1 DEMOCRATIC TRAJECTORY IN COMPARATIVE PERSPECTIVE By NICHOLAS D. KNOWLTON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FU LFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2015
2 Â© 2015 Nicholas Knowlton
3 To Ava and Jordan
4 ACKNOWLEDGMENTS The origins for the topic of this dissertati on reside through a series of conversations I enjoyed with my advisor , Dr. Leonardo VillalÃ³n. I had long been interested in the study of de mocratization in Africa following the end of the Cold War . I n particular, my interests concerned how effective intern ational assistance had been in facilitating democratization , including the question as to why, democracy that purportedly swept across the African continent in the early 1990s, many countries failed in maintaining democratic systems o f rule (or at least failed to be widely regarded as democracies). While such a top ic was fascinating to me, it was rather complex , involved researching numerous countries, and could have easily regressed into a cumbersome project that would have required a n extens ive amount of time to arrive at what I would have found to be a sufficient and satisfactory conclusion . However , my past experiences in Ghana provided a useful entry point for a case study in democratization. As Ghana was widely considered to have democratized through the assistance of the international community, the country presented itself as a rather straightforward dissertation project. As I was finalizing my research topic intending to simply produce a case study perspective emphasizing historical institutionalism Dr. VillalÃ³n questioned why I chose to limit myself to one case, despite my known interests in the politics of democratization across the continent. Instead, he wondered, why not investigate the larger q uestion while couching the research under a long term perspective emphasizing regime trajectories ? In allaying my initial hesitancy, Dr. VillalÃ³n assured me of his commitment to guide me through the process . What followed was the t rajectory of democracy couched within a comparative context.
5 While I am grateful for all of the help and guidance Dr. VillalÃ³n provided through the course of this dissertation project. I will always be particularly grateful to him for challenging me to tak e a risk and undertake the kind of project that was more difficult, but provided a greater sense of personal satisfaction, while additionally providing a much greater contribution in return. In addition, I would like to express my gratitude to my dissertat ion committee, McLaughlin. Their willingness to provide additional guidance and support, to serve as additional academic mentors, and to go above an d beyond their offici al duties as c ommittee members will always be remembered. I am also grateful for the funding I received for this project. Receiving pre dissertation travel grants from the University of Florida Department of Political Science, as well as the University of Florida Center for African Studies enabled me to pursue and complete my dissertation prospectus, and provided me the opportunity to obtain the necessary contacts and logistics for subsequent dissertation research. In addition, a National Security Education Program Boren Fellowship provided funding for seven months of dissertation fieldwork in Ghana, including summer language training in Twi (Asante) through the 20 13 African Languages Initiative . Lastly, funding provided through a University of Florida Gradu ate School Dissertation Fellowship provided six months of support towards completing the composition of this dissertation. I would like to thank all of the research participants that were willing to be interviewed for this project. Their responses and answ ers were very insightful, and were
6 furthermore instrumental in producing I would not have been able to complete this pro ject had it n ot been for the guidance and support offered through my in country advisor, Dr. Richard Asante. Though I was an international graduate student conducting research for a limited amount of time in Ghana , Dr. Asante treated me as his own student and became an additional mentor to me . In addition, I would like to acknowledge the help and support provided by the University of Ghana , including the Institute of African Studies, the Office of Research, Innovation and Development, and the International Programmes Of fice. I would like to thank Dr. Sebastian Elischer for all of his help and guidance on previous drafts and article length versions of this dissertation. His comments and feedback were additionally incisive and very constructive in improving the dissertatio Ghana case study and its comparative analysis. In addition, I would like to thank Rachel MacDonald, an undergraduate Junior Fellow in the University of Florida Department of Political Science, for her assistance in managing the data for this project. Lastly, I will always be grateful to my wife for her eternal patience during my time in the doctoral program at the University of Florida. Her unwavering encouragement and positive attitude always succeeded in getting me through the many challenges experie nced in this process.
7 TABLE OF CONTENTS page ACKN OWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGURES ................................ ................................ ................................ ........ 11 LIST OF ABBREVIA TIONS ................................ ................................ ........................... 12 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION: TRAJECTORIES AS A TOOL FOR EXPLAINING DEMOCRACY IN AFRICA ................................ ................................ ...................... 16 The Argument ................................ ................................ ................................ ......... 25 A Note On Democracy and Democratic Consolidation ................................ ........... 37 Research Design: Nested Comparative Analysis ................................ ................... 40 Plan of the Dissertation ................................ ................................ ........................... 45 Conclusion ................................ ................................ ................................ .............. 50 2 DEMOCRACY ................................ ................................ ................................ ........ 51 Democratic Transitions ................................ ................................ ........................... 51 Transitions in Africa ................................ ................................ ................................ 54 ................................ ............................... 58 Foreign Aid ................................ ................................ ................................ ....... 58 Elections ................................ ................................ ................................ ........... 60 Competitive Authoritarianism in Africa ................................ .............................. 61 Other Explanations ................................ ................................ ........................... 62 Limits of Cu rrent Understandings on the Subject ................................ ............. 63 ................................ ................................ ..... 65 Brief Historical Background ................................ ................................ .............. 66 .............. 73 Conclusion ................................ ................................ ................................ .............. 80 3 INTERNA TIONAL AID AND LEVERAGE IN GHANA ................................ ............. 83 International Aid in Ghana ................................ ................................ ...................... 84 The Economic Recovery Programme (ERP) ................................ .................... 86 ................................ .................. 88 The End of the Cold War and Its Importance to Ghana ................................ .... 94
8 Ghana After the Cold War: Limited Leverag e, Legitimacy, and Continued Aid Dependence ................................ ................................ ................................ ........ 97 Fallout From the Inaugural Elections ................................ ................................ 9 8 International Leverage in Ghana ................................ ................................ .... 102 104 Con clusion ................................ ................................ ................................ ............ 108 4 DOMESTIC POLITICAL OPPOSITION ................................ ................................ 112 Origins of the Opposition ................................ ................................ ...................... 116 The New Patriotic Party ................................ ................................ ........................ 126 The 1992 Election and Afterward ................................ ................................ .......... 130 ajectory ...................... 135 Conclusion ................................ ................................ ................................ ............ 138 5 EXECUTIVE AGENCY ................................ ................................ ......................... 142 The Rawlings Fac tor: Pragmatism ................................ ................................ ........ 146 Factionalism ................................ ................................ ................................ .......... 150 .............. 154 Repression: Costs and Benefits ................................ ................................ ..... 155 Toleration: Costs and Benefits ................................ ................................ ....... 158 Rawlings Chooses to Tolerate ................................ ................................ .............. 159 Locking ................................ ........................... 162 Conclusion ................................ ................................ ................................ ............ 166 6 .. 169 Ghana as a Case of a Much Larger Phenomenon ................................ ................ 170 Outcome o f Interest: Regime Trajectory ................................ ......................... 175 Democratic trajectories ................................ ................................ ............ 176 Authoritarian trajectories ................................ ................................ .......... 176 Mixed trajectories ................................ ................................ ..................... 176 Condition Variables ................................ ................................ ........................ 177 Domestic political opposition ................................ ................................ .... 177 Foreign aid dependence ................................ ................................ .......... 178 International leverage ................................ ................................ .............. 179 Executive agency: tolerance or repre ssion ................................ .............. 181 A Theory of Regime Trajectories in Africa ................................ ............................ 183 Analysis ................................ ................................ ................................ ................ 185 Democratic Trajectories in Africa ................................ ................................ .......... 195 Conclusion ................................ ................................ ................................ ............ 201 7 .. 204 ........................... 213 ................................ ...................... 217 Ghana at a Crossroad: The 2016 Elections ................................ .......................... 222
9 Prospects For Democracy In Africa ................................ ................................ ...... 225 Implications ................................ ................................ ................................ ........... 227 APPENDIX A SUMMARY STATISTICS OF DEMOCRATIC OPENINGS IN AFRICA, 1990 1994 ................................ ................................ ................................ ...................... 231 B DEMOCRATIC OPENINGS IN AFRICA AND THEIR PREDISP OSED TRAJECTORIES ................................ ................................ ................................ .. 232 LIST OF REFERENCES ................................ ................................ ............................. 233 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 247
10 LIST OF TABLES Table page 1 1 Democratic Transitions in Africa, 1990 1994 ................................ ...................... 17 1 2 Successful Democratic Transitions in Africa and Their Contemporary Regime Status, 2010 ................................ ................................ ................................ ....... 18 1 3 Flawed Democratic Transitions in Africa and Their Contemporary Regime Status, 2010 ................................ ................................ ................................ ....... 19 1 4 Democratic O penings in Africa, 1990 1994 ................................ ........................ 37 6 1 Regime Trajectories in Africa ................................ ................................ ............ 189 6 2 Truth Table of Regime Trajectories in Africa ................................ .................... 195 6 3 Membership Within Democratic Trajectories in Africa ................................ ...... 199 A 1 Measuring Domestic Political Opposition, Relative Aid Dependence, and Int ernational Leverage ................................ ................................ ...................... 231 B 1 Structural Conditions and Their Influence on Regime Trajectories in Africa ..... 232
11 LIST OF FIGURES Figure page 1 1 Model of Regime Trajectories ................................ ................................ ............. 35 1 2 Summary of Nested Analysis ................................ ................................ .............. 43 2 1 Paths of Transitions in Africa, 1988 1994 ................................ ........................... 57 3 1 Inflation, 1980 1992 ................................ ................................ ............................ 89 3 2 Foreign Direct Investment, 1980 1992 ................................ ............................... 90 3 3 Economic Growth, 1980 1992 ................................ ................................ ............ 90 3 4 Per Capita Income, 1980 1992 ................................ ................................ ........... 91 3 5 Per Capita Income in Ghana Compared to Africa, 1980 1992 ........................... 92 3 6 International Aid in Ghana, 1980 1996 ................................ ............................... 93 3 7 International Aid in Ghana Compa red to Africa, 1990 1994 ............................. 103 5 1 Repression Versus Toleration ................................ ................................ .......... 159 5 2 Political and Civil Rights in Ghana, 1992 2000 ................................ ................. 161 6 1 Democratic Trajectories, 1990 2012 ................................ ................................ . 173 6 2 Authoritarian Trajectories, 1990 2012 ................................ .............................. 173 6 3 Mixed Trajectories, 1990 2012 ................................ ................................ ......... 174 6 4 Democratic Trajectories ................................ ................................ .................... 184 6 5 Authoritarian Trajectories ................................ ................................ ................. 184 6 6 Mixed Trajectories ................................ ................................ ............................ 185 6 7 Plot of Degree of Membership Within a Democratic Trajectory Against the Degree of Membership in L * A * O * T. ................................ ............................ 200 7 1 Diagrammatic Summary of Regime Trajectories in Africa ................................ 212
12 LIST OF ABBREVIATIONS C O E Committee of Experts CDR Committee for the Defense of the Revoluti on EC Electoral Commission ECG Electoral Commission of Ghana EGLE Every Ghanaian Living Everywhere (Party) ERP Economic Recovery Programme E.U. European Union GBA Ghana Bar Association GDP Gross Domestic Product GSS Ghana Statistical Service ICC Inter Part y Coordinating Committee IMF International Monetary Fund INEC Interim National Electoral Commission IPAC Interparty Advisory Committee LNA Large N Analysis MFJ Movement for Freedom and Justice mb SNA Model Building Small N Analysis mt SNA Model Testing Sm all N Analysis NCD National Commission for Democracy NCP National Convention Party NDC National Democratic Congress NPP New Patriotic Party NUGS National Union of Ghanaian Students PAMSCAD Programme of Action to Mitigate the Social Costs of Adjustment
13 PNDC Provisional National Defense Committee QCA Qualitative Comparative Analysis SAP Structural Adjustment Program SNA Small N Analysis TUC Trades Union Congress U.S. United States
14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE LEGACIES OF DEMOCRATIC EXPERIMENTS IN AFRICA: GHAN DEMOCRATIC TRAJECTORY IN COMPARATIVE PERSPECTIVE By Nicholas Knowlton December 2015 Ch air: Leonardo VillalÃ³n Major: Political Science In many ways, Ghana represents a puzzling and deviant case of democracy in Sub Saharan Africa. Despite experiencing problematic inaugural elections in 1992, the country has since developed into a widely reco gnized liberal democracy. Explanations for this outcome remain rather limited, with prior accounts suggesting this outcome may be explained either as an idiosyncratic case or an outlier. This dissertation investigates the case of Ghana emocracy and explains how the country became a democracy despite experiencing initially inauspicious conditions. Furthermore, I argue that rather than serving as an outlier, Ghana instead represents a paradigmatic example of what will be called a democrati c trajectory in Africa, with insights leading to a broader understanding of democratization in Africa after the Cold War. Comprised from a confluence of domestic and international factors, including access to foreign resources and significant domestic comp etition between political forces, an explanation that favors a concept of democratic trajectories , rather than democratic transitions, not only accounts for how this country developed into a widely recognized democracy through a
15 steadfast and incremental p rocess, but additionally contributes to the study of post Cold War democratization elsewhere in Africa.
16 CHAPTER 1 INTRODUCTION : TRAJECTORIES AS A TOOL FOR EXPLAINING DEMOCRACY IN AFRICA M ore than a quarter of a century has passed since the end of the Cold War between the United States and Soviet Union, yet in many ways the world continues to experience its aftershocks. The increase in the number of democratic regimes in the world in the years since has often been attributed to both the decline of great powe r rivalry, as well as the increased conditionality of foreign assistance. As Soviet support dissipated , and as future financial assistance became further conditional upon measures of good governance and democracy, many regimes were compelled to both libera lize and democratize in order to survive. Within Sub Saharan Africa (Africa hereafter), such conditions were all the more important, as many regimes were still reeling from the global economic shocks experienced in the 1970s (Bratton and v an de Walle 1997, 272; Nugent 2004; Fraser 2009). As a result of the increased difficulty in with democracy (Table 1 1) . Between 1990 and 1994, no fewer than 27 countries underwent proces reforms and/or elections, others instead experienced democratic movement through their of political reform and willingness to submit their rule to competitive elections. Bratton and van de Walle (1997) classified 15 countries as experiencing where incumbents accepted political reform and followed with holding free a nd fair inaugural elections. While the remaining 12 experienced less than free and fair elections (largely due to incumbent manipulation) ,
17 transition, as incumbents in thes e instances were nevertheless unable to halt or control the processes of political reform (1997, 120 1). To be sure, while standard conceptualizations of democratic transitions would likely preclude any instance of elections that were manipulated or otherw ise less than free and fair , Bratton and van de cases from other instances where any substantial democratic change was prevented by incumbent interference. Table 1 1: De mocratic Transitions in Africa, 1990 1994 1 Successful Transitions Flawed Transitions Precluded Transitions Blocked Transitions Benin Burkina Faso Angola Liberia Cape Verde Cameroon Burundi Sudan Central African Republic Comoros Chad Rep. of Congo Djib outi Dem. Rep. of Congo (Zaire) Guinea Bissau Equatorial Guinea Ethiopia Lesotho Gabon Guinea Madagascar Ghana Nigeria Malawi Ivory Coast Rwanda Mali Kenya Sierra Leone Mozambique Mauritania Somalia Niger Swaziland Tanzania SÃ£o TomÃ© and PrÃ ncipe Togo Uganda Seychelles South Africa Zambia Source: Reproduced From Bratton and Van de Walle 1997, 120 . However, in spite of these democratic openings, subsequent coups, crises, and institutional stagnation continued to challenge the dem ocratic gains made during the inaugural period. 1 this analysis as its democratic trajectory occurred prior to the end of the Cold War in 1989.
18 as a widely recognized and reasonable assessment of democracy on the continent, b y 2009, only half of the countries previously experiencin g democratic transition s could still democracies (2010, xxvi). Of the 15 successful cases of democratic transition, only four succeeded in producing a liberal democratic regime in the time since their i nitial transitions (Table 1 2). As more than twenty years of observation demonstrates that the most likely cases of democracy in Africa that is, those countries that experienced the most propitious conditions for a dvancing democra cy at the time of their transitions have since resulted in contemporary regimes stretching across the democracy authoritarianism spectrum, and have furthermore only produced five regimes that can claim to be more than nominally democratic, such results sug gest that further consideration ought be given to understand why some countries experienced democratic breakdown, as well as why others remained fragile and tenuous well after their initial democratic transitions . Table 1 2: Successful Democratic Transiti ons in Africa and Their Contemporary Regime Status, 2010 Liberal Democracy Electoral Democracy Competitive Authoritarian Electoral Authoritarian Politically Closed Benin Guinea Bissau Central African Republic Republic of Congo Cape Verde Lesotho Madagas car S Ã£ o Tom Ã© and Pr Ã ncipe Malawi South Africa Mali Mozambique Seychelles Zambia However , the case of Ghana presents a rather puzzling case of democratization, in that this country experienced what many would have considered an in auspicious
19 democratic transition in 19 93 (Nugent 1995; Bratton and v an de Walle 1997), yet nevertheless emerged as one of the most democratic regimes on the continent twenty years later. Indeed, of Bratton and van de Walle (1997) , including the holding of problematic inaugural elections, possessing a dominating executive , a restricted freedom of the press, and an overall blurred line between state and political party resources at the time of tran sition, Ghana has since managed to follow what may be best characterized as an incremental and steadfast path of democratic improvement. How is it that Ghana, a country with inauspicious democratic features at the time of its transition, nevertheless manag ed to follow a steadfast path of democratization? Whereas all other c ountries in Africa with similar inaugural conditions have since produced non democra tic contemporary regimes , what explains why Ghana took a different path in producing (as well as mainta ining) its current democratic status? As Table 1 3 demonstrates, among all other countries classified as flawed by Bratton and van de Walle (1997), that is, countries that experienced some amount of political reform bu t did not experience fully free and fa ir inaugural elections, Ghana stands out not only as the only country to have been widely recognized as a contemporary democracy, but additionally stands out for being recognized as a liberal democracy , based on . Table 1 3 : Flawed Democratic Transitions in Africa and Their C ontemporary Regime Status, 2010 Liberal Democracy Electoral Democracy Competitive Authoritarian Electoral Authoritarian Politically Closed Ghana Djibouti Burkina Faso Ivory Coast Kenya Cameroon Equat orial Guinea Gabon Mauritania Togo Swaziland
20 In answering these question s , this dissertation additionally speaks to the much larger question concerning c ontemporary democracy in Africa; that is, why did some democratic transitions continue to deve lop for more than twenty years after their initial democratic transitions in the immediate post Cold War period explain little with regard to contemporary regime status, the investigation of Ghana and its particular trajectory of democracy contains the potential to provide additional insights into what has continued to remain a rather under exp lored area. I investigate this case as a deviant case of democracy, that is, as an outlier that does not accord with prior explanations, and as an outcome that was not to be expected fro m a country with a flawed transition. At the same time, my research shows that while the case of Ghana initially appears as a deviant case of democracy in Africa, this deviance is largely dependent on a perspective emphasizing democ ratization through trans itions and as occurring at single points in time. As a result , by discounting the role of process , current explanations on democratization in Africa generally remain limited in their ability to account for regime change over time. In brief, I argue that Gh by the factors present at the time of its democratic opening , defined as the period in which an authoritarian country experiences a critical juncture that produces a crisis of maintaining the authoritar ian status quo, and where the political leadership consciously embarked on democratic reform in order to manage the crisis , including subjecting itself
21 to nominally democratic inaugural elections . Utilizing a perspective favoring historical institutionalis m, including elements of critical junctures, path dependence, and unintended consequences (Collier and Collier 1991; Pierson 2000; 2004), ultimate regime trajectory was influenced by a combination of factors present at the time ocratic opening , i ncluding significant amounts of international aid , significant amounts of international leverage held by pro democracy Western donors , significant domestic political opposition, as well as the national tolerate add In turn, the presence of these factors placed Ghana on a self reinforcing democratic path that continued to improve in a steadfast, incremental process. As I will argue in de mocratic transition did not occur at a single point in time, but rather over a span of years following its initial democratic opening. To be sure, this dissertation adopts the term, democratic opening, for what many would largely consider a democratic tran sition (Bratton and van de Walle 1997, 10). T hough the two concepts may seem analogous to one another , a distinction is democratic transitions emphasizes authoritarian re gimes transforming into democracies following their inaugural elections, a democratic opening instead emphasize s the contingent nature of these events , in spite of the pro democratic sentiment that initially i nstigated them. In other words, democratic open ings refer to periods of political change where democratic reform precipitated the collapse of a previous authoritarian regime, but where the resultant equilibrium (whether democracy or otherwise) remained undetermined .
22 In addition to explaining both why a nd how Ghana proceeded to follow a democratic trajectory, this dissertation contributes to the wider literature on democratization and regime change by positing a theory of regime trajectories. In leveraging the insights generated from this case, the findi ngs from a comparative analysis of Ghana and the wider set of countries experiencing democratic openings in the democratic wave following the end of the Cold War (1990 1994) assist in identifying the types of regime trajectories that emerged, and explain w hy different trajectories came to be associated with certain states. This project sheds light on an under explored area within comparative politics and African studies. As prior analyses have largely accounted for the particular and individual country leve l experiences of democracy in Africa, there largely appears a gap concerning the cross case examination of these experiences since their initial openings (Diamond and Plattner 1999 ; 2010 ; Villalon and VonDoepp 2005 ; Mustapha and Whitfield 2009 b ) . In additi on, as much of the predominant literature has tended to emphasize the concept of democratic transitions as events occurring at single points in time, such a perspective conspicuously omits the types of legacies these events produce, thus leaving much to be explained. Lastly, there also remains the issue as to how the experiences of these cases stand in relation to the broader context of democratization in Africa, including the question as to whether any shared factors or conditions exi s t that may assist i n explaining why some countries appeared to follow a steady trajectory of democratic institutionalization, while others did not. 2 2 Notable exceptions to this observation, however, include edited works by VillalÃ³n and VonDoepp (2005), as well as by Mustapha and Whitfield (2009b).
23 Moreover, this project contributes to an ongoing conversation concerning the Bratton and Van de Walle 1997). In their work, Bratton and Van de Walle provided a seminal contribution to the study of democracy in Africa by highlighting the conditions precipitating democratic transitions, including popular protests, political and econo mic liberalization, and de mocratic elections (1997, 116 7). By identifying and elaborating upon these factors, the authors showed how certain countries succeeded in undertaking democratic change, while others did not. However, as noted above, when comparin transition outcomes with more recent assessments of democracy in Africa, it becomes its contemporary regime status . If we compare the type s of democratic transitions that Bratton and van de Walle labeled 2010 assessment of democracy in Africa (xxvi), upon review, there is little correlation between the type of transition a country ex perienced and its contemporary regime type, even when controlling for geography and colonial legacy. Serving as the point of departure for this study, this project argues in favor of a theory of regime trajectories that emphasizes that the conditions prese nt at the time of such democratic openings explain why Ghana, as well as other countries experiencing such change in the same period, experienced divergent trajectories after their initial democratic openings. ) typology of democratic transitions in Africa, the concept of a democratic opening encompasses what would be considered successful and flawed transitions, but re conceptualizes them into a single
24 category that emphasizes the commitment to political change : not necessarily a change to democracy, as an emphasis on transitions suggests, but rather a change ultimately resulting in either a democratic or non democratic outcome . It is important to note that while both this dissertation and Bratton and van de Wal Bratton and van de Walle treat what ought to have been regar ded as a democratic opening as a democratic transition instead . In doing so, their analysis unfortunately explains a phenomenon that did not occur at that particular time, but instead emerged As a result, their investigation misses what could have been an opportunity for explaining the long term prospects for democracy among the countries in their sample . Furthermore, by mistaking an opening for a transition, subsequent investigations of this period in African politics have continued to treat political change during this period within a perspective of transitions and its associated perspective of consolidation, thus leaving an important area within African politics unexplored, and the important question concerning the legacies of these democrat VonDoepp 2005; Mustapha and Whitfield 2009 b ). To be sure, while the authors note their findings to be contingent on whether democracy , the ir analysis nevertheless typologizes the kinds of transitions experienced across the continent within a hierarchy, with instances of successful transitions listed at the top of this hierarchy and thus being the kinds of cases most likely to succeed in maintaining democracy than
25 their flawed counterparts. T he study of regime change has generally held explanations for regime change within a perspective emphasizing transitions Schmitter 1986) . Bratton and van de Walle similarl y couched their (1997) analysis of regime change within this dominant perspective. However, as subsequent scholars have noted the limits to what the transitions literature can provide to studies of regime change (Collier 1999; Carothers 2002 ; Diamond, Fuku yama, Horowitz, and Plattner 2014 ) , the challenge remains how to properly explain the events and subsequent political change that took place in Africa following the Cold War. 3 The Argument The end of the Cold War brought about a period of significant chan ge for Africa. The decline of great power politics on the continent, compounded with decreased access to international resources, created conditions of crisis for many authoritarian African regimes (Schraeder 1994; Bratton and Van de Walle 1997; Ake 2000; Fraser 2009). This event, and the period immediately following it, presents a distinct period worthy of comparative analysis in how each state simultaneously experienced the shock from the collapse of the of great power rivalry between the United States (U .S.) and the Soviet Union that followed (Huntington 1991) . In unexpectedly entering an environment of scarcer international resources, including increased conditionality to foreign assistance, these countries ex perienced resulting crises of the status quo. As governments within many of the then authoritarian regimes already saddled with significant amounts of foreign debt found themselves 3 To be sure, while Carot literature within the policy community (2002b), a close reading of his argument nevertheless applies to the transitions literature within the academic community. This criticis m, in turn, prompted a response from
26 unable to rely upon relatively unconditional access to foreign assistance t o reinforce their political and economic systems, and furthermore found themselves under significant pressure from domestic pro democracy movements, such governments were compelled to change in order to survive. Positing the period immediately following th project highlights the actions and experiences of one state that engaged in democratic reform in response to such a crisis of regime maintenance, and furthermore examines how the conditions during this period of change affected its regime trajectory. In other words, as Ghana, much like many other African states that stood poised between the threat of disintegration and the unknown ter Cold War (VillalÃ³n 1998, 8), this project addresses how this state weathered this crisis, and investigates how its response shaped its regime trajectory in the years that followed . In addition, as the actions t aken during this period of high contingency produced distinct paths that became self reinforcing and increasingly difficult to change over time (Pierson 2004), countries in Africa, such as Ghana, that experienced initial democratic movement embarked upon d istinct regime trajectories that served to influence their likelihood of maintaining a democratic system of rule. For the purposes of this dissertation, regime trajectory is defined as the overall direction that a regime traveled since it s inaugural democr atic opening. R egime trajectories provide temporal cont inuity to institutions, guiding them through processes of evolutionary change, rather than through institutional rupture . Precipitated by democratic openings, regime trajectories
27 emerged as a function of a combined presence of factors influencing the direction the a prolonged period of time, until such trajectories were altered by future critical junctures, or other exceptional circumstances. While, in re ality, regime trajectories followed a range of paths along a democratic authoritarian spectrum, in general, regime trajectories fell within three kinds of outcomes: democratic, authoritarian, and mixed trajectories. By adopting this tripartite typology, th ese categories provide a representative range of outcomes that account for the variety of regimes that emerged in the course of the 1990s and 2000s . 4 First, democratic trajectories are defined as cases that have (1) held regularly scheduled, national, mult i party elections since their initial opening; (2) experienced an overall continuous expansion o f political and civil liberties without any significant setbacks; (3) had no exp erience of democratic breakdown, including occurrences of coups or other types o f extralegal transfers of power; and (4 ) had met the basic requirements to be regarded as an electoral democracy by 2010. In contrast, authoritarian trajectories are defined as those cases that failed to meet the criteria for a democratic trajectory , despi te experiencing initial democratic movement during their respective openings . I ndicators of authoritarian trajectories include the following: (1) a opening; (2) an occurrence of regime breakdown that never fully recovered; and (3) the failure to hold regularly scheduled multi party elections, or the presence of such an amount of active political manipulation by the incumbent government that such elections and politic al competition become practically meaningless. Lastly, mixed trajectories are 4 authoritarian, and competitive authoritarian regimes (2010, 365 369).
