point for our model. Following standard welfare theory, we can further define different welfare measures for consumers, producers and the government ( in the case that there are tax revenues from a controlling policy). (Schmitz, Moss, Schmitz, & Furtan, 2010; Just, Hueth, & Schmitz, 2005; Jechlitschkaj, Kirschke, & Schwarz, 2007) Thus we define the measure of consumer welfare from goods consumption as Consumer Surplus, ( 5 6 ) the measure of producer welfare from the sale of goods as Producer Surplus, ( 5 7 ) which lead to the aggregated we lfare indicator of a nation in the market being
117 ( 5 8 ) Applying these conce pts graphically to the Figure 2 4, we can determine that to measure the welfare of the two countries under the scenario presented, we must be able to calculate the equilibrium of the Excess Demand and Excess Supply schedules. This can easily be done if we know the [own price] elasticity of individual supply and demand schedules in the two nations Since throughout the model we assume that all such curves are linear and derived from Cobb Douglass functions, and if we generally define elasticity as ( 5 9 ) Then we can show 7 that the elasticity of Excess Supply/Excess Demand is the weighted sum of domestic supply and demand elasticities without regard to sign such that: ( 5 10 ) where is the elasticity of the Quantity Traded for a given nation. Once we have calculated the above elasticities f or a given marketplace (say pineapple trade from Nicaragua to the United States), and met the equilibrium condition using an optimization program 8 to determine the initial conditions of free trade, we can continue with the next 7 Consult Appendix C for proof. 8 Which, in the case of this project, will be an adjusted mathematical model in Microsoft Excel employing its Solver add in. Other programs, such as GAMS, are a lso an option. However, because this project is also partially focused on developing tools for policy analysts some of whom are not advanced in economic programming we implement the most simplistic and widely available method.
118 steps of the welfare analys is by applying the price wedge method to find the tariff equivalent of non tariff measures. Interestingly, Calvin and Krissoff (1998) complete an analysis via a focus solely on the import market. We reproduce their graphic al conception here, in Figure 5 2. Again, we see a representation of how the price wedge, tariff equivalent method works; they decompose the prevailing [wholesale] price in the domestic import market, 9 into portions, with representing the effective tariff rate on imports of the product, and representing the tariffized equivalent of the costs of compliance to a phytosanitary regulation. Their world price is calculated based on CIF (cost of freight and insurance) prices. Similarly, Nimenya, Ndimira and Henry de Frahan (2012) provide an alternative price decomposition which, with some minor changes in not ation, we present in Equation 5 11. ( 5 1 1 ) where in the freight on board price of a good, IFC are the insurance and freight costs 10 and TC are the internal transactions costs within country j associated with a product shipped from country i to j If we evaluate based on wholesale prices or main port pri ces, we can ignore these internal transactions costs. Further, c ombining, or extending, these work s with the world encompassing model provided by Nolte and Rau 9 Graphically they show this being at autarky, but this need not be the case. However, this is a useful conceptualization here as it is an SPS measure which directly bans imports of fresh produce from Vietnam into the United States still. During the referen ce period, 2000 2005, an SPS measure also created an effective ban on imports of pineapple from Nicaragua into the US. Additionally, Figure 5 2 allows us to conceptualize welfare changes resulting from the elimination of SPS measure compliance costs. Here, area a is a loss to domestic suppliers but a gain for domestic consumers, thus it is welfare neutral; consumers gain area b from increased imports; the government accrues customs tax revenues of area d ; and the society as a whole experiences a dead weight welfare loss of the sum of area c and e 10 These can be approximated by taking the difference of the CIF and FOB prices.
119 (2006) we can see that WP and P i n Calvin and Krissoff (Figure 5 2) correspond conceptually to P w 1 and P A 2 respectively Alternatively, in their later work for the USDA, Calvin and Krissoff provide a gr aphical representation of the i mport market model with the decomposed price, whic h we replicate here in Figure 5 3 Thi s is crucial as it facilitat es two main things. First, we know that we need not calculate the curve ES B 2 for which data may be difficult to find, as has been noted throughout the literature referenced for this work. Instead, we can note that, following Calvin and Krissoff we can calculate the effective internal price in the importing nation, decompose it and carry these parts over to the international market as indu ced inflection points or kinks in the Excess Demand curve. Reapplying our initial condition that the intern ational market is in equilibrium, we can determine the new world price which will be faced by the exporting country where the now kinked schedule ED A crosses the schedule ES B 1 This new price can be mapped over to the domestic market of the exporting count ry and welfare changes calculated under the new conditions. The second advantage is that, because our focus in this project is solely on losses to farmers and not to the society as a whole, we need not necessarily compute the domestic demand curve within t he exporting (and in our case developing nation). Without this curve, we must recognize that some portion of the losses in Producer Surplus will be recaptured by domestic consumers, which in turn could be redistributed through transfer programs, but assumi ng no such programs exist or function poorly in a developing Nation is not far from reality. This is an advantage in our application because consumption data, especially on non staple, specialty products or at high
120 levels of specification are not readily a vailable, or even collected, by these governments. Calculating Model Parameters Briefly, let us recall standard e conomic theory Generally, theory tells us that quantity demanded is affected by the price of a good price of close substitutes for that goo d and income of consumers and demographic variables which act as proxies for preferences and tastes. Likewise, quan tity supplied varies with the price obtained in the marketplace the price of other outputs, input prices (such as land rents, labor costs, seeds and fertilizer) and producer characteristics (including education, risk preferences, farm size, etc.) 11 Mathematically we can express these as general form functions: Q uantity D emanded of a Good, j QD = d (P | Pj, Y, D) = d (own good price, price of other goods, income, demographic factors) Q u antity S upplied of a Good, j QS = s (P | Pj, Pi, O, D) = s (output price, price of other outputs, input prices, characteristics). From the discussion above, we know that it is crucial to know the elasticities of the required functions. In many studies of welfare economics, researchers do not undertake the task of calculating their own elasticity parameters. Instead they often borrow from the econometric literature where this has already been completed or, when there are assumed measure. For the purposes of this project, however, we will undertake our own, overly simplified, likely biased analysis and co mpare our results to those found within the literature as a benchmark. As a basis for c omparison, however, we provide 11 This microeconomic theory provides little help regarding the actual choice of specific demand or supply shifters. What is importan t is that other factors that can be included be exogenous that is not determined by the model and be relevant.
121 Appendix D, which presents a table summarizing all the previously calculated and relevant elasticities that could be found in the availabl e literature. Previously, in Equation 5 9 we defined the own price elasticity of a good in the g eneral case, and in Equations C 1 and C 2 extended this to define the own price elasticity of supply and demand 12 ty is the percentage change that occurs in one term of a function due to a percentage change in some related variable. Under the conceptual econometric equations above, and under the assumptions applied in this work as to the functional form of our models being derived from Cobb Douglass CES functions of utility and profit, it can be shown, as in the following equation taken from Perali (2003) that the relationship is given via natural logarithms 13 : ( 5 12 ) With this in place, we know that we can adopt a log log econometric model whose resultant coefficient will provide us with the own price elasticities for use in calculating applicable schedules of domestic and exces s demand and supply. We rely heavily upon a training manual from the United Nations Food and Agriculture Organization ( FAO ) (Perali, 2003) who presents an illustration of a practical method for estimating these parameters using time series data. This is su pplemented by consulting other, similar 12 At times these are referred to simply as supply elasticity and demand elasticity, although here we attempt to avoid confusion and tend toward s being specific in the terms applied. 13 It is important to note that the e lasticity is unit free and that i f the absolute value of the elasticity is greater than one (|E | >1 ) we say that it is elastic and if the absolute value of the elasticity lies betw een zero and one (0<|E| <1 ) it is inelastic.
122 endeavors in the literature All regressions are run under an ordinary least squares ( OLS ) system. Domestic demand p arameters Considering the general form expressions provided in the previous section, here we present a specific econometric regression model to estimate the own price elasticity of demand. Data are to be transformed into logarithms, both to ensure a better fit and f or analytical convenience. In a Cobb Douglas demand specification expressed in terms of quantity, the parameters are directly interpreted as elasticities. The model is to be applied in all countries, whether developed or developing. Following the example of Perali (2003) we define quantity demanded in log log form as the following expression: ( 5 1 3 ) Thus w e use price, consumption and income data from the data set s explained in the following section that have been assembled from various sources plus two demographic variables. The first of these, N is simply the national population while the second, d is the ratio of the population in terms of urban versus rural. This second de mographic variable acts as a proxy for taste preferences. (Regmi & Dyck, 2001) It s justification is greatest for the pineapple market, for which the literature states that consumption growth has been greatest in urban areas. Ho wever, it generally carries weight for all FFV products where transaction costs to rural markets are high and consumer preferences less differentiated. The reader will also note the presence of the standard variable, squared income. As Perali (2003) summar izes the well know n issue, onlinearities in demographic effects may imply the existence of household economies quadratic income term may be justified by the shape of the Engel
123 14 (p. 23) The greek lower case letter rho signifies the stochastic error term. Models similar if not identical to the one employed here are plentiful in the literature where simple estimation methods are used. (Berry & Pakes, 2003; Ferris, 2005) We, of course, recognize the caveat that more complex and robust estimation techniques are available and thus results here may be biased. Domestic supply p arameters Similarly, we recall the general form model of domestic supply, rely on the FAO technical manual and other literature for guidance, and present our own regression equation, again in log log form. However, first, it is beneficial to review two concepts via definition as the supply response for crops can be stu died eith er in the short or in the long run. Short run : producers can alter only the level of variable factors. Land and household labor supply, for example, cannot be changed. Long Run : producers can vary all factors of production and producers and resources can enter or leave the industry. In the short run, the elasticity of supply response is expected to be smal ler as fixed factors are not a decision variable. Long run elasticities of supply response can be very high since fixed factors become increasingly v ariable and can be reallocated. When choosing the dependent variable, total output is always preferred to area planted or yield. Further, we must recognize that agriculture is a sector with high amounts of uncertainty; f armers operate under uncertainty wit h respect to yields, and prices of inputs and outputs. A failure to incorporate the fact that farmers make most production decisions based on expectations within the econometric modeling strategy 14 A quadratic income term would make the elasticity change across the income distribution: the same good can be a luxury at low income levels and a necessity at high income levels.
124 may result in inadequate analysis. More importantly for our discussion, the farmer in developing countries is typically more risk averse. Here then, in maximizing their profit function, the farmer do es not equate marginal cost to [average] price but instead invokes an inequality such that they will produce at a poi nt where marginal cost is less than that [expected] price. Because the farmer cannot observe current year prices, which are met after production, they make decisions based on lagged prices. A similar argument also applies for production. This facet was wel l examined by Nerlove in the 1950s and has since been widely criticized and widely applied. (Askari & Cummings, 1977) Therefore we specify our model as: ( 5 14 ) In this model P t 1 is th e producer price in the previous period and q t 1 was the output in the previous period. The variable h is the total area under cultivation and t is a time trend included as a catch all variable to account for conditions such as weather, socio political cha nge and macroeconomic factors. We stress that, because of data unavailability, we are unable to include more refined parameters in the estimation such as a set of relative input prices or other exogenous shifters suc h as weather, farmer education, or technical change Equilibrium points In order to calculate the equilibrium points we recall that, as stated in Equations 5 2 and 5 3, Quantity Supplied or Demanded is a linear function wit h some slope and some intercept. Under the equilibrium condition, i n autarky we know that these two functions are to be equated, enabling us to solve for P at equilibrium such that:
125 ( 5 15 ) Once we have solved for the equilibrium price, it simply becomes a matter of inputting this price into the original Equations 5 2 and 5 3 to solve for Q However, to this point we have only calculated econometrically the own price elasticities for the necessary curves in the model. Therefore, we further note the general expression of own price elasti city presented in E quation 5 9. In this relation, we can identify two components: the first, is the slope 15 of the curve, Q with respect to price, P while the second, is the ratio of price to quantity. Therefore, if we have calculated eco nometrically the own price elasticity of some curve, Q we can decompose it and solve for the slope: ( 5 16 ) Likewise we must also determine the intercept of the linear curve. Via similar algebraic manipulation we find that the intercept is ( 5 17 ) From the data points used to calculate the curve we can either (1) find some mean price and mean quantity to approximate the quantity and price or (2) use some selected static figure from a single observation (year) to calculate the quantity price ratio With elasticity know and the price to quantity ratio approximated, it becomes a simple algeb raic matter to solve Equation 5 9 and determine the slope. In this study we will 15 and respectively, fr om Equations 5 2 and 5 3.
126 implement the first method of calibration and create what we will term an Average Year comprised of a price and quantity which are the arithmetic means of the respective observations for the years 2000 through 2005. Because we are implementing the Excel modeling framework from Jechlitschkaj, Kirschk e and Schwarz (2007) the only calculat ions we might possibly need so that we can then easily determine the new world price that accounts for the SPS measure. However, wit hin their Excel model, this is not needed if we make a slight adaptation to the model by defining the Effective World Price as ( 5 18 ) where lowercase Greek letter psi represents the tariff equivalent measure and is the world price determined at free trade equilibrium. The tariff equivalent measure includes all of the components from Equation 5 11 expressed, piece meal, as percentages of the free trade equilibrium price. Thus, calculating welfare becomes as simple as inputting the appropriate calibration measures for the two domestic markets and then solving the market at the varying tariff equivalent levels. The resulting measures can then be compared to determine welfare changes resultant from each stepwise addit ion of the price components, ending with the price residual which is the tariff rate equivalency associated with SPS measure compliance. Calculating Rate of Protection Equivalent To calculate the tariff rate equivalent of the SPS regulations, we compare the world price, calculated as the equilibrium price in the international market plus insurance and freight costs and the customs tax (tariff) applied at the border, with the domestic wholesale price in the importing market. For some markets, namely Vietna m to the
127 United States, current SPS regulations explicitly or effectively ban the import of some products. This would be the worst case scenario and thus we adopt it as the scenario for study. W e follow examples set forth in the literature to approximate t he insurance and freight costs by examining the difference between the CIF and FOB prices for a similarly situated homologue. (Calvin & Krissoff, 2005; Calvin & Krissoff, 1998; Yue, Beghin, & Jensen, 2005; Yue & Beghin, 2009; Nimenya, Ndimira, & Henry de Frahan, 2012) As becomes clear w ith an appropriate solving of Equation 5 11 as well as through the discussion surrounding it at the end of the section on welfare calculatio ns, the tariff equivalent of the SPS measures is the residual component of the prevailing importer domestic price, after accounting for the different real costs of trading in goods. Therefore to determine the protection rate afforded by the SPS measures, a ll one need do is calculate the residual as a percentage of the world price calculated by the market model. Data In the following sub sections, we briefly present the data sources for each market and demand/supply/excess supply schedule. Before this, it is worthwhile to note a number of general transformations which were completed to normalize data sets into standard units of measure. The data used for these transformations is discussed in the final sub section. Prices and values will always be expressed in real United States Dollars with 2005 as the base year. Quantities are to be measured in metric tonnes and areas in hectares. For each set of data after a brief discussion of its sourcing, we also engage in an examination of data rationality as proposed by Perali (2003) via graphical analysis. In an effort to conserve space, the graphs are not presented in this work, but are available upon request.