28 defined as those cases that did not succeed in maintaining a democratic trajectory, yet have still managed to avoid falling within the authoritarian category. Utilizing the insig hts derived from the Ghanaian experience, I posit that the observed variation in regime trajectories among countries experiencing democratic openings between 1990 and 1994 may largely be explained through the combined presence of three structural condition s : (1) s ignificant political opposition; (2) reliance on foreign aid ; and (3) international leverage . This, in turn, was mediated through a fourth factor: during their inaugural administrat ion. 5 forces as influencing the long term probability of success of democratic reform (1991, 79 88), the presence of significant domestic political opposition during a dem ocratic is hypothesized to affect the direction of a given . This is due to the ability o f such groups to pose a formidable obstacle to any non democratic manipulation of the political system by the incumbent administration . In order to empirically assess for the presence of such opposition, this variable is operationalized according to three conditions: that the opposition was relatively unified, minimally competitive, and able to mobilize voter s to turn out. Parties that were unified and minimally competitive obtained at least 30 per cent of the vote in 5 To be sure, tolerati on of democratic demands does not necessarily mean that an executive desired or otherwise willingly accepted further democratic reform. Tolerance of additional democratic reforms could include an executive being against reform, yet utilizing measures to co ntrol the process so as to minimize the end result.
29 the inaugural presidential election, or otherwise lost the election by 15 per cent or less. 6 In addition, opposition parties demonstrated their mobilization capacity by encouraging a sufficient number of their members and sympathizers to vote so as to obtain a minimum turnout rate of more than 50 per cent of registered voters. 7 Secondly, the impact of foreign aid is also hypothesized to have aff ected the long term trajectories of many African countries. In a period of increased conditionality and where aid was often attached with democratic demands, for countries like Ghana whose share of foreign aid exceeded ten per cent of their annual GDP, the re is significant reason to believe that there would have been a significant interest in continued liberalization, not necessarily because such aid precipitated change in a kind of policy quid pro quo, but more likely because recipient governments wanted t o continue receiving the benefits of foreign aid into the future (Wright 2001; Dunning 2004; Levitsky and Way 2010). In order to control for the impact of foreign aid on this gural period, this condition has been operationalized according to the average share of aid as constituting ten per cent or more of their GDP would reasonably be considered to h ave felt more pressure to democratize than other countries receiving less than ten per cent. 8 6 It is important to note that the latter condition losing with less than 30 per cent total but being within 15 percentage points of the winning party inherently means that the winning party won with less than 50 per cent. 7 See Kaya and Bernhard (2013) for discussion on voter mobilization capacity. In addition, while the cutoff rate of 50 per cent turnout may appear low, this is to control for instances where voter rolls were inflated due to outdated vote r registers, which were widely believed to have been a frequent occurrence during this period. 8 In short, the reasoning behind this ten per cent threshold is due to a natural inflection point observed between countries with democratic and authoritarian tr ajectories. Addition details on this condition are provided in Chapter 6.
30 However, while foreign aid may incentivize governments to democratize, it may also hinder such efforts by promoting corruption, or even by strengthening the capac ity of the government to repress. 9 As aid inherently contains the potential to help and/or obstruct democratization, it is worth considering the role of international leverage as a third factor, and the manner in which it enhances ty to promote democratic reform. international leverage is composed of elements cal power in the international system, including gross domestic product (GDP), level of oil production, possession of nuclear weapons, as well as whether the country retained any geostrategic or security interest among Western countries or whether the stat e was a particular benefactor from any non Western major powe r (Levitsky and Way 2010, 372 3). In other words, not to be misconstrued as international pressure (where one country exerts influence over another), international leverage instead is the latent susceptibility to international pressure that a given country has with respect to Western countries interested in promoting democratic reform. In short, countries receiving a high level of international leverage would be expected to be more susceptible to democratic reform via international aid as opposed to other countries receiving less international leverage. Together, I argue that these three factors serve as the structural conditions for explaining why different countries proceeded to follow divergent regime trajectories after their initial democratic openings. For democratic trajectories, conditions exhibiting the 9 relationship with democracy promotion.
31 combined presence of significant domestic political competition, high international leverage, and foreign aid dependence produced the ideal situations largely favoring the continued advancement of democratic behavior over time. In contrast, in countries where significant political competition was absent at the opening , and where the state was largely resilient to Western democratic pressure wh ile also non reliant on foreign aid, these factors produced conditions largely inimical to the long term advancement of democracy, thus predisposing a country to follow an authoritarian trajectory. Lastly, for those countries with mixed or ambiguous combin ations of these factors , such countries are hypothesized to have followed an inconsistent trajectory that would exhibit neither clearly democratic nor authoritarian tendencies. W hile such structural conditions incentivized countries towards democracy, it i s important to recall that during periods of democratic opening, the environment remain ed highly contingent and open to multiple outcomes . As a result, national executives in these inaugural periods retained a high degree of agency in determining the ultim ate path their respective countries would follow. Yet, w hile executives possessed a range of policies to pursue, such decisions largely fell along one of two choices: whether to of rule . Utilizing the reasonable assumption that national executives sought to obt ain as much power as they could and retain it for as long as possible, these executives in their inaugural administrations faced the choice as to whether they would seek to revert their respective countries back to the authoritarian past, or whether they would instead fare better by maintaining the post opening democratic environment. On the one hand, executives in the post democratic opening environment could have sought to
32 consolidate their power and restore the kind of discretionary authority previously seen prior to have chosen to cede the possibility of restoring authoritarian rule and accept ed a new democratic status quo. In this sense, executives did not necessarily have to be converted to democracy, but could have instead sensed that they would likely be more successful at maximizing their interests by working within the confines of a democrat ic system. In addition , when faced with the choice between maximizing their power or their duration of rule, executives would have valued their tenure in office more than the pursuit of absolute authority. This is a reasonable assumption, as it stands to reason that executives would have likely preferred to sacrifice authoritarian control if it meant that they could remain in office In turn, the decision to either work within the confines of a democratic system or abroga te it altogether was guided by the presence and absence of the three previous ly highlighted factors. In short, where executives recognized the presence of significant domestic political opposition, high international leverage, and significant amounts of co nditioned foreign aid, they were less likely to try to consolidate their powe r and instead opt for policies that would serve to prolong their rule in the post democratic opening environment . This would have likely included tolerating additional democratic reforms, including removing various restrictions on the media and press, allowing for greater tolerance of political dissent and political demonstrations, as well as supporting and abiding by decisions made from the judiciary, even if such decisions were c ontrary to executive positions . In other words , as these actions occurred in a period where the
33 pro democratic sentiment was already riding high following the collapse of the former authoritarian regime, many of the policies designed to prolong the executi office would likely have coincided with actions facilitating further d emocratic reform in the country . In contrast, in instances where the opposition was insignificant, where international aid was relatively small, and where leverage was lim ited or inconsequential, executives possessed a higher probability of success in consolidating their rule. In other words, the decision to repress or tolerate democratization largely depended on whether executives recognized the constraints to their rule a nd acted accordingly . To be sure, executives could still have acted contrary to what the structural circumstances suggested (due to either misperception by the executive or even a personal interest in promoting democracy or authoritarianism) , though to do so would have likely resulted in an outcome below their desired expectations , due to the previously highlighted structural obstacles . In addition, while it is possible that executives may have been generally inclined to try to consolidate their authority , believing that they could retain both discretionary authority and indefinite tenure as chief executive reminiscent of the democratic opening, it is important to remember that all of the countries in this sample had previousl y experienced a political crisis that challenged the authoritarian regime. In ord er to resolve these crises, authoritarian leaders and their governments realized they would be required to engage in democratic reform, including holding inaugural elections. For incumbents or (in cases of executive turnover) newly elected executives t o have imm ediately withdrawn such reforms in their first administration would have been
34 problematic for obvious reasons. I t would have likely provoked protests similar to the ones experienced prior to the democratic opening , and would have additionally likely resulted in international aid upon which many African countries were dependent being suspended or withdrawn. 10 In sum, these factors provide the model for explaining why differ ent countries proceeded to follow divergent regime trajectories after their initial democratic openings. As levels of foreign assistance, opposition presence, and the degree of leverage exerted by external donors influenced the decision for national execut ives to pursue term r egime prospects (Figure 1 1 ). In addition, it is important to note that the relationship between international aid and leverage was rather dependent in that the tw o worked in concert with one another in order to provide effective international (democratic) pressure on a given country. Where high amounts of international aid coincided with high amounts of international leverage, foreign aid was likely to be effective in providing democratic pressure. In contrast, in instances where aid was insignificant, despite donors possessing high amounts of leverage, any aid provided was likely to remain ineffective. Lastly , in other instances where aid was significant but levera ge was not , the effectiveness of aid was similarly likely to be uncertain . With regards to the production of regime trajectories, w here executive actions accorded with structural conditions favoring democratization, including significant political oppositi on, significant receipts of foreign aid, and high international leverage, 10 This argument concerning the nature by which African executives were limited to two choices, concerning the repression or toleration of democratic demands, will be discussed in further detail in Chapter 5.
35 such countries proceeded along democratic trajectories. In contrast, in countries where the political opposition failed to establish a significant presence, where democratic leverage remained low, and where foreign aid was a relatively small percentage of state income, such structural conditions generally failed to constrain executive interests in consolidating power, thus providing the ideal circumstances for an authoritarian traject ory. Lastly, for those countries where variations in structural conditions produced rep ress democratic reform, such countries subsequently followed inconsistent (mixed) trajectories exhibiting neither clear democratic nor authoritarian tendencies. Figure 1 1 : Model of Regime Trajectories A theory of regime trajectories contributes to the s cholarly study of democracy in Africa by answering the unresolved question as to why some democratic experiments succeeded and why others did not. In order to execute this plan of research, this project adopts a historical perspective highlighting the role s of critical junctures, path dependence, and institutions when explaining the trajectories of many African regimes. Regime Trajectory Executive Agency: Repress or Tolerate? International Pressure Foreign Aid International Leverage Domestic Political Opposition
36 and processes that resulted in influencing the long term trajectories of African states experiencing democratic reform in the early 1990s (Pierson 2000; 2004). the results from the Ghanaian cas e against the trajectories o f 26 other African countries that experienced democratic openings in the period immediate ly following the end of the Cold War (Table 1 4 ). Using Bratton and Van de Walle's initial typology of democratic transitions, I posit that each country in which the a uthors labeled as a successful or flawed democratic transition after 1990 can be considered to have ex perienced a democratic opening . As previously highlighted, countries in these categories experienced a crisis of the authoritarian status quo that require d the implementation of democratic reforms and inaugural elections. Furthermore, unlike instances of democratic openings occurring after 1994, this set of cases is unique, in that national executives in this period did not have the benefit of conditioning their openings based on the previous experiences of other countries so as to manipulate the democratic process in an authoritarian manner (Levitsky and Way 2010). 11 By comparing the historical legacies of countries that undertook significant efforts at dem ocratization in the immediate aftermath of the Cold War, this project contributes to the broader literature on democracy in Africa by not only providing a typology of regime trajectories that emerged more than two decades ago, but by additionally identifyi ng and explaining the reasons that brought about such trajectories. By comparing the case of Ghana with the wider range of cases experiencing democratic openings in this period of significant change, 11 Specifically, Levitsky and Way note that by 1995, countries undergoing democratic tr ansitions had manipulating the electoral rules of the game (2010, 19; see also Carothers 2002).
37 this project provides both theoretical and practical ins ights relevant to those with scholarly and/or professional interests on the subject of democratization and regime change in Africa. Table 1 4 : Democratic Openings in Africa, 1990 1994 Countries Benin Burkina Faso Cameroon Cape Verde Central African Repu blic Comoros Congo Rep. Djibouti Equatorial Guinea Gabon Ghana Guinea Bissau Ivory Coast Kenya Lesotho Madagascar Malawi Mali Mauritania Mozambique Niger S Ã£ o Tom Ã© and Pr Ã ncipe Seychelles South Africa Swaziland Togo Zambia A Note On D emocracy and Democratic Consolidation While this dissertation investigates the phenomenon of regime trajectories in Africa following the Cold War, it remains inherently necessary to provide some clarification with regards to the concepts of democracy and a uthoritarianism as they are used in this investigation . First, it is worth acknowledging the nature of democracy as being a longstanding contested concept within the discipline. On the one hand, there are debates concern ing maximalist and minimalist defini tions of democracy. Minimalists often refer to being arriving at political decisions in which individuals acquire the power to decide by means [1950 ] 1962 , 269 ). By conceptualizing democracy with as few attributes as possible, minimalists generally argue, it enables
38 analysts to efficiently discriminate between democratic and non democratic regimes. In contrast, maximalist definitions go beyo nd the Schumpeterian baseline and extend definitions of democracy to include items beyond the ability of citizens to elect their leaders and representatives. A practical example of this reasoning would include constituting politi cal rights and civil liberties. On the other hand, there are other debates concerning the conceptualization of democracy as a dichotomous concept or as a matter of degree, that is, whether democracy can be conceived as something which a country either poss esses or does not, or whether it is possible to conceive of democracy as something that countries can 184; cited in Collier and Adcock 1999, 538). 12 In contrast, other scholars argue that democracy can, in fact, exist in varying levels and that countries can exhibit democratic behavior to higher and lesser degrees. Perhaps the most notable p roponent among this democracy as an ideal type by which all other democratic countries (polyarchies, in Another example of this vie w may include the quality of their democratic and authoritarian regimes. (1999) pragmatic approach for conceptual izing democratization. Noting that various perspectives of democracy each contain inherent strengths and weaknesses, the 12 Other proponents of this view would include Linz (1975), Huntington (1991), and Geddes (1999) (as identified by Collier and Adcock 1999, 538).
39 authors argue that the decision for conceptualization (and subsequent operationalization) ought to be predicated on the needs derived f rom the research question and the objectives of the investigation. In particular, they argue methodological choices are often best understood and justified in light of the theoretical framework, analytic goals, and context of research involv ed in any particular study. As 539). (1971) conceptualization of democracy largely defined as the maximi zation of public contestation and the right of participation as an ideal type that countries move towards or away from along their respective regime trajectories. most useful avenue for conceptualizing regime trajectories, as it is sufficiently flexible to be employed in an investigation of regime change over time, and as it provides a common foundation for a wide audience . In addition, a s I conceptualize political change in Africa through a lens of regime trajectories, rather than through a framework of transitions, it becomes inherently problematic to discuss consolidation as it relates here . As popularly defined in the literature, democratic consolidation occurs at the moment when the regime is internalized by all aspects of However, as the concept of regime trajectory inherently rejects the notion that political change occ urs at a moment in time, so too does it reject a perspective that sugg ests a regime has be en consolidated otherwise immune from breakdown after either meeting a certain threshold of internalization among members of society (Linz and Stepan 1996) or experiencing a
40 number of executive turnovers (Huntington 1991). In a ddition, it should also be noted that the concept of consolidation is not without its critics ( ; Carothers 2002 ), who generally argue that inconsistencies with regards to the formal rules and actual behavior of countries , as well as the conce longstanding difficult concept to define) make the concept of consolidation inherently problematic , and who further note the problematic view of consolidation occurring as a result of an inherently natural, forward moving pro cess. Furthermore, e ven its proponents note the difficulty in conceptualizing the term for a wide audience (Schedler 1998) , let alone measuring for its presence (Schedler, 2001) . Nevertheless , because the concept of regime traj ectory conceptualize s politi c al change as occurring through a democratic opening experienced during critical juncture, and that the observations of this change occur over time (as opposed to a particular moment in time), it becomes inherently problematic to incorporate a discussion of consolidation into this investigation. In other words, regime trajectories never decision to tolerate or r epress democratic demands made from society . To be sure, trajectories may be altered due to exceptional circumstances (as discussed in Chapter 6), but otherwise , countries remain firmly established within their respective trajectories until another crisis or critical juncture occurs, thus establishing a new path. Research Design: Nested Comparative Analysis Serving to explain the origins of regime trajectories in Africa, as well as the reasons why the democratic experiments produced str ikingly distinct regimes twenty years after their initial occurrence, with little correlation between
41 this dissertation adopts a nested research design that leverages the i nsights generated through the in depth investigation of a single case against the experiences of a larger set of cases. As previously elaborated by Lieberman (2005), the mission of nested analysis is to provide a multi method framework for inquiry that com bines the logic of quantitative and qualitative approaches for the purposes of assessing existing theories, as well as the building of new ones . Rather than debating the purported superiority or limitations of either approach to inquiry, nested analysis in stead highlights the complementarities between the two, as well as the manner in which each approach, categorized as large N analysis (LNA) and small N analysis (SNA), may be sequentially utilized in order to maximize the benefits of each approach while mi nimizing their limitations. 13 As perhaps best explained by Lieberman, nested analysis provides a stronger basis for causal inference than the sum of small N and large N parts. Rather than emphasizing the common inferential logic of qualitative and quantit nested analysis approach emphasizes the complementary distinctiveness in these two modes of analysis and strategies f or causal inference (2005, 450 ). Briefly summarizing the nested approach to inquiry, nested analysis begins with a preliminary large objective of assessing for any general relationship observed among the variables of interest (Lieberman 2005, 438). After assessing for an y observed general relati onship from the results of the initial LNA, researchers then decide if the results from the LNA shoul d be used within either a model testing SNA (mt SNA) or a model building SNA 13 For reviews on the debate, merits, and limitations between quantitative and qualitative approaches, see Lijphart (1971); Ragin (1987); King, Keohane, and Verb a (1994); Gerring (2001); Brady and Collier (2010); Mahoney and Goertz (2006).
42 (mb SNA). On the one hand, i f the results from the LNA appear to accord with t he may validity through mt SNA, where the intent is to verify whether the variables of interest are indeed acting in accordance with what the given theory would provide, and wher e be in accordance with what the initial model would predict (Lieberman 2005, 442, 444). On the other hand, should the results of the LNA not appear to support the baseli ne theory, researchers instead engage in a mb SNA, where, through the selection of deviant cases, the goal is to explain why the baseline theory is limited, and whether the theory can be improved upon or replaced (2005, 443, 444 445). In sum, nested analys is provides a means of explaining political phenomena and their causes in a framework that maximizes the benefits of large and small N approaches, while minimizing their limitations. Constituting more than a simple combination of approaches, nested analysi s instead leverages each component against its counterpart in order to produce a strategy of inference that is greater than the sum of its constituent parts (Lieberman 2005, 450). Furthermore, as this dissertation engages in a model building exercise using a case that has been previously regarded as a deviant case of democracy in Africa, this research design provides an updated explanation that shows the case of Ghana not to be an outlier (as previously suspected), but rather a paradigmatic example of a muc h broader phenomenon. Figure 1 2 provides a diagrammatic summary of the nested approach.
43 Source: adapted from Lieberman 2005, 437. Figure 1 2 : Summary of Nested Analysis E xplanations of democratization in Africa remain limited insofar as the conditions on the continent do not generally accord with many of the standard theories of democratization, including economic explanations (Przeworski , Alvarez, Cheibub, and Limongi 2000 based explanations ( Moore 1966; Rueschemeyer, Stephens , and Stephens 1992 ), or even more recent explanations of democratization vis Ã vis competitive authoritarianism (L evitsky and Way 2010). T his dissertation highlights this gap between existing theoretical explanations and the actual experiences of the majority of countries in Africa and engages in a model building exercise through the in depth investigation of the seemingly deviant case of democracy in Ghana. LNA Satisfactory an d Robust? Yes No mt SNA mb SNA Results Results Congruent with LNA Problems Revised model with mt LNA Unable to revise model; end analysis Problem with Case Problem with Theory
44 Data for the in depth investigation of Ghana was derived from sev en months of fieldwork during which I collected semi structured, elite interviews among in country experts within government, politics, academia, and civil society. 14 For the purposes of es were most likely to be familiar with the governmental processes occurring over the course of the past 20 years, and served to assist the dissertation by either corroborating prior ors and mechanisms that would have otherwise remained unknown. 15 Following the logic of nested analysis, the results from the Ghanaian case are used to build an alternative explanation of democratization in Africa. Using qualitative comparative analysis (Q CA), I then test the findings from the Ghanaian example against the wider body of cases experiencing democratic op enings between 1990 and 1994. In were operationalized into a series of conditions (aid, leverage, political opposition, executive decision) that assess ed for the presence or absence of these conditions operating under a perspective highlighting conjunctural causation and the types of outcomes resulting from such interactions. 16 14 For this dissertation I interviewed approximately 30 respondents, some of which resulted in multiple es of protecting confidentiality, occupation and sector, so as to provide a sufficient amount of information to the reader while maintaining participant conf identiality. 15 The format for the elite interviews was semi structured, and questions within the interviews were posed in an open ended manner so as to allow respondents the opportunity to share the full extent of their knowledge and expertise (Bernard 200 6, 212). 16 While the research design deviates from the standard execution of a (crisp set) QCA, it is important to note that the comparative analysis that this dissertation utilizes adopts many of the same assumptions of causation help by adherents of QCA, including conjunctural causation, equifinality, and multiple equilibria (Ragin 1987).
45 Plan of the Dissertation Through the following six chapters, this dissertation will explain how Ghana succeeded in following a democratic trajectory. Building on the Ghana example, it will additionally explain what a theory of regime trajectories provides for the study of democra cy and democratization in Africa. Following the introduction, Chapter 2 provides a survey of the scholarly literature on democratization in Africa. M ore importantly, the chapter Furthermore, situating the case of Ghana as an outlying case of democracy in Africa, the chapter situates the Ghanaian example within the wider body of literature concerning demo cratization in Africa. Lastly, through acknowledging the contributions of previous research in this area, including works by VillalÃ³n and Huxtab le (1998) , VillalÃ³n and Von Doepp (2005), as well as Whitfield and Mustafa (2009), this chapter shows how this d issertation fits within the ongoing conversation regarding contemporary democratization in Africa and how the dissertation serves to contribute to this conversation by investigating an unexplained area of the discipline . Chapters 3, 4, and 5 provide the e trajectory by analyzing in depth regime trajectory in a democratic direction. Chapter 3 highli ghts the international factors namely, international aid an d leverage that served to constrain the national along the lines of what was done in the era. In short, through the combination of the state receiving a significant amount of foreign aid from Western donors, as well as the state being in a particularly vulnerable bargaining position against the conditions attached to such aid
46 from Western donors, the Ghanaian government was incentivized to tolerate continued democratization in exchange for continued access to such funds. By explaining the economic status of the country at the time of its democratic opening, as well as fragile nature of its economy, Chapter 3 shows how the international dimension contributed to facilitating a long term regim e trajectory favoring democracy. Following the explanation of the international factors, Chapter 4 continues by highlighting the domestic factors that encouraged the state to follow a democratic regime trajectory. In brief, Chapter 4 explains the role of t he domestic political opposition in the country and highlights how its continued presence as being a credible, competitive alternative to the National Democratic Congress (NDC) (even when formally boycotting the inaugural parliamentary elections) served to incentivize the government of President Jerry J. Rawlings to tolerate continued democratic demands. In other words, although the domestic opposition had frag mented from its peak as the Movement for Freedom and Justice (MFJ) , the pro democracy movement enc democratic opening, the domestic political opposition primarily manifested through the New Patriotic Party (NPP) nevertheless maintained the ability to frustrate government policie s through its ability to encourage protests, strikes, and other measures designed to impede the ability of government to execute its policies in a seamless manner. 17 More importantly, the political opposition possessed the ability to delegitimize the NDC go vernment. A s the NDC government continued to rely upon international assistance to 17 6). In addition, civil society is distinct from other alternative forms, including political and economic society (Linz and Stepan 1996).
47 maintaining legitimacy remained a core interest for the NDC during the inaugural adminis tration of President Rawlings. In effect, though operating outside of government, the political opposition retained a functional veto over government that served to discourage the outright repression of democratic demands from the opposition. While Chapter s 3 and 4 survey the structural conditions that influenced the any account of the process would be incomplete without reference to the actions taken by the executive in response to the structural conditions present at t he time of the democratic opening . In short, while international pressure and domestic political opposition served to incentivize the country to follow a democratic trajectory, it is important to remember that the environment at the time of the democratic opening nevertheless remained highly contingent and open to a wide range structural incentives favoring continued democratization, President Rawlings and his NDC led gove rnment ultimately retained the ability to execute polic ies that could further continue democratization, or repress it altogether. The decision to either allow or prohibit further democratization, in essence, depended on whether Rawlings and his government that Rawlings wanted to stay in power for as long as possible, his administration faced a two level decision regarding the question of de mocratization and the nature of his executive authority, with the outcome of the latter influencing the decision of the former. In short, Rawlings faced the question concerning the scope and durability of his
48 authority: in an environment where his administ ration could no longer take for granted the benefits of discretionary authority and indefinite rule, Rawlings and his administration needed to decide whether they would seek to reestablish the kind of authority reminiscent of the authoritarian era, or whet her they would instead work within the confines of the nascent democratic system . In rightly sensing that international and domestic pressure would likely prevent his administration from consolidating power in an authoritarian manner and would furthermore raise the significant possibility of his ouster were he to attempt such an act Chapter 5 argues that Rawlings and his administration were aware of such constraints and instead pursued an agenda that ultimately served to prolong his duration of rule while w orking within the confines of a democratic environment . In other words, by reading the proverbial writing on the wall, Rawlings avoided outright repression of democratic demands and tolerated continued reforms in return for an increased duration of rule. H owever, while these actions were initially conceived to serve the immediate interests of the NDC, they ultimately sowed the seeds of its future electoral defeat by allowing the unabated growth of an electoral environment where the political opposition coul d freely compete and eventually win the 2000 general election. Through the survey of the Ghanaian example, including the manner in which continued democratic demands, Ch apter 6 leverages the results from this in depth case analysis by evaluating them with in a comparative analysis of all other countries experiencing democratic openings in the same period as Ghana. Comparing the experience of Ghana against 27 other countrie s, Chapter 6 demonstrates
49 deviance as a democracy emerging from unlikely origins is not deviant at all, but rather is a paradigmatic example of something else. By identifying Ghana as an example of a democratic trajectory, the chapter aids in resolving the much larger question as to why some countries maintained democratic systems of rule, while others did not. A theory of regime trajectories that investigates regime change over time, and does so in a manner that combines elements of agency and structure, provides both academics and policymakers an explanation concerning the contemporary status of democracy in Africa while also providing a tool for future investigations of this issue in Africa and abroad. Through each of the preceding chapters h ighlighting the Ghana case study and its relevance towards a broader understanding of democratization in Africa , the concluding chapter summarizes the insights ascertained from this study, and considers the implications with regard to the study and practic e of democracy in Africa . While providing the chapter additionally provides a competition, international leverage, and foreign ai d dependence may serve to account for the pattern of regime change in Africa for the past two decades. Moreover, speaking to both the scholarly community and practitioners of African politics, the conclusion elaborates upon both the theoretical and policy relevant contributions ascertained from this exercise. By identifying the long term trajectories of regimes in Africa, as well as theorizing about their causes and mechanisms of reproduction, the results of such an analysis not only assist in responding to the question framing this project, but additionally assist towards understanding which states are likely to maintain their
50 current regime status, which states are likely to change, and what factors, if any, may account for why this is the case. Conclusion experiments produced significantly divergent legacies more than twenty years after their initial occurrence, this project puts forth an agenda that not only aims to identify and describ e the different types of trajectories produced from this period, but additionally explains both why and how each type of trajectory came to be associated with each country within this study. In addition to the anticipated practical and theoretical contribu tions, this project utilizes an original research design combining elements from nested analysis and QCA. In short, as there exists no other type of work investigating the post Cold War legacies of democratization in Africa within a comparative, medium N f ramework, this project provides further clarity within an under explored area in the discipline, and does so while providing practical insights for the field of African politics.
51 CHAPTER 2 T hi s chapter examines the literature on democratization as it applies to post Cold War Africa. Beginning with a summary of early research on democratization, the discussion shows how such research was adapted to the African context through Bratton and van de how subsequent research responded to findings. Following this discussion, the chapter next surveys the status of democracy research as it concerns Africa, but more importantly, it identifies gaps in the extant literature that have yet to be explained while using the case of Ghana and its particular path to democracy to illustrate the point . In sum , by showing where past scholarly advancements have taken us, we may thus come to k now the areas remaining to be explored . Democratic Transitions In general, when speaking about democratization scholars have tended to adopt a perspective emphasizing political change as occurring through democratic transitions. It is important to establis h at the outset , however, that in both highlighting and explaining regime trajectories in Africa following the initial wave of democratic openings in the early 1990s, this dissertation inherently critiques the transitions literature . Yet , before elaboratin g on the limits of how such a conceptualization of democracy affects scholars , it is first necessary to explain the origins of this paradigm, as well as the contributions it has brought to the study of democratization and regime change. The origins of the transition paradigm are often 1986 work, Transitions From Authoritarian Rule ( Collier 1999; Carothers
52 2002 ; Lindberg 2006 ) . Explaining the processes of change that were taking place largely in Lati explanation of democratic transition s that centered on the roles of context and agency . In particular, the authors advanced an explanation centered on the agency of elite groupings acting w ithin a given political system. While careful to not discount the existence of macrostructural factors, the author the high confusion about motiv es and interests, plasticity, and even indefinition of political determinin g the outcomes [of transitions] . Furthermore, they argued that transitions occurred through the interaction of five groups representing authoritarian (hard liners and soft liners) , democratic (maximalists and minimalists) , and military interests. In a type of two level game the authors posited that outcomes of transitions were a function of the strategic competition between democratic and authoritarian factions, as well as a function of the competition within such factions, with the military serving as a latent third force . In sum, t he outcome of this process depended on the manner in whic h each faction not only succeeded in facilitating a kind of negotiated agreement with their competitors, but also succeeded in facilitating agreement within their own faction. the impact of founding elections as the period when a democratic transition ended and when a democracy began to emerge (1986, 61). Though the authors carefully acknowledge d e
53 importance o f founding elections suggest ed that elections were nevertheless a constitutive feature towards meeting their definition of democracy . To be sure, while any definition of democracy inherently includes elections as a necessary condition, the point here is th transition from authoritarianism to democracy. This conceptualization of democratization as occurring through the strategic interaction between pro and anti democratic forces influenced subsequent investigations of regime change. Most notably, (1991) work may be text , with Przeworski additionally highlighting the strategic interaction among domestic p olitical actors during periods of transition . Unlike O'Donnell and Schmitter, however, Przeworski incorporated game theory to explain and further specify actor interests and probable outcomes. In particular , Przeworski emphasized the concept of a competing political groups as significant in determining the long term prospects for democracy. In short, Przeworski hypothesized that should the relation of forces among democratic and authoritarian factions remain unknown to everyone at the time of transition, this situation would produce the most propitious conditions for democracy, as incentives would inherently exist to construct post transition institutions according to systems of checks and balances . This, in turn, would serve to inh ibit any one group from obtaining and/or exercising disproportionate power in the newly democratic system (87 88). While careful to avoid constructing a caricature of the transitions literature , these texts nevertheless provide the basis for analyzing demo cratization through the
54 transitions paradigm , as well as serve as reference points for its popularization . However, r esearch concerning the political changes observed in Africa during the early 1990s struggled to conceptualize such phenomena within this fr amework . The challenges in explaining such changes within an African context ultimately led to a new way of conceptualizing democratic transitions on the continent, with Bratton and van de text serving as an important demonstration of such thinking . Transitions in Africa The first half of the 1990s was a remarkable period for democracy in Africa; increases in political turbulence and domestic unrest resulted in a wave of political reform on the continent. The question as to why these changes occurre d, why such democratic reform was so widespread, as well as why some countries underwent democratic reform while others did not prompted many scholars, including Bratton and van de Walle , to investigate this emergent political phenomenon. B ratton and van d (1997) work stands out as one of the most often referenced by scholars interested in democratization in Africa . Unlike prior models of democratization emphasizing pacted transitions among elite pro and anti democracy factions er 1986), Bratton and van de Walle posited a different explanation for the changes witnessed in the African context in the first half of the 1990s . First , the authors highlighted the impetus for such reform s as largely originating from the bottom up throug h mass protest movements ( 269); second, rather than account ing for such changes as derived exclusively from either structure or agency, the authors propose d a structured contingency approach ( 45), in which historical/institutional legacies structure d prese nt conditions while simultaneously
55 al lowing the role of agency to influence the future course of political development ( very much analogous to historical institutionalism ). 1 In addition , the authors discount ed the explanatory significance of economic and i nternational causes of regime change in favor of an approach that instead emphasized the significance of domestic politics and the manner in which institutions shaped such interactions and activities . Though the authors were careful to recognize internatio nal and economic explanations as contributory , the authors nevertheless ascribed their role essentially secondary or supporting role in relation to the ex planation of regime ) . Unlike democratic transitions elsewhere, Bratton and van de Walle argued, transitions in Africa remained distinct due to the resilient legacy of n eopatrimonialism (61). Because informal rule had been institutionalized within rational legal structures and had furthermore promoted institutional features of presid entialism, clientelism, and the utilization of state resources for non state uses , Bratton and van de Walle argued that it was inherently problematic to explain the processes of change occurring on the continent within the same framework that had character ized much of what had previously occurred in Europe and Latin America. In accounting for such differences, the authors highlighted six features that were distinct when accounting for democratic transitions under neopatrimonial rule: (1) that transitions in such contexts originated from domestic political protest; (2) that outcomes of transition depended on the ultimate fate of the incumbent dictator; (3) that rather than elite fracture occurring over ideological schisms (i.e. hardliners and softliners ), eli tes 1 On historical institutionalism, see Sanders (2008), as well as North (1990) and Pierson (2004 ).
56 instead divided over access to patronage; (4) that, consequently, elite pacting was unlikely in neopatrimonial contexts; (5) that transitions in neopatrimonial regimes centered on struggles over reestablishing formal rules, as well as the rule of law; and (6) that in transitions in neopatrimonial contexts, the middle classes often tended to align with opposition forces (83 9). For these reasons, the authors argued, a new approach was necessary in order to properly understand what was happening in the Af rican context, as well as explain why some countries experienced democratic transitions, while others did not. In a similar manner to the transitions paradigm , the authors found that democratic transitions in Africa occurred through a process of sequential steps ultimately leading to the holding of an inaugural election, that is, what the authors defined as the necessary condit ion of a democratic transition. However, the authors advanced an alternative explanation for this process , positing the impetus for these transitions as originating from bottom up protest movements predicated upon prior ec onomic grievances (Bratton and v an de Walle 1997, 98, 269). As many countries in Africa experienced economic decline during the course of the 1980s, the authors argue d , by the end of the decade such declines began to manifest themselves in the form of popular protests that many governments found themselves unable to constrain (101). As such protests turned political in many cases, some governments were prompted to assu age protestors through limited political reforms. Where such reforms were insufficient to diffuse protests , governments were then compelled to democratize by holding competitive inaugural elections . Bratton and van de Walle identify 11 instances in which t his process occurred. Figure 2 1 provides a reproduced
57 transition in Africa. Source: Reproduced From Bratton and Van de Walle 1997, 117. Figure 2 1: Paths of Transition s i n Africa, 1988 1994 To be sure, not every country followed this course: Bratton and van de Walle identify five cases that experienced transitions without prior political protests, while other regimes succeeded in obstructing, preempting, or even blocking t ransitions through a variety of institutional means, including influencing or otherwise subverting inaugural elections (24 cases) . Interestingly, in contrast to prior models of transition emphasizing 986; Przeworski 1991), the authors found that in cases where democratization was driven by elites, the results more often resulted in failure than success (Bratton and van de Walle 1997, 119). In sum, Bratton and van de Walle conclude with a qualified outl ook for democracy in Africa. The authors rightly note that democratic transitions do not end at Democratic Elections Before 1995? Liberalization Before 1993? Political Protests? (1988 1992) 42 Countries Yes (28) Yes (28) Yes (11) No (17) No (0) No (14) Yes (12) Yes (5) No (7) No (2) No (2)
58 founding elections, yet must be truly realized through consolidation , that is, being both institutionalized and legitimized by all with in the state ( 235 6). A dd ressing the prospects for the consolidation of democracy, Bratton and van de Walle identify the role of d omestic political institutions including the judiciary, l egislature, and civil service the military, as well as political and civil society as instrume ntal ( 242 55). However, rather than providing projections of the future, the authors conclude that democracy in Africa will ultimately depend on whether democracy becomes fully internalized among the political elite and citizenry , as well as whether democr acy succeeds in providing the benefits both political and economic that it has been purported to deliver ( 279). Beyond Democratic Experiments argument explaining the reasons for the dramatic political change s occur ring in Africa following the end of the Cold War, factors that have arguably altered the course of d emocracy on the continent. In general, subsequent investigations of democratic chang e in Africa have centered around three issues concerning the role of foreign aid and assistance, the democratizing influence of continuous elections, and the emergence of competitive authoritarianism. In addition, o ther studies have engaged in more explora tory research by analyzing the individual experiences of countries that attempted to democratize during the democratic wave . Foreign Aid First, while Bratton and van de Walle discounted the role of the international environment in influencing the course of democratization in Africa, subsequent research has challenged this assertion by providing additional insights into the mechanisms in which international factors, such as foreign aid, serve to influence the course of regime
59 change. For example, Wright (200 9), as well as Bueno de Mesquita and Smith (2009) , have addressed the manner in which foreign aid disbursements affect the likelihood of democrat ization through the size of an authoritaria n Wright, if a leader believes he or she will likely retain their executive position after submitting to democratic reforms based upon the breadth and robustness of their distributional coalition they will likely choose to reform in order to continue receiving the benefits provide d from in ternational aid (554 5). If, however, they do not believe they will remain in office after allowing for a democratic opening, then , Wright argues, foreign aid will be less likely to succeed in precipitating regime change. In contrast, Bueno de Mesquita and Smith (2009) similarly note the importance of elite interests, yet differ by positing that smaller coalitions are more likely to produce the intended policy objectives for foreign aid donors , including increased democratic reform (315, 331). In spite of t he differences between these works, they nevertheless highlight the importance of the relationship between donors and recipient s , as well as the need to consider recipients interes ts with respect to foreign aid, including ability to enc ourage democratic reform depends on factors other than simply the amount of a given disbursement. In addition , Dunning (2004) emphasizes the significance of geop olitics with respect to foreign aid and its impact in Sub Saharan Africa. Building on prior res earch from Goldsmith (2001) concerning a small but positive relationship between foreign aid and democracy in Africa bet ween 1975 and 1997 , Dunning argued that this effect was instead temporally d ependent. That is, the effects of foreign aid were influence d by the manner in which superpower rivalry between the U.S. and Soviet Union was widely
60 known to include the distribution of foreign assistance to African states on the basis of establishing either pro Western or pro Soviet alignment . In s plitting Goldsmi dataset into Cold War and post Cold War periods, Dunning demonstrated that the positive (and significant) correlation between foreign aid and democracy was limit ed to the post Cold War period ( 418). First, prior to the collapse of the Soviet Union, there was significant reason to believe that allocations of foreign aid were likely driven by geostrategic interests, as the United States and the Soviet Union were both well known to have engaged in competitive alliance building in Africa . Secondly also explained that the reasons by which former Soviet client states shifted towards democracy after the Soviet collapse may be largely explained through changes in U.S. aid distribution, including elements of conditiona lity and democr atic reform ( 419). In the nature of the international environment, in an absence of great power rivalry, foreign aid may , indeed , possess the potential for influencing d emocratic change . Elections In addition to international factors that were previously unaccounted for in past explanations of democratic change in Africa, subsequent explanations have emphasized the role of elections as constituting a causal step on the wa y towards democracy . Lindberg ( 2006; 2009) is an important scholar and his work on democracy in Africa exemplifies this view . Whereas Bratton and v an de Walle (1997) highlight the significance of elections as constituting a necessary condition for a democr atic transition, Lindberg provides an intriguing extension to this claim by positing that elections in Africa may instead serve as a causal factor through the successive impact of multiple electoral cycles (2006, 3). Positing that elections socialize the p opulation to
61 democracy and democratic norms, and additionally deepen such norms through their repetition (111 argument not only challenged the then prevalent view of political liberalization preceding democratization (145; see also Bratton a nd v an de Walle 1997, 27 7), but additionally contributed by conceptualizing elections as a performative step towards a democratic ideal type ( see Dahl 1971). In other words, Lindberg de monstrated that it may be more fruitful to regard them as processes that over time reinforcing, self improving quality of Competitive Authoritarianism in Africa R ecently, attenti previous democratic transitions have stalled in the ambiguous zone between democracy stands out in this regard though their investig the end of the Cold War. T hough designed to provide a general explanation for this phenomenon , Levitsky and Way nevertheless offer insights amenable within an African context that accoun t for levels of democracy in Africa, including reasons why certain countries digressed into non democratic outcomes following their inaugural transitions. In short, though the authors agree with Bratton and van de Walle that the end of the Cold War produce d a global shockwave resulting in a wave of democratization around the world , they differ by arguing that in Africa (and elsewhere) the net result was not an overall expansion of democracy, but rather the emergence of a new form of authoritarian rule. P osi ting the relative lack of i nternational linkage between most African countries and the West (i.e. the U.S. and the E.U.) , the authors argue that in
62 response to the wave of democratization that swept across the continent in the aftermath of the Cold War, Af rican incumbents realized that an international emphasis on electio ns as a benchmark for democracy enabled them to manipulate the legal rules to their advantage while continuing to receive international assistance and foreign aid , thus producing political environments in which multiparty competition was real, but nevertheless remained unfair ( 2010, 5). 2 Other Explanations To be sure, not all studies of democratic change in Africa have utilized the three previous approaches. However, other notable studies ha ve nevertheless used Bratton as a baseline for research on the subject. Eschewing general explanations in exchange for more nuanced view s of democratization in Africa, these studies utilize individual case studies or two case comparisons to better understand why some countries remained democratic since their transitions in the early 1990s, and why others did not. For example, w experiments ten years after their first occurrence, Vill edited volume investigates what has transpired since many African regimes first cases of democratic transition experienced differing regime o utcomes a decade after their inaugural elections, and questioned what factors, processes, or events contributed sense of the varied experiences of these countries that, at least initially, took parallel 2 organizational) cross border flows (of capital, goods, services, people, and information) between particular countries and the United States and the [Euro
63 the influence of institutions and elites on democratic processes, the edited volume comprises ten case studies that investigate how each country fared in the ten years following their initial transition, and explain how each case either succeeded or failed in maintaining a democratic regime. Mustapha and Whitfield (2009) expand on the issue concerning the legacies of eriments through an eleven case edited volume, including cases selected for purposes of representing varying colonial legacies, geography, and outcomes from their initial d emocratic openings. Furthermore, each contributing chapter significantly altered the regime trajectory of a given state. From the chapter case studies, plus a concluding chapter engaging in paired comparisons of selected cases, the volume concludes by emphasizing the roles that presidentialism, clientelism, state society relations, and elections have had in influencing each countr regime status. Limits of Current Understandings on the Subject While these works have provided noteworthy contributions towards a greater understanding of democracy in Africa, there n evertheless remain under explored area s . Since Bratton and v an de Walle tigating reasons for the wave of democracy that swept across Africa after the Cold War (as well as why the wave produced ostensible transitions in some countries but not others) , there have been multiple responses concerning the issue of what has transpire d since many of these states first attempted democratic systems of rule. While the dialogue produced from
64 these investigations has been contributory for explaining the particular political histories of those countries that undertook democratic reform , it r emains to be explained why these countries experienced such divergent outcomes despite experiencing what may be reasonably considered similar starting points two decades earlier . In other words, while we know the particular reasons why individual countries maintained democratic regimes and why others did not, general explanations for this divergence remain elusive. In addition, explanations of democratization (both in Africa and elsewhere ) tend to emphasize factors based at different levels of analysis, tha t is, that they identify factors from either the international or domestic environment as the primary determinants. While any explanation of reality must invariably engage in some amount exclude one level over another, such analyses risk missing a complete account of regime change. As such, i t other, and if so, whether such comparisons may yield additional insights towards explaining why some states in Africa remained de mocratic, while others did not. Furthermore , explanations of democracy in Africa tend to treat democratization through a lens of transition, that is, as democratization occu rring at a single point in time rather than through a long term process of incremental change. As a result , few works have since taken on the challenge to explain democracy in Africa through an alternative perspective , particularly one that analyzes regime change over time. Lastly, phenomenon of competitive authoritarianism has provided a noteworthy explanation for
65 the limited success of contemporary democracy in Africa, it should be noted that while their model largely explains authoritarianism, it unfortunately fails to explain the cases where democratization has, in fact, occurred . Instead, instances of democracy are largely explained through case specific , non comparable factors, including the personal preferences of national executives. As a result, of competitive authoritarianism provides and explanation to why democracy is the exception rather than the rule among their Africa n cases . In sum, in spite of extant research on the subject of de mocracy in Africa, there still exists a gap in explaining the long term changes in regimes between democracy and authoritarianism , as well as accounting for such changes in a comparative context . An explanation that emphasizes regime change over time, and does so while providing generalizable insights provides a greater understanding of which factors assist the long term prospects for democracy, and additionally explains why some countries succeed in maintaining democratic regimes after their initial openin gs while others do not. Gh Curious Case of Democracy As previously highlighted in Chapter 1, Ghana provides a puzzling case for analysis due to its distinctive route to a contemporary liberal democracy. As an unlikely ctiveness as the only contemporary African democracy to emerge from a problematic democratic opening provides an opportunity to engage in a model building exercise . The goal is to better ascertain the reasons why some countries, despite experiencing auspic ious initial conditions at the time of their democratic openings, failed to maintain democratic systems of rule, while others with inauspicious initial conditions nevertheless succeeded in spite of such obstacles. In other words, as the case of Ghana remai ns a deviant case of democracy in Africa , its
66 deviance provides additional information on a subject that has since remained unexplained. Brief Historical Background Since independence, Ghana has had about equal experience operating under both authoritaria nism and democracy. Prior to the 1992 elections, however, military rule had been the most common form of government. And yet, in spite of these origins Ghana currently boasts one of the most highly regarded democracies on the African continent. To apprecia status requires first explaining the journey the country has taken since its independence. In 1957, Ghana became the first Black African colony to gain independence. At the time, Ghana boasted on e of the best economies on the continent, as well as a additionally served as a leading ideological voice for Pan Africanism and decolonization. However, despite such auspici ous beginnings, politics in Ghana took a non democratic Preventive Detention Act that granted wide latitude to the state security forces to imprison individuals deemed a threat to state interests for up to five years without trial . 3 of the First Republic passed in a referendum widely considered to have been rigged effectively allowed Nk rumah to rule by presidential decree (Amenumey 2011, 228). By 3 In practice, this law was widely perceived as having been largely applied against the political opposition 212; Amenumey 2011, 228).