128 Domestic Supply United States Service presented the most reliable and refined data source for US production data. usable s et spans the years 1980 to 2006 and came from an amalgamation provided by NASS of historical data published in the annual Non citrus Fruits and Nuts Summary and Statistics of i Agriculture For the aggregated crucifers market, we employ a data set co mprised only of the broccoli market (no cauliflower) for all purposes (i.e. fresh and frozen) from the USDA. The set is also from NASS, compiled by USDA from issues of the reports Vegetables Final Estimates and Vegetables Summary It spans 1950 to 2010. All data had to be transformed because the USDA reports volumes in 10 00s of hundred weights (cwt), values in 1000s of nominal US Dollars and area in US acres The set for each product was a whole data set (it contained all of the necessary variables) and t hus no supplementation was required. Graphical analysis of data plots for price and production indicate that data meet expectations. Chiefly, prices for both crucifers measured as broccoli and for pineapples are increasing and take on a logarithmic sha pe. Production quantities for crucifers are found to be similarly positioned. Finally, matching the expectation based on output volume trends downward in a linear fashion. Europe T he ability to produce pineapple within Europe, of course, is minimal at least, if one considers only continental Europe. However, over seas states of Portugal and
129 France are in prime growing locations and have been actively engaged in production. These ex tra continental territories are fully part of the European Union, are administered by its sanitary and phytosanitary laws and other agricultural standards and receive the same access to EU and national farm programs. Unfortunately, reliable price data are not available for the French territories 16 for which production and area data was found. Theref ore, much like we did with the U nited States, we consider a pineapple supply market for the EU comprised of Portuguese production D ata on production and area are ta ken from the FAO STAT database and span the years 1961 to 2010. Price data in 1988 and ends in 2011, thus the estimation will be carried out over the period 1989 to 2010. For the crucifers markets in Europe, we consider France and Germany independently. The prices for the cauliflower market are used as the appr oximator for both products combined into a single market and the set utilized is specified as the total value of production at producer prices Data here is taken from the EuroStat datab ase available online for no fee P rice data for France was available for the entire period 1974 to 2005 Price and a rea data for Germany did not become available until 1990 ( after reunification), therefore the regression will only consider this period. Production data for both nations was available beginning with the year 1965 while area data for France was available starting in the same period. Turning to data analysis, we see a relative amount of price volatility for P ortugese Portuguese pineapples. The overall slope is upwards, but likely to be 16 The French over seas territories of French Guiana (South America), French Polynesia (South Pacific), Guadeloupe and Martinique (Caribbean) and Runion (Indian Ocean) were considered for inclusion.
130 relatively flat, indicating that we can expect the price elasticity of supply within this production market to be relatively inelas tic. Historically, the supplier price has remained relatively flat on the whole, although this is also influenced by the large historical swings in price. The price volatility is also found when plotting prices over production quantities in the crucifers m arkets. For France, we do see a generally linear, upward sloping historical trend for real prices, but we must also note the way in which these prices were calculated which may have an arbitrary stabilizing effect on the trend. French historical production saw volatility as well, with a general upward trend until the mid 1990s followed by a downturn in recent decades. Expectations about elasticity results, thus, are difficult to make. In the German supply market, the supply graph appear s less volatile, alth ough with a large jump in price between 80,000 and 140,000 metric tonnes supplied. This is likely associated with simultaneous changes born out of German reunification in the 1990s as the Federal German Republic absorbed the eastern, former communist DDR (Democratic German Republic). Nonetheless, the curve is expected to be positively sloped. Examination of the historical price trend appears relatively calm, also with positive slope and possibly logarithmic in shape. The production curve indicates, as one would expect, a large jump in production once the DDR is annexed into the nation. However, on either side of the historical unification mark, the historical trend in production quantity appears relatively flat. Vietnam and Nicaragua Data for all th e variables within the domestic supply regression models for the two 1961 to 2010. No further transformation of the data was necessary as the FAO reporting standards are the normalized units applied in this work. First, a simple review of the
131 data shows that there will likely be concerns as the level of detail is almost non existent. Observations on monetary values of production are truncated after the first digit (i.e. 6 m illion US dollars is given as the value, nothing more specific). A search for more detailed price data was fruitless. G raphics for Vietn am indicate from the very start that we are likely to encounter problems later on with our regression analysis; when p rice and quantity are graphed together, we see prices staying relatively flat within a given band despite dramatic rises in total quantity. Combining the historical analysis with findings within the literature, this is likely due to state subsidization and communitarianism typical of Communist political philosophy. This is bolstered by viewing price data over a time trend, where again, it stays within a flat band, typically between 136 and 140 USD per metric tonne for pineapple and 200 and 240 USD per me tric tonne for crucifers Production over time, however, does not exhibit the same trend. For pineapples i t increases at a rapid rate, plummets quickly in the mid 1990s and then rebounds even faster while for crucifers it exhibits a standard logarithmic shape Thus, while quantities have risen, prices have remained relatively steady, possibly indicating the aforementioned stabilization policies of the government and/or changes in technology, neither of which are included within the estimation model due t o data availability. Extraordinarily similar observations of data plots are made for Nicaraguan pineapple. In terms of the price spread, it stays flat (oscilates) within a band of 170 to 200 USD per metric tonne. H owever, here, the same argument over dome stic policies is not as likely. Data on production volume over time, however, appear to be almost
132 normally linear and upward sloping. Therefore, we will conclude that there are dramatic issues with the data reporting on prices, with quantities likely to b e more accurate. Domestic Demand United States Data on consumption of broccoli are used to approximate the crucifers demand Broccoli Statistics database, compiled by th e National Agricultural Statistics Service. This set covers the period from 1960 to 2011, but because of missing data from the World Bank on GNI, the set is cut down to 1970 to 2010. Consumption quantities (disappearance) are net of exports and reported in millions of pounds (lbs.) which are converted to metric tonnes. Prices are reported as seasonal averages in constant (base year 2005) USD per hundred weight (cwt.) and are therefore also converted into USD per metric tonne. Pineapple consumption data also are from the USDA, published in various versions of the Fruit and Tree Nut Yearbook compiled by NASS. The set begins with 1980 and ends in 2010. It only provides total quantity utilized, in millions of pounds (lbs.). Prices must therefore be approximated using a separated data set, also from USDA, which provides wholesale values of fresh product in mainland ports. This price data is reported in nominal USD per short ton (US measurement) and is transformed into real USD per metric tonne using price deflator s and volumetric conversions. Then, to simulate retail prices, we calculate retail markup percentage to be applied to the wholesale prices. This mark up is calculated by taking the gap between the retail and wholesale price as a percentage of the wholesale price. The gap is estimated based on a report from the USDA which included Nielson Homescan Data. (Stewart, Hyman, Buzby, Frazo, & Carlson, 2011)
133 Turning our attention to graphical observations, the price consumption respo nse plot holds the theoretically expected shape, with consumption relatively flat at low prices and then declining rapidly in a logarithmic function as price rises. Prices over time have been decreasing and, conversely, consumption has been increasing. As incomes rise, so too does the consumption of broccoli, in a relatively linear fashion. The data for pineapple are not as neat. Notably, the price response curve is upward sloping. The historical price trend appears to be almost quadratic in its shape, sl oping upwards but concave down, while historical consumption trends have a similar shape, but concave up. Finally, the income response plot appears to be linear and with the expected positive slope. Europe The crucifers market for the two European nation s is approximated using the total consumption of cauliflow er This data is available for both France and Germany from the EuroStat database. database. The data sets span 1974 to 2005, however for France there is a six year gap in consumption data between 1990 and 1995. Alternative sets could not be found. In analyzing the German data via graphical methods, we found a rather oddly shaped relationship in the price response (demand) schedule; it appea rs to have somewhat of a positive overall slope which would go against economic theory for Normal Goods. The historical price trend is upward sloping and relatively linear as is the historical consumption trend. Finally, when examining the income consumpti on response plot, we see the standard relation of consumption increasing with income, with a logarithmic shape. In the French market, the price response plot shows dramatic volatility and no general conclusion can be made on its shape. Prices over time hav e risen in a
134 seemingly linear fashion, but the consumption trend over time is indeterminate. The income response plot also provides little ability to make a determinate statement as to the general relationship of the data. Importantly, detailed hi storical consumption data was not available in a full set for EU pineapples It is likely that the EuroStat data exists, as a factsheet from the government of the Netherlands publishes some of it, but it could not be made available when requested for this project Because of this availability, we choose not to implement a regression analysis to calculate our own elasticities for EU demand. However, using the Food Balance Accounts data of FAO Stat, we are able to compile enough data to calculate an Average Year in order to calibrate the Partial Equilibrium models. Price proxies will be computed as they were for the American data. The retail markup for combined fruit and vegetables has been estimated at 290%, following analysis by Tschirley, Muendo and Weber (Tschirley, Muendo, & Weber, 2004) who identified ranges between 290 and 150 percent for European horticultural markets In fact, Kleemann (2011) completed a survey analysis within the Europea n market for pineapple and determined that between 2007 and 2010 the retail price premium attached to pineapple import wholesale prices in the EU ranged between 0 USD per kilogram and 1.17 USD per kilogram, with an average of 0.76 USD per kilogram. We adop t the maximum value computed there. Vietnam and Nicaragua Likewise, and as is common with developing nations, data sets on consumption of either pro duct in either case market were limited In fact, in examining potential markets for this project, no con sumption datasets could be accessed which were suitable enough to run a regression analysis on We were able to compile enough proxy data
135 from the FAO Food Balance Sheets sets in order to calculate an Average Year quantity for use in calibrating models. Pi neapple consumption data are reported directly. For the crucifers market however, approximation must take place. Theoretically, this consumption data could be estimated via a number of methods should one be able to access data on say, average shares of a ggregated consumption. Another example is provided by Corzine (2008) in her thesis work where she used household level survey data in Mozambique to estimate total consumption. Alternatively, a number of authors such as de la Cruz Medina and Garca (2005) have noted the percentages of supply which are typically been consumed in country. Because of the focus and nature of this project, we will estimate consumpt ion by determining weights equal to the percentage of total vegetable imports accounted for by broccoli and cauliflower. These weights are then multiplied by the total vegetable consumption data available from the FAO database. These data sets also do not include prices the refore we will adjust them using a selected representative retail mark up taken from the literature. In Vietnam, recent years have seen a growth in the modernization of the food retail sector and overall supply chain, but this is still new and full libera lization did not take place until 2007. (Maruyama & Trung, 2011) Specific retail price data could not be found for Vietnam for a comparison and own calculation, therefore we borrow from a value chain analysis report which exami ned three nations in the region, including Vietnam. (Asia DHRRA, 2008) T he markup of 253.5% applied is calculated from data on calamansi, a citrus fruit, in the Philippines. It is an average over the period 2000 to 2005. No sim ilar study could be identified for use in the case of Nicaragua, however the food retail market there appears to be similarly situated to that in Vietnam. (Reardon &
136 Berdegu, 2002; Reardon, Barrett, Be rdegu, & Swinnen, 2009; Reardon, Codron, Busch, Bingen, & Harris, 2001; Berdegu, Balsevich, Flores, & Reardon, 2005) This leads us to adopt the same domestic retail mark up. Additional Data Used to Transform, Calculate Tariff Rate and Transport Costs For all models, data on population and the proportion of it which is urban and rural National Income per capita using the Atlas accounting method, which is also from the deflated using the CPI to create real values with a base year of 2005. To take prices within the United States from nominal to real values we utilize the annual Consu mer Price Index data provided by the US Bureau of Labor Statistics. Applying 2005 as the base year, we calculate the annual price deflator which is multiplied by the nominal price, resulting in the real (deflated) price. Exchange rates to convert currencie s into US Dollars are taken from the EuroStat database and cross referenced with annual averages calculated from the set supplied by the US Federal Reserve Bank in New York and certified for trade usage. The final data needed are the tariff rates and the i nsurance and freight costs, which will be used under the decomposed price equation to find the residual which is the assumed SPS measure equivalency. The calculate d data are presented in Table 5 3. Tariff data are extraordinarily straight forward. The WT O provides a database which provides detailed historical data on ad valorem specific and special tariff rates applied by importing countries. The set begins with the year 1996, provides detail at the same HS 6 product level and on a reporter partner basi s. From this we are able to calculate the Total Average Tariff Rate, composed of all the en vigeuer customs taxes applicable
137 in the Average Year range of 2000 to 2005. Data on IFC costs, however are not as straight forward. Th e most straightforward method for including IFC costs would be to have them observed and reported directly. Of course, this is very rarely the case. Logically then it should be easy to calculate an estimation of the IFC costs; simply take the difference b etween the CIF price per units for imports and the FOB price for exports between a selected pairing of trade partners. However, as may well often be the case in economic studies, nothing is ever as simple as it may seem. The problems associated with trad e data reporting and validity are well recorded. (Hummels & Lugovsky, 2003; Gaulier, Mirza, Turban, & Zignago, 2008; Gehlhar, 2006) Therefore, when available, we adopt alternative methods. For US products we u tilize data from the US Census Bureau, in line with recommendations made by Gehlhar (2006). This sets reports the total CIF value of landed imports and the Customs Import Value (CIV), which is analogous to the FOB. For European markets, we are not so lucky and must rely instead only on the data from ComTrade. Also, because Nicaragua did not report export values to the ComTrade database for pineapples, we use Costa Rica as an approximator as it is the s of the product. The spread between the total CIF value and total CIV [FOB] value is taken as a percentage of total CIV [FOB] for each year between 2000 and 2005 and then averaged to create an Average Year tariff equivalent rate of IFC. Results Elasti cities and Model Parameters A summary of all the calculated elasticities and key statistical figures related to their regressions is presented in Table 5 1. Generally, model outcomes showed very
138 little significance on the price variables, whose coefficien ts are the elasticity measures required for the PE models. In many instances, goodness of fit was moderate to high, but these are likely false positive results. Domestic supply elasticities There were two supply models in which results were satisfactory crucifers and pineapples in the US. Afte r running the Nerlovian model regression on the identified data set, we find an own price elasticity of supply for US broccoli of 0.41. Each of the variables in the model has the expected sign, and all are statisti cally significant at or below the 10% level, with the exception of the trend variable. The model has an extremely high goodness of fit, with an R squared of 0.9946 and an adjusted R squared of 0.9941 With a Durbin Watson result of 1.815 we conclude that w e need not worry about first order (auto ) correlation being present within the model. Because our data set is sufficiently large (i.e. there are more than p*10 observations), we will accept the validity of the D W statistic. The White Test outcome, with a p value of 0.4046, leads us to conclude that there is likely no heteroskedasticity problem, that the errors are independent of the regressors and that the model is correctly specified. (White, 1980) Likewise, the calculated s 0.87. The model had good fit at 97% unadjusted r squared and an almost ideal D W statistic of 2.031. Based on p o heteroskedasticity, errors are independent and model specification is appropriate. However, the caveat here is that the model has a low number of observations; with 4
139 total parameters and only 26 points, we may not be able to assuredly accept these results 17 The own price ela sticity term for French cauliflower was also statistically significant. However, it carried a sign contrary to standard expectations (the coefficient was negative). Goodness of fit in the model was low, at 42.37 percent. However both the D W statistic and value were almost ideal. For all other supply functions, the own price elasticity values were not statistically significant, some being just barely outside of the 10% level, others being far, far beyond it. Three supply markets return v alues with unexpected signs (crucifers in France and Germany and pineapples in Nicaragua). Interestingly the first of these showed a near ideal D W value, but its number of observations was in and it had an abysmally low R squared of 0.3923. Both supply models for Vietnam showed incredibly low elasticity values, which, given our observations in the data and the historical background, is not surprising 18 Nicaraguan pineapple and Vietnamese cruci fer supply models are likely to suffer from positive autocorrelation. On a positive note, all of the calculations indicated that supply for the two goods was price inelastic and their relative magnitudes matched expectations from the graphic analysis of th eir data (in absolute value terms). 17 Although, to add an even fu rther caveat, N=26 > p*5, therefore we are not likely to encounter the same problems of micro model is in the grey zone. 18 It might be interesting to add a dummy variable to account for the period before and after the political reform in the country, but we refrain from doing so here as this is beyond the intended scope of this project.
140 Domestic demand elasticities Moving into the realm of demand, three of the models held coefficients of elasticity which were found to be statistically significant. Again, two of these were both of the US markets. Here it is important to make a specific note about the results reported in Table 5 1 ; for US crucifer demand the outcome reported is based on an altered model from that presented in Equation 5 13 as the variable of Income squared was dropped to attain this es timate. Because of uncertainty about the presence of income economies of scale in these markets, for each demand equation we ran a linear regression that both included and omitted this variable. For this case of US broccoli, with the income removed the p value on the own price coefficient went from 0.1466 to 0.0139, a large change relative to all the others observed in our modeling endeavors. Removing the term had only a m inute negative impact on goodness of fit (less than 0.01 change) and on measures for autocorrelation and correct model specification. Both of these significant elasticity coefficients showed that demand for the two goods in the US is inelastic and they were relatively similar in magnitude with 0.51 for crucifers and 0.57 for pineapples. However, both also show signs of heteroskedasticity and autocorrelation. The crucifers model, with 49 observations, can be said to likely suffer from these issues, while the pineapple model, with only 29 observations, is inconclusive as it is not a model with severe micro numerosity yet is still a small sample. Due to data issues, we could not estimate a demand regression for pineapple I nthe EU, however for crucifers the elasticity values between France and Germany were similar, at 0.11 and 0.13, respe ctively, but only the German value was statistically significant. Based on the Durbin Watson value and the observation size, we conclude that there may be an autocorrelation problem but we cannot say so conclusively.
141 General discussion on elasticity e stimation results In the points laid out in the previous section, it becomes painfully clear that the estimation results were greatly biased, principally by the data quality, as well as by the invoked constructs of the regression models. Only half of the e lasticities calculated were statistically significant, and several on the supply side had signs which defied theoretical expectations. Therefore, it would be imprudent to proceed with the analysis on welfare using these results only Recognizing this challenge, we adopt a methodology mentioned previously and common within the literature, including by Calvin and Krissoff; we shall assume standardized elasticity values for use in defining the partial equilibrium models and calculating tariff equivalents and welfare impacts. The imposed values are reported in Table 5 2 The choice of these values is based upon previous estimations found within the literature and informed by gen eral observations of the data used to conduc t our own estima tions as well as those results which held to be significant 19 For American supply elasticities we will adopt both of the values for crucifers and pineapples, as they are similar to those found within the literature. Additionally, these measures approximat ely map to expectations given the nature of the production processes and the structure of the markets. These same elasticities will be utilized (imposed upon) the European markets. For the EU crucifer supply market, the decision to disregard our own estim ation results is based on the fact that (1) the results had appropriateness based on statistical tests and (3) the absolute values calculated lie at 19 As an example of the later, we know from our graphical analysis of the price output response (supply) of crucifers in Germany that a roughly 100% increase in the price was associated with approximately a 130% increase in production. If these numbers were the exact ones, the own price elasticity would be around 1.3.
142 the very lower extremity of wh at can be found within the literature. [See, for example, Tiffin et al (2011) and Durham and Eales (2006) ] The decision with regards to the European pineapple market is similarly ju stified, however, we do note that the US measure is likely larger than what European producers actually face. For Nicaraguan pineapple and Vietnamese crucifers and pineapple producers we adopt a value of 0.2. This measure is informed by early works such a s Behrman (1968) and Islam and Subramanian (1989) as well as a direct measure published by Vesdapunt (1990) for Thailand, which is one of the leading producers of pineapple This measure accounts for the likely risk aversion that developing nation producers have and their uncertainty about prices as well as their inability to easily adjust production methods due to restrictions on access to the means of production and credit. Again, this may be larger than what producers might face, especially in an until recently planned economy (Askari & Cummings, 1976) but it facilitates our central focus. Demand measures to be utilize d are determined similarly. The US own price elasticity estimated for pineapples and crucifers fits well within the standard range found for other crops. In fact, our calculation for crucifers is almost on par with that found by Acharya and Molina (2004) for the same product grouping. For uniformity, we also apply these to the European case. Finally, we will employ a domestic own price demand elasticity of 1.35 for pineapple in Vie tnam and Nicaragua. This figure w a s computed for the world by Hallam et al (200 4). They also calculated values for Vietnam, Honduras and Guatemala which were massively large but we shall adopt the more conservative value for the two developing nations nonetheless. For Vietnamese demand elasticity of crucifers, we apply the aggregated measure calculated by Seale, Regmi and Bernstein
143 (2003) for vegetables and fruits in that nation. It is simply a matter of coincidence that it is equal to the result for the developed nations. Selecting o ther model parameters As mentioned in the methodological section, the final two parameters needed to calibrate the model and thus define the supply and demand curves are representative price and quantity values. Because of concerns over data reporting and uniformity, we elect to implement an Average Year calculation. From each set of price and quantity data we take the arithmetic mean of the observations for the years 2000 through 2005. The range is selected based on (1) data availability, (2) improved data reporting and tracking procedur es, and (3) its historical position, being a sufficient period after the establishment of the WTO but before the onset of the recent economic crisis. In the data section we considered separately several issues related to those [consumption] markets for wh ich enough data was not available to run a regression model to attempt auto calculation of own price elasticity. There we also laid out how an appropriate micro data set was compiled for the purposes of determining the price and quantity values of the Aver age Year. Note that for several of the demand markets retail price data was not available and thus proxies were used. This is a known bias in our model parameters. Final calculated values are presented in Table 5 2. Here we make note of three observations in particular First, that for all markets except the German crucifers market the average demand price was generally greater than the average supply price. Second, the average supply quantity in the United States and France outstrips the average quantity demanded in those nations respectively. When running the market models, it is likely that we will find that these two nations will actually be exporters to the world market and not importers, eliminating them from any further
144 analysis. Finally, Nicaragua shows a supply price quantity ratio ten times larger than that of Vietnam. This will translate into a significantly steeper slope of its supply curve (and may help indicate why the calculated supply elasticity in our regression analysis held the wrong sign ). Partial Equilibrium Results After defining all the parameters, we finally turn to the central analytical focus of this work. Inputting the data into the Excel market models and using the Solver function to induce the equilibrium condition, we determin e prices, quantities and welfare measures in stages: first autarky, then free trade, followed by trade with standard barriers (transactions costs and normal price policies). After this third stage, we are able to calculate the price residual and then the t ariff equivalent rate equivalent of presumed SPS measures. The final stage is to solve the equilibrium for this tota l equivalent rate of protection. To measure the effects of the policies on producers, we calculate the change in producer surplus between th e SPS equivalent equilibrium solution and that of the scenario with none (i.e. IFCs and normal tariffs only). Because the total values of Producer Surplus may be relatively ambiguous, we report the changes as percentages, those with negative signs being l osses and those with positive, gains. All r esults are reported in Tables 5 4 and 5 5. Unintended results First, a rather odd note; when running the spatial PE models with the data reported earlier, we found that, under these parameters, trade in crucifer s goes the opposite direction. Instead of being an exporter, as expected, Vietnam is shown to be a large importer. France showed an incredibly low equilibrium price at autarky of 179.62 USD per metric tonne with Germany a distan t second at 511 USD per met ric tonne and then
145 the United States at 595.28 USD per metric tonne. Under the free trade only scenario, producers in Vietnam lost anywhere from 44.66% to 84.32% of their welfare. Clearly, this was not an expected result from the outset t of this project. However, it was less surprising give n the data presented in Table 5 2. More importantly, it means that this product market is eliminated from further analysis in under the auspices of this project. This is most regrettable because it was the market where w e hoped to examine differences among EU member states. After finding these results, the author did engage in research to determine if applying the tariff rate equivalent in reverse would be feasible, but upon reviewing relevant literature such an exercise proved not to be justifiable, as Vietnam has relatively low SPS standards on imports. Some potential causes of this outcome will be discussed in Chapter 7 Tariff rate equivalents Continuing our analysis, our focus now turns, necessarily, to the pineapp le market exclusively. Calculated tariff equivalent rates of SPS measures for the four possible world market configurations considered varied wildly. From 14% of the base world price for exports from Nicaragua to the United States, to 5,565% for exports or iginating in Vietnam and destined for the agglomerated European Union. This extreme value is, without doubt, due to the underlying assumption that the EUs only production source of pineapples with similar safety characteristics would be its severely limite d own production. To exacerbate the situation, because of data issues, this was calculated only with data from Portugal, excluding other areas chiefly the several French overseas territories in tropical climates. While large variation is to be expected even for the same commodity across different years in the SPS equivalent rate, it is difficult to accept this number as given. Similarly, the equivalent rate for SPS in a world market
146 composed of j ust the EU and Nicaragua is more acceptable, although sti ll in the high er rang e s at 384% As a comparison, this value is nearer to th at found by Nimenya, Ndimira and Henry de Frahan (2012) for frozen fish filets than their similar calculations for horticultural crops. When looking at fire blight restrictions in Japan for US apples, high as 500%. In the US market, the 14% for trade with Nicaragua may be due to the closer ties and technic al assistance afforded to the country by the developed trading partner since it is in close geographic proximity and benefits from special trade programs under American Caribbean Basin Economic Recovery Act, the Caribbean Basin Trade Partnership Act and ot hers. Subjectively, the 49% value for US Vietnamese pineapple trade is on par with the majority of similar estimates in the literature. Producer s urpluses of developing n ations First, let us examine the general welfare improvements that developing nation producers might accrue under purely free trade. These results are available in Column 7 of Table 5 5. Movement to an open economy within the European world models showed drastic increases in welfare as the supplier rushed to fulfill the massive European e xcess demand for pineapples just over 726,000 metric tonnes at its largest, when the only world supplier is Vietnam. In this world, as one would naturally expect given the average price and production data, European producers almost entirely stop their pr oduction. The same is true for when the supplier is Nicaragua, although production economies here mean far fewer imports at a dramatic world price of 59,347.03 USD per metric tonne. The results for the opening of the US market from
147 autarky are much more re asonable, but producers in the export suppliers of Vietnam and Nicaragua still see producer surplus gains of 93.89% and 105.83%, respectively. Accounting for insurance and freight costs, as well as domestic price policies in the import nations, closes th e gap a little bit. As we can see in column 10 of Table 5 5, these barriers have in terms of absolute magnitude a greater negative impact on nation farmers accrue. The ad justed world prices invoked by a kink in the Excess Demand curve of the large, importing markets are lower; around 10% lower for all markets except the EU Nicaragua world, where the price falls by 50%. These prices allow us to calculate the residual. Fina lly then, with the SPS equivalents calculated, we solved the models one final time to determine welfare measures. Loss rates to developing nation producers as a result of SPS compliance were between 12.62% and 89% of the surplus which would have been gaine d in the absence of such measures. Nicaraguan and Vietnamese producers lose most when faced with the European SPS regimes at 60.71% and 89.67% Losses in the American market are less, at 12.62% and 23.61% respectively. Sensitivity Analysis on Elasticitie s To conclude our analysis, we examined the sensitivity of welfare losses to developing nation producers at increasing own price supply elasticity. For the analysis we selected the originally induce 0.2 elasticity, as well as 0.5, 0.8 and perfect price ela sticity of 1. When conducting the market models, we found that at the very first selected level and for every level thereafter, the model could not be solved in an economically meaningful way for the two European worlds. Therefore, we omit these results fr om Table 5 6 and instead focus solely on the American markets. Because
148 Table 5 6 is data intensive, we also present Figure s 5 4 and 5 5. The first of these, Figure 5 4, presents changing ad valorem protection equivalents of SPS regimes over the range of elasticities. Here we see that the protection afforded American producers by the compliance costs imposed on imports grows at a much faster rate for Vietnamese producers than for Nicaraguan homologue s as price responsiveness increases. Between perfect elasticity of supply and 0.2, the effective rate on Vietnamese imports almost triples, but for Nicaragua it only doubles. Likewise, Figure 5 5 graph s the losses for the two supplying nations across the d ifferent elasticities. Generally, we see that as the responsiveness of producers to price increases, the expected welfare losses from SPS measures to foreign firms also increase. For Nicaragua, a movement from 0.2 to 1 more than doubles the loss rate, wh ile for Vietnam the losses increase by a factor of 1.69 (just less than one and two thirds). We can also note that the 0.8 measure is very near that imposed for domestic suppliers in developed nations. At this near equality of price responsiveness, Nicarag uan producers lose almost 5 times more (or an absolute difference of19.66%) than American producers gain from the protection afforded by SPS regime compliance. Conversely, American producers gain 3 times more (or an absolute difference of 75.55%) than wha t Vietnamese farmers lose. Discussion Overall, the analysis completed within this pillar can be said to confirm the proposed hypotheses stated in Chapter 1 The results presented indicate that, within the case study of the pineapple market presented here European SPS regimes as measured by tariff equivalents are larger than those in the United States. As such, welfare of firms in less developed trading partners producing for export lose more in percentage terms that they do under compliance with the American standards. As of
149 yet, however, what cannot be determined with complete certainty are the actual magnitudes of these differences. In large part, the reduced certainty level about magnitudes is because of concerns over data availability and reliabi lity. Additionally, it is clear that these magnitudes can be greatly influenced by the assumptions made regarding the underlying structure of the market, namely the responsiveness of supply to movements in prices; t he more elastic these curves ar e, the greater the calculated losses in producer surplus. What is more, the choice about what elasticity measure is correct can be difficult. In attempting to compute our own model parameters, the results were highly dependent upon the quality of underlyin g data and the theoretical assumptions made. These problems were more pronounced in regression models run for developing nations, further complicating the ability to feel confident about the total magnitude of spatial partial equilibrium outcomes. Further, it is important to note here our concerns over robustness of the results on the tariff equivalents and producer surpluses regarding the European Union. As pointed out in the section presenting these results, the magnitude of the outcomes are relatively h igh, notably a tariff equivalent of SPS measures at some 5,565% for pineapple imports of the EU from Vietnam. While previous works have shown large amounts of variation in outcomes, such a drastic number is concerning. We can identify a number of factors w hich likely bias the result in this work, creating such a large magnitude. First, the assumption that the world is comprised of only two nations, without a Rest of the World (ROW) market or market share accounting scheme introduced to deflate the outcomes modeling assumption applied which sets the reigning price under compliance at the
150 autarky price as adopted from the Calvin and Krissoff works (1998; 2005) on US Japa nese apples. I n the real world this is highly improbable; the EU would likely supplement from some other supplier (se e previous point ) or simply not consum e at such a high autarky price. This has an even great er impact when combined with the third element likely causing bias: the underlying structure of the imposed domestic market for Vietnam. As an example, compare the world prices under free trade with the EU between Nicaragua and Vietnam, found in Column 5 of Table 5 5. The Vietnamese result is some 13 t imes smaller than in the case of a world with Nicaragua as the exporter. Because we assumed identical supply elasticities for the two nations, in our model this difference then is due to the calibration results. Fourth, we recall the fact that the entirety of the European domestic production market was highly limited; other tropical regions such as the plethora of French overseas territories were not included due to data availability. If they were, it is likely that the resulting calibration parameters in t his market would not be so exaggerated. This stresses the importance then of all three parameters (elasticities and calibrating prices and quantities). With these four issues in mind, one should take the actual magnitude of our results here with a grain o f salt. Clearly further work is needed to improve the robustness of the results, although the general outcome supports what is already in the literature about the stringency of EU SPS policies.