67 instituted one party rule through another rigged referendum. for the popularly well received coup by the National Liberation Council in 1966, thus inaugurating the period of military rule in Ghana. Although the armed forces twice returned power to civilian leadership in 1969 and 1979, none of the civilian administr ations survived a full term in mismanagement, corruption, increased foreign debt, as well as declines in productivity and foreign exchange. By the time of Flight Lieutenant Jerry Rawlings second coup in were over 116 per cent, exports had fallen to 31 per cent of their 1966 levels ( Oquaye 2004, 99; World Bank 2014), its local currency (the cedi) was overvalued, and prices for basic commodities, including foo d , were significantly elevated above what had been considered reasonable to the average Ghanaian . with enacting such reforms a Limann was in political gridlock, thus complicating matters further. It was this environment that surrounded the return of Rawlings to power through Rawlings suspended the Constitution, dissolved the national legislative and executive institutions, and inaugurated a new military government, the Provisional National Defense Council (PNDC), with himself serving as the Chairman (Oquaye 2004, 97). During the period of PNDC rule, was effectively instituted (largely enforced through arbitrary arrests and detentions) in
68 order to suppress political dissent and opposition to military rule. Rawlings and the PNDC moved to further consolidate their rule through the institutionalization of local administrative units throughout the country (initially called Defense Committees, though later renamed Committees for the Defense of the Revolution). These committees, though designed to incr ease the ability of Ghanaians to decide local matters, additionally served to increase the national penetration of the PNDC government through a type of top down, paternalistic relationship (Crook 1999; Crawford 2009). political opposition continued to exist, largely through various private clubs and professional organizations, such as the Ghana Bar Association (Nugent 1995). However, many credit the formal awakening of the opposition to have begun in 1988, after Profes sor Adu Boahen delivered a series of lectures that had openly addressed the issue of the culture of silence pervading the country, and had furthermore openly criticized the government for the 1981 coup and its undemocratic administration (Boahen 1992; Oqua ye 2004, 488; Elischer 2013, 143). Opposition continued to grow as the PNDC made some unpopular decisions in 1989, including the passage of the Religious Bodies Registration Law and the Newspaper Licensing Law, which largely alienated the religious communi ty (including the influential Christian Council of Ghana) and the media from the PNDC. In addition, the National Union of Ghanaian Students (NUGS) and the ranks of the Trades Union Congress (TUC) began to voice formal opposition to the PNDC and its policie s. In 1990, the different sectors of opposition coalesced and formed the Movement for Freedom and Justice (MFJ), an ostensibly non
69 ideological spectrum and was united by the desire for a return to constituti onal rule and multi party democracy. Such a challenge may have been controllable for the PNDC, however the rise of political opposition coincided at a time when international politics had experienced a tectonic shift. First, the fall of the Berlin Wall in 1989 , and the increasingly apparent demise of the Soviet Union as the only alternative non Western superpower , informed the PNDC as well as many other national governments on the continent that their ability to seek Soviet financial and/or military assista nce was no longer possible . The successful expansion of democratic regimes in Eastern Europe additionally bolstered domestic constituencies advocating for reform. Secondly, with the decline of superpower rivalry on the continent, so too did overall foreign assistance. No longer would countries like the U.S. provide aid on a simple basis of alignment with the West; instead, aid became increasingly narrowed. As Western donors obtained increased leverage as being the primary providers of aid, it increasingly a rrived with democratic conditions attached (Schraeder 1994; Bratton and van de Walle 1997, 135; Dietrich and onditional on political reform (World Bank 1989). 4 In sum, the end of the Cold War decreased the geostrategic importance of Africa and largely left Western donors in a privileged position against their African counterparts. Governments in need of continued access to 4 Although th analogous to basic democratic principles. In addition, the report w as popularly regarded as a proxy for democratic reforms (Boafo Arthur 1998, 170 1; Oquaye 2004, 499).
70 assistance including the PNDC would find it difficult to continue to obtain assistance without making any democratic concessions (Oquaye 1995, 261). Facing a formidable reform movement at home and an international environment no longer amenable to assisting wholly non democratic regimes, the PNDC faced a crisis of the status quo. Sensing authoritarian rule to be increasingly untenable, Rawlings and the PNDC led government initiated a process of democratization. In 1991, after the National Commiss ion for Democracy (NCD) through a series of regional seminars held the previous year had submitted a report showing wide public interest in returning to multiparty politics, Rawlings announced that he and the PNDC would begin a process of constructing a ne w constitution, ultimately returning the country to a multiparty system of rule (Nugent 1995, 199 200). 5 Though essentially capitulating to democratic demands, the PNDC nevertheless controlled the pace and extent of reforms leading up to the 1992 elections (Jonah 1991; Bratton and van de Walle 1997, 170; Riedl 2014; Bates 1994; Levitsky and Way 2010, 301; Lyons 1999, 161; Ninsin 1998, 214; Whitfield 2009 b , 50; Nugent 2004, 418 ). In a largely top down process, the PNDC composed a Committee of Experts (CoE) c harged with drafting the new constitution, and additionally controlled the composition of a Consultative Assembly, a body representing various sectors of society that was charged with vetting (and altering, if the national referendum. By the time the 5 as the regional seminars were initially intended to assess the performance of the newly created District Assemblies, not a general discussion on the return to democracy (Nugent 1995, 198). It was further remarkable that the NCD even allowed discussion of a return to multiparty politics to be in its fina l report to the PNDC government, particularly as Rawlings had openly expressed his disdain for multiparty politics being one in which the public was not against, rather than in favor of a return to multiparty politics (NCD 1991, 36; Jeffries and Thomas 1993, 335 6; Nugent 1995, 199; Dadzie and Ahwoi 2010, 76 7).
71 draft constitution was sent for a national referendum in May, 1992, it was generally accepted that the Constitution had been the product of a controlled process favoring the incumbent government, as well as Chairman Rawlings (Jonah 1991; Nugent 1995, 216 7; Oquaye 2 004, 490 1; Riedl 2014, 134 7). 6 In spite of its misgivings on the process and outcome of the constitution building exercise , the New Patriotic Party (NPP), an opposition party formed f ollowing the legaliz ation of political parties shortly before the referendum, accepted the draft constitution and advocated its members to vote in favor as well (Nugent 1995, 220). With the Constitution overwhelmingly approved by voters, the general elections remained the fin al act before the inauguration of the Fourth Republic. Although the MFJ had dispersed itself into various competing political parties, two parties emerged as the main contenders: The National Democratic Congress (NDC), a largely renamed version of the PNDC , with Rawlings as its presidential candidate; and the NPP, with Adu Boahen as its presidential aspirant. Presidential and parliamentary elections were scheduled separately, with the former preceding the latter. 7 The results from the November 3 presidentia l election provided Rawlings and the NDC a resounding victory 6 Although this general view of PNDC dominance of the constitution building process is not universal (Jeffries and Thomas 1993), a rather notable example of PNDC influence over the process included the the Constitution, which essentiall y absolved PNDC officials from any offenses committed prior to the Fourth Republic. It should also be noted that the PNDC stipulated that the Constitution would not take effect until January 1993, that is, until after the inaugural elections had taken plac e (Oquaye 1995, 274; Riedl 2014, 136 7). 7 Presidential elections were scheduled for November 1992, with parliamentary elections occurring the following month.
72 over all other competing parties, capturing over 58 per cent of votes, and obviating the need for a run off election. 8 In the aftermath of the inaugural presidential election, most of the politi cal opposition in the country had become incensed at the outcome, alleging that the election had been rigged in favor of the NDC (NPP 1993; Nugent 1995, 234). As a result of perceived fraud in the presidential election, as well as having a lack of faith in l Commission (Oquaye 2004, 529 30), most of the opposition boycotted the parliamentary elections in December 1992, thus providing the NDC near total control of government (ECG 2014). Thus, the Ghanaian Fourth Republi c began with a very problematic start: the absence of political opposition from government made the country a de facto one party state, and the powers granted to the President including the appointment of a majority of the cabinet ministers from Parliament exclusive ability to introduce legislation imposing any costs to the Ghanaian state allowed for the executive to dominate Parliament ( Whitfield 2009 b , 56 7 ; Gyimah Boadi  2010, 146 ). 9 As a result, initi al forecasts of Ghanaian democracy were not promising, and were tentative at best (Nugent 1995; Oquaye 1995; Bratton and van de Wal le 1997, 120; Green 1998, 205 6; Gyimah Boadi 1999, 171 ). 10 Lyons perhaps best 8 The rules for the 1992 elections, as well as current elections in Ghana, stipulate that a pres idential candidate must obtain more than 50 per cent of votes before being declared the winner; otherwise, the top two candidates are placed in a subsequent run off election to determine the winner. 9 To be sure, Ninsin has also argued that the Ghanaian pa rliament was perhaps more accountable under its inaugural period (1993 1997), as NDC MPs were more likely to challenge the president on matters of appointments and legislation (2008, 71). 10 y free in 1993 (Freedom House 2014a).
73 captured the sentiment when noting that the res ults of the 1992 election and its improve in a steadfast process, with no setbacks or lapses. How is it that this country, despite an inauspicious start to its Fourth Republic, succeeded in developing a democratic system of rule, ultimately becoming widely regarded as one of the few liberal democraci es on the continent? influence of one of the following factors (or the combination thereof): civil society, the Interparty Advisory Committee (IPAC), international aid, and Lindbe regarding the continuation of uninterrupted elections. First, civil society has often been invoked as furthering Ghanaian democracy for its role in advocating for the practical implementation of democratic ideals, as well as for its influence in the policymaking process (Gyimah Boadi 1999; Haynes 2003; Gyekye Jandoh 2006; Abdulai and Crawford 2010; Arthur 2010). Not only has civil society been previously credited with contributing to the count civil society organizations have been active in promoting civic education, advocating for continued democratic reform, el ection monitoring , as well as sponsoring presidential and parliamentary debates (Asante 2013; Gyimah Boadi  2010, 144; Gyi mah Boa di and Yakah 2013, 274 8). 11 It is in this vein that civil society has been credited with 11 Civil society organizations in Ghana are additionally well known to assist in election monitoring.
74 pushing democracy in Ghana forward since the democratic opening in 1992 (Gyekye Jandoh 2006). 12 Secondly, t he establishment of the Interparty Advisory Committee (IPAC) has been another source credited with having (Asante 2013). This argument generally rests on the notion that the founding of IPAC provided an avenue for all of the political parties to come together and provide rec ommendations for reforming the electoral rules in a transparent and consensus based manner. 13 More importantly, IPAC is credited with establishing elite consensus among the main parties, largely defined as the general agreement on participating in an electorally democratic environment, as well as the commitment by all parties to uphold and enforce the laws maintaining such a system (Frempong 2007, 141; Whitfield 2009b, 54). Above all, the creation of IPAC established a network of trust among the po trajectory (Ninsin 1998, 215; Agyeman Duah 2008, 167). acknowledging the influence of the international communit democratic opening occur in a period of increased donor conditionality, but subsequent reforms occurred with the help and funding of donors and their respective agencies. Assistance measures have included projects funding local deve lopment, civil society organizations, parliamentary training and empowerment, as well as aiding in the 12 Civil society has also been credited with aiding the political transitions in the 2000 and 2008 general elections ( Gyimah Boadi 2001;  2010, 144). 13 This argument coincides with others favoring the reform of the Interim National Electoral Commission into an independent Electoral Commission (EC), as IPAC augmented the ability of the EC to enact reforms (Gyekye Jan doh 2006).
75 execution of general elections (Gyimah Boadi and Yakah 2013). Since the early 1980s, Ghana has enjoyed a position as constituting a success story for int ernational donors. This relationship by means of Western donors having effectively conditioned their aid on continued democratic reform not only influenced the decision of the PNDC to democratize in 1992, it has furthermore continued to influence subsequen t administrations (Whitfield 2009b; Whitfield and Jones 2009). 14 Lastly, the hypothesized nature of elections serving as a democratizing influence in Ghana (as well as elsewhere on the continent) has generated a productive debate on the potential causal eff democratic standing (Lindberg 2006; 2009). As an inherent critique of the transition inaugural elections do not ne cessarily indicate the occurrence of a democratic transition but are more often just a step in the transition process As a rather paradigmatic example of this view, despite experiencing such a problematic start to its Fourth Republic, Ghana continued to democratically improve by continuing to conduct regularized and uninterrupted elections. Through this process, it is argued that citizens became increasingly familiar ized with and socialized to democratic norms, including free and fair competition, broad based participation, and legitimacy in the governing body as previously noted facto rs, elections in Ghana served to propel the country along its democratic trajectory. 14 This particular issue will be discussed in more detail in Chapter 3.
76 While these contributions have certainly aided in explaining how and why beginnings, the re are limits to these explanations. For example, it is well known that civil society had significantly fragmented in the post opening period and did not return as a significant factor until the 1996 elections (Oquaye 1995, 261; 2004, 500 1; Gyimah Boadi 1 999, 173). As previously noted, following the return to multiparty politics the MFJ had fragmented with various factions gravitating to the various political parties formed prior to the 1992 elections (Agyeman Duah 2008, 163; Frempong 2012, 64). 15 Following those elections, and the return of Rawlings to power, the general environment remained rather unfriendly to non state sanctioned civil society groups: there remained tense political sentiment following the boycott of the political opposition from parliame ntary elections, the state still maintained what were considered strict rules over the media, and much of the civil society organizations that did operate during this period only did so out of close relations with Rawlings and the NDC (Debrah 2005, 139). Furthermore, prior to the founding of the Institute of Economic Affairs in 1996, no other independent, pro democracy civil society organization had been in operation since the inauguration of the Fourth Republic (Agyeman Duah 2012, 147). As the reliability of civil society has been inconsistent, it is problematic to argue that this sector operated as a constant force propelling democracy forward in a continuous fashion. In other words, it presupposes the question as to what factor in the absence of civil so ciety pushed the government to continue to democratize, thus enabling civil society to reassert itself. 15 Some of which were incorporated into the renamed government party, the NDC (Nugent 1995, 243).
77 The formation of IPAC in 1994 is similarly problematic. In short, the challenge to the efficacy of arguments favoring the establishment of IPAC concerns the issue of endogeneity, and the manner in which IPAC had been initially established. To begin, IPAC had not been created sui generis: it was the result of a political crisis that emerged from the aftermath of 1992 inaugural elections , including the oppo parliamentary elections . As the political environment had remained rather tense following the elections, and as opposition parties had formally expressed their lack of oral environment, the government faced what many have described as a crisis of legitimacy that not only threatened its ability to govern, but additionally threatened its ability to obtain future international aid (Boafo Arthur 1998; Ninsin 1998; Levitsky and Way 2010, 303). 16 In an effort to provide transparency to the electoral process, as well as to try to reincorporate the political opposition into the institutions of government (and satisfy concerns of domestic and international legitimacy in the proce ss), Rawlings and his administration enacted a series of electoral reforms. Among the most notable reforms was the to reform the Interim National Electoral Commission into an independent Election Commission (EC), as well as the govern on instituting IPAC upon the recommendation of donors (COG 1992). It is in this respect that arguments that attribute the establishment of elite consensus to IPAC neglect to incorporate the reasons why the committee was initially c reated . it represents the moment where the government recognized the limits to its governing 16 That the parliamentary elections only attained a 29 per cent turnout only a dded to the concerns of legitimacy (Jeffries and Thomas 1993, 363; Frempong 2007, 137).
78 ability, and it furthermore represents the moment where elite consensus may have opposition could electorally compete while using a common and agreed upon set of rules. In other words, the issue with arguments treating IPAC a s precipitating elite consensus in Ghana, and thus serving to propel the country in a democratic direction, mistakenly treat an effect (i.e. IPAC) as the cause of democracy . While arguments favoring the role of the international environment rightly note th e influence of donors over the democratization process in Ghana, they tend to miscalculate the influence of aid without referencing whether such aid was significant enough to actually encourage governments to democratically reform. To be sure, with few exc eptions, Western donors generally possessed a significant amount of leverage d Levitsky and Way 2010, 40) over African recipient countries . 17 However, the amount of lever age that donors could apply to recipient countries was a function of the amount of aid they were willing to provide. In other words, given a situation in which two countries were targeted by Western donors, where each country possessed the same political a nd economic conditions (including similar levels of GDP), where donors possessed a high amount of leverage over both countries, but where one country received aid equaling 20 per cent of its GDP while the other country received aid totaling approximately f ive per cent of its GDP, it stands to reason that the country 17 international system, including gross domestic product (GDP) , level of oil production, possession of nuclear weapons, whether the country retains any geostrategic or security interest among Western countries, as well as whether the state was a particular benefactor from any non Western major powe r (Levitsky and Way 2010, 372 3).
79 receiving more aid would be more likely to reform than its less funded counterpart. In addition, it is also reasonable to argue that the country receiving more aid would be less likely to misapp present a far greater cost than the country receiving less aid. With respect to Ghana , it is n ot just that donors possessed a high amount of leverage over the country, it was also that Gha na received a substantial amount of aid averaging more than 10 per cent of its GDP during its democratic opening . In short, international aid in Ghana possessed a higher probability of producing the desired policy changes following the 1992 elections not o nly because the country was vulnerable to international pressure ( as the country remained in need of financial assistance before, during, and after its democratic opening), but also because donors provided a sufficient amount aid to the country so that the government would have an incentive to continue with democratic reform. Lastly, while an argument favoring the causal effect s of unimpeded and repeated elections d emonstrate some salience with respect to explaining the Ghanaian case, such an argument inher ently takes for granted the events and factors producing not only the first round of inaugural elections, but additionally the subsequent elections thereafter. That is, even assuming that repeated elections worked to strengthen or deepen democracy, we are still left with need to explain why elections were even held. This concern is all the more important as such an explanation rests on a perspective emphasizing path dependence and long term change over time. In other words, much like the challenge to argume nts favoring IPAC and civil society, if explanations emphasizing the occurrence of multiple, regularized elections argue that elections
80 produced steadfast changes over time, such an argument inherently depends on a prior factor or condition to have produce d such a situation. In short , it is not just that a country like Ghana held multiple elections; more importantly, it is why Ghana held elections at all. In sum, previous research has aided in answering the question as to why Ghana became a democracy, yet s uch explanations are not without their limitations. These include concerns over endogeneity, as well as a general tendency to treat the effects of previously unaccounted factors as causes of democracy in Ghana. In addition, there still exists a gap in expl aining the long term changes in regime type, as well as accounting for such changes with in a comparative context . As much of this literature utilizes a perspective emphasizing democratization as an event, rather than a process, it is necessary to reevaluat e such arguments as new scholarship challenges the transition paradigm and advocates for a perspective that alternat ively explains political change , including regime change, as occurring over prolonged periods of time and through path dependent historical trajectories brought about by periods of significant change (Capoccia and Ziblatt 2010; Mustapha and Whitfield 2009a; Pierson 2000; 2004). To be sure, this is n ot to be misconstrued as a dismissing pr ior scholarship on this subject; rather, this dissertati on builds upon such research in an effort to aid in increasing our knowledge and understanding of this subject that synthesizes international and domestic explanations in a manner that can be used towards broader comparative inquiry. Conclusion This surve y on the state of research concerning democratization in Africa has intended to show how prior research has , indeed, contributed to the general study of
81 democratizatio n, and has additionally illuminated the particular experiences of countries that had prev iously undertaken democratic reform. In addition, this chapter has highlighted what may be regarded as the most prev alent theories of regime change as they apply in an African context. However, in spite of the abundance of information available on this sub ject, this chapter has shown that there remains a gap concerning the question as to why, after a period of significant democratic opening, countries previously experiencing such openings proceeded to follow divergent paths , with some resulting in contempor ary democratic regimes, while others followed authoritarian or otherwise mixed courses . While past research has investigated the reasons as to why the continent experienced such dramatic change in the first half of the 1990s, it remains to be explained wha t has happened since such changes occurred, at least within a comparative context. To be sure, while recent investigations have, indeed, identified broad patterns that account for the wide emergence of semi authoritarian regimes on the continent ( Diamond 2 009; Levitsky and Way 2010), research has yet to explain why some countries, despite predispositions to authoritarianism (including Ghana and Benin) nevertheless resulted in democratic outcomes. Fortunately the historical record provides an intriguing case through Ghana s democratic trajectory . How is it that this country experiencing a problematic democratic opening including a flawed inaugural election, a near one party government resulting from an opposition boycott of p arliamentary elections, and a domi nating executive nevertheless became a widely recognized liberal democracy? 18 Operating as a deviant 18 contemporary democracy, in general (Bratton and van de Walle 1997, 120; Diamond 2010, xxvi).
82 comparative perspective promises to provide additional insights to th e study of regime change in Africa that have yet to be explained . As subsequent chapters will show, three factors, including reliance on foreign aid, a general vulnerabil ity to Western external democratizing pressure, and a sustained and significant domestic political opposition . In short, the significant presence of all three conditions served to influence the decision of the executive, President Jerry John Rawlings, to t olerate ongoing democratic demands in his first administration, thus locking Ghana onto a democratic trajectory. repressed ongoing democratic demands elections , constitute the subject of the next three chapters of this dissertation, and will be explained in further detail.
83 CHAPTER 3 INTERNATIONAL AID AND LEVERAGE IN GHANA This chapter begins the portion of what will constitute the in depth investigation of would certainly not have occurred without the contribution of a robust pro democracy movement operating within the country recent political history would neverthele ss be remiss by not accounting for the influential role of the international environment and the manner in which foreign assistance guided the democratization process. Ghanaian experts in politics and government generally agree that international aid and p ressure influenced the country to proceed along a democratic course. 1 T his chapter interrogates these attitudes and shows how , in fact, international pressure constrained President inaugural administration through the combined effects of not only receiving substantial amounts of foreign aid, but additionally being in an environment where the state was in significant need of such aid with only Western donors willing and available to help, provided that the Government of Ghana (i.e. the PNDC) was wi lling to enact reform. At the same time, however, while such pressures incentivized the country towards democratic reform, it would be incorrect to argue that international aid caused democratization. R ather, in conjunction with domestic pressure (discusse d in subsequent chapters), international aid and the leverage that Western donors had with respect to the terms of their aid constrained the G overnment of Ghana so as to continue to tolerate ongoing demands for democratic reform . In the same way that diffe rent 1 Interview with lecturer at the University of Ghan a School of Law, Legon, Ghana (September 21, 2013); interview with professor at the University of Ghana Institute of Statistical, Social, and Economic Research, Legon, Ghana (October 14, 2013); interview with senior NDC parliamentarian, Accra, Ghana (Febru ary 20, 2014); interview with senior NPP official, Accra, Ghana (November 18, 2013); Interview with former CPP parliamentarian, Accra, Ghana (January 20, 2014).
84 governments across Africa implemented the minimum amount of reform required to continue to receive international However, in doing so, and ultimately avoiding outright repression of democratic demands f rom the political opposition became locked into a self rein forcing trajectory of democracy that was buttressed and supported through key events, including the reform of the Interim National Electoral Commission (INEC) into the independent E lectoral Commission (EC) , the institutionalization of the Interparty Advisory Committee (IPAC), the reemergence and growth of political civil society, and the uninterrupted and si multaneous holding of national presidential and p arliamentary elections. The is generally well known, though the story of its full impact remains inco mplete. As recent research has begun to investigate the ways in which the effectiveness of international aid in promoting democratic reform was conditioned by the amount of leverage donors possessed over recipient countries (Levitsky and Way 2010), this chapter contributes by providing an updated account as to how international aid and pressure succeeded in not government to democratize, but more importantly, how both factors inaugural electe d government from relapsing into past authoritarian practices . International Aid in G hana In order to fully explain how Ghana became dependent on external assistance it is first necessary to explain why it came in need of such assistance in the first place. 2 2 istory during this period.
85 First, at the time of 1981 coup, the Ghanaian economy had been in longst anding decline. Corruption was generally seen as entrenched, inflation had risen to levels above 100 per cent, exports continued to be in decline, and prices for basic commodities continued to increase (Oquaye 2004, 99 ; see also Jebuni and Oduro 1998, 25 8 ) . As the civilian government of Dr. Hilla Limann was perceived as both politically ineffective (due to legislative gridlock in Parliament), and subservient to the International Monetary Fund (IMF) by enacting currency devaluations, the political atmospher e in Ghana became ripe for change by any means necessary , thus facilitating popular coup on December 31, 1981. As part of professed mandate for his coup, he and the PNDC pledged to reform the economy . Almost immediately, the PNDC began combatting corruption in the country, including kalabule , or the widespread practice of profiting at the expense of the state (Nugent 1995, 27). In the first years of his rule, Rawlings and the PNDC investigated suspected instances of widespread corruption and adopted price controls in the marketplace. In short, while state policies maintained affordable commodity prices for the urban population, the policies were generally harmful to producers and sellers who were compelled to accept prices below what woul d otherwise be considered reasonable. For instance, Nugent notes that cocoa farmers received only 35 per cent of what prices were in 1975 ( 1995, 81). Traders caught selling above state mandated prices faced a variety of consequences including, but not lim ited to , confiscation of goods and property , eviction from the market place , or even imprisonment . proved ineffective . Budgetary deficits and inflation remain ed unaddressed, and anti -
86 c orruption measures in the marketplace brought about increased smuggling and black markets for traders to find ways of making a sustainable livelihood. Further compounding the situation, the first months of 1983 witnessed severe droughts that contributed to rampant bushfires across the country. government to expel all foreign nationals, including appr oximately 1.2 million Ghanaians who returned home, created an economic crisis for the country . Food shortages became so commo n that it became part of the national lexicon to refer to the protruded collarb ones from starving citizens as (Nugent 1995, 108). As a result, in order to avoid outright collapse the PNDC was compelled to engage in discussions with the IM F and World Bank for financial assistance. The Economic Recovery Programme (ERP) It should be noted that engaging in negotiations with the IMF and World Bank was rather extraordinary for the PNDC, as the perceived threat of donor directed initiatives being imposed on the country (most notably, currency devaluation) served as ) . Discussions between the Government of Ghana and the IMF and World Bank produced an Economic R ecovery Programme (ERP) in 1983 . In exchange for donor support, the PNDC agreed to institute a number of economic reforms commonly associated with structural adjustment programs (SAPs), including currency devaluation, privatization of state owned companies, retrenchments in the state civil service, and removal of various food and service subsidies. Though such reforms were politically difficult for the PNDC, the government successfully navigated the terrain of economic reform by not only improving overall economic performance , b ut also avoiding the same type of civil unrest that brought about the
87 conditions facilitating Rawlings coup. in this regard to a with soci ety after the onset of adjustment, simultaneously building ties with its new interest in encouraging economic growth remained paramount, even if it meant resultant economi c hardships for some sectors of society . behavior was (including curtailing corruption) , beli eving that economic progress and growth would be sufficient in legitimizing their rule to the population, despite its authoritarian nature. Whitfield and Jones (2009) highlight the degree of influence that external actors possessed during this period towar ds the policy and technical work for the [ERP], and implementation of reforms relied heavily is sense, the World Bank and IMF regarded the PNDC government as weak, in terms of its administrative effectiveness . T hey thus became interested in supplanting the implementation of the Programme with their own staff in order to help ensure compliance with conditions. In doing so, the country thus became widely regarded as a darling for the donor community and as an example of a successful case of structural adjustment ( Herbst 1993; Gyekye Jandoh 2006; Whitfield and Jones 2009; Levitsky and Way 2 010, 301) . In sum , the ERP established a critical link between the country and the (Western) donor community that would continue for decades . I t thus became imperative that Rawlings
88 and the PNDC did not disrupt this linkage in order to continue receiving i nternational financial support. In addition to the adoption of the ERP by the PNDC, another important event includes the W were largely un successful in producing the kind of economic growth in Africa that had been initially projected, executives at the Bank redirected their attention to the manner in which such target countries, including Ghana, were being governed. Noting that a which was largely characterized as measures that served to reduce corruption, l imit the amount of discretionary authority by African national executives, and increase overall public accountability in their administrations . As the Bank , Barber B. Conable, noted in the report, a root cause of weak economic performance in th e past has been the failure of public institutions. Private sector initiative and market mechani sms are important, but they must go hand in hand with good governance a public service that is efficient, a judicial system that is reliable, and an administrat ion th at is accountable to the public ( 1989, xii). To be sure, World Bank policies expressly forbade the Bank from advocating for democratization; however, in practice (as well as a close reading of the report ), the effectively r egarded as an implied democratic condition ality for future economic assistance , particularly as democratic norms and institutions naturally coincided with many of the characteristics associated with good governance (Boafo Arthur 1998, 1 70 1; Gyekye Jandoh 2006, 117).