151 Figure 5 1 Conceptual framework of partial equilibrium modeling in agricultural m arkets. Source: Author. Adapted from : Parappurathu 2007. Partial equilibrium models for agricultural policy analysis. (pp. 2; Figure 1). National Centre for Agricultural Economic and Policy Research, Government of India. New Delhi. Output prices Input (factor) prices Investments o Research & development o Irrigation o Equipment o Education Weather Others Area Yield Production Prior stock Supply Price Imports Demand Exports Income Population Preferences o Changing habits o Culture o Information o Etc. Prices Others Human consumption (food) Seed Feed Other uses
152 Figure 5 2 Import market showing trade and welfare eff ects of removi ng an SPS measure. Source: Author. Adapted from Calvin and Krissoff December 1998. Thechnical Barriers to Trade: A Case Study of Phytosanitary Barriers and U.S. Japanese Apple Trade. (pp. 358; Figure 1(a)). Journal of Agricultural and Resource Economics.
153 Figure 5 3 Import market showing decomposed p rice wedges with an SPS measure. Source: Calvin and Krissof f. October 2005. Resolution of the U.S. Japan Apple Dispute: New Opportunities for Trade. (pp. 13; Figure 2) Economic Research Service of the USDA Washington, D.C. FTS 318 01. Original source: ERS of the USDA.
154 Table 5 1 Results of econometric models estimating elasticities Country Function Elasticity Standard Error N F stat p val. R squared Adj. R squared Var. not stat. sig. Durbin Watson d stat White's Test p value Crucifers France Demand 0.109 0.18255 26 0.7804 0.109 0.1138 ALL 2.468 0.4142 Supply 0.144 0.07368 31 0.005 0.4237 0.3351 1.963 0.717 Germany Demand 0.134 0.07691 32 0 0.6745 0.6119 ln(Income), Population, Urban ratio, Income^2 1.119 0.8123 Supply 0.187 0.11367 13 0.0049 0.8162 0.7243 Production, Price, Year 2.052 0.4063 USA Demand 0.506 0.1958 41 0 0.9823 0.9804 ln(Income), Population, Urban ration (dropped Income^2) 0.998 0.473 Supply 0.410* 0.08904 49 0 0.9946 0.9941 Year 1.815 0.4046 Vietnam Supply 0.064 0.07476 49 0 0.9917 0.991 Price, Year 1.247 0.354 Pineapples EU Supply 0.178 0.11482 18 0 0.8527 0.8074 Prod, Price, Year 2.185 0.515 Nicaragua Supply 0.160 0.16577 49 0 0.9704 0.9677 Price 1.381 0.6062 USA Demand 0.568 0.27274 27 0 0.9719 0.9652 ln(Income), Population, Urban ratio 1.095 0.2535 Supply 0.871 0.26541 26 0 0.9703 0.9646 2.031 0.4967 Vietnam Supply 0.053 0.9663 49 0 0.9573 0.9534 Price, Year 1.232 0.2828 *Indicates statistically significant at or below the 10% level.
155 Table 5 2 Imposed elasticities used for base welfare analysis and Average Year prices and quantities for calibration calculations. Product Country Function Elasticity Average Price Average Quantity Crucifers France Dem and 0.51 412.37 246133.33 Supply 0.41 250.22 430800 Germany Demand 0.51 349.22 217000 Supply 0.41 420.86 165040 USA Demand 0.51 720.83 713760.7 Supply 0.41 654.31 818044.9 Vietnam Demand 0.51 785.3 4 44487.77 Supply 0.2 222.16 29213.17 Pineapples EU Demand 0.57 2096.2 9 1135333.33 Supply 0.87 1958.8 6 1047.1 4 Nicaragua Demand 1.35 664.33 40166.6 7 Supply 0.2 187.93 47000 USA Demand 0.57 1395.95 524816.81 Supply 0.87 343.72 261420.4 1 Vietnam Demand 1.35 488.9 3 352000 Supply 0.2 138.31 369633.33
156 Table 5 3 Tariffs and tariff equivalents of insurance and freight costs own calculations. Product Exporter Importer IFC tariff equivalent Tariff Sum, as a percentage Crucifers Vietnam France Not calculated 0.116 11. 60% Germany 0.472868 0.116 58.89% United States 0.063458 0.088 15.15% Pineapples Nicaragua European Union 1.324038 0.058 138.2 % USA 0.100687 0 10.07% Vietnam European Union 0.185568 0.058 24.36% USA 0.169361 0 16.94% Costa Rican data used as a proxy
157 Table 5 4 Results of initial partial equilibrium analysis in crucifer markets Autarky Free Trade Trade Partners Price Quantity World Price Quantity Traded Change in PS France 179.62 376051.81 212.48 57 699.58 26.73% and Vietnam 995.07 39429.00 84.32% Germany 511.00 178708.60 567.81 17246.17 16.02% and Vietnam 995.07 39429.00 48.99% United States 595.28 786938.30 607.73 14975.94 2.96% and Vietnam 995.07 39429.00 44.66%
158 Table 5 5 Results of spatial partial equilibrium analysis of SPS measures in pineapple market s calculations. Autarky Free Trade IFC + Tariffs SPS Trade Flow (2) Price (3) Quant. (4) World Price (5) Quant. Traded (6) Change in P S [%] (7) World Price (8) Quant. Traded (9) Change in PS [%] (10) Tariff equiv. [%] (11) World Price (12) Quant. Traded (13) Change in PS [%] (14) EU 257858 73090 4563 726540 99.95 3914 700086 12.83 5565 590 221174 3603.7 3 importing from Vietnam 402 457676 1742.39 16.82 89.67 US 971 645372 698 293781 45.95 628 249730 9. 95 49 502 139056 26.46 importing from Vietnam 402 457676 93.89 11.92 23.61 EU 257858 73090 59347 148492 93.59 29678 129118 38.70 384 13625 110024 40.46 importing from Nicarag. 510 57387 30026.7 2 56.46 60.71 US 971 645372 930 39249 7.64 845 34470 0.98 14 755 28304 1.26 importing from Nicarag. 510 57387 105.83 10.93 12.62
159 Table 5 6 Results of sensitivity analysis, changing price elasticities of supply for Vietnam and Nicaragua Source: Free Trade IFC + Tariffs SPS measure Trade Flow World Price Change in PS World Price Change in PS Tariff equivalent World Price Change in PS 0.2 United States 698.92 45.95% 628.77 9.95% 49% 502.36 26.46% importing from Vietnam 93.89% 11.92% 23.61% United States 930.82 7.64% 845.23 0.98% 14% 755.38 1.26% importing from Nicaragua 105.83% 10.93% 12.62% 0.5 United States 571.52 62.90% 526.53 14.95% 78% 406.95 60.62% importing from Vietnam 119.39% 11.57% 32.05% United States 894.44 14.28% 815.32 1.70% 17% 716.13 2.49% importing from Nicaragua 196.07% 12.97% 17.68% 0.8 United States 474.55 73.80% 445.90 19.26% 111% 342.93 113.21% importing from Vietnam 130.01% 10.60% 37.66% United States 844.60 22.99% 774.95 2.96% 23% 660.03 5.43% importing from Nicaragua 309.35% 14.35% 25.09% 1 United States 426.45 78.54% 404.66 21.48% 133% 313.20 156.32% importing from Vietnam 132.29% 9.96% 40.10% United States 804.85 29.63% 742.83 4.09% 28% 620.13 8.77% importing from Nicaragua 391.41% 14.82% 30.31%
160 Figure 5 4. SPS regime tariff equivalents at differing supply elasticities Source: Author, based on own calculations. 0.2 0.5 0.8 1 Vietnam 49.00% 77.81% 110.70% 132.86% Nicaragua 13.54% 17.43% 23.24% 28.38% 0.00% 20.00% 40.00% 60.00% 80.00% 100.00% 120.00% 140.00% Tariff equivalent of SPS regime Tariff equivalent measures of SPS regimes at different assumed elasticities of domestic supply in developing nations
161 Figure 5 5 Percentage losses in consumer surplus at differing supply elasticities Source: Author, based on own calculations. 0.2 0.5 0.8 1 Vietnam 23.61% 32.05% 37.66% 40.10% Nicaragua 12.62% 17.68% 25.09% 30.31% 10% 15% 20% 25% 30% 35% 40% 45% % Loss of Producer Surplus Losses of producer surplus from SPS measures at different assumed elasticities of domestic supply in developing nations
162 CHAPTER 6 ANALYSIS OF PILLAR I I: POTENTIAL FOR MIC RO LEVEL ANALYSIS OF STRUCTURAL CHANGE IM PACTS ON HOUSEHOLD W ELFARE Methodological Background Before continuing we make one very spe cific note, for the sake of transparency: the central focus initially intended for this work was, in fact, this pillar and an investigation of structural change using household level data. Due to access issues much like the farmers we hoped to examine a n applied examination had to be abandoned. However, it is still a meaningful discussio n to begin here and thus, in Chapter 6 we initiate what hopefully with better resources can on e day come to fruition. With that said we have contend ed Chapter 2 that a fuller compl e ment of outcomes must be consider ed; including those that consider the oft en times conflicting values orientations expressed within and adopted by the body politic of a given n ation As we have discussed in Chapters 1 and 2 of this thesis in the case of trade and development, sanitary and phytosanitary ( SPS ) measures are one realm where a balance must be found and thus more detailed information to inform decision making must be available. To put it differently, cost benefit analysis of an SPS policy should, we argue, incorporate an analytical approach which examines impacts beyond the home market and at a deeper level. This thesis focuses on the first half of this equation: the costs. In Chapter 5 we investigated the initial level of this cost analysis via the welfare impacts aggregated at a group level. However, economic theory per se gives us a clue that the effects are likely to reach a level deeper than this; much of the literature on welfare economics reiterates the micro economic conc ept that as the barriers to market access increase, firms who cannot adapt exit the market. (Oyejide, Ogunkola, &
163 Bankole, 2000) In the development context, this can mean the difference between sustained growth and economic sta bility and continued poverty. Before continuing, it is worth stating a preliminary set of objectives for such a research project: Examine what factors (variables) hold a significant impact on household welfare in selected export production markets; Determine who wins and who loses from structural changes (ie aspects of welfare distribution) in developing nations and to what magnitude do they win/lose; Test whether impacts differ across different measures of welfare; Discuss the implications of SPS m easures in relation to issues of development, namely vulnerability and poverty; Proposing a Methodological Approach engaging in analysis beyond the market level. Due to the natu re of the theorized impacts, it seems most appropriate to examine changes within the agricultural firm Many times in the developing world, the greatest concern for development outcomes the ones which must be balanced with the need to afford protection to home consumers and environment through standards regimes lies then with the agricultural household and with small farmers. One key piece of literature has been identified which has initiated such an analysis within the field, that by Maertens and Swinnen (2009) no substantially similar investigations could be found, especially no ne with a focus on SPS compliance with the exception of Minten, Randrianarison and Swinnen (2009) In man y respects, this rarity and need for expansion was a driving force behind the desire to implement and examine their methodological approach and testing their results with
164 noted improvements. Based on the discussion presented in Chapters 1 and 2 we propose the inclusion of the following hypotheses to test the following hypotheses: Following Maertens and Swinnen (2009), that engaging in the supply chain for Fresh Fruit and Vegetable exports increases welfare measured on income, consumption and expenditures of the farm household; Also following Maertens and Swinnen (2009) that the welfare of those engaged in the supply chain as wage laborers is improved more under higher SPS measures in export markets than that of small farmers working to produce themselves; That initial endowments of capital in land, labor, human and social terms have a significant and positive impact on welfare measures; That there is no statistically significant difference on impacts evaluated across different measures (i.e. that impacts on income, consumption and expenditure are relatively similar) In many respects, these hypotheses are informed by the keystone work upon which this methodological prospectus is based. However, the final two are specific additions which are reactionary to that work. The general empirical framework to be implemented is rather simplistic in its econometric form; a standard treatment effects model is adopted. In its simplest form of this this model type, an OLS regression is run on the following equation: ( 6 1 ) where Y is t he welfare measure for a household subject to treatment i W 1 and W 2 are the treatment variables, in this case the structural role the household plays in the production chain (either a contract farmer or a wage laborer) and X i is a vector of covariate shif ters including capital endowments, household demographics and locality. Thus, Maertens and Swinnen provide a simple yet effective work around for the selection bias problem that could be encountered with the econometric methodology. Additionally, they pro vide a patch of similar quality for addressing the fact that in this
165 analysis, the treatments hold the potential to be non mutually exclusive (i.e. an agricultural household might engage in the selected product market through both activities simultaneousl y). From this observation the y develop and implement two altered models, one a regression which integrates propensity scoring results from a bivariate probit model and a second using a propensity score matching technique. Because the focus of this thesis shifted to the welfare phase, we will refrain from any more in depth discussion of econometric methodology here. What is important to note is the non econometric methods used in both the Maertens and Swinnen paper (2009) a nd in Minten, Randrianarison and Swinnen (2009) In both cases analysis was completed based upon surveys administered to households However, this necessarily limited their analysis to welfare measured by income and als o meant that anything beyond the year in which they were administering surveys was recall data. For differing reasons, both of these elements present avenues for improvement if appropriate data can be found. In the following subsection we consider an alte rnative possibility for field methodology that might allow research to transcend these issues: the use of historical, household panel data. Further, in Appendix E, we present a summary of potential variables about which data is either required or would be useful. Identifying Potential Data Sources Finding such data sets presents a challenge in and of itself. The most likely source are the Living Standards Measurement Surveys coordinated by the World Bank. In fact, after an extensive search, no other suita ble sources could be identified other than these as most other household panel surveys do not span a long enough period of time in regards to important dates for SPS issues or lack pertinent variables. In reviewing the LSMS data sets a number of nations we re considered. Peru d id n ot span
16 6 sufficient time horizon and the case specifics of Bosnia and Herzeg ovina proved untenable (the set spanned the armed conflict there and the nation has limited arable land and role of agriculture in the economy ). Nicaragua was almost ideal. It was requested but there was no response from the Nicaraguan national statistics office, even when these requests were made in Spanish, to release the final data set needed. Vietnam also appear ed useable, but Vietnam required payment b eyond the means of the researcher. Under the right circumstance and with appropriate funding, these two sets are most likely the best avenue for implementing the project. Additionally, the World Bank and partner organizations are undertaking a massive pro ject within Africa to create panel data sets. The project, commonly titled the Integrated Surveys on Agriculture (ISA), is under the Living Standards Measurement Surveys program and is being conducted in E thiopia, Malawi, Mali, Niger, Nigeria, Tanzania and Uganda These data sets would still present a problem of appropriate time horizon coverage however they may prove appropriate especially if the scope of analysis is further narrowed onto adoption of, say, a very specific change in SPS regimes i nstead of the broader scope of this paper. Discussion If an appropriate data set can be found, and after what is likely to be a significant amount of data transformation to form the appropriate and condensed variables, applying the general econometric methodology identified above would be a relatively easy process. Granted, a more specific and robust examination of the econometric and statistics literature is still needed to meet the challenges posed by expanding into different m easurements. However, the method would empower the broader aspirations of this work to be fulfilled. This second pillar holds promise of informing not simply what
167 the overall welfare changes might be from SPS measures, but also how they happen. Should the results show causality that certain production and value chain structures have a meaningful impact on rural household welfare, then the policy maker might be able to understand that not only will an SPS regime carry aggregated impacts, but also that it wil l have significant and identifiable impacts on modalities. Importantly, the work by Maertens and Swinnen (2009) provided evidence that the actual policy targets are then important. In their investigation of the Senegalese green bean supply market for expo rt, they found that compliance with stricter SPS measures might be good news for wage laborers often the poorest and most insecure members of society while bad news for small holders engaged in the production chain (who exit the market when they cannot c ompete under the new standards with increasingly vertically integrated agro industrial firms). Even if this policy maker is not engaging the results, development professionals might, and this second pillar might enable them to target reactionary programs based on new trends in SPS regimes in the developed world.
168 CHAPTER 7 SUMMARY, IMPLICATION S AND CONCLUDING REM ARKS Sanitary and phytosanitary requirements by government and private industry are increasingly stringent in the developed world. A in developed nations have risen, their demand for differentiated, high quality products has increased and they look increasingly towards imports to meet this demand. An increasing share of these imports come s from the developing world and economies which are in transition. Farmers and other firms within the supply chain there often lack access to the various forms of capital needed to implement sanitary and phytosanitary (SPS) requirements from developed nations causing the cost of com pliance to be high and often prohibitive in developing nations This is still in spite of the fact that trade has played an increasing role in the policy orientation for global development. To compound this dissonance, for agriculturists in the developing world, fresh fruits and vegetables hold relative promise as a means for improving livelihoods as they fetch a higher price than core commodities and allow farmers to engage the comparative advantages present in their particular ecosystems and climates. D espite the attractiveness of this product targeting, fresh produce is also more prone to face sanitary and phytosanitary measure impacts because of their raw and perishable form. The regimes these products face differs across developed nation markets. The purpose of this project was, at its core, to begin an exploration into methodologies and related challenges for analyzing welfare impacts of SPS measures which act as trade barriers and drive structural changes in the marketplace. Its focus shifted ou t of necessity to the first of these, with hopes that the second can be explored at a later date. To evaluate the aggregated welfare impacts, we have applied a
169 price wedge methodology which enabled the calculation of SPS regimes in the form of a tariff eq uivalent. Applying this rate showed that the losses to developing nation farmers can be relatively large. For the pineapple market we found that the residual associated with non tariff measures, where SPS impacts lie, is larger in the European market than in that of the United States. Sensitivity analysis showed that the magnitude of both welfare impacts and the tariff equivalent protection rate can change significantly based upon the assumed price elasticity faced by suppliers. One point must be mad e very clear: this project does not hold as its aim to argue against those SPS measures which are legitimate, scientifically based protections of consumers, ecosystems and values. What the author does advocate is engaging in a comprehensive investigation o f impacts so that policy responses are tailored to meeting the sometimes competing objectives of safety assurance and development promotion. Limitations of the Adopted Methodology In conducting the welfare analysis of Chapter 5, a number of limitations c an be identified and are worth a moment of discussion. These areas offer avenues for future improvement upon this work and also hold important significance for accepting the strength of the results presented previously. Perfect Substitutes Assumption: Add ressing Imperfect Product S ubstitut ion Several studies indicate that consumers form opinions about the general quality of goods based upon the place in which they were produced, thus providing the basis for their perceptions that goods of different origin are not the same (i.e. are not equal substitutes). (Johnecheck, Wilde, & Caswell, 2010; Chisik, 2002) As a number of aforementioned authors note, new investigations have been making rapid improvements in the applicab ility and robustness of the method to welfare analysis. One
170 such contemporary example are studies by Nimenya, Henry de Frahan and Ndimira (2008; 2009; 2012) who investigate the impacts of SPS measures in the EU on the expo rts of avocados, fresh peas, fresh green beans, and frozen fish fillets from selected African nations. Their analysis includes some integrated or multi market analysis and, most importantly, it applies Constant Elastic ity of Substitution derived functions that provide a work around to what had previously been mentioned as a problematic assumption: the perfect substitutability of foreign products for domestic ones. Their work is largely based on two extensions of the art icles published by Yue and colleagues (Yue & Beghin, 2009; Yue, Beghin, & Jensen, 2005) but represent a departure from this literature in a number of important ways. Namely, they focus on applying the price we dge method to determine tariff equivalents in a North South or developed importer developing exporter trade relationship. These applications, and others like them, are rooted in a seminal piece from Armington ( 1969 ) after which the resulting elasticity me asures are commonly named. enabled contemporary researchers to endogenize what is essentially a cross price elasticity of demand for like goods but of different origin. The technique enters the analysis at the utility function stage, applying a CES form such that: ( 7 1 ) where the greek lower case sigma is the constant elasticity of substitution between the like good of domestic origin j and foreign origin(s) i
171 It is becoming clearer that these extensions are important to appropriately calculating SPS impacts especially, and trade issues, generally. The method could not be employed in this work due to complexity and data requirements Shift P ara meters in Calculating Supply C urves In Chapter 6 we very briefly highlight ed the need to conduct research on the household lev el in an effort to conduct an analysis induced structural change of the production market caused by changing SPS requirements However, access to this type of data set might also carry implications for the analysis presented in Chapter 5, for the first an alysis pillar Chiefly, this relates to the robustness of the adaptations of the profit function proposed by Nimenya, Henry de Frahan and Ndimira (2008) who included land, labor and capital measures, and then implement ed the full model to derive the supply curve. In their study, these authors lacked crucial data on certain key variables of this altered framework; variables which many datasets examining agricultural households hold promise of addressing, including the Li ving Standards Measurement Surveys highlig hted in this work. In our work we currently lack and have duly noted appropriate supply shifters. While Nimenya et al included endogenous long run variables, other authors cited previously provide differing ex amples of appropriate supply shifters for short run models, notably weather and input costs, which are likely to be of great import to the production of these sensitive crops. Also not included are changes in production and post harvest technologies which are likely central to the recent increase in imports of fresh fruits and vegetables from developing nations. Without these shifters, the supply schedules used in the model lead us to conclude that the tariff equivalents and their welfare impacts are likely to be over estimates of current circumstances.
172 Issues Related to the PE Market M odel Nothing can summarize this point better than a statement from the technical manual used to develop the Excel world market price (Jechlitschkaj, Kirschke, & Schwarz, 2007, p. 89) This is, perhaps, the greatest methodological concern we have with this work. Our worlds are restricted to overly simplified, binary structures, with one exporter and one importer. However, we know that it is an extraordinarily rare situation where this i s the case in reality. Given adequate resources, a more appropriate approach to the equilibrium model would be to include, at a minimum, a Rest of the World Market from which the international market could draw. Alternatively, other methods could be adopte d to adjust for market share issues. Calvin and Krissoff (2005) did this via adjust ments to the i r price series. Without market share adjustments, we cannot safely reject the notion that our calculations are likely overestimates of the true impacts. Apart from this, the model implemented does not account for other theoretically viable changes that may actually take place in the market. In the Chapter 2 we discussed the main iterations of these. Given our hope for expanding into the second pillar, the most logical next extension of the market model is to examine long run shifts of the supply curve in the exporting nation. This model held technology and other capital issue s constant, but the actual impacts of SPS policies are now approaching the long run. Pu t more colloquially, SPS regimes are here to stay and analy tical tools need to adapt.
173 Data Intensity and A pproximation It has likely become clear throughout the preceding work that the greatest limiting factor was data, in terms of both availability and reliability. The author is acutely aware of the severe limitations imposed upon more substantive outcomes by the lack of data and other necessary resources. In most cases, it is clear what type and source of data is needed, it just simply does not exist. O ther times when it does exist, it presents a large number of reliability obstacles (such as the ComTrade data discussed previously) and work arounds were required. Some of these patches are sturdier than others and those which are prima facie weak we made in an effort to examine the greater analysis tools, which we feel are valuable per se The analysis completed here proves to be highly data intensive, at least as far as developing nation markets are concerned. The spatial partial equilibrium framework req uires a seemingly endless series of calibrations in order to normalize data to ensure outcomes are economically meaningful. For an application of the treatment effects model to test welfare implications of structural change induced by SPS measures, the int ensity of required data is less, but in most sets identified for potential use, a large amount of additional calculations will be required to formulate the more simplistic variables for the econometric analysis. Despite it all, t he exercise, even with its data flaws which cast a modicum of doubt, has still proven fruitful as it provides a launching point for improved, more in depth and robust research work in the field s of both trade and welfare economics and development policy analysis Despite grievous, a nd usually grave, challenges, we attempted at every turn to fill the gaps in an appropriate and justifiable manner.
174 Implications Despite these limitations, the analysis completed here did show the asymmetrical nature of SPS regimes. This is crucially impo rtant as the United States moves further towards more stringent regimes which many in the media and in academia contend are more reactionary than necessary. While the Food Safety and Modernization act does roach, it is clear from our modeling that increased stringency has deleterious results for producer surplus. Although the methods for estimating benefits of SPS regimes can be dubious, most notably in terms of health impacts, any cost benefit matrix which does not include even a portion of these external effects is limited. These costs must be included within a ny analysis of policy, especially in n ation s whose governments have committed themselves to the more altruistic notion of development facilitation. Using the models here, an astute policy analyst might be able to reformulate SPS regimes to target more prec isely national needs while abating the lo s ses to farmers in their trading partners. In fact, this leads us to the final point to be made in this work. We very purposefully adopted an approach for analysis which developed and implemented tools which are widely available. This included partial equilibrium models done in the almost omniprese nt Microsoft Office Suite. Of course we could have considered more example of how this analysis can be completed in a technologically minimalist environment. The issues w e encountered and the points of concern we have over results robustness were influenced not by limitations of theory or technology, but of information. Given the appropriate base, an analyst could theoretically conduct the appropriate analysis in a matter of days.