89 In sum, Ghana continued to economically improve as a re sult of the ERP adopted in 19 83. Inflation rates stabilized and even began to decline (Figure 3 1) , foreign direct investment returned (Figure 3 2) , economic growth rebounded and hovered ar ound five per cent (Figure 3 3) , and per capita income rose as well (Figure 3 4) . Measured in terms of stabilizing the overall economy and preventing an outright economic collapse, the ERP could have been considered a success. Source: World Bank 2014; Qu ality of Government 2013. Figure 3 1: I nflation, 1980 1992
90 Source: World Bank 2014; Quality of Government 2013. Figure 3 2: Foreign Direct Investment , 1980 1992 Source: World Bank 2014; Quality of Government 2013. Figure 3 3: Economic Growth, 1980 1992
91 Source: World Bank 2014; Quality of Government 2013. Figure 3 4: Per Capita Income, 1980 1992 However, in spite of these gains, many challenges remained. First, while the economy did improve during this period, rates of prosperity still remained below p revious peaks. For example, as per capita income reached as high as over $1100 in the mid 1970s before the arrival of economic crisis (Summers and Heston n.d.; cited in Herbst 1993, 18), and as per capita incomes remained below the continental average (Fig ure 3 5), the PNDC knew it still had a long way to go before it could declare an economic victory.
92 Source: World Bank 2014; Quality of Government 2013. Figure 3 5: Per Capita Income in Ghana Compared to Africa, 1980 1992 Secondly, despite increases in f oreign investment since its collapse in 1984, levels of investment unfortunately remained below the level necessary to wean the country off of international aid (Whitfield and Jones 2009 , 191 ). As a result, the country continued to rely on th e Bank and the IMF for support. In turn, the World Bank and IMF narrowed the terms of their assistance to be increasingly condition based, with some conditions needing to be met prior to disbursement. 3 Thirdly , the government continued to depend on external assistance , much of which in this period went primarily to infrastructure and other measures largely designed to alleviate hardships caused by the reform program, thus allowing the government to weather the political turbulence associated with the ERP (Nugent 1995, 17 4; Gyekye 3 To be sure, though conditions became increasingly tightened the Bank and IMF nevertheless tolerated slippage in some areas (Tsikata 2001; cited in Whitfield and Jones 2009, 191).
93 Jandoh 2006, 99). 4 As Figure 3 6 shows, foreign aid in Ghana continued to increase (as a share of its GDP) throughout the period of the ERP, as well as into its democratic opening. Source: World Bank 2014; Quality of Government 2013. Figure 3 6 : International Aid in Ghana, 1980 199 6 Furthermore, as funds from the World Bank (delivered through the International Development Association) and the IMF constituted approximately 58 per cent of total aid received by Ghana between 1984 and 1991 , reaching as high as 68 per cent in 1987 (Jebuni and Oduro 1998, 41), it is reasonable to conclude that it would have been a primary interest of both Rawlings and the PNDC to at least satisfy to a level that would have allowed aid disbursements t o continue. In deed , interviews among Ghanaian elites intimately familiar with this time period argue that Rawlings and the PNDC did choose to democratize in 1992, in part, due to growing international 4 The implementation of the Programme of Action to Mitigate the Social Costs of Adjustment (PAMSCAD) was instrumental in this regard (Nugent 1995, 174).
94 pressure. 5 As a result, despite any political objection s that may have occurred as a PNDC led government was thus obliged to do what it could to remain in good standing with donors. In addition, it shoul d also be noted that World Bank and IMF lending policies hel d a significant degree of influence over the terms of other donors, including the U.S. (Boafo Arthur 1998, 171). In other words, to lose good standing with the World Bank would have likely jeopardized aid from other donors and agencies. The End of the Cold War and Its Importance to Ghana T he end of the Cold War was significant in Ghana for two reasons: it meant the loss of an alternative, non Western source of international aid, and it additionally meant the subsequently diminished geostrategic importance o f the country with respect to the West , including the U.S . First, pr ior to the end of the Cold War , African countries had generally enjoyed relatively unconditional international assistance from either the United States or the Soviet Union (and their allie s ) , with the only requirement being international alignment with the either of the two donor superpowers. Countries declaring themselves to be either pro Western or pro Soviet enjoyed military and development aid with what were la rgely considered generous terms. I n some cases, notably Zaire (Wrong 2002) , countries would essentially bargain between the U.S. and Soviet Union in an effort to receive as much aid as each superpower was willing to provide in exchange for their support . Ghana was unexcept ional in this regard, as Rawlings and the PNDC in its initial years attempted to avoid Western (conditioned) aid by seeking assistance from the 5 Interview with senior NPP official, Accra, Ghana (November 18, 2013); Interview with former CPP parliamentarian, Accra, Ghana (January 20, 2014).
95 Soviet Bloc , including Cuba, Libya, and the Soviet Union (Nugent 1995) . However, as as its credit rating low , such sources were either unwilling or unable to provide assistance to the PNDC ( Herbst 1993; Jebuni and Oduro 1998, 40) . Thus, by 1983 the PNDC was left with the World Bank and IMF as its donor of last resort. Nevertheless, despi te its commitments to economic reform through the IMF and World Bank, both donors and the PNDC recognized the possibility of Ghana obtaining alternative assistance from the Soviet Union , thus providing the PNDC an inherent amount of leverage when bargainin g with its donors , as well as a certain amount of slippage when implementing their desired economic reforms (Gyekye Jandoh 2006 ; Whitfield and Jones 2009, 190 1 ) . By the end of the 1980s, however, it became increasingly clear th at Ghana would no longer hav e a potential alternative source of financial assistance: the economic collapse of the Soviet Union and increasingly apparent dissolution of the Soviet Bloc meant that Ghana lost its main source of leverage in negotiating with its donors. Furthermore, as t good governance, the PNDC was thus compelled to politically reform or face what would have been a likely economic crisis reminiscent of the one it previously experienced in 1983. Secondl y , the end of the Cold War entailed the generally diminished geostrategic importance of Africa to the U.S. and the West. Schraeder notes that in the immediate period following the end of the Cold War, the U.S. decreased its overall diplo matic presence on t he continent, including a significant reduction in foreign aid (1994, 251 ; see also van de Walle 2009 ). As the Soviet Union no longer represented a threat to
96 U.S. interests on the African continent, the U.S. was thus free to emphasize other foreign policy interests, such as promoting the spread of democracy in former Soviet aligned countries in Eastern Europe and elsewhere . As a result of this reorientation of priorities , many African countries, including Ghana, lost a substantial amount of leve rage in the terms of international aid . To be sure, Ghana remained strategi cally insignificant to the U.S. throughout the Cold War. Though the country was an informal leader because it was the first Black African country to gain independence, its mili tary remained rel atively small , and it possessed few resources of strategic interest. As American foreign policy was primarily oriented towards Eastern Europe, African affairs generally remained a low priority (Schraeder 1994). 6 However, in spite of this , it would be incor rect to conclude that the country was irrelevant to U.S. interests . During the PNDC era Ghana remained as one of the most important countries within U.S. foreign policy towards Africa . Former U.S. Ambassador Robert Fritts noted that the Reagan administrati on remained concerned viewed Ghana as a potential springboard for future Soviet expansion in West Africa and elsewhere on the continent ( Kennedy 1999, 101 ). In particular, Ambassador Fritts acknowledged that at the time, We [the U.S.] were concerned over an expanding wedge of Russian, Chinese, Libyan and Cuban influences and that Ghana could become a platform to destabilize West Africa. Key members of Rawlings's entourage, including his chief of security, had fough t with Samora Michel in Angola. The idea of radical revolution expanding in Africa and affecting our access to strategic resources and to military bases was all part of Cold War tensions. We also thought that Ghana had a special history and Ghanaians prove n skills, which if freed and supported, could reverse its downward 6 O bvious exceptions would have included sudden crises that would have either required the attention of the U.S. President, or prolonged ones requiring Congressional intervention (Schraeder 1994).
97 economic spiral and create a more open political system. Ghana was thus an integral part of U.S. interests in Africa (101) . Thirdly, despite regarding the expansion of Soviet influence in West Africa, by the latter part of the 1980s Ghana was increasingly experiencing domestic unrest and political protests, manifested largely through the Movement for Freedom and Justice (to be discussed in more detail in Chapter 4). In short, Ghana remained in a general state of fragility between 1988 1992, and the country continued to experience growing dome stic unrest with military rule ( with few resources to satiate or constrain it ) . T he government thus f aced a crisis of the status quo where political change became all but certain. Furthermore, it was additionally understood that without continued government would have likely fail ed (Nugent 1995, 187). Ghana After the Cold War: Limited L everage , Legitimacy, and Continued Aid Dependence As a result of the combined pressures of domestic political protest, increased conditionality for international assistance, and a n overall scarcer environment for obtaining such international assistance fol lowing the end of the Cold War, the PNDC began a process of democratization and a return to multiparty politics. To be sure, while National Commission for Democracy (N CD) to st udy the return to democracy as early as 1982, the pace and extent of reforms had conspicuously accelerated as the Cold War was coming to its end (Gyekye Jandoh 2006). 7 In addition, it is noteworthy that the 7 In particular, Gyekye Jandoh argues that the origins of Ghan transitions occurring in former communist countries coinciding with increased donor conditions accelerated the decision of Rawlings and the PNDC to return the country to a multiparty electoral democracy (2006, 129; on demonstration effects, see Huntington 1991).
98 itutional rule occurred just four days before an important conference of World Bank donors was held in Paris on May 14 15, 1991 (Oquaye 2004, 499). However, while Rawlings and the PNDC acquiesced to political reform, it would be incorrect to state that suc h a political opening was without challenges: the process was largely considered as having been controlled by the PNDC to its own advantage ( Gyimah Boadi 1994; Riedl 2014 ). In addition, the inaugural elections proved problematic after the main opposition p arty, the New Patriotic Party (NPP), accused the government of electoral fraud in the November 1992 presidential election and subsequently boycotted the parliamentary election the following month . Fallout From the Inaugural Elections The fallout from the 1 992 elections was significant: despite being declared a n overall successful election with notable discrepancies by international election monitors ( Nugent 1995, 236 ), accusations of electoral cheating and fraud from the opposition challenged the democratic credentials of the National Democratic Congress (NDC), the political party formed from the PNDC. 8 To be sure, most analyses of the 1992 elections show that despite some irregularities, such instances were either insignificant or did not pose enough of a c oncern to have undermined what would otherwise still have been a likely NDC victory (Jeffries and Thomas 1993; Nugent 1995 ) . 9 Nevertheless, the political environment remained rathe r tense following the 8 Results from the 1992 presidential elections showed a victory for Jerry Rawlings (NDC), receiving approximately 58 per cent of votes, while the main challenger, Adu Boahen (NPP), received approximately 30 per cent (ECG 2014). 9 To be sure, however, Nugent does note the plausibility that the amount electoral discrepancies experienced during the preside ntial election could have prevented Rawlings from winning in the first round (1995, 242).
99 elections. Op position parties formally expressed their ability to provide a free and fair electoral environment; the NPP even went as far as to document their particular grievances in The Stolen Verdict ( 1993). In particular, the NPP alleged that the presidential election had been flawed due to numerous irregularities, including the use of a defective and out of date voter register, ballot stuffing (allegedly occurring while ballots were being transported to the INEC headquarters for official tabulation), voter intimidation, a nd participation of those deemed ineligible to vote . Perhaps the most damaging accusation was the charge that the INEC had intentionally colluded with the (P)NDC in order to obtain a predetermined victory for Rawlings (1993, 67 8) . While the veracity of th substantiated (Nugent 1995), the charges nevertheless sent political shockwaves throughout the country and the report served to demonstrate how aggrieved the opposition was in regarding the election of R awlings as being wholly fraudulent. As a result, the government faced a crisis of legitimacy that not only threatened its ability to govern, but additionally threatened its ability to continue to receive international aid (Boafo Arthur 1998; Ninsin 1998; L evitsky and Way 2010, 303). At stake was the regime democratic legitimacy. Without recognition of its legitimacy to govern on the basis of the was imperiled, as it was generally understood that Western d onors would be monitoring . While international election observers ultimately judged the elections to have been conducted without fatal problems, the political from the process and boycott of the ina ugural parliamentary elections subsequently resulting in a de facto one party government produced an
100 unavoidable challenge to the legitimacy of the Furthermore, the low turnout in the inaugural parliamentary elections ( onl y 29 per cent ) raised additional legitimacy concerns (Jeffries and Thomas 1993, 363; Frempong 2007, 137). Indeed, interviews with the political elite confirm rejection of the legitimacy of the 1992 elections produced a two part proble m for Rawlings and his administration: not only was his ability to govern challenged on the basis of claims of electoral fraud, such claims additionally threatened status with international aid donors . 10 Furthermore, as the country continue d to rely upon international assistance throughout this period, and as funding for the 1992 elections had been provided largely through the donor com munity, maintaining legitimacy as well as the sense that such funds w ere being utilized effectively meant t hat the NDC led government would need to find a means for overcoming the political crisis, lest funding for future projects (including future elections) be reduced or withdrawn entirely (Boafo Arthur 1998, 174). As a result of the damage done to the regime and the NDC agreed to reform the electoral system. First, as part of the transition process to the Fourth Republic in 1993 , the government dissolved the INEC , and in its place re constituted an independent election comm ission (EC). To be sure, while the creation of the EC was provided through provisions in the Constitution of the Fourth Republic, the reconstituted EC nevertheless enjoyed its constitutional independence without any general challenge or interference from t he exe cutive (in contrast to its 10 Interview with senior lecturer (1) at the University of Ghana, Department of Political Science, Legon , Ghana (October 1, 2013); interview with senior lecturer (2) a t the University of Ghana, Department of Political Science, Legon, Ghana. (October 4, 2013); interview with Researcher from the Ghana Institute of Economic Affairs, Accra, Ghana (March 7, 2014).
101 predecessor , the INEC) . 11 Furthermore, as the INEC elections was generally viewed unfavorably (Nugent 1995, 239), and was consequently viewed with suspicion thereafter (Gyekye Jandoh 2006), the reconstitution of the EC provided a fresh start to the electoral process and allowed the EC to pursue additional reforms without the political baggage of its predecessor. 12 Secondly, Rawlings and the NDC led government also adopted certain recommendati ons from donors and election monitoring groups. In particular , the government (in conjunction with the EC) adopted a recommendation initially provided by the Commonwealth Observer Group (COG) that suggested the institutionalization of political dialogue am . What transpired was the creation in 1994 of what would be known as the Interparty Advisory Committee political parties, the EC, and leading international donors, IPAC (chaired by the Chairman of the EC) served as a means for these groups to engage in consultative could be improved (Frempong 2007, 141 ; Asante 2013, 63) . Although the body had no formal, legal authority, because the nature of such discussions were consultative, and because recommendations brought from the Committee were consensus based, many of the recommendations from IPAC resulted in a ctual implementation by the EC. These included the holding of presidential and parliamentary elections on the same day, 11 To be sure, while the chairman of the EC enjoys constitutio nal independence from executive 12 To be sure, however, some remained suspicious of the EC as its inaugural Chairman, Dr. Kwadwo Afari Gyan, was a previous executive member of the INEC (Frempong 2007, 141).
102 providing photo voter identification cards, and transparent ballot boxes (Frempong 2007, 141). Furthermore, many such reforms were suppo rted through funds provided by international donors, including the provision of voter identification card s to registered voters before the 199 6 elections, (Boafo Arthur 1998, 182). In sum, s continued access to foreign aid due to the damage done to rule , the government nevertheless succeeded in repairing such damage by reforming the electoral system and implementing a consul tative body that provided the politic al opposition a stake in the electoral system. As Ghana continued to be dependent on international aid throughout its democratic opening ( receiving aid constituting mo re than ten per cent of its GDP) , and as such aid was generally conditioned on political reform, the inaugural government had strong incentives to proceed with democratic reform, or at the very least, to constrain itself demands for further reform. International Leverage in Ghana While Ghana re ceived significant amounts of international aid before, during, and after its democratic opening in 1992 (Figure 3 7) (Jebuni and Oduro 1998 ; Gyekye Jandoh 2006; World Bank 2014 ), the effectiveness that such aid had in Ghana was not simply due to the amoun t of funds that were given. More importantly, its effectiveness was a function of the leverage international donors could exercise through the conditions attached to their aid. In general, throughout the post Cold War period donors maintained significant l everage over the Ghanaian government (Whitfield and Jones 2009, 192 5 ) .
103 Source: World Bank 2014; Quality of Government 2013 . Figure 3 7: International Aid in Ghana Compared to Africa, 1990 1994 (Levitsky and Way 2010, 40), Western donors possessed a high amount of int ernational leverage over Gha na for a variety of reasons. 13 First (as discussed above), Ghana had a weak economy and was highly indebted. Secondly, its economy was sma ll, as GDP remained less than $50 billion between 1983 1996 (World Bank 2014) . 14 In addition , its narrow economic base based on cocoa production, timber, and gold mining made the economy vul nerable to market shifts in commodity prices . Thirdly , at the time of its democratic opening, Ghana was neither a major oil producer, nor was it a secondary oil producer (USEIA 2014). Fourthly , because of the end of the Cold War, the U.S. and European Union did not have any significant security or geopolitical interests i n Ghana. 13 372 3). 14 Thus, the impact of aid was likely to be significant in this regard.
104 Lastly, unlike other African countries with special relationships to more wealthy countries (such as Cameroon and Gabon historical relationship with France ), Ghana assistance or access to aid. In sum, coup in 1981 thus placed Ghana in a rather disadvantageous position relative to Western don ors . Furthermore, because donors were already in a rather reluctant m ood to provide support and would, in general, only contribute under conditions of political reform, the Government of Ghana had little option but to accept such aid under the terms that donors stipulated . 15 To be sure, as Ghana enjoyed a privileged status f or its prior achievements under structural adjustment (Gyekye Jandoh 2006 ; Whitfield and Jones 2009 ), the government had marginal latitude in implementing such reforms, including the ways in which aid was utilized and the extent to which reforms would be e nacted. 16 Putting the Pieces Together: In general, although the Ghanaian economy continued to improve as a result of international assistance and the ERP , the country continued to depend on foreign aid up to a nd through its democratic opening in 1992 (Herbst 1993) required continued access to international assistance, the state pursued economic and political liberalization in order to continue receiving the benefits provided from intern ational aid (Lyons 1999, 158, 160; Green 1998, 203; Whitfield 2009b, 65 6; 15 situation where donors effectively set the economic policy agenda at the expense of the Ghanaian government (Whitfield and Jones 2009, 193). 16 led government selectively targeted donor funds for electrification projects in Ghana in order to benefit the NDC in subsequent elections.
105 Whitfield 2009c; Levitsky and Way 2010, 301). This relationship was challenged following the boycott of opposition parties after the inaugural presidential elections in 1992, as the Maintaining legitimacy was very important to Rawlings and the NDC government , as donors had largely turned apprehensive to providing aid to authoritarian regimes following the collapse of the Soviet Union and t ld Bank 1989; Green 1998, 193 4; Levitsky and Way 2010, 303; Boafo Arthur 1998). Although the country enjoyed good standing with the donor community due to its prior successes in the 1980s (Levitsky and Way 2010, 301), the country remained limited in its ability to negotiate (Armstrong 1996; Whitfield 2005): it continued to need donor assistance, and the end of the Cold War not only decreased its geostrategic importance, it additionally removed the possibilit y of obtaining non Western sources of aid in the near term. As a result, Rawlings managing the post election political crisis. unresolved would h international aid. Ultimately, history shows that Rawlings and his government undertook additional reforms in an ef fort domestic and international c ommunity. Such reforms included lifting media restrictions in the country (albeit in a controlled fashion), as well as allowing the newly reformed EC to impl ement IPAC at the request of the donor community . government expanded the playing field by which opposition parties could compete
106 against the NDC. What is further notable is that even in periods where Rawlings was vocally indignant a gainst political reform, such as Supreme Court rulings that were decided in favor of the NPP, his government maintained restraint and allowed such reforms to proceed without directly interfering with the process or the state institutions mandating such cha nge. To be sure, such reforms could have been made entirely cy (Levitsky and Way 2010, 305); however , it stands to reason that even in the event that Rawlings and his government were not converted democrats a nd continued to harbor visions of consolidating their rule in a fashion reminiscent of the pre democratic opening ruling style of the PNDC , to have done so would surely have aggravated international donors and would have likely resulted in international ai d being suspended or withdrawn. 17 In addition , the influence of the international donor community in the non governmental sector has been well known, and has been previously argued to have aided the effectiveness o f these groups in advocating for continued reform after the inaugural elections of 1992 (Gyimah Boadi 1999, 180; Gyimah Boadi and Yakah 2013, 275). Such organizations have been well known to engage in civic education, report on democratic progress in the country , produce policy recommendations, and even (more recently) facilitate presidential and parliamentary debates. At the same time, however, many of the democracy promotion activities from such pro democracy institution s including t he Ghana Centre for Democratic Development (CDD), the Institute o f Economic Affairs (IEA), and the Institute for Democratic Go vernance (IDEG), among 17 Rawlings was well known to have been outwardly dismissive of multiparty democracy prior to the 1992 inaugural elections, and was also known to have called for a review (and possible revision) of the Fourth Republic Constitution (Gyimah Boadi 1994, 80).
107 others are significantly financed through funds provided by international donors (Gyimah Boadi and Yakah 2013). Furthermore, t he independence and effectiveness of such NGO s operating within the country may be partially explained by the financial support they receive from the international community (Gyekye Jandoh 2006 , 158 9 ). Furthermore, while Ghana enjoyed a rather privileged status among donors for its past success from the ERP, and additionally enjoyed a certain amount of slippage when it came to enforcing donor conditions, it would be incorrect to assume that the government was allowed to ignore political reform, or that donors were unwilling to hold the government acc ountable when it failed to meet certain conditions. For example, the World Bank suspended its disbursements in 1992 following government overspending (Devarajan, Dollar, and Holmgren 2001, 77 ), and the IMF suspended its aid in 1996 for similar reasons ( Whi tfield and Jones 2009, 193) . 18 In addition, Green (1998) notes how particular, Green notes how , ion of multiparty democracy drew disparaging remarks from Western diplomats. The United Kingdom pressed for government consultations with Ghanaians on the future of the political system and the U.S. assistant secretary of state for Africa, Herman Co h en , wa rned that pressures on the PNDC for some kind of political opening tipped the balance in favor of long suppressed societal demands for political freedoms 1991, the PNDC had given up one of its key political positions, abandoning the nonpartisan True Democracy project and Consultative Group meeting with donors, 14 15 May) (1998, 199; see also EIU 1990, 16). 18 To be sure, while bilateral donors disbursements aided in alleviating the full impact of such aid suspensions, such contributions did not cover the full cost (Whitfield and Jones 2009, 193). As a resul t, the country experienced immediate and significant fiscal deficits (Devarajan, Dollar, and Holmgren 2001, 77).
108 international aid government acted as a constraint to government through its emphasis on refo rming the system and maintaining enough of a democratic environment to show donors that the NDC represented the legitimate government of Ghana. In addition to successfully pressuring the government to maintain democratic standards, the international commun it y additionally funded democracy promotion initiatives within civil society, as well as largely funded subsequent general elections (Gyimah Boadi and Yakah 2013). To be sure, though these additional activities operated in an indirect regard, they neverthe less augmented the ability of civil society to advocate for positive change while supporting an electoral process that would minimize the potential for manipulation and fraud (Gyekye Jandoh 2006) . Conclusion nternational donors , this chapter has shown how the international environment constrained the Government of Ghana from engaging in behavior that would have undermined its democratic credentials. Despite experiencing what may be widely regarded as a flawed inaugural election, as well as a significantly problematic democratic opening in 1992, the country nevertheless succeeded in maintaining a democratic trajectory. The international community was instrumental in this regard, as donors successfully constraine overt undemocratic behavior. One respondent even noted the role of international aid and
109 19 Not only was Ghana in need of international assistanc e due to its longstanding debts, fragile economy, and prior commitments with the IMF and World Bank, its ability to negotiate the terms of such loans, as well as its ability to evade its commitments remained limited as well. Furthermore, Nicolas Robe rtson, a former U.S. diplomat who served in Ghana , notes in an interview that while Rawlings complained about the excessive influence of the United States, Rawlings nevertheless acknowledged made us do it. We never would ha ve become democratic had it not been for the State Depa He [Rawlings] was complaining but I thought it was the nicest thing anybody said about my institution To be sure, not all respondents agreed on th e degree of influence of the in the country. 20 Others were more circumspect, not ing that the influence of foreign aid was only successful due to the domestic demands for democracy from the citizenry. 21 Nevertheless, in spite of these criticisms, these respondents still agree d that foreign aid and international pressure contributed in p romoting democracy in the country. Furthermore, these responses international and domestic pressure as necessary for establishing a democratic trajectory, as they acknowledge the complementary role o f dome stic political pressure 19 Interview with professor at the University of Ghana Institute of Statistical, Social, and Economic Research, Legon, Ghana (October 14, 2013) . 20 Interview with senior official at IMANI Center for Policy & Education, Accra, Ghana (February 13, 2014). 21 Interview with World Bank official (1), Accra, Ghana, March 21, 2014; interview with World Bank official (2), Accra, Ghana (March 28, 2014).