175 APPENDIX A KEY DEFINITIONS RELA TED TO THE SPS AGREE MENT OF THE WTO 1. Technical regulation Document which lays down product characteristics or their related processes and production methods, including the applicable administrative pro visions, with which compliance is mandatory. It may also include or deal exclusively with terminology, symbols, packaging, marking or labeling requirements as they apply to a product, process or production method. Explanatory note The definition in ISO/I EC Guide 2 is not self contained, but based on the so 2. Standard Document approved by a recognized body, that provides, for common and repeated use, rules, guidelines or characteristics for products or related proces ses and production methods, with which compliance is not mandatory. It may also include or deal exclusively with terminology, symbols, packaging, marking or labeling requirements as they apply to a product, process or production method. Explanatory note The terms as defined in ISO/IEC Guide 2 cover products, processes and services. This Agreement deals only with technical regulations, standards and conformity assessment procedures related to products or processes and production methods. Standards as defin ed by ISO/IEC Guide 2 may be mandatory or voluntary. For the purpose of this Agreement standards are defined as voluntary and technical regulations as mandatory documents. Standards prepared by the international standardization community are based on conse nsus. This Agreement covers also documents that are not based on consensus. 3. Conformity assessment procedures Any procedure used, directly or indirectly, to determine that relevant requirements in technical regulations or standards are fulfilled. Explanatory note Conformity assessment procedures include, inter alia procedures for sampling, testing and inspection; evaluation, verification and assurance of conformity; registration, accreditation and approval as well as their combinations.
176 APPENDIX B PRINCIPLES OF EU FOO D SAFETY LAW The guiding philosophy for the new system was set out in the 2002 Regulation which identifies key principles: 1. Food safety is to be ensured by focusing on the food value chain as a whole: "In order to ensure the safety of food, it is necessary to consider all aspects of the food production chain as a continuum from and including primary production and production of animal feed up to and including sale or supply of food to the consumer because each element may have a potential impact on food safety" (paragraph 12). A later summary of import conditions for seafood state this in simpler language: "the food law of the European Union implements the principle of quality management and process oriented controls throug hout the food chain from the fishing vessel or aquaculture farm to the consumer's table. Spot checks on the end product alone would not provide the same level of safety, quality and transparency to the consumer" (European Commission n.d.: 1). 2. Food saf ety is based upon the principle of containing risk. Elimination or avoidance of risks to health requires risk assessment, risk management and risk communication (paragraph 17). In other words, food safety is to be achieved through identification and contro l of risk, which means a process control approach, rather than a testing approach. The HACCP (Hazard Analysis and Critical Control Point) methodology is central to this. 3. The precautionary principle is established as part of food law: "The precautionary principle has been invoked to ensure health protection in the Community, thereby giving rise to barriers to the free movement of food or feed. Therefore it is necessary to adopt a uniform basis throughout the Community for the use of this principle. In tho se specific circumstances where a risk to life or health exists but scientific uncertainty persists, the precautionary principle provides a mechanism for determining risk management measures or other actions in order to ensure the high level of health prot ection chosen in the Community" (paragraphs 19 and 20). 4. Traceability. Control of risk and rectification of problems requires traceability. This is secured through identification of suppliers and customers for all products incorporated into food or feed (paragraph 29). This in turn requires record keeping and labeling (article 18, paragraphs 1 and 2), so that information can be supplied to the authorities on demand. The EU insists on having traceability in place: a documentary trail through which food (co mponents) can be traced back to the farm/field/pond from which they originated (if originating from within the EU).
177 5. Private sector responsibility. Paragraph 30 of the preamble to the General Food Law legislation states that: "A food business operator is best placed to devise a safe system for supplying food and ensuring that the food it supplies is safe; thus, it should have primary legal responsibility for ensuring food safety" (CEC 2002). Furthermore, "feed and food business operators at all stages of production, processing and distribution within the businesses under their control are responsible for ensuring that feed and food satisfy the requirements of feed and food law which are relevant to their activities" (CEC 2004d: preamble, para. 4). The prec ise impact of this legislation varies from country to country. The Guidance Principles on the General Food Law note that the way in which this liability is implemented in particular EU countries depends upon national legal orders: "Though the requirement l aid down in Article 17 (1) is directly applicable from 1 January 2005, the liability of food business operators should flow in practice from the breach of a specific food law requirement (and from the rules for civil or criminal liability which can be foun d in the national legal order of each Member state). The liability proceedings will not be based on Article 17 but on a legal basis to be found in the national legal order and in the specific infringed legislation" (CEC 2004e: 6). Source: (Humphrey, 2011)
178 APPENDIX C PROOF OF EXCESS SUPP LY/DEMAND ELASTICITY Recall the general definition of ow n price elasticity provided in E quation 5 9. Thus we define the own price elasticities of supply and demand respectively as: ( C 1 ) ( C 2 ) Further, the elasticity of the Quantity Traded Curve (ES/ED) is ( C 3 ) Via substitution then we find that ( C 4 ) Taking the absolute value of the final step of Equation C 4 above, finding like terms, rearranging and noting that the product of two real numbers is itself always real 1 yields : ( C 5 ) Thus, by the produc t rule, we arrive at Equation 5 10 1
179 APPENDIX D META DATA SUMMARY OF ELAS TICITY MEASURES P roduct Region Function Elasticity Source Years covered Notes Broccoli US Supply 0.12 Buxton in (Lohr & Park, 1995) Not specified (short run) California Supply 0.56 (Lohr & Park, 1995) 1985 1990 (organic, long run) California Supply 0.4 (Lohr & Park, 1995) 1985 1989 (organic, short run) Crucifers US Demand 0.4626 (Acharya & Molina, 2004) 1970 2003 Not stat. sig. (Hicksian) US Demand 0.5389 (Acharya & Molina, 2004) 1970 2002 (Marshallian) Fruit and vegetable Vietnam Demand 0.51389 Seale, Regmi & Berstein 1996 Pineapples Guatemala Demand 66.66 (Hallam, et al., 2004) 1970 2002 Honduras Demand 93.43 (Hallam, et al., 2004) 1970 2003 Thailand Demand 0.0845 (Vesdapunt, 1990) Not specified Vietnam Demand 23.75 (Hallam, et al., 2004) 1970 2001 World Demand 1.35 (Hallam, et al., 2004) 1970 2000 Germany Import demand 1.8605 (Vesdapunt, 1990) Not specified US Import demand 1.2276 (Vesdapunt, 1990) Not specified Thailand Supply 0.2616 (Vesdapunt, 1990) Not specified
180 APPENDIX E LISTING OF POTENTIAL VARIABLES FOR USE IN HOUSEHOLD LEVEL ANALYSIS Variable Status Priority Ranking HH Income Required, very important 1 HH composition across age and sex (relationship ties if possible) Required, very important 1 Age of HH Head Required, very important 1 Advance sales/contract Required, very important 1 Work as agricultural laborer / Employment in agroindustry Required, very important 1 Location of household (general) Required 2 Gender of HH Head Required 2 Education of HH Head Required 2 Units of livestock Required 2 Value of non land assests Required 2 Crops grown Required 2 HH consumption Preferred 3 HH Expenditures Preferred 3 Geographic location identifiers for farm plots (e.g. town, postal code, region, etc.) Preferred 3 Farm size Important 4 HH own consumption Important, not necessary 5 Education of other HH members Important, not necessary 5 Ethnicity Important, not necessary 5 Union / producer's organization member Important, not necessary 5 Other group participation Important, not necessary 5 Maternal language Important, not necessary 5 Participation in technical assistance programs Important, not necessary 5 Participation in development programs Important, not necessary 5 Total production Important, not necessary 5 Total cultivation Important, not necessary 5
181 Health Optional 6 Education of mother and father of HH members or HH Head Optional 6 Per capita land holdings (calculable) Optional 6 Agro chemical use Optional 6 Production practice(s) Optional 6 Special certifications (fair trade, organic, etc.) Optional 6 Irrigation Optional 6 Changes in farm size (sale or lease of land) Optional 6 Agricultural problems (pests, droughts, fires, etc.) Optional 6 Place of work Optional 6 Type of work Optional 6 Hours worked Optional 6 Income from work Optional 6 Secondary job Optional 6 Time allocation Optional 6 Prices Optional 6 Education Optional 6 Number of farm laborers Optional 6 Who purchased farm product (product destination) Optional 6
182 LIST OF REFERENCES Adelman, I. (2001). Fallacies in Development Theory and Their Implications for Policy. In G. M. Meier, & J. E. Stiglitz, Frontiers of Development Economics : The Future in Perspective (pp. 103 134). New York: The World Bank and Oxford University Press. Ale sina, A., & Dollar, D. (2000, March). Who gives foreign aid to whom and why? Journal of Economic Growth, 5 33 63. Alvarez, R. (2006). The transnational state and empire: U.S. certification in the Mexican mango and Persian lime industries. Human Organizati on, 65 (1), 35 45. APHIS. (2012). Fruits and Vegetables Import Requirements (FAVIR) Retrieved October 13, 2012, from FAVIR Database: https://epermits.aphis.usda.gov/manual/inde x.cfm?ACTION=pubHome Asia DHRRA. (2008). Value Chain Analysis Report: Cambodia, Phillipines, Vietnam. Quezon City, Phillippines: Asian Partnership for the Development of Human Resources in Rural Asia and the ASEAN Foundation. Retrieved from http://asiadhrra.org/activityblogs/2ndlsfmrtw/lsfmvaluechain2.pdf Askari, H., & Cummings, J. T. (1976). Agricultural supply response. A survey of the econometric evidence. Westport: Praeg er Publishers Inc. Askari, H., & Cummings, J. T. (1977). Estimating Agricultural Supply Response with the Nerlove Model: A Survey. International Economic Review, 18 (2), 257 292. AVRDC The World Vegetable Center. (2007, September 9). Challenge Program on High Value Crops Fruit and Vegetables. Baldwin, R. E. (1991). Measuring the Effects of Nontariff Trade Distorting Policies. In J. d. Melo, & A. Sapir (Eds.), Trade Theory and Economic Reform: North, South, and East (pp. 25 42). Cambridge, Mass.: Basil Bl ackwell. Beghin, J. C., & Bureau, J. C. (2001). Quantitative Policy Analysis of Sanitary, Phytosanitary and Technical Barriers to Trade conomie Internationale 107 130. Behrman, J. (1968). Supply Response in Underdeveloped Agriculture. A Case Study of Four Major Annual Crops in Thailand, 1937 1963. Amsterdam: North Holland. Berdegu, J. A., Balsevich, F., Flores, L., & Reardon, T. (2005). Central American fruits and vegetables Food Policy 254 269.
183 Berry, S., & Pakes, A. (2003, September 18). Classic Empirical Models of Static Equilbrium. Yale University. Retrieved from http://www.econ.yale.edu/~steveb/Econ6 00/chapter2.pdf Bhagwati, J., & Srinivasan, T. N. (2002, May). Trade and Poverty in the Poor Countries. The American Economic Review, 92 (2), 180 183. Bignebat, C., Colen, L., Humphrey, J., Lemeilleur, S., Maertens, M., Subervie, J., Vagneron, I. (20 11). The US Food Safety Modernisation Act : public regulations and private standards for fresh fruit and vegetable imports. Assessment of the impacts of non tariff measures NTM on the competitiveness of the EU and selected trade partners. Bittman, M. (20 11, 21 June). No Food Safety in These Numbers. New York. Retrieved November 2011, from http://opinionator.blogs.nytimes.com/2011/06/21/no food safety in these numbers/ Borraz, O., Besanon, J., & Clergeau, C. (2006). Is It Just About Trust? The Partial Reform of French Food Safety Regulation. In C. Ansell, & D. Vogel (Eds.), Cambridge, Mass., USA: The MIT Press. Bottemiller, H. (2011, June 21). Taylor Talks FSMA Implementation, Funding. Washington, D.C.: Food Safety News. Retrieved November 2011, from http://www.foodsafetynews.com/2011/06/taylor talks fsma implementation funding/ Broberg, M. (2009). European Food Safety Regulation and the Developing Countries. Copenhagen: Danish Institute for International Studies. Caduff, L., & Be rnauer, T. (2006). Managing Risk and Regulation in European Food Safety Governance. Review of Policy Research, 23 (1), 153 168. Calvin, L., & Krissoff, B. (1998). Technical Barriers to Trade: A Case Study of Phytosanitary Barriers and U.S. Japanese Apple Tr ade. Journal of Agricultural and Resource Economics 351 366. Calvin, L., & Krissoff, B. (2005). Resolution of the U.S. Japan Apple Dispute: New Opportunities for Trade. Washington, D.C.: Economic Rsearch service of the USDA. Canadian International Develop ment Agency. (2012, August 13). Vietnam Retrieved October 12, 2012, from http://www.acdi cida.gc.ca/acdi cida/ACDI CIDA.nsf/En/JUD 217143241 QWY CBI. (2008, February) The EU market for pineapple. Ministry of Foreign Affairs of the Netherlands. Retrieved from http://www.cadexco.bo/actual/boletin inteligencia/Pineapple%20Survey.pdf
184 CCITD. (2012, August). Specialty Commodity Opportunity Outlook Cauliflower and Broccoli. Fresno, California, USA: California Centers for International Trade Development in cooperation with the California Department of Food and Agriculture. Retrieved from http://fresnocitd.org/wp content/uploads/2012/08/Specialty Commod ity Opportunity Outlook_Cauliflower Broccoli.pdf?f_name=Jason&l_name=Entsminger&email=j.entsminger%40ufl.ed u Central America Data. (2008, September 12). Europe agreed to maintain limits on agrochemicals on pineapple Retrieved from Central America Data: T he Regional Business Portal: http://en.centralamericadata.com/en/article/home/Europe_Reduces_Agrochemic al_Limits_for_Pineapple 1/ Centra l America Data. (2009, September 9). Europe Reduces Agrochemical Limits for Pineapple Retrieved from Centra lAmerica Data: The Regional Business Portal: http://en.centralamericadata.com/en/article/home/Europe_Reduces_Agrochemic al_Limits_for_Pineapple Central America Data. (2012, October 3). Europe is Not Easy for Non Traditional Retrieved from Central America Data: The Regional Business Porta l: http://en.centralamericadata.com/en/article/home/Europe_is_Not_Easy_for_Non Traditional_Exports Chisik, R. (2002). Reputational Comparative Ad vantage and Multinational Enterprise. Economic Inquiry, 40 (4), 582 596. CIA. (2012). Retrieved August 1, 2012, from CIA The World Factbook: https://www.cia.gov/library/publications/the world factbook Commission of the European Communities. (2000, February 2). Communication from the Commission on the Precautionary Principle. Brussels, Belgium. Retrieved from http://ec.europa.eu/dgs/health_consumer/library/pub/pub07_en.pdf Corzine, M. N. (2008). An Analysis of Import Tariff Escalation: A Case of Maize Trade Between South Africa and Mozambique. Michigan State Univer sity. Cruz Medina, J. D., & Garca, H. (2005, November 13). Pineapple: Post harvest Operations. (D. Meja, Ed.) Deardorff, A., & Stern, R. (1998). Measurement of Nontariff Barriers. Ann Harbor: The University of Michigan Press. Deodhar, S. Y., Rastogi, S. K., & Krissoff, B. (2007, January 6 9). An Approach Paper on Indo US Trade in Wheat and Mango. Washington, D.C.