110 (whether from the citizenry or from NGOs) in encouraging democracy, or at the very least, preventing a return to authoritarianism in the country. and as the Government of Ghana remained limited in its ability to negotiate terms of aid in its favor, Rawlings had little choice but to govern in a way so as to avoid being perceived as opposing democratic development in Ghana. To be sure, even if Rawlings a nd the NDC led government were committed democrats, to have followed any alternative course that may demands for political reform, or would have otherwise been regarded as obstructing democratic progress w ould have likely resulted in restrictions to ability to obtain continued access to international aid. It is no exaggeration that had the Government of Ghana lost access to international aid , the country would have likely experienced an eco nomic crisis of the same magnitude it experienced a decade earlier, and would have furthermore made the ousting of Rawlings and his NDC government a distinct possibility (Nugent 1995) . By expanding current explanations of foreign aid and its role in promot ing democracy in Ghana, this chapter contributes by providing an updated account of this process that includes an assessment on the role of leverage possessed by international donors, as well as the manner in which such leverage was exercised over the Rawl ings administration. Indeed, aid and the leverage that international donors had with respect to the terms of their aid ating an electoral environment where the opposition could operate without being overtly
111 repressed. However, in spite of the assistance provided by donors and the manner in which the Ghanaian state remained vulnerable to external pressure, it is equally imp ortant to explain how the domestic opposition remained as a significant force in Ghanaian politics, despite its formal withdrawal from government following the inaugural elections. In short, without the domestic opposition comprising a credible alternative to discrediting the inaugural s democratic credentials ( obtain needed international finance ) resulted i trajectory will be explained in further detail.
112 CHAPTER 4 DOMESTIC POLITICAL OPPOSITION This chapter trajectory by highlighting the role of the domestic environment, and in particular, the role of the domestic political opposition during the country critical juncture between 1992 and 1996 . In brief , this chapter re examines the role that civil society , previously defined 6), democra cy. P rior works (IPAC) contributed to producing the . In this chapter I interrogate such claims an d find them to be incomplete, as they omit the common denominator that remained present : a sustained and significant domestic political opposition that served to constrain Rawlings and the NDC government from consolidating thei r rule . W hile civil society has certainly been a proactive force in contemporary Ghanaian politics, its presence was not always reliable . As will be explained in further detail in this chapter, civil society ical opening as various factions gravitated to each of the contending political parties during the run up to the 1992 election . As such, it becomes problematic to theorize that this sector consistently pushed Ghanaian democracy throughout its Fourth Republ ic, as its presence remained generally absent (or at least, insignificant) until the time of the second general elections in 1996.
113 marginal presence during the democratic opening, the political opposition nevertheless survived a nd demonstrated its potency to the incumbent government during the inaugural elections as well as its aftermath. Even though the NPP experienced a resounding defeat in the inaugural elections, it still managed to be minimally competitive against the ND C by garnering a respectable 30 per cent of the total vote share in the presidential election. In other words, while the NDC achieved a near two to one victory against its nearest competitor for the presidency, it nevertheless had to regard the NPP as a seriou s contender and as a credible challenge to its authority . formidable opposition party was additionally buttressed by three supporting factors: international pressure, historical political cleavages, and an ethnic association with the Akan . First, as highlighted in the preceding chapter , the international community aided the political opposition through its insistence that acquire legitimacy at both the domestic and international level . By legitimacy, donors want ed the government, as well as its leaders, to be a representation As the opposition boycotted parliamentary elections following irregularities experienced during the earlier presidential election, the NDC led government faced a crisis of legitimacy that . Maintaining access to foreign aid was vital for Rawlings and the NDC, as without it the country and the economy wo uld likely have relapsed to levels not seen since the days of the ERP and structural adjustment. In order to overcome this challenge, the government agreed
114 furthermore agreed to institutionalize IPAC . In sum, as a result of the reforms the government not only regained the sense of legitimacy that it needed to continue to enjoy access to financial assistance, it additionally produced two important factors that aided in maintain ing democratic trajectory (to be discussed in further detail below) . A second factor concerns what have general ly been considered the Nkrumah and Danquah Busia traditions. In brief, these t wo traditions are named after some of the most prominent political figures right political axis, with the Nkrumahist tradition tending to advocate socialist oriented policies and prog rams (including an increased presence of the state in the economy), and the Danquah Bu sia tradition favoring what would be considered neo liberal, market oriented policies. During the democratic opening, the NPP emerged as the only party advertising its adh erence to the Danquah Busia tradition, while numerous other parties contested the Nkrumahist tradition . Ultimately, the NDC succeeded ( perh aps haphazardly) in appropriating the Nkrumahist mantle, or at least the largest share of it (Nugent 1995) . 1 In other words, while the NPP experienced a clear defeat at the 1992 presidential polls, it nevertheless claimed Nkrumahist orientation. tionship between ethnicity and political association. Throughout the Fourth Republic the NDC and NPP have had a well known and generally recognized affiliation with two of the 1 Africa Confidential , 14 November, 1992, 33(16): 2 5.
115 (Fridy 2007; Elisc her 2013) . This ethnic attachment has gone as far as to have many political and government officials openly refer to the Volta and Ashanti regions home to majority Ewe and Akan ethnic groups , respectively votes. H owever, in spite of these ethnic attachments , Ghana still avoids having politics defined solely on an ethnic basis as all parties maintain national coverage and as no ethnic group possesses a political majority (Fridy 2007 ; Arthur 2009 ; Whitfield 2009c ) . A s such, all parties must acquire votes across multiple ethnic dimensions in order to win national elections. 2 Interestingly, it is in this sense that ethnic politics aided formidability as an opposition party particularly as Akans have constitute d at least 45 per cent while still enabling the NPP to conduct its affairs as a non ethnic party. In other words, though the NPP never became an exclusively Akan party, it nevertheless re tained a substantial electoral base from which it could derive sustained support. Furthermore, it is reasonable to suggest that the NDC would have recognized the potential for the NPP to politicking , and would have furthermore acted in ways so as to prevent its occurrence , including (but not limited to) recruiting Akan elites into the NDC (Elischer 2013) . As a result, contrary to what previous research regarding ethnicity and democracy would have suggested (Horowitz 2000), ethnicity arguably aided democracy in Ghana , rather than undermined it. 2 To be sure, in spite of these ethnic affiliations, it would be incorrect to consider the NDC and NPP as recent work, the NDC would be categorized as a personalistic party, while the NPP would best be considered as an ethnic alliance (Elischer 2013, 177). Since the inaugural elections, both parties have transformed into non ethnic parties (see also Fridy 2007 ).
116 In sum, this chapter shows that the domestic political opposition in Ghana proved to be a sustainable and significant presence during the democratic opening, which se rved to only increase in size over time. Because the opposition achieved minimally competitive standards against the incumbent government at the inaugural elections , the government was compelled to regard the opposition as a credible political threat, and in turn, was additionally compelled (through international pressure) to reform the electoral process and to open up the electoral rulemaking process in a more transparent manner. In other words, despite any possible interests in using their incumbent statu s to marginalize the opposition, Rawlings and the NDC knew that to do These actions become all the more extraordinary when considering that these reforms took place in the formal absence of a ny opposition within government . To be sure, while these reforms and actions likely would not have occurred if not for the pressure placed on the government by the international community, the success and viability of such reforms would similar ly not have been achieved if not for the significant and viable presence of the political opposition operating within the country. As the case of Ghana shows, neither domestic nor international factors alone can wholly account for the trajectory; only when these factors are considered in concert with each other does a more accurate explanation emerge that accounts for the structural incentives favoring a democratic trajectory. Origins of the Opposition In order to talk about the success of the domestic political opposition in maintain ing origins of the opposition, including its ideological basis, as well as its role in producing
117 success in combatting corruption, by the late 1980s large segments of the Ghanaian population began to feel disaffection with the regim e . As highlighted in the previous chapter, the ERP, while benefitting the overall economic situation in the country, left many losers in the process, including urban workers and civil servants who had either been laid off, or had seen their working conditi ons deteriorate throughout the course of the decade (Shillington 1992) . Further compounding the situation, Nugent notes how the regime lost its sense o f Nugent 1995, 187), that is, due to previous successes in governance, the PNDC began to feel a sense of invulnerability and began to believe it could govern without fear of consequences . In doing so, the PNDC enacted a series of laws that both angered and alienated various influential segments of society. When the government enacted restrictions on religious organizations in the country, including passing the Religious Bodies Registration Law, as well as restricting the activities of Mormons and Witnesses in the country, the religious community responded in an uproar against the government by openly protesting the measures in their churches and through interfaith organizations, including the Christian Council and Ultimately, the PNDC was forced to concede to the demands of the religious community in the country , however the damage done to the relationship between the two would endure . In addition, the government created another controversy when in 1989 it revoked all licenses for print publishers and required newspapers to re re gister with the
118 government under new, more stringent regulations (Nugent 1995) . media had already operated in a restricted environment, and while there existed only one well known newspaper that provided an alternative perspective to th position, the increased restrictions were regarded by the population as excessive and unnecessary . In addition, w hile the alleged purpose of these measures was to increase the professionalism of the press and media in the country, the conseq uence of such restrictions nevertheless resulted in the proliferation of smaller, tabloid like publications with much less journalistic integrity (Karikari 1998, 195). More importantly, however, as sports coverage remained acceptable and generally unrestri cted by the government undertones, sometimes cou ched in aesopian/satirical terms, and aimed at sending a Thirdly, acc usations of corruption within the PNDC leadership in the latter portion of the decade rattled the regime and its credibility with the population (Nugent 1995). This was especially troubling, particularly as largely b uilt on an anti corruption and anti elitist platform. 3 Thi s, in turn, led to reshuffling within the PNDC leadership, as well as to increased factionalism within the PNDC that became all the more visible to the general population. The climax of such faction alism became evident when the government arrested Major Courage Quarshigah in 1989 on charges of plotting a coup against Rawlings. The arrest of Quarshigah was quite a shock to Ghanaians, as he had been popularly considered a staunch loyalist to 3 Indeed, Rawlings was well as he would often wear his military uniform in public (as opposed to wearing expensive suits), and would also engage in manual labor at publicized n ational work projects (Shillington 1992; Nugent 1995; Oquaye 2004).
119 Rawlings, and was considered a national hero for previously preventing at least two coup attempts ( Shillington 1992, 145; Nugent 1995, 194). By 1988 resentment against the PNDC reached its peak. Major civil society organizations, including t he churches, the media, t he National Union of Ghanaian Students (NUGS), the Ghana Bar Association (GBA), large segments of the Trades Union Congress (TUC) , as well as many other organizations all became united in their y democracy. One event that has been generally attributed as having instigated opposition forces to unite against the regime was the Adu Boahen lecture series , in which Professor Adu Boahen delivered a series of lectures that highlighted many of the grieva nces felt in Ghanaian society, as well as other alleged injustices perpetrated by the PNDC during its tenure in government ( Boahen 1992; Nugent 1995; Oquaye 2004). More importantly , by addressing these political grievances in public, generally regarded as having broken a there existed a general understanding among the population that political speech, especially speech in opposit ion to the PNDC government, was effectively prohibited in the country by means of arbitrary arrest , indefinite detention, and intimidation. 4 To be sure, whether this was an explicit policy by Rawlings and the PNDC, or whether it was a side effect of the ge neral disorder and absence of clear lines of authority that prevailed amongst the state security forces and the local defense committees remains unclear (Shillington 1992; Nugent 1995; Oquaye 4 freedom of speech and movement in the country were rather constrained by the capricious use of authority by state security forces (2013, 216).
120 2004). Nevertheless, the ultimate effect of this ambiguity resul ted in a de facto policy of (self) censorship in order to avoid harassment, arrest, imprisonment, or worse. This pervading sense of silence became all t o o real after the mysterious murders of three High Court judges and a retired army major in 1982 that we re widely believed to have been politically motivated (Nugent 1995, 87; Oquaye 2004, 308 9). While Rawlings was rather outspoken in his condemnation of the killings, noting the murders as an act of this revolutionary process is b , 38), because the investigation of the murders resulted in the arrest and conviction of Amartey Kwei, a member of the PNDC who had additionally implicated Captain Kojo Tsikata, one of Rawl (though was never arrested , and had later been cleared through a retraction made by Kwei prior to his execution in 1983 ), conspiracies of the murders being wholly orchestrated by the PNDC persisted (Shillington 1992 , 92 3 ; Nugent 199 5; Oquaye 2004 ). Two years following the Adu Boahen lectures and the subsequent dissolution of the culture of silence, the political opposition came together and formed the Movement for Freedom and Justice (MFJ) , incorporating elements of society from all sides of the political spectrum. 5 The MFJ immediately went to work staging non violent protests and rallies throughout the country, though mostly in the urban centers of Accra and Kumasi. In general, the goals of the MFJ included the return of multiparty d emocracy to Ghana, as well as the repeal of what were considered to be repressive laws, including but not limited to the Preventive Custody Law and the Newspaper Licensing Law (Oquaye 2004, 348 51). served to publicize it s opposition to 5
121 PNDC rule, its most dramatic success may arguably be attributed to its efforts in (NCD) regional seminars on local governance in the country (discussed bel ow) . Officially inaugurated in 198 government provided for the decentralization of its authority through the creation of District Assemblies throughout the country. According to Shillington, The District Assembly was to be the highest political and administrative authority on th guide, su pervise, administer all other political authorities in the district. It would have the power to raise local finances from specified sou rces, such as levies on individuals, on entertainment and on trade. It would initiate development projects and ultimately, when central government departmen ts were decentralized to the districts, it would have more or less full control over those departmen ts that most affected them, especially health and education (1992, 160). In essence, District Assemblies were to allow Ghanaians to govern their local affairs , including directing their own local development ( supplemented by resources disbursed from the ce ntral government) , while the national government could instead focus on domestic and foreign policy . Two thirds of the District Assemblies would be filled through non partisan elections (as political parties were still banned) , while the remaining one thir d would be filled by appointment from the PNDC. In addition, the central government appointed a district secretary that served as the chief executive of the district, as well as an ex officio member of the District Assembly. 6 As the NCD held regional semin ars throughout the country to assess the effectiveness of the decentralization exercise, the MFJ inserted itself into these 6 While the stated purpose of enacting District Assemblies was to devolve power and decentralize authority in the country, in practice, the District Assemblies were argued to have strengthened the central
122 seminars, using the feedback process to instead voice their desires of return ing the country to a multi party democracy. Participati ng in the regional seminars as a critic of the regime was not easy, as many participants in the seminars were invited by the government (reaching as much as two thirds of participants in some meetings), including sitting assembly members, as well as repres entatives from Committees for the Defense of the Revolution Nana Konadu Agyeman Rawlings) (Shillington 1992, 166). In other words, the regional seminars contained an inherent degree of bias to wards the PNDC, as many participants had a structural interest in perpetuating the status quo. In spite of such obstacles, including a refusal to recognize the MFJ as an organization at the regional seminars, the MFJ nevertheless succeeded in transforming the discussion away from a District Assembly validation exercise towards a broader discussion concerning the democratic future of the country. The result was rather extraordinary, with the NCD essentially forced to acknowledge the widespread sentiment amo ng the population in its final report to the PNDC . What was initially overshadowed instead by the regional seminars being turned into a forum for discussing the most appropr iate form of governance in the country, as well as the return to a multiparty system of demo cracy (Oquaye 2004, 489) . noted how n is not against political parties as an ideal instrument manner in which the NCD framed this statement expressing the issue in terms of not
123 being against multiparty democracy rather than favoring it more than the current system desires while framing them in a manner policies opposing the forma tion of political parties , and to avoid provoking an unintended crisis of governance as a result (Dadzie and Ahwoi 2010, 76 7) . Shortly after the NCD submitted its report to the PNDC, as a sur prise to many in the opposition the government responded by anno uncing that it would begin implementation of a program to return the country to a multiparty system of rule (Shillington 1992, 173) because it amounted to the total abandonment of an alternative vis (1995, 199). Rawlings and the PNDC were well known to have detested multiparty politics, claiming that political parties were inherently elitist and marginalized ordinary Ghanaians from participating in decision making processes (Nugent 1 995, 47 ; Oquaye 2004 ). Instead, Rawlings and his cohort had advocated for a no party system of governance reminiscent of the Libya n model under Gadhafi, where local politics would be handled through direct citizen participation within citiz en committees , while national policy would remain the prerogative of the national government . 7 While the MFJ claimed success in prompting Rawlings and the PNDC to institute a return to multiparty democracy, the MFJ was less successful in its efforts to inf luence In an effort to get ahead of the tide of change, t he PNDC directed the process of regime change in the country, including the 7 Despite appe did not include mechanisms of accountability for decisions made at the national level, as well as for national officeholders (Nugent 1995, 48).
124 composition of the Committee of Experts (CoE) that was charged with draftin g a new constitution, as well as largely controlled the composition of the Consultative Assembly, the body designated to deliberate and adjust the constitutional proposals prior to holding a national referendum in regarding its adoption (Nugent 1995; Oquay e 2004; Riedl 2014). 8 The opposition had many misgivings about the process, including the fact that the constitution ( pending adoption ) would not enter into force until January 1993 , after ial and parliamentary elections . I n addition, the opposition was incensed at the controversial in clusion of an indemnity clause uni laterally inserted by the PNDC that absolved the PNDC from any illegal and/or criminal activities during its rule. 9 In the end, however, fearing that opposing the draft constitution in the national referendum would inherently mean prolonging PNDC rule , the opposition acquies ce d and urged their supporters to vote in favor of the constitution. After the constitution was over whelmingly ratified in May 1992, the PND C followed by lift ing the ban on political parties and inaugurating the contest for who would become the first president of the Fourth Republic. Consequently, the MFJ largely disbanded as factions rushed to either start or join one of the various political parties created in the newly legalized environment. Whether the MFJ the PNDC to capitulate and return the country to 8 It is important to emphasize that the Consultative Assembly was effectively consultative in its input into the constitution building process, in that the PNDC retained the prerogative in overriding its decisions (Nugent 1995, 217). 9 Another notable change of the draft const itution prior to its ratification included the Consultative Assembly consolidating the roles of head of government and the head of state within the presidency, rather than having such duties respectively divided between a prime minister and president, as w as initially envisioned from the draft submitted by the Committee of Experts (Afari Gyan 1995).
125 Ãª tre was explicitly provisional, as stated in its title. In addition, the NCD had been established years earlier under the auspices of bringing the country back to a democratic system of rule, however different from other multiparty models (Shillington 1992). Some within the PNDC even regard ed the formation of the district assemblies in 1988 89 as the precursor to a logical conclusion in returning the country to democracy ( Nugent 1995, 200 ; see also Jeffries 1992, 224 5 ). At the same time, however, it is important to acknowledge that Rawlings and the PNDC retained a particular view as to what democracy meant, which often did not include multiparty politics. Furthermore, one would be remiss to not include how international pressure encouraged the PNDC to reconsider its view on multiparty democr acy (Gyekye Jandoh 2006). As had occurred throughout Africa following the end of the Cold War , the decline of the Soviet Union and expansion of democracy in former Soviet satellite states precipitated a shift in how Western countries and donors would provi de foreign aid to countries like Ghana without some sort of democratic reform. conference in P In sum , as an explanation likely includes a combination of the abovementioned factors (Jeffries 1992, 225; Nugent 1995, 202) , the point is that the opposition and the MFJ nevertheless regarded the event as a victory for their cause ( Oquaye 2004, 490 ). This perceived victory subsequently served to encoura ge the opposition, including the prospec t that they could not only succeed in
126 ousting Rawlings (either through preventing his future candidacy or ou sting him electorally) , but that they could even capture the presidency in the process . The New Patriotic Party While the New Patriotic Party (NPP) was officially established in August 1992, its origins date back to the Danquah Busia Memorial Club establis hed in the late 1980s ( Nugent 1995, 221 ). 10 Despite the PNDC banning formal political party activity during its period of rule, it nevertheless allowed the operation of ostensibly non political clubs. 11 The NPP comprised leftovers from the Progress Party (th e party in power during the Third Republic), with most of its followers residing within dominant Akan regions. The success and wealth from business and entrepreneurship. I ts party symbol, the elephant, was considered a symbol of wealth and power among the Akan (Nugent 1995, 233). Despite beginning the inaugural presidential and parliamentary campaign at a disadvantage, as the ( P ) NDC was generally known to have begun campaig ning on a n unofficial basis prior to its official certification in June , the NPP nevertheless remained confident that it could compete and win. 12 First, t he NPP believed that by campaigning on the basis of democracy and freedom, and by reminding voters of a ll of the alleged the NPP could 10 While the ban on party activity was officially lifted in May 1992, it was not until August when most opposition parties received their certification from the INEC. In addition, it should be noted that accounts differ on the exact date of the founding of the Danquah Busia Memorial Club, as Nugent (1995) suggests the club formed in the late 1980s, while Fordwor claims its founding in 1991 (2010, 188). 11 While man y clubs, including the Danquah Busia Memorial Club, that operated during the PNDC era were arguably political in their discussions and practical operations, such clubs were allowed to continue as other pro government clubs had been previously allowed to op erate (Fordwor 2010). 12 All political parties were required to register with the INEC and receive certification prior to conducting official campaign activities. The NPP did not receive their official certification until August 1992 (Nugent 1995, 218).
127 overcome the incumbency advantages of Rawlings and the NDC (Elischer 2013, 147 ; Riedl 2014 ). This pro democracy platform was further reinforced through the pa choice of Adu Boahen as its presidential candidate, the man who had been previously heralded as breaking th . Second ly , the NPP also relied on its elitist credentials to support its campaign . As obtaining funds from corpo rate and international donors were banned, the party thus drew upon its network of wealthy members and supporters to independently finance its party operations (Jeffries and Thomas 1993, 338) . In contrast, it was well known that the line separating state a nd party activity remained blurred, as the NDC and its candidates enjoyed a certain advantage by accessing state resources for use in its campaign activities , including monies and vehicles to travel around the country ( Gyimah Boadi 1994, 79; Nugent 1995, 2 18 9; Levitsky and Way 2010, 302). Third ly , the NPP also counted on its ethnic affinity with the Akan as an asset when competing against the NDC. In 1992, the Akan constituted at least 40 per cent of (GSS 2002) . A s the NPP was wide ly recognized as being associated with the Akan, and had largely installed Akan officers into its party leadership, the NPP believed it could court the Akan vote while persuading enough non Akans to vote for the party in order to win the presidency (Jeffri es and Thomas 1993, 359). At the same time, however , this is not to suggest that politics in Ghana was conducted on an ethnic basis. To be sure, while some scholars have emphasized the importance of ethnicity in Ghana ian politics ( Chazan 1982 ), they are ca reful to note the fluidity with which its salience ebbs and flows as a function of politics and power in the Ghanaian political sphere . In contrast, other scholarship challenges ethnicity as a
128 primary determinant in Ghanaian politics by acknowledging the c omplex manner by which ethnicity is compelled to compete with a range of other equally salient issues in the country ( Lentz and Nugent 2000; Nugent 2001; Oelbaum 2004; Fridy 2007; Arthur 2009 ). 13 Indeed, at the time of the 1992 presidential election the pol itical dynamics of the NPP and NDC encompassed a wide range of political cleavages, including geography (South/North), socioeconomic status (rich/poor ), occupation (labor/professional services), and place of residence ( urban/rural ) . In addition, ideology traditions, the Nkrumah and Danquah Busia traditions , also helped mitigate the hazard that elections would be solely determined by ethnic politics . 14 Indeed, Whitfield argues that the reason why elec tions in Ghana have been generally competitive and stable is across social cleavages such as ethnicity, region, urban/rural and 623 ; see also El ischer 2013 ; Lindberg and Morrison 2008; Levitsky and Way 2010, 304 305 ). Nevertheless , in spite of all of these cleavages, the ethnic association between the NPP and the Akan, as well as between the NDC and the Ewe remained inseparable (Fridy 2007 ; see al so Elischer 2013 ) . As a result, while it would be disingenuous to discuss these parties without acknowledging their ethnic proclivities , it would likewise be 13 I t is also important to note that the Akan in Ghana are not a homogenous group, but rather a collection of sub groups sharing certain common foundations including (but not limited to) history, lineage, language, and mythology from which they identify. As a result, while such sub groups identify themselves under this overarching ethnic category, it is problematic to assume that the Akan can be politically mobilized as a single ethnic group. 14 To be sure, while Elischer classifies the NPP in 1992 as an ethnic alliance, he is careful to acknowledge the limits to this assessment, particularly as the NPP possessed a rather programmatic party platform, and as the party characteristics fell just short of being classified as a (non ethnic) catch all party (2013, 150) .
129 . An illustrative analo gue to this relationship may be seen with regards to the role of ethnicity in elections in the United States , where particular ethnic groups are generally associated as voting with either of the two main political parties. E thnicity in Ghana may be similar ly characterized, as the Akan and Ewe provide an electoral base from which the NPP and NDC draw their respective support . In sum, sustained and significant presence in Ghanaian politics . T he NPP was able to secure electoral support from (thus providing a core constituency for the party) , and in turn, the government was inherently compelled to remain restrained in its approach to challenging the NPP , as a ny overt or extra legal acts of repression against the NPP would risk provoking widespread Akan opposition against the government . 15 Fourth ly , the NPP also benefitted from being the only opposition party with an ideologically distinct platform from the NDC . As previously highlighted, p olitics in Ghana had long revolved a round two political traditions named after prominent political figures from t the Nkrumah and Danquah Busia traditions (Nugent 1995; Oquaye 2004; Whitfield 200 9b). As the NPP became the sole claimant of the Danquah inherently meant that it would have to compete with other (though smaller) parties, the 15 Rawlings never went as far as to equate the two together and furthermore maintained a public abhorrence to tribalism in politics (2013, 152).
130 NPP emerged as the only opposi tion party w ith an ideological alternative to the status quo . 16 In sum, because the NPP had a history of advocating for democracy, had elected as its presidential candidate silence, had sufficient domestic and i ndependent finances so as to fund party operations, was associated (though not synonymous) group, yet all the while possessed the only alternative ideology to that of the incumbent party, the NPP had strong reasons to believe that they could compete against the NDC in the inaugural elections and win . Furthermore, because many of these factors were readily apparent to the population and outside observers, it is reasonable to argue that the NDC recognized the potential with which the NPP could have competed against them, and in doing so, would have acted so as to overcome such challenges. Indeed, this appears to have been the case as the NDC engaged in what Elischer describes as the NDC incorporated prominent Akans into positions of party leadership so as to mitigate the threat of Akans voting en masse for the NPP (2013, 151, 254) . As all parties campaigned up until the November presidential election , tensions remained high, as al l sides believed they had credible chances of claiming victory. The 1992 Election and Afterward The final results of the inaugural presidential elections proved otherwise. With an approximately 54 per cent turnout of registered voters, Rawlings and the NDC captured approximately 58 per cent of the vote (obviating the need for a run off election), while 16 These pol two party system (Whitfield 2009b).
131 the NPP and Adu Boahen received approximately 30 per cent (ECG 2014) . 17 Incensed at the outcome, the NPP alleged fraud on the part of the NDC and INEC, claimi ng, among other things, that final tallies had been manipulated, that voters had been intimidated, that ballot boxes had been tampered and stuffed with NDC votes, and that the voter register was both outdated and inflated (NPP 1993). While election monitor s had been present and had given their approval of a relatively free and fair election, with some noted irregularities (COG 1992; Jeffries and Thomas 1993; Nugent 1995), the NPP, as well as other opposition parties, remained unmoved. Claiming that the pres the parliamentar at the polls, the NDC and other pro government parties captured 198 out of 200 parl iamentary seats. 18 Thus, when Ghanaian Fourth Republic was inaugurated on January 7, 1993, it began as an effective one party state, leaving many analysts circumspect regarding its democratic future (Gyimah Boadi 1994; Nugent 1995; Oquaye 1995; Bratton and van de Walle 1997; Green 1998; Gyimah Boadi 1999; Lyons 1999). 19 Despite its formal absence from government, the NPP remained active in Ghanaian politics. Shortly after the elections, the NPP, in conjunction with other opposition parties in the country, for med an Inter Party Coordinating Committee (ICC) that would effectively serve as a kind of shadow cabinet outside of government 17 Elischer argues that ethnic bandwagoning appears to have had an influence on the electoral outcome, as returns from the 1992 p residential election saw the NDC incorporating Akan votes from the Central and Brong Ahafo regions (2013, 151; see also Nugent 1995, 271). 18 Specifically, the NDC won 189 seats. 19 s of electoral fraud, such irregularities were not significant enough so as to alter the final outcome of the presidential election (Jeffries and Thomas 1993; Nugent 1995).
132 (Fordwor 2010). Fearing that the party would lose relevance in its absence from government, as well as would aid in perpetuating the rule of the NDC, the NPP additionally remained active by holding frequent press conferences, as well as by cultivating additional visibility through the media ( Jonah 1998, 100; Frempong 2012, 57) . In addition, the NPP remained active in challenging NDC policies, with its most dramatic victories occurring in the courts. Notable instances included successful challenges publicity in the state media, limitations on the right to assemble, and a proposal to declare Rawling national holiday ( Frempong 2007, 138 9). To be sure, the court victories of the NPP did prompt the Rawlings government to contemplate actions to bring the judiciary under executive control (Nugent 1995, 275). However, to have done so would likely have triggered a public backlash, as well as would have created factional divisions within the NDC. In the end, Jonah notes, The effect of these and other rulings of the Supreme Court was to establish clearly that the NPP while not in Parliame nt, still possessed the capacity to influence or check the political conduct of government. It was as if the NPP You got us at the pol you in the court ( 1 998, 101). In other words, while the NPP had little say in the crea and legal framework to its advantage. Furthermore, it is interesting to note that was effectively a one party legislative institution (with 198 of 200 members aligned with the government), the NDC nevertheless remained rather moderate in affairs, in terms of avoiding any constitu tional re engineering or passage of other laws
133 20 To be sure, while o engage in any substantive amending of the Constitution, despite near total control of government. Explanations for this reticence vary: some respondents suggested this may have been due to the inexperience of the Parliament, as the majority of the cham ber was composed of first time p arliamentarians; in addition, others noted that this may have been attributable to the lack of necessity in rewriting the Constitution, as the NDC had largely controlled the constitution writing process, and as a res ult, obtained most (if not all) of the provisions that it desired (Nugent 1995; Oquaye 2004; Riedl 2014) ; even others attributed the role of leadership within Parliament, as the first Speaker, Justice Annan, was believed to have been a moderating influence within the institution . 21 Lastly, part of the reason may additionally include that any change to the Constitution would have required a national referendum with 75 per cent assenting (RoG 1992 , 290 .4 ) . Nevertheless, the point remains that absence from Parliament, and despite the ability of Rawlings and the NDC to have initiated sweeping constitutional change to its favor, the Government of Ghana exercised restraint in its 20 19 96) was more likely to challenge executive policies than subsequent Parliaments that increasingly became characterized as party dissent and cross party voting became increasingly absent. 21 Int erview with lecturer at the University of Ghana School of Law, Legon, Ghana (September 21, 2013); interview with lecturer at the University of Ghana, Department of History, Legon, Ghana (January 30, 2014); interview with senior NPP Minister of Parliament, State Parliament House, Accra, Ghana (November 21, 2013); interview with senior NPP Minister of Parliament, State Parliament House, Accra, Ghana (November 26, 2013); interview with senior official at the Ghana Institute for Democratic Governance, Accra, Gh ana (November 25, 2013); interview with senior official at Ghana Center for Democratic Development, Accra, Ghana (November 25, 2013); interview with senior researcher at the Ghana Institute of Economic Affairs, Accra, Ghana (March 7, 2014); interview with senior NPP official, NPP Party Headquarters, Accra, Ghana (November 18, 2013).
134 range of actions for executing policy. As the inaugural elections pr oved problematic for between the NPP and NDC remained high and unresolved, President Rawlings was in need of a solution for this situation that threatened to undermine his r ule. 22 As highlighted in the previous chapter, Ghana continued to rely on international aid following the inaugural elections, and it was widely known that without such assistance, the government would have likely failed, or at the very least, would have le d the country back into a recession reminiscent of the kind experienced a decade earlier during the initial years of the PNDC (Nugent 1995, 187) . In addition, as the NDC government was eyeing the 1996 elections and wanted to retain as many seats as possibl e in the Parliament, it was imperative that the government find a means to overcome the political impasse so that it could campaign free of controversy. 23 In sum, though not formally in power, the NPP maintained the ability to 24 As the NPP effectively known to have constituted a significant portion of the electorate (or at least in the eyes of the NDC), Rawlings and his governmen t knew they needed to govern carefully as international donors continued to pressure the government to reconcile the political impasse, and as the NPP retained the ability to frustrate government policies through protests, court 22 Interview with senior NPP Minister of Parliament, State Parliament House, Accra, Ghana (November 5, 2103); interview with senior lecturer (2) at the University of Ghana, Depar tment of Political Science, Legon, Accra (October 4, 2013). 23 Interview with lecturer at the University of Ghana, Department of Political Science, Legon, Ghana (October 30, 2013). 24 Interview with senior lecturer (1) at the University of Ghana, Department of Political Science, Legon, Ghana (October 1, 2013).
135 judgments , and threats of g eneral strikes. To be sure, while the NDC retained the prerogative to dispel opposition activities through use of state security services, pressure from international donors nevertheless served to constrain any repressive tendencies within the government, thus securing a rather permissive environment in which the opposition could not only operate, but thrive as well. 25 In an effort to overcome the st Co mmission agreed to form an Interparty Advisory Committee (IPAC) composed of representatives of each of the main political parties of the country that would be headed by the Chairman of the Electoral Commission (Asante 2013) . Initially proposed by the Commo nwealth Observer Group following the inaugural presidential elections, IPAC served to facilitate interparty dialogue and coordination regarding the electoral rules of the game. Through this process, IPAC facilitated elite consensus among the political part ies, and served as the impetus for initiating electoral reform (Frempong 2007) . , as well as representatives from the donor community, were free to propose possible reforms to the ctoral system, discuss their proposals with other party and donor representatives, and have their recommendations considered by the Chairman of the Electoral Commission. While IPAC remained an informal institution and did not possess any formalized legal a uthority, because any such recommendations made from the committee came as a result of consensus based dialogue among the main participants in the political process, and because the Electoral Commission possessed 25 Interview with senior lecturer (3) at the University of Ghana, Department of Political Science, Legon, Ghana (October 1, 2013).
136 constitutionally guaranteed independence fr om the executive over its day to day e instituted and put into force. Examples of IPAC recommendations being instituted by the E lectoral C ommission include holding presidential and parliamentary elections on the same day, using transparent ballot boxes, improving the voter registration process and registry , and providing voter identity cards to reduce concerns of electoral fraud. Above all, the position in to the electoral system, t hus giving the opposition a stake in the system and an incentive to participate in future elections. By incorporating the opposition into the electoral system and enacting additional c trajectory became further secured. Despite still losing the subsequent presidential election in 1996, the political opposition accepted the results and acknowledged its defeat at the polls. More importantly , the opposition succeeded in obtaining a signif icant presence within Parliament , capturing 67 seats (61 for the NPP) (ECG 2015) . By and large, the 1996 elections were considered to have been conducted free and fair democracy (Lyons 1999). By this time, G political climate became conducive for fostering alternative and critical perspectives about politics in the country (Agyeman Duah 2012). To be sure, this is not to suggest that prior to 1996 civil society had completely vanished, nor is it to suggest that civil organizations, such as the Ghana Bar Association democratic opening civil society and its constituent civil organizations were considered
137 to have been significantly marginalized, not only as many pre opening organizations redirected their energies within political parties and political society following the democratic opening in 1992, but also as the NDC succeeded in creating and coopting and the Ghana Private Roads Transport Union (Gyimah Boadi 1994b). 26 Furthermore, Gyekye Jandoh (2006) additionally notes that civil society did not emerge as a fact or for democratic change until at least som etime after the 1992 elections . Nevertheless partially fostered by international support from donors and Western aid agencies (Gyekye Jandoh 20 06, 158), its overall successes would not have occurred without the emergence of a facilitative political environment brought about through the prior efforts of the domestic political opposition. As the citizenry observed the opposition achieve significant judicial victories against the government (resulting in substantive policy changes), and furthermore saw t he NDC led government exhibit restraint by holding itself to the laws prescribed in the Constitution, the return of civil society as a force for demo cratic change became inevitable. democracy posit the role of civil society and IPAC (including electoral reforms) as pushing the country in a democratic direction (Ninsin 1998; Gyimah Boad i 1999; Haynes 2003; Gyekye Jandoh 2006; Frempong 2007; Agyeman Duah 2008; Abdulai an electoral environment in which all major players had come to a cons 26 regarding the general dissipation of civil society following a democratic opening (1986, 55 6).
138 of the game. Furthermore, this process proceeded forward through the efforts of a vibrant and increasingly active civil society that was eager and willing to re insert itself into the political dialogue for the sake of advancing democracy in the country. The point is, however, that rather than serving as the primary drivers of democracy, both IPAC ng , and in rapprochement with the opposition . Conclusion Not to be misconstrued as disputing previous arguments emphasizing the influence that IPAC and civil society had i , instead, this chapter demonstrates that at the core of these arguments rests the common denominator of the political opposition as facilitating the emergence of such factors in the first place. The political opposition in Ghana survived the era of PNDC rule and party political tradition. I ocracy benefitted from an interesting association between the NPP and the Akan. Though it would be disingenuous to characterize opening, ethnicity in Ghana in conjunction with the political tradition s arguably prevented democratic regression by means of ma intaining a potential popular base of opposition support , contrary to what would be expected from traditional explanations (Horowitz 2000) . In addition, while the NDC successfully mitigated the threat of ethnic politics by incorporating a sufficient number of Akan votes in order to win the 1992 presidential election, this does not mean that the NDC had effectively
139 instead served to compel the party to become more programmatic in its party platform and to increase its national appeal across all ethnic communities. (Elischer 2013; see also Riedl 2014). It is t his consistent opposition presence, even when formally absent from government , that helps explain why IPAC had been initially instituted, as well as helps account for the return of civil society by 1996. I nterviews among the political elite part problem for Rawlings and his administration: not only would his ability to govern be challenged on the notion that his presidency was based on a perceived act of electoral challenged as well. 27 In order to overcome this challenge, the newly reformed Electoral Commission agreed to recommendations from the Commonwealth Observer Group to institute IPAC in order to bring back a sense of legitimacy to the political process. Funded in large part by the donor community, IPAC succeeded in serving as a venue for open discussion of electoral reform among the major parties, the chairman of the E lectoral C ommission , and donors (Asante 2013). More importantly, IPAC provided the trust neede d to establish elite consensus electoral rules and laws. Thus, it the parliamentary elections, IPAC would not ha ve been established. Though opposition 27 Interview with senior lecturer (1) at the University of Ghana, Department of Political Science, Legon , Ghana (October 1, 2013); interview with senior lecture r (2) at the University of Ghana, Department of Political Science, Legon, Ghana (October 4, 2013); interview with Researcher from the Ghana Institute of Economic Affairs, Accra, Ghana (March 7, 2014).
140 parties may have succeeded in generating a presence in Parliament had they participated in the inaugural parliamentary election, because government retained a heavy executive dominance over Parliament, and because the ir presence would have likely been as a minority, their ability to have brought about electoral reform would have been considerably less than through participation in IPAC as an equal member among political parties, the E lectoral Commi ssion , and the donor community. 28 In this respect, it would appear that the opposition boycott provided certain democratic dividends for the country and its citizens . In sum, by continuing to operate as a significant political force in opposition to the NDC , regime in a democratic direction. 29 To be sure, while initial actions from the opposition brought political tensions to a climax and fueled further distrust between the governmen t and the opposition, such a strategy nevertheless succeeded in producing the institution that was instrumental in obtaining reconciliation between the main parties, as well as consensus regarding the management of the electoral rules of the game. Moreover , well as its successful challenges to government policy, maintained an environment where political dissent was not only possible, but was occasionally successful in changing government policy. To be sure, while it would be disingenuous to claim that it reasonable to argue, 28 It is generally understood that the NDC victory in th e November, 1992 presidential election would have likely influenced the outcome in favor of the NDC in the December parliamentary elections. 29 goals were second ary to obtaining power is debatable (Riedl 2014). The point is, however, that despite dividends for the country.
141 however, that in conjunction with pressure exerted by the int ernational donor community constrained tendencies, push ed for continued electoral improvement , and ultimately assist ed in the
142 CHAPTER 5 EXECUTIVE AGEN CY While Chapters 3 and 4 demonstrate the extent to which the international environment via international leverage exerted by international donors, as well as the degree to which recipient countries were dependent on such assistance and the tic political opposition constrained the inaugural government from engaging in any potential undemocratic behavior that would have served to undermine nevertheless remained dependent on an additional factor. Were international aid, leverage, and a sustained and significant domestic political opposition sufficient for producing a democratic traject ory, such a wholly structural account would have neglected the amount of contingency present at the time . I t would trajectory, especially as African elites and their actions on political processes have often been instrumental in influencing politics on the continent (Bratton and van de Walle 1998; VillalÃ³n and VonDoepp 2005; Hyden 2006; 2013; Mustafa and Whitfield 2010). In short, while international pressure and domestic political opposition incentivize d the country to follo w a democratic trajectory, the environment at the time of democratic opening nevertheless remained highly contingent and open to a wide range of actions affecting As a new institutional equilibrium had juncture that produced the democratic opening in 1992 , President Rawlings and his NDC led government ultimately retained the ability to decide which path the country
143 would take by means of executin g policies that could further democracy in the country , or repress it altogether. The decision to either allow or prohibit further democratization, this chapter argues , depended on whether Rawlings and his government recognized the limits to their rule tha t had been previously provided through international and domestic pressure , including the significant amount of leverage possessed by international donors in negotiating the terms of their aid, and the pre sence of a sustained and significant domestic political opposition . Utilizing the reasonable assumption that governments wish to stay in office for as long as they can , I argue that the Rawlings administration faced a two level decision regarding the quest ion of democratization and the nature of its authority, with the outcome of the latter influencing the decision of the former. At the core of the argument , the Rawlings administration in its first term faced an important question concerning the desired sco pe and duration of its authority: would Rawlings seek to (re)establish absolute authority reminiscent of the PNDC era, or would he instead allow for continued democratic reform? In answering this question, however, it is important to note that Rawlings and his government faced certain constraints in the first months of their administration (in addition to the constraints detailed in Chapters 3 and 4) . First, as the country had recently undertaken a democratic opening led by Rawlings and the (P)NDC , it would have been inherently problematic to have pursued an agenda of reconsolidating commitments to democracy and decentralization of authority (Rawlings 1990; 1991;
144 Shillington 199 2, 165 ), it would have threatened to fractionalize the NDC, and would have furthermore likely reignited the domestic and international forces that had recently pushed for regime change in the first place . 1 Secondly, it is also important to recall that des pite initiating a democratic opening and a return to constitutional rule in 1993, Rawlings and the NDC continued to receive appeals for further democratic reform from domestic and international forces, including Western donors (Frempong 2007; Whitfield 200 9 b ; Gyimah Boadi and Yarkah 2013, 263) and the domestic political opposition (Oquaye 2004) . In addition, as the NPP challenged the democratic legitimacy of the NDC government by alleging that the 1992 presidential lings and his government faced immediate demands for further democratic reform. To have reconsolidated his authority in a manner similar to w hen he was Chairman of the PNDC would have likely resulted in suspensions of international aid, would have provoked domestic unrest, and would have furthermore risked precipitating a political and economic breakdown (Nugent 1995, 187). In sum , for either democracy or absolute rule , the environment following the democratic open ing constrained the initial and his government . Reestablishing executive authority reminiscent of the PNDC era including concentrated power and indefinite rule thus remained a nonstarter as a matter of practicality . As a result , as the political 1 Justice Annan, for example, was well known to have been an influential political moderate within the (P)NDC that supported the return to democracy in 1993 (Dadzie and Ahwoi 2010).
145 absolute authority and indefinite rule, Rawlings was thus encouraged to pursue his interests by working within the confines of an ostensibl y democratic system. S ensing that international leverage and domestic political opposition would likely prevent his administ ration from consolidating power, and sensing that such an act would have additionally served to destabilize the country to the point where the possibility of a coup would have been both likely and successful, I argue that Rawlings and his administration recognized such constraints and as a result pursued an agenda that served to prolong his tenure as chief executive by working within a democratic system, including tole rating continued democratic reform . In other words, by recognizing the challenges to his authority and the likely consequences were he to disregard them, Rawlings avoided repressing the democratic demands from the politica l opposition and tolerated continued reform in return for political stability and an increased duration of rule , including a likely reelection in 1996 . To be sure, this is not to suggest that Rawlings never had agency in this process and was merely reactin g to the structural circumstances he was given. In reality, all choices remained available to Rawlings, as well as to those in his service. At the same time, however, the structural circumstances that Rawlings faced nevertheless made some decisions more li kely than others, as they would have likely affected his executive interests highlighted above. Thus, by recognizing the challenges to his authority and choosing to work within the confines of a democratic system, Rawlings exercised his own agency by pursu ing the path that he believed was most likely to maximize his interests concerning executive power and tenure.
146 However, by pursuing an agenda that was initially designed to serve immediate interests , such actions ultimately sowed th e seeds of their future electoral defeat by allowing the unabated growth of an electoral environment where the political opposition could freely compete . By the 2000 general elections it was too late for the NDC : the path that had been laid through Rawling produced a new set of structural incentives that vested all political actors to continue participati ng within the electoral system. As the polls demonstrated an opposition victory in the presidential election , Rawlings, the NDC, and the country had too much to lose through abrogating the election . In sum, by sacrificing executive authority in allowing for continued democratic reform, that is, by pursuing domestic stability as the means to maximize his duration of rule , actions u ltimately placed Ghana on a democratic trajectory. The Rawlings Factor: Pragmatism In making the argument that Rawlings and the PNDC tolerated democratic reform in exchange for an increased tenure of rule as president , it is important to note the history o f pragmatism expressed by Rawlings throughout his tenure as Chairman of the PNDC prior to the democratic opening . In short, Rawlings had rather revolutionary aspirations when coming to power in 1981 as the leader of the PNDC (Rawlings 1983 a ) , how ever in practice, he was rather flexible in his ideology and style of governance (Nugent 1995) . Shillington additionally observes how This flexibility helps explain why Rawlings and t he PNDC accepted democratization in 1992, and it also helps explain why Rawlings continued to tolerate democratic reform in his inaugural administration.
147 First, t he at the time of the 1981 coup and the actual out come th inauguration of the Fourth Republic in 1993 is rather striking . Coming to power for a second time in 1981, Rawlings 1983a, 1; Nugent 1995; Oquaye 2004). Claiming at the time was rooted in inherent contradictions of multiparty democracy (including corruption, elitism, and cronyism) , Rawlings and the PNDC sought to implement a n ew form of governance that was, in th eir words, based councils throughout the country that could, in essence, govern their local affairs in an autonomous fashion through a form of direct democracy. Shortly aft er coming to power, the PNDC created Defense Committees ( sub divided into Workers and Peoples Defense Committees) whose objectives were to largely socialize and organize local communities into the structure of the PNDC. Largely influenced lution 1981 would result in a concrete change restructuring the attitudes and concepts and a redefinition and implementation of the norms and goals of society (Oquaye 2004, 1 43). In practice, however, the Defense Committees proved to be problematic . In urban areas , Defense Committees were attributed to have provoked social unrest and increased labor disputes between workers and managers (Nugent 1995) . Workers Defense Committee s, for example, came into conflict with trade unions as the two bodies claimed to be the legitimate representatives of workers and their interests (Shillington 1992, 88 9). In addition, the governance of Defense Committees became
14 8 increasingly arbitrary and unsystematic, resulting in confusion among the citizenry as to who had legitimate authority. In rural areas, Defense Committees struggled to penetrate the countryside , resulting in conflict with traditional authorities (i.e. chiefs and local elites) that had remained influential. As early as 1984 the PNDC realized that the continued practice of having semi autonomous Defense Committees operate throughout the country was proving too problematic to handle. Furthermore, c oncerned that the c ommittees would jeo pardize agreement s the regime had made with the IMF and World Bank in regards to the count (ERP) , the PNDC recalled the Defense Committees and reorganized them under Committees for the Defens e of the Revolution (CDRs) that wou ld be more accountable to the central government (Nugent 1995, 136 9). While much of the same architecture remained in the makeup of the CDRs, a noteworthy change included the fact that rather than being semi autonomous structures, the CDRs were now subser vient to the PNDC and were subsequently regarded as an extension of their authority. Secondly, it is important to note that the PNDC engagement with the IMF and World Bank, as well as its agreement to instituting the ERP , was a significant compromise fro m its initial goals. Indeed, part of the reasons as to why Rawlings staged the 1981 coup included the then Limann government engagement with international donors on the issue of economically reforming the country , including devaluing the national currenc y , reducing the civil service, and implementing other structural adjustment measures . Such a move was controversial for Rawlings: prior to signing the ERP in December 1992, Rawlings was accused by previously close allies, Chris Atim
149 and Alolga Akata Pore, of having betrayed the revolution by even considering devaluation as part of the economic reform program ( Shillington 1992, 103 5). This animosity culminated in a conspiracy to oust Rawlings in late October, and an actual coup attempt a month later in Nove mber. 2 For Rawlings , as the chief executive of the PNDC, to have negotiated with international donors and to have enacted many of the same policies that ( at least in part ) precipitated the 1981 coup demonstrates that when pushed, Rawlings was willing to je ttison ideological principles for pragmatic solut ions if it would perpetuate his tenure as chief executive . In sum, throughout the PNDC era Rawlings and his colleagues often altered their strategy to accommodate changes in the international and domestic en vironment. Those that did not agree with such a strategy often did not remain in the PNDC, whether by resigning in protest or being removed by Rawlings. The PNDC made it widely known that it was willing to maintain flexibility in its style of governance, i n spite of the revolutionary rhetoric espoused by its leaders at the time of Rawlings 1981 coup. willingness to reinstitute multiparty politics in 1992, a condition to which Rawlin gs had previously been vehemently opposed , and was reluctant to implement . Nevertheless, through continued domestic and international pressure, as well as the need to stay dem ocracy was manageable . A s Nicolas Robertson, a former U.S. diplomat who served in Ghana, recounts in his experience interacting with President Rawlings and his administration, 2 Subsequent to the coup attempts in 1982, Rawlings purged all six other founding members of the PNDC by the end of 1983 ( Shillington 1992, 106, 135).
150 He [Rawlings] was, for one reason or another, open to small suggestions and we just kept forcing him, step by step, do this, have an election, have a fair election, You have a press which is increasingly free, increasingly vocal. You h you had lots of organizations that with a little b it of support from us began playing independent roles and building a society that politically and economically line though, you saw a country that politically and economically was increasingly open (Kennedy 2009, 55). In other words, as Rawlings was ultimately willing to sacrifice core ideology for the sake of prudence and longevity, this pragmatism influenced the manner in which Rawlings chose to tolerate demands for continu ed democratic reform, and continued through his tenure as leader of the NDC and president of the country . Factionalism noting the degree of factionalism within the NDC that prevented Rawlings from exercising unilateral authority in any extreme direction. While the NDC (between 1992 1996) would be reasonably characterized as a personalist party built upon persona (Elischer 2013), the NDC remained composed of many ide ological elements that spanned the Danquah Busia and Nkrumah political traditions (Bob Mil l iar 2012, 584). In addition, despite possessing a super NDC allied itself with smaller Nkrumahist parties, including the Na tional Convention Party (NCP) and the Every Ghanaian Living Everywhere (EGLE) Party. Dubbed the retained its influence over the alliance by having Rawlings fill the presidential slot, while the NCP filled the position of vic e president with its candidate, Kow Nkensen Arka a h. On the other side of the political spectrum, the NDC also incorporated non Nkrumahist elites . Most notably, Justice Daniel Francis Annan was a rather influential
151 figure who was held in high esteem by a wi de sw ath of Ghanaian society. He was additionally well known to have been a political moderate within the NDC, as well as within the PNDC before the 1992 democratic opening (Dadzie and Ahwoi 2010) . tendencies in its early years, Justice Annan remained a party stalwart while also being a well known adherent to the Danquah Busia political tradition (Dadzie and Ahwoi 2010, 88). Justice Annan possessed a long relationship with Rawlings. A former High Co urt and Appeals Court judge, he had served with Rawlings in the PNDC and was widely regarded to have been the de facto deputy chairman of the PNDC (Dadzie and Ahwoi pragmatism i multiparty democracy, Justice Annan was most notably remembered for heading the NCD, and for leading its regional seminars in 1990 that culminated with the NCD report enco uraging a return to multiparty politics. In addition, Justice Annan was credited with convincing Rawlings that would he return the country to multiparty commitment to regim e change ( Dadzie and Ahwoi 2010, 128) . 3 With the help of an ostensible political alliance between himself and Captain Kojo Tsikata (another popular figure within the PNDC ), Justice Annan was known to have acted to curb radical tendencies (Dadzie and Ahwoi 2010, 95). As the first 3 Interview with senior lecturer (1) at the University of Ghana, Department of Political Science, Legon, Ghana (October 1, 2013); interview with senior NPP Minister of Parliament, Parliament House, Accra, Ghana (November 5, 201 3).
152 executive excess. As best explained by Dadzie and Ahwoi, Justice Annan was clearly of the view that Parliament should play a complemen tary role to the Executive and should not be a stumbling block in d not have its way all the time (2010, 116). In sum, while Rawlings ret ained a generous amount of discretionary authority as president of the country, and additionally enjoyed being the leader of the NDC, his conduct was nevertheless not without its own internal constraints. As the NDC largely inherited the factions previousl y united under the PNDC banner, and additionally incorporated other factions from the two main political traditions, the party had to conduct its affairs carefully in order to retain all of its disparate elements, all the while maintaining a general orient ation towards the Nkrumahist tradition. 4 The challenges in victory in 1992 as , over time, the NCP began to feel marginalized by factors including its lack of cabinet appointments, amba ssadorial appointments, and general lack of influence in the President). 5 These differences manifested themselves as Vice President Arkaah began to openly express his policy differences with Rawlings and the NDC (Bob Mi l liar 2012, 585). 6 Such disagreements economic policies, including the attempt at instituting a 17.5 per cent value added tax (Ayee 2007, 174; 4 Milliar 2012, 584; see also Dadzie and Ahwoi 2010). 5 Africa Confidential , 18 November, 1994, 35(23): 3 4. 6 Africa Confidential , 19 November, 1993, 34(23): 5 6.
153 cited in Bob Mil l iar 585). 7 Th e relationship between Rawlings and Arkaah degenerated so much so that Rawlings even assaulted Arkaah during a cabinet meeting in 1995 (Frempong 2007, 143). As a result, the NCP withdrew from the Progressive Alliance, whereby the NCP sided with the NPP . In terestingly, Arkaah subsequently went on to serve as the running mate for the NPP in the 1996 presidential election . While the differences between Rawlings and Arkaah demonstrate the difficulty Rawlings faced in mana ging the various factions within the NDC and its allies , the point is that the episode shows how Rawlings needed to avoid alienating influential figures within the NDC, such as Justice Annan. In addition, Rawlings also needed to manage the internecine disp utes within his own party, including rivalries between Kwesi Botchwey ( finance minister) and Tsatsu Tsikata (chief executive of the Ghana National Petroleum Corporation and cousin to Kojo Tsikata, a close ally of Rawlings), as well as between Kojo Tsikata and his wife, Nana Konadu Agyeman Rawlings. 8 As a result of these challenges , Rawlings needed to govern in a delicate manner and avoid pursuing policies that would be viewed as radical or excessive to the factions comprising the NDC , as well as the Progressive Alliance . Fortunately for Rawlings, the Arkaah Affair did not result in an NDC loss in the 1996 elections. H owever, it is nevertheless reasonable to argue that Rawlings would have been careful to not repeat such a mistake, as t o have don e so would likely have been disastrous for his administration, and would have benefit ted the political opposition in the process . 7 Africa Confidential , 26 Ma y, 1995, 36(11): 1 2. 8 Africa Confidential Africa Confidential , 17 November 1995, 36(23): 4 5.
154 Determining the Nature of Executive Power m and factionalism, t he decision to either allow or prohibit further democratization in the period following the , in essence, depended on whether Rawlings and his government recognized and acknowledged the likely constraints to co nsolidating executive authority, manifested through international and domestic pressure. As the country had entered a critical juncture following the PNDC and reconstitute a multiparty democracy in the form of the F ourth Republic, undefined . This uncertainty was further to boycott the inaugural parliamentary elections. As a result, contingency remained high throughout this period and executive agency retained an influential role in determining the ultimate direction the country would follow . As Rawlings possessed a preponderance of discretionary authority as the chief executive officer, and as the executive enjoyed a constit utionally 58), trajectory. Assuming that Rawlings wished to stay in power as long as possibl e , I argue that at the beginning of his term, Rawlings faced a decision concerning the nature of his authority as c hief executive of the country . In short, in an ostensibly democratic environment where executive authority no longer retained both concentrat ed power and indefinite rule (for reasons explained above) , and where a dis satisfied domestic opposition continued to call for further democratic reform , Rawlings and his administration needed to decide whether they ought to resist such demands for greater
155 democratization from the opposition and international donors (and retain as much executive authority as possible ) , or whether they should instead satiate such demands by allo wing further democratic reform and reincorporating the opposition into the electo ral system , thus regaining domestic and international legitimacy in the process . In other words, the decision came down to whether Rawlings and the NDC would either repress democracy in their first term, or tolerate it instead. Repression: Costs and Benefi ts enjoyed a rather privileged position within government, for reasons stated above . Beyond instituting a constitution that afforded certain political rights and civil liberties, with few exceptions the composition of the inaugural government remained largely the same as its non democratic predecessor in the eyes of many Ghanaians in the country . By p as well as control ov Rawlings and the NDC retained the ability to enact any piece of le gislation they desired, with the exception of any overt constitutional reengineering. However, maintaining such a status quo would come at a cost. As highli ghted in previous chapters, Ghana remained dependent on international aid throughout the period of the democratic opening. In addition, the outcome of the inaugural elections, including the boycott of opposition parties, produced a crisis of legitimacy for Rawlings and his government . As international donors were weary over extending funds to a government that was perceived to have been elected under problematic circumstances, and as many within the country (particularly among the political opposition) beli eved that the election had been fixed in favor of the NDC , maintaining a high degree of
156 discretionary authority , especially through force, would come at a cost: the likely suspension (or even withdrawal) of international aid, and the unabated growth of an unstable political environment. First, the international environment at the time of the opening of the Fourth Republic remained highly favorable to democratization. With the fall of the Berlin Wall in 1989, the collapse of the Soviet Union occurring shortl y thereafter, and democratic transitions occurring in former Soviet blocs, the issue of democracy promotion became highly salient for the international donor community. turn in 1989 suggested that dem ocratization would be a likely against advancing regime change. Furthermore, a s the continent became strategically marginalized through the collapse of the Cold War, count ries like Ghana could no longer enjoy international aid from Western donors (most notably, the U.S.) without offering some kind of democratic improvement in exchange (Schraeder 1994; Lancaster 2007; Levitsky and Way 2010) . Secondly , the domestic political environment remained rather tense following the subsequent withdrawal from government. Despite losing the main opposition party, the NPP, still provided a credible threat to the future the NPP was well known to have been associated with the group, the Akan (Elischer 2013). In addition, the NPP succeeded in capturing at least 30 per cent of the electorate in the ina ugural presidential election . As a result, it is reasonable to suggest that the threat of domestic disorder and unrest remained
157 significant in the minds of the national government , especially as the opposition saw them as an illegitimate occupier of the pr post election manifesto, The Stolen Verdict (1993). 9 In sum, maintaining the status quo (that accorded a high amount of discretionary authority and resources to the president) would have inherently required active repression of the political opposition, and as a result, would come at a cost for Rawlings and the NDC. Without continued access to international aid the country would have likely fallen into economic recession, which would have only been made worse by th e challenges in governing a population consisting of a significant proportion believing the government to be illegitimate. 10 In other words, while Rawlings and the NDC could have lls for further democratic reform , to have done so would have been at great peril . N ot only would it have likely eroded any political and economic gains made , it would have additionally threaten tenure as president by reintroducing an unstable environment that, in the past , had facilitated coups and the breakdown of democratic rule . 11 9 In addition, as significant rioting occurred in Kumasi, Sunyani, and Tama le following the results of the presidential election, and as no fewer than four bombing events occurred within Accra, the threat of political violence remained significant for Rawlings and his administration (Nugent 1995, 234). 10 Though not to suggest tha t every person who voted for the NPP in the 1992 presidential election subsequently regarded the Rawlings administration as illegitimate, as the NPP nevertheless garnered no fewer than 30 per cent of votes in that election, it is reasonable to suggest that have been regarded as constituting a significant proportion of the population. 11 for Rawlings (Nugent 1995; Agyeman Duah 2012). As a result, it is reasonable to suggest that any actions taken that would have brought the country into economic decline would have likely increased the Africa Confidential , 26 May, 1995, 36(11): 2).
158 Toleration: Costs and Benefits At the same time, however, Rawlings and the NDC faced an alternative choice. tolerat e the such demands (or at least a sufficient amount thereof ), the government would abate the amount of political tension permeating throughout the country and regain a sense of legitimacy both to the democratic process and to their rule . Furthermore, in doing so the government would continue to enjoy uninterrupted access to needed international financial assistance. y require sacrificing executive power to some degree , such an act would likely serve to prolong his tenure as president of the country by promoting an increasingly stable and less coup prone political environment, as well as effect tenure through his first term. Furthermore, by maintaining access to international aid, Rawlings also maintained a strong influence in directing development aid that served to enhance his prospects for reelection in 1996 (Brigg s 2012). In sum, by choosing to tolerate the op would effectively exchange his potential for exercising absolute authority in the country in return for a longer term as president. F igure 5 1 provides a summary of the costs and be nefits associated with each of
159 Consequences of Action Benefits Costs Decision by Rawlings Repress High Executive Power Less Aid Domestic Instability Coup prone environment Tolerate Continued Access to Aid Do mestic Stability International and Domestic Legitimacy Loss of Executive Power Figure 5 1: Repression Versus Toleration Rawlings Chooses to Tolerate In much of the same manner that Dahl hypothesized democratic transitions as occurring when the costs of r epression outweighed the costs of toleration (1971, 15 6), democratic trajectory depended on whether the costs of repressing the ed the costs of tolerating them. In other words, valued maximizing his discr etionary authority as president or whether he valued maximizing his tenure instead. 12 Sensing that international donors and the domestic political opposition would likely act to prevent his administration from consolidating power and that attempting so would furthermore create a situation in which his extra legal removal as president would not only be possible but would also be plausible Rawlings ultimately pursued an agenda t o prolong his duration of rule through the toleration o f continued democratic reform 12 To be sure, maximizing executive authority in this instance would be based on the assumption that one would prolong their tenure through overt repression.
160 (Agyeman Duah 2012, 167). As early as his inaugural presidential address, Rawlings was noted to have been rather conciliatory to the opposition, which was a notable departure from his rhetoric during the campaign and the 1992 transition (Frempong 2012, 55 Rawlings expressed his commitment to work within the bounds of the law and the Constitution, and to build consensus with the opposition. In spite o f initial failures to build consensus with the opposition through the Interparty Coordinating Committee, (Jonah 1998, 100), the government nevertheless remained notably restrained in its actions to oppose the NPP. Indeed, t n, the country experienced a gradual but continuous expansion of civil and political liberties ( Figure 5 2 ) . In particular, the country experienced increased freedoms with respect to the media and civil rights, and additionally oversaw a dramatic expansion of private media, including newspapers and radio stations (Gyimah Boadi 1999, 174). Furthermore, by allowing subsequent elections to be monitored by both domestic and international election observers, subsequent elections became further secured from elect oral manipulation, should the results have ever turned against the incumbent party.
161 Source: Freedom House 2014 Figure 5 2: Political and Civil Rights in Ghana, 1992 2000 In addition, through the recommendation of the Commonwealth Observer Group and the a a rapprochement with the domestic political opposition by reincorporating them into the electoral system through the Interparty Advisory Committee . As an in dication of the major pa rties representing the two main political traditions participating in a consensus the incorporation of the political opposition into IPAC signified a new start to the relationship between the governm ent and opposition . Furthermore, by incorporating tions into actual electoral law, this process signified Rawlings widespread legitimacy and political stability in the country.
162 In sum, when faced with the choice between preserving his high amount of discretionary authority or his tenure of office, Rawlings valued the security of office more and chose to retain it through tolerati ng the opposition and its demands for political reform, thus regain ing domestic and international legitimacy in the process . To be sure, w hile consolidated executive authority could have provided a longer term of office by means of overt repression , such a course of action would not have been likely to succeed , for reasons explained above . As a result, the choice for Rawlings came down to a matter of which executive interest he would prefer to pursue: executive tenure or executive authority. In addition, t h is is not to be misconstrued as arguing that Rawlings had been converted to democracy or had otherwise embraced democratic values, nor is it to surmise that Rawlings secretly desired to reinstitute authoritarianism . Rather, it is only to demonstrate that e ven if Rawlings had maintained any undemocratic intentions while serving as president, to have executed any such desires would have likely had grave consequences for both himself and the country, and would have additionally threatened the security of his t enure as chief executive. In doing so, Rawlings pursued a rational strategy that valued his tenure as pre sident in exchange for sacrifices to the scope of his presidential authority by accommodating demands for additional democratic reform. Locking In Ghan By choosing to tolerate demands of democratization in exchange for an increased duration of rule as president of Ghana, Rawlings locked in the country to proceed along a democratic trajectory for the following two decades. Followi ng the facilitated through participation in IPAC the main political parties reestablished elite consensus among
163 the electoral rules of the game. In addition, they established a sense of trust a mong each other regarding the manner in which they would conduct their political affairs, population would select their leaders and that any inconsistencies or grievance s produced from this process would be resolved through legal recourse. By the time of the 1996 elections it was clear that the critical juncture democratic opening in 1992 had closed , and in its wake had produced a new set o f political incentives that had been internalized by all of the main political parties in the country, including the NPP and NDC ( Ayee 1997 ; Jeffries 1998 ; Gyimah Boadi 1999 ; Lyons 1999 ). To be sure, while incumbent advantages in publicity and finance cont inued to challenge the democratic credibility of the regime, opposition parties nevertheless recognized the outcome of the election as legitimate and continued to press for their desired reforms through the courts and parliamentary process. In addition, op , and even publicly congratulated him on his victory inaugural address continued to emphasize reconciliation and his tolerance of differing political views (Ninsin 1998, 217). In addition, the 1996 elections also coincided with the return of civil society as an active voice for democracy in Ghanaian politics (Gyimah Boadi 1999, 173; Agyman Duah 2012, 147), whose presence continued to expand thereafter. As civil society organizations sponsored voter education initiatives, organized community workshops on improving democratic governance in the country, engaged in election monitoring , provided policy analysis , reported on government activity , and even spon sored political
164 debates between parliamentary and presidential candidates, civil society acted as a positive force in propelling the country along its democratic trajectory (Gyekye Jandoh 2006; Abdulai and Crawford 2010; Arthur 2010) . However, in contrast to previous ion to tolerate additional democratic reform . widely recognized until its trajectory was tested during t he 2000 general elections. In the run up to the election, questions emerged as to whether Rawlings would adhere to the term limit as president of the country (Frempong 2007, 151 152). As other African leaders had attempted to abrogate preside ntial term limits, and had even succeeded in countries as Gabon, Burkina Faso, Guinea, Togo, Uganda, and Chad, many wondered whether Rawlings would attempt to do the same. term limit or abrogate it altogether remain a mystery. H owever , the idea of extending his term as president was nevertheless (Frempong 2007, 151 2). In response , public reaction was immediate and harsh. In the end, Rawl ings chose to step down and abide by the term limits. Nevertheless, his influence remained as he effectively appointed his own successor, John Atta Mills, by pressuring the NDC to nominate Mills at a public rally in Swedru in 1998 . Mills was widely believe d to have been an attempt by Rawlings to maintain influence within the
165 d eclaration to consult Rawlings 24 hours a day were he to be elected president of the country ( Ayee 2 002, 172 ; Agyman Duah 2012, 168, 212 ). 13 The 2000 elections produced an NPP victory and defeat of the NDC. For the first executive power had been peacefully exchanged from one political party to another. To be sure, reports su ggest that Rawlings sympathizers had attempted to circumvent the electoral results by causing a state of emergency, but were prevented by a refusal of senior military officials to carry out the order ( Agyeman Duah 2008, 181 2; 2012, 168 9). Nevertheless, t he NDC was compelled to acknowl edge its defeat and relinquish the p residency: to have done otherwise would have likely failed, given the factional ri valries in the NDC at the time. In addition, a failed military coup in neighbo ed the risks and dubiousness of executing such a n act (Smith 2002, 525; Haynes 2003, 65). More importantly, a refusal to recognize the results of the 2000 election would have plunged the country into economic and political turmoil, and would have furthermo re nullified all that which Rawlings and the NDC had worked to accomplish in the eight years prior , including Raw accomplishment: the 1992 Constitution. 14 All of the political actors had a vested interest 13 r backfired, exacerbating fissures and factionalism within the NDC that would be attributed in its 2000 electoral defeat (Ayee 2007, 174). 14 Interview with Senior Lecturer, University of Ghana, Legon Center for International Affairs and Development, Legon, Ghana (September 11, 2013); interview with Senior Lecturer (2) at the University of Ghana, Department of Political Science, Legon, Ghana (October 4, 2013); interview with Research Fellow, University of Ghana, Institute of African Studies, Legon, Ghana (Ma rch 30, 2014); interview with senior NPP Minister of Parliament, Parliament House, Accra, Ghana (November 21, 2013); interview with Researcher from the Ghana Institute for Economic Affairs, Accra, Ghana (March 7, 2014); interview with former CPP Minister o f Parliament, CPP Party Headquarters, Accra, Ghana (January 20, 2014).
166 in maintaining the political system, and as a result, the 2000 elections demonstrated the . Conclusion As by 2001, s uch an achievement could not have occurred had it not been for the con tributions from an active civil society, a thriving media, an independent electoral commission, and the elite consensus provided through IPAC. However, when distilled to their initial causes , all of these factors would have likely been absent if not for th e presence of an active and sustained domestic political opposition, as well as an effectively leveraged aid regime conditioned on democratic progress. In addition , not have occurred without the executive recognizing suc h constraints and opting to pursue policies designed to maximize the duration of his rule. By valuing security and tenure of office more than the amount of discretionary authority at his disposal , and by subsequently choosing to tolerate continued democrat ic reform instead of repress ing it, Rawlings ultimately locked in Ghana on a democratic trajectory. This chapter has shown how executive agency operated in facilitating the n the their demands for reform established a precedent by which similar decisions would be based, and set in motion a new set of institutional logics that would be rei nforced through path dependence, sunk costs, and positive feedback (Pierson 2004). By the opening had undoubtedly closed and produced a new equilibrium, this process e nabled the regime to proceed along an incremental path where dome stic and international
167 pressure continued to push the country and its leaders to expand political rights and civil liberties without any significant instances of regression or suppression. To be sure, factors such as civil society (Gyekye Jandoh 2006) , IPAC (Frempong 2007; Asante 2013), journey to democracy dem ocratic trajectory had been firmly established. Again , this is not to suggest that Rawlings had either become a converted democrat , or had alternatively harbored authoritarian intentions; such a question is beyond the scope of this dissertation. The point intentions , by the actions he initiated in the first term of his administration he beca me what Levitsky and Way describe as being 299 300). As political institutions create d in a process dominated by Rawlings and the (P)NDC generated increasing returns through their consolidation and socialization among , an would have been too costly to bear for any of the political parties, even if it included losing the presidency . As both the NPP and the NDC demonstrated their long term interests in the through their acceptance of electoral defeat in 1996 and 2000 (respectively), as wel l as through the belief they could still compete and win in a future election, they demonstrated one of the major signifiers of sustainable demo cratic politics (Przeworski 1991 ) . Nevertheless, none of these developments would have occurred had it not been for the conscious decision of Rawlings and his administration to pursue a rational strategy along the path of least resistance in order to retain the presidency for as long as possible. In doing so,
168 however, such actions (whether intended or otherwise) ult imately confirmed the country democratic trajectory , as well as the legacy of it s democratic experiment .
169 CHAPTER 6 Up to this point this dissertation has provided an updated account proc ess of democratization that explains its transformation as a product of conjunctural causation occurring within a critical juncture. As a result, the equilibrium that emerged following the close of the critical juncture placed the country onto a path of de mocracy where, over time, path dependent incentives, such as positive feedback and the internalization of the electoral rules by all major political actors (including chief executive) produced one of the most widely recognized democracies on the African continent. By providing an alternative explanation for couches its political transformation as a process rather than an event occurring at a single point in time, and does so while appreciating the interactions between st ructure and agency , this chapter contributes to the ongoing conversation concerning the reason as to why Ghana was the only country with a flawed democratic transition to become a widely recognized liberal democracy (Bratton and van de Walle 1997; Diamond 2010, xxvi). In addition, however, this chapter provides another contribution through its assessment of this case in comparison to all other African countries attempting regime change in the same time period. In brief, this chapter formalizes the findings derived from the Ghana case study by proposing a theory of regime trajectories in Africa, composed as a function of the combined presence and absence of international leverage, international aid, domestic political opposition, and executive tolerance for c ontinued democratization. U tilizing qualitative comparative analysis (QCA) for the large ,
170 this chapter assesses how these factors interacted and produced democratic and non d emocratic trajectories. T he results from this assessment demonstrate that the lessons account for the vast majority of democratic, authoritarian, and mixed regime trajectori es on the continent , and furthermore illuminate an area of study that , with few exceptions, has generally remained under examined . The insights generated from this exercise provide an answer to the question as to why, in a period of widespread democratizat ion on the African continent, some countries succeeded in developing democratic systems of rule while others did not. Ghana as a Case of a Much Larger Phenomenon Through previous chapters I have argued occurred not at the moment of its inaugural elections, but instead emerged through a long term process of incrementa l change and repeated elections. I have additionally shown that such a trajectory was a function of international and domestic pressure influencing the exec ess continued democratic reform. T he question remains , however, as to what, if any, insights this case may provide towards a wider understanding of democratization in Africa. Since Bratton and ) study that investigated the origins of democratic transitions in Africa in the immediate post Cold War period, few studies have explored the legacies of countries maintained dem ocratic regimes while others did not. 1 In addition, as previous 1 VillalÃ³n and VonDoepp (2005), Mustapha and Whitfield (2009b), and Levitsky and Way (2010) are notable exceptions.
171 transitions based literature cannot explain , and thus generally regards the case as an outlier ( Bratton and van de Walle 1997; Levitsky and Way 2010), such appare nt deviance provides a useful point of departure for rethinking contemporary regime change in Africa. 2 Using Ghana as an example of what will be considered a democratic trajectory, this section leverages the insights from the Ghanaian example towards build ing a new explanation for the v ariation in democratic outcomes since their initial openings over 20 years ago . Positing that the end of the Cold War brought about a period of significant change for Africa (Bratton and Van de Walle 1997; VillalÃ³n 1998), and that the actions taken during this period of high contingency produced distinct paths that became self reinforcing and more difficult to change over time (Pierson 2004), I argue that countries in Africa experiencing democratic openings