185 Disdier, A. C., Fontagn, L., & Mimouni, M. (2008, May). The Impact of Regulations on Agricultural Trade : Evidence from the SPS and TBT Agreeme nts. American Journal of Agricultural Economics, 90 (2), 336 350. Durham, C., & Eales, J. (2006). Demand Elasticities for Fresh Fruit at the Retail Level. Retrieved from http://www.ftc. gov/be/seminardocs/061012DurhamEales.pdf Echols, M. A. (1998). Food Safety Regulation in the European Union and the United States: Different Cultures, Different Laws. Columbia Journal of European Law 525 544. ECLAC. (1998). Non Tariff Trade Barriers to E xports from the LAC Region. United Nations Economic Commission for Latin America and the Caribbean. Santiago de Chile: ECLAC. EFSA (Director). (2012). EFSA explains risk assessment of food and feed risk [Motion Picture]. Retrieved July 31, 2012, from http://youtu.be/5xUhPmthgZk EFSA. (2012, June 15). Reasoned opinion on the modification of the existing MRLs for ametoctradin in various commodities. EFSA Journal Parma, Italy. Retrieved from http://www.efsa.europa.eu/en/efsajournal/pub/2771.htm ERS. (2011, June). US Broccoli Statistics Retrieved from Economics, Statistics and Market Information System: http://usda.mannlib.cornell.edu/MannUsda/viewDocumentInfo.do?documentID=1 816 FAO. (2005). European Food Safety Control Systems: New Perspectives on a Harmonized Legal Basis. Bangkok: Second FAO/WH O Forum on Food Safety Regulators. Retrieved April 5, 2011, from http://www.fao.org/docrep/meeting/008/y5871e/y5871e0l.htm FAO. (n.d.). Nicaragua Retrieved July 31, 2012, from FAO Initiative on Soaring Food Prices: http://www.fao.org/isfp/country information/nicaragua/en/ FAS. (2009). Agricultural Economy and Policy Report: Vietnam. Washington, D.C.: FAS of U SDA. Retrieved from http://www.fas.usda.gov/country/vietnam/Vietnam%20Agricultural%20Economy% 20and%20Policy%20Report.pdf FAS. (2012). Nicaragua Retrieved August 1, 2012, from http://www.fas.usda.gov/country/Nicaragua/Nicaragua.asp FAS. (2012). Vietnam Retrieved October 12, 2012, from http://www.fas.usda.gov/country/vietnam/vietnam.asp
186 FAS of USDA. (1999). Food and Agricultural Import Regulations and Standards Report : United States of America. Government of the United States of America, U.S. Depa rtment of Agriculture, Office of Food Safety and Technical Service, Foreign Agricultural Service. Washington, D.C.: USDA. Retrieved July 31, 2012, from http://www.fas.usda.gov/itp/ofsts/us.html FDA (2007, November). An integrated strategy for protecting the nation's food supply. Food Protection Plan Washington, D.C.: US Food and Drug Administration. Retrieved July 31, 2012, from http://www.fda.gov/Food/FoodSafety/FoodSafetyPrograms/FoodProtectionPlan2 007/ucm132565.htm FDA. (2009, Fall). Produce Safety Regulation (U. G. Administration, Ed.) Washington, D.C.: Office of Information and Regulat ory Affairs, Office of Management and Budget, Executive Office of the President of the United States. Retrieved July 31, 2012, from http://www.reginfo.gov/public/do /eAgendaViewRule?pubId=200910&RIN=0910 AG35 Ferrantino, M. J. (2003, September 17). Analytical Approaches to Quantifying Economic Effects of Non Tariff Measures. Washington, D.C. Ferris, J. (2005). Agricultural Prices and Commodity Market Analysis. East Lansing: Michigan State University Press. Findlay, R. (2009). The Trade Development Nexus in Theory and History. United Nations University World Institute for Development Economics Research 13th Annual Lecture. Helsinki, Finland: UNU. Retrieved May 31, 2012, from http://youtu.be/_wFsT3C ExQ Frohberg, K., Grote, U., & Winter, E. (2006, August 12 18). EU Food Safety Standards, Traceability and Other Regulations: A Growing Trade Barrier to Exports? Gold Coast, Australia: International Association of Agricultural Economists Conference. Gaulier, G., Mirza, D., Turban, S., & Zignago, S. (2008, March). International Transportation Costs Around the World: a New CIF/FoB rates Dataset. CEPII. Gebr ehiwet, Y., Ngqangweni, S., & Kirsten, J. (2007, March). Quantifying the Trade Effect of Sanitary and Phytosanitary Regulations of OECD Countries on South African Food Exports. Agrekon, 46 (1), 23 39. Gehlhar, M. (2006). Chapter 11 Merchandise Trade data. G TAP Database Documentation West Lafayette, Indiana, USA: Center for Global Trade Analysis, Purdue University.
187 General Assembly of the United Nations. (2000, September 18). United Nations Millennium Declaration. Retrieved from http://www.un.org/millennium/declaration/ares552e.pdf Goldberg, P. K., & Pavcnik, N. (2004). Trade, inequality and poverty: what do we know? Evidence from recent trade liberalization episodes in developing countries. Cambridge, MA: National Bureau of Economic Research. Retrieved from http://www.nber.org/papers/w10593 Golub, F., & Finger., J. (1979). The Processing of Primary Commodities: Effects of Developed Country Tariff Escalation and Developing Country Export Taxes. Journal of Political Economy 559 577. Graffham, A. (2006). EU Legal Requirements for Imports of Fruit and Vegetables (a Supplier's Guide). Government of the United Kingdom, Department for I nternational Development (DFID), London. Retrieved from http://www.agrifoodstandards.net/en/filemanager/active?fid=91 Greenaway, D., & Milner, C. (1993). Trade and Industrial Po licy in Developing Countries. London: Macmillan. Greenaway, D., Morgan, W., & Wright, P. (2002). Trade liberalisation and growth in developing countries. Journal of Development Economics, 67 229 244. Grilli, E., & Riess, M. (1992). EC Aid to Associated Co untries: Distribution and Determinants. Weltwirtschaftliches Archiv 202 220. Hallam, D., Liu, P., Lavers, G., Pilkauskas, P., Rapsomanikis, G., & Claro, J. (2004). The market for non traditional agricultural exports. Commodities and Trade Division, FAO, R aw Materials, Tropical and Horticultural Products Service. Rome: Food and Agriculture Organization of the United Nations. Retrieved from ftp://ftp.fao.org/docrep/fao/007/y5445e/y5445e00.p df Head, K. (2000, October 22). Gravity for Beginners. Vancouver, British Columbia, Canada. Henson, S. (2006, May 26 30). The Role of Public and Private Standards in Regulating International Food Markets. Bonn, Germany. Retrieved from http://www.ilr.uni bonn.de/iatrc2006/iatrc_program/Session%204/Henson.pdf Henson, S., & Caswell, J. (1999). Food safety regulation: an overview of contemporary issues. Food Policy, 24 589 603. Henson, S., & Jaffee, S. (2006). Developing Country Responses to the Enhancement of Food Safety Standards. In U. B. Grote, New Frontiers in Environmental and Social Labeling. Heidelberg: Physica Publisher.
188 Henson, S., & Jaffee, S. (2008). Unders tanding Developing Country Strategic Responses to the Enhancement of Food Safety Standards. The World Economy 548 568. Henson, S., & Loader, R. (2001). Barriers to Agricultural Exports from Developing Countries: The Role of Sanitary and Phytosanitary Requ irements. World Development, 29 (1), 85 102. Henson, S., Brouder, A. M., & Mitullah, W. (2000). Food Safety Requirements and Food Exports from Developing Countries: The Case of Fish Exports from Kenya to the European Union. American Journal of Agricultural Economics 1159 1169. Hertel, T., Ivanic, M., Preckel, P., & Cranfield, J. (2003). Trade Liberalization and the Structure of Poverty in Developing Countries. Center for Global Trade Analysis, Purdue University. West Lafayette, IN: Global Trade Analysis Project. Hillman, J. (1991). Technical Barriers to Acgricultural Trade. Boulder: Westview Press. Hooker, N., & Caswell, J. (1999). Framework for Evaluating Non Tariff Barriers to Trade Related To Sanitary and Phytosanitary Regulation. Journal of Agricultural Economics 234 246. Hummels, D., & Lugovsky, V. (2003, May). Usable Data? Matched Partner Trade Statistics as a Measure of International Transportation Costs West Lafayette, Indiana, USA: Purdue University. Humphrey, J. (2011). Standards, Food Safety and Trade: The European Union and The United States. Assessment of the Impacts of Non tariff Measures (NTM) on the Competitiveness of the EU and Selected Trade P artners. European Commission's NTM IMPACT Research Programme. Iacovone, L. (2005). Analysis and Impact of Sanitary and Phytosanitary Measures. Harvard University, Center for International Development. Cambridge, Mass.: Integration and Trade. Retrieved from http://www.cid.harvard.edu/cidtrade/Papers/iacovone.pdf IFAD. (2008). The role of high value crops in rural poverty reduction in the Near East and North Africa. Near East and North A frica Division. Rome: International Fund for Agricultural Development. IFAD. (2012). Rural poverty in Nicaragua Retrieved August 1, 2012, from Rural Poverty Portal: http://www.ruralpovertyportal.org/web/rural poverty portal/country/home/tags/nicaragua Islam, N., & Subramanian, A. (1989). Agricultural exports of developing countries: estimates of income and price elasticities of demand and supply. Journal of Agricultural Economics 221 231.
189 Jaffee, S. (2005, September 15). Food Safety and Agricultural Health Standards and Developing Country Exports: Rethinking the Impacts and the Policy Agenda. Washington, D.C.: The World Bank. Jaffee, S. M., & Henson, S. (2005). Agro food Exports from Developing Countries: The Challenges Posed by Standards. In M. A. Aksoy, & J. C. Beghin, Global Agricultural Trade and Developing Countries (pp. 91 114). Washington, D.C.: The World Bank. Jaffee, S., & Henson, S. (2004). Stan dards and Agro Food Exports from Developing Countries: Rebalancing the Debate. World Bank. Jechlitschkaj, K., Kirschke, D., & Schwarz, G. (2007). Microeconomics using Excel : integrating economic theory, policy analysis and spreadsheet modelling. London: R outledge. Johnecheck, W. A., Wilde, P. E., & Caswell, J. A. (2010). Market and Welfare Impacts of COOL on the U.S. Mexican Tomato Trade. Journal of Agricultural and Resource Economics, 35 (3), 503 521. Jolly, C. M., & Gadbois, M. A. (1989, Summer). Foreign aid as a promotional strategy. The Review of Black Political Economy, 18 (1), 59 74. Josling, T., Roberts, D., & Orden, D. (2004). Food Regulations and Trade: Toward a Safe and Open Global Food System. Washington, D.C.: Peterson Institute. Just, R. E., Huet h, D. L., & Schmitz, A. (2005). The welfare economics of public policy: a practical approach to project and policy evaluation. Noryhampton, Mass.: Edward Elgar Publishing, Inc. Karova, V., Roberts, D., Granta, J. H., & Peterson, E. (2009, June 10 12). An E mpirical Assessment of Phytosanitary Regulations on US Fresh Fruit and Vegetable Imports. Santiago, Chile. Kleemann, L. (2011, January). Organic Pineapple Farming in Ghana A Good Choice for Smallholders? Kiel, Germany: Kiel Institute for the World Econom y. Retrieved from www.ifw kiel.de Knowles, T., Moody, R., & McEachern, M. G. (2007). European food scares and their impact on EU food policy. British Food Journal, 109 (1), 43 67. Lang, R. (2005). A partial equilibrium analysis of the impact of the ECOWAS EU Economic Partnership Agreement. United Nations Economic Commission for Africa, Trade and Regional Integration Division. Addis Ababa: UNECA. Retrieved from https://www.gtap.agecon.purdue.edu/resources/download/2728.pdf
190 Lee, V. (2008). Food safety mechanism and food labelling requirements in the European Union,the United Kingdom and France. Government of the Autonomous Region of Hong Kong, Research and Library Services Division, Legislative Council Secretariat, Hong Kong. Leibovitch, E. H. (2008). Food SafetyRegulation in the European Union: Toward an Unavoidable Centralization of Regulatory Powers. Texas International Law Journa l 430 450. Lohr, L., & Park, T. (1995). Supply Elasticities and Responses to Relative Price Changes in Organic Produce Markets. Journal of Sustainable Agriculture 23 45. Lucier, G. (1999, April). Broccoli: Super Food for All Seasons. Washington, D.C.: Economic Research Service of the USDA. Maertens, M., & Swinnen, J. F. (2008). Trade, Standards, and Poverty: Evidence from Senegal. World Development 161 178. Maertens, M., & Swinnen, J. F. (2009). Trade, Standards, and Poverty: Evidenc e from Senegal. World Development 161 178. Maruyama, M., & Trung, L. V. (2011, March 10). Modern Retailers in Transition Economies: The Case of Vietnam. Kobe, Japan: Kobe University. Retrieved from http://www.rieb.kobe u.ac.jp/academic/ra/dp/English/DP2011 08.pdf Maskus, K., Otsuki, T., & Wilson, J. (2001). Maskus, K.E., T. Otsuki, and J.S. Wilson. In K. E. Maskus, & J.S. Wilson (Eds.), Quantifying the Impact of Barriers to Trade: Can it be Done? Ann Arbor: The University of Michigan Press. McCulloch, N., & Ota, M. (2002). Export Horticulture and Poverty in Kenya. Institute of Development Studies. Mehmet, (1999). Westernizing the Third World: The Eurocentricity of Economic Development Theories. London: Routledge. Merrill, T. (1993). Nicaragua: A Country Study. Washington, D.C.: GPO for the Library of Congress. Retrieved August 1, 2012, from http://countrystudies.us/nicaragua/ Minten, B., Randrianarison, L., & Swinnen, J. F. (2009). Global Retail Chains and Poor Farmers: Evidence from Madagascar. World Development 1728 1741. Mitchell, L. (2003). Economic Theory and Conceptual Relationships Between Food Safety and International Trade. In J. C. Buzby, International Trade and F ood Safety: Economic Theory and Case Studies (pp. 10 27). Washington, D.C.: USDA. Retrieved from http://www.ers.usda.gov/publications/aer agricultural economi c report/aer828.aspx
191 Nicholson, W. (2004). Microeconomic Theory: Basic Principles and Extensions (9th ed.). Mason, Ohio: South Western College Publishing. Nimenya, N., Henry de Frahan, B., & Ndimira, P. (2008). Quantifying non tariff measures in internati onal agricultural trade: a tariff equivalent of technical barriers to trade on African horticultural exports to the European markets. People, Food and Environments: Global Trends and European Strategies. Gent. Nimenya, N., Henry de Frahan, B., & Ndimira, P. F. (2009, December 13 15). A Tariff Equivalent of Non Tariff Barriers on European Horticultural and Fish Imports from African Countries. Fort Meyers, Florida, USA. Nimenya, N., Ndimira, P. F., & Henry de Frahan, B. (2012). Tariff equivalents o f nontariff measures: the case of European horticultural and fish imports from African countries. Agricultural Economics, 43 1 19. Nolte, S., & Rau, M. L. (2006). Incorporating food quality standards into equilibrium models: a review and discussion of pos sible approaches. Food Regulation and Trade: Institutional Framework, Concepts of Analysis and Empirical Evidence. Bonn. OECD. (2003). The Impact of Regulations on Agro food Trade: The Technical Barriers to Trade (TBT) and Sanitary and Phytosanitary Measur es (SPS) Agreements. Paris: Organization for Economic Co operation and Development. OECD. (2008). Is it ODA? Paris: OECD. Retrieved from http://www.oecd.org/investment/aidstatistics/ 34086975.pdf Okello, J. J., & Swinton, S. M. (2007). Compliance with International Food Safety and a Large scale Farm Producing for Export. Review of Agricultural Economics, 29 (2), 269 285. Okello, J. J., Narrod, C., & Roy, D. (2007). Food Safety Requirements in African Green Bean Exports and Their Impact on Small Farmers. Washington, D.C.: International Food Policy Research Institute. Otsuki, T., Wilson, J., & Sewadeh, M. (2001). Saving two in a billion: quantifying the trade effect of European food safety standards on African exports. Food Policy, 26 (5), 495 514. Overton, B., Beghin, J., & Foster, W. (1995). Phytosanitary Regulations for U.S. and Agricultural Trade Flows: Tobacco Inputs and Cigarette Output. Agricultural and Resource Economic Review 221 231. Oyejide, T. A., Ogunkola, E. O., & Bankole, S. A. (2000, April 37). Quantifying the Trade Impact of Sanitary and Phytosanitary Standards: What is Known and Issues of Importance for Sub S aharan Africa. Washington, D.C., United States of America.
192 Parappurathu, S. (2007). Partial Equilibrium Models for Agricultural Policy Analysis. New Delhi: National Centre for Agricultural Economic and Policy Research (India). Retrieved from http://www.iasri.res.in/sscnars/socialsci/13 Partial%20Equilibrium%20Model.pdf Perali, F. (2003, December). Partial Equilibrium Analysis of Policy Impacts (part I). Damascus, Syria. Rau, M. L., & Schlueter, S. W. (2009, August 16 22). Framework for analyzing regulations and standards in the NTM Impact project. Beijing, China: International Association of Agricultural Economists. Ravallion, M. (2003). The Debate on Globalizatio n, Poverty, and Inequality: Why Measurement. Washington, DC: World Bank. Reardon, T., & Berdegu, J. A. (2002). The Rapid Rise of Supermarkets in Latin America: Challenges and Opportunities for Development. Development Policy Review 371 388. Reardon, T., Barrett, C. B., Berdegu, J. A., & Swinnen, J. F. (2009). Agrifood Industry Transformation and Small Farmers in Developing Countries. World Development 1717 1727. Reardon, T., Codron, J. M., Busch, L., Bingen, J., & Harris, C. (2001). Global Change in Agrifood Grades and Standards: Agribusiness Strategic Responses in Developing Countries. International Food and Agribusiness Management Review 421 435. Regmi, A., & Dyck, J. (2001). Effects of Urbanization on Global Fo od Demand. In A. Regmi, Changing Structure of Global Food Consumption and Trade (pp. 23 31). Washington, D.C.: Economic Research Service of the USDA. Roberts, D., Orden, D., & Josling, T. (1999, October 1 2). WTO Disciplines on Technical Barriers to Agricultural Trade: Progress, Prospects and Implications for Developing Countries. Geneva, Switzerland. Saunders, P. (2010). The Precautionary Principle. Policy Responses to Societal Concerns in Food and Agriculture (pp. 47 58). Paris: OECD. Schmitz, A., M oss, C. B., Schmitz, T. G., & Furtan, H. (2010). Agricultural Policy, Agribusiness and Rent Seeking Behaviour. Toronto : University of Toronto Press. Schraeder, P. J., Hook, S. W., & Taylor, B. (1998). Clarifying the Foreign Aid Puzzle: A Comparison of Ame rican, Japanese, French, and Swedish Aid Flows. World Politics 294 323.
193 Scott, S., & Riggs, N. J. (2011). The Top 5 Pineapple Producing Countries Retrieved October 13, 2012, from Top 5 of Anything: The Top 5 Pineapple Producing Countries Seale, J., Regmi A., & Bernstein, J. (2003). International Evidence on Food Consumption. Washington, D.C. : Economic Research Service of the USDA. Seelke, C. R. (2008). Nicaragua: Political Situation and U.S. Relations. Congressional Research Service, Foreign Affairs, De fense, and Trade Division. Washington, D.C. : CRS. Shames, L. (2010, May 6). FDA Could Strengthen Oversight of Imported Food by Improving Enforcement and Seeking Additional Authorities. Washington, D.C.: US Government Accountability Office. Steiner, B. E. (2006). Governance Reform of German food safety regulation: Cosmetic or real? In C. Ansell, & D. Vogel (Eds.), What's the Beef? The Contested Governance of European Food Safety (pp. 181 210). Cambridge, Mass., USA: The MIT Press. Stewart, H., Hyman, J., Bu zby, J. C., Frazo, E., & Carlson, A. (2011). How Much Do Fruits and Vegetables Cost? Washington, D.C. : ERS of the USDA. Stiglitz, J. E. (1996, August). Some Lessons from the East Asian Miracle. The World Bank Research Observer, 11 (2), 151 177. Tvora, F. L. (2008). Developments in the World Soybean Market: a Partial Equilibrium Trade Model. Federal Senate, Government of the Federal Republic of Brazil, Braslia. State largest in world, despite conventional wisdom. Sacramento, California, USA. Retrieved July 31, 2012, from http://www.msnbc.msn.com/id/16 600877/ns/business us_business/#.UBwuLk1lTEU The Old Farmer's Almanac. (2012). Broccoli Retrieved October 13, 2012, from The Old Farmer's Almanac USA: http://www.almanac.com/plant/broccoli Thompson, G. D. (2003, July 27 30). Retail Demand for Greenhous Tomatoes: Differentiated Fresh Produce. Montreal, Quebec, Canada. Tiffin, R., Balcombe, K., Salois, M., & Kehlbacher, A. (2011, November 23). Estimating Food and Drink Elasticities. University of Readin g. Retrieved from http://www.defra.gov.uk/statistics/files/defra stats foodfarm food price elasticities 120208.pdf
194 Tjeertes, P. (2004). Brassi ca oleracea L. (cauliflower and broccoli). Protabase (G. Grubben, & O. Denton, Eds.) Wageningen, Netherlands: Plant Resources of Plant Resources of Tropical Africa Database: http://database.prota.org/dbtw wpd/exec/dbtwpub.dll?AC=QBE_QUERY&BU=http%3A%2F%2Fdatabase.prota. org%2Fsearch.htm&TN=PROTAB~1&QB0=AND&QF0=Species+Code&QI0=Bra ssica+oleracea+cauliflower+and+broccoli&RF=Webdisplay Tomek, W., & Robinson, K. (1977). Agriculture Price Analysis and Outlook. I n L. Martin (Ed.), A Survey of Agricultural Economics Literature: Volume 1 (pp. 329 409). Minneapolis: University of Minnesota Press. Tschirley, D., Muendo, K. M., & Weber, M. T. (2004). Improving Kenya's Domestic Horticultural Production and Marketing Sy stem. Horticultural Marketing Nairobi, Kenya: Tegemeo Institute Of Agricultural Policy and Development, Egerton University. UF/IFAS Extension. (2012). Crucifers Retrieved October 13, 2012, from Electronic Data Information Source: http://edis.ifas.ufl.edu/topic_crucifers UNCTAD. (2012, April 26). Infocomm Commodity Profile Pineapple Retrieved from All ACP Agricultural Commodities Programme of UNCTAD: http://www.unctad.info/en/Infocomm/AACP Products/COMMODITY PROFILE --Pineapple/ United Nations. (2002, March 22). Monterrey Consensus of the International Conference on Financing for Development. Monterrey, Mexico. United Nations Statistics Division. (2012). ComTrade Retrieved October 13, 2012, from United Nations Commodity Trade Statistics Database: comtrade.un.org US Department of State. (2012, February 24). Background Note: Nicaragua Retrieved August 1, 2012, from Background Notes and Country Fact Sheets: http://www.state.gov/r/pa/ei/bgn/1850.htm USAID. (2012). Market Trends for Pineapple: Market Survey. USAID, Tanzania Agriculture Productivity Pr ogram. Retrieved from http://www.fintrac.com/cpanelx_pu/tapp/13_41_99_TAPP%20 %20Pineapple%20Market%20Survey.pdf Vesdapunt, R. (1990). Analysis of d emand for and supply of Thai pineapple. Bangkok, Thailand: Kasetsart University Vogel, D. (2003, January 7). The Politics of Risk Regulation in Europe and the United States. The Yearbook of European Environmental Law Paris, France: Institut du dveloppe ment durable et des relations internationales. Retrieved from http://www.iddri.org/Publications/Collections/Idees pour le debat/id_0301_vogel.pdf
195 White, H. (1980). A Heteroskedasticity Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity. Econometrica 817 838. Williamson, J. (2000, August). What Should the World Bank Think about the Washington Consensus? The World Bank Research Observer, 15 (2), 251 264. Williamson, J. (2004). A Short History of the Washington Consensus. From the Washington Consensus towards a new Global Governance. Barcelona, Spain. Retrieved May 31, 2012, from http://www.iie.com/publications/papers/williamson0904 2.pdf Wobst, P. (2000). Why the Poor Care About Partial Versus General Equilibrium Effects Part I: Methodology and Country Case. International Food Policy Research Inst itute, Trade and Macroeconomics Division. Washington, D.C: IFPRI. World Bank. (2005, January 10). Food Safety and Agricultural Health Standards: Challenges and Opportunities for Developing Country Exports. Washington, D.C.: The World Bank. World Bank. (200 5). Food Safety and Agricultural Health Standards: Challenges and Opportunities for Developing Country Exports. Washington, D.C.: The World Bank. World Bank. (2012). Nicaragua Country Brief Retrieved August 1, 2012, from http://www.worldbank.org/en/country/nicaragua/overview World Bank. (2012). Vietnam Overview Retrieved October 12, 2012, from http://www.worldbank.or g/en/country/vietnam/overview World Bank Group in Vietnam. (2005). Accelerating Rural Development in Vietnam. The World Bank, Rural Development and Natural Resource Sector Unit East Asia and the Pacific Region. Hanoi: The World Bank. Retrieved from www.worldbank.org.vn Yapp, C., & Fairman, R. (2006). Factors affecting food safety compliance within small and medium sized enterprises: implications for regulatory and enforcement strategies. Food Control 42 51. Yu e, C., & Beghin, J. C. (2009). Tariff Equivalent and Foregone Trade Effects of Prohibitive Technical Barriers to Trade. American Journal of Agricultural Economics 930 941. Yue, C., Beghin, J. C., & Jensen, H. H. (2005). Tariff Equivalent of Technical Barr iers to Trade with Imperfect Substitution and Trade Costs. Ames: Center for Agricultural and Rural Development at Iowa State University.
196 BIOGRAPHICAL SKETCH Jason Entsminger completed his graduate work under the Atlantis Dual Degree Programme in Agricultural Economics and International Develop ment. The program is a high mobility, internationally cooperat ive partnership under an accord between the US Department of Education and the European Commission's Direct orate General for Educ ation and Culture. It awards a Master of Science degree from the US home i nstitution [here, the University of Florida] and an EU Erasmus Mundus joint degree, the International Master of Science in Rural Development [the IMRD]. Mr. Ent mobility program included: the University of Florida in Gainesville, United States; the Universit degli Studi di Pisa in Pisa, Italy; Agrocampus Ouest (formerly the cole Nationale Suprieure Agronomique de Rennes) in Rennes, France; and the Hum boldt Universit t zu Berlin in Berlin, Germany. Prior to his graduate studies Mr. Entsminger completed studies at the University of Nevada, Reno. In his five years there he was awarded a Bachelor of Science in Agricultural and Applied Economics, a Bachelo r of Arts in International Affairs with a specialization in Law, Diplomacy and Comparative Political Organization and a Bachelor of Arts in French Language and Literature. He graduated under the honorarium With Distinction. In addition he completed m inor c ertificat e s in Applied Statistics and Italian Studies. While at the University of Nevada, Reno and the University of Florida he held positions working within the offices of state specialists for the respective state Cooperative Extension Services. He is a native of Reno, Nevada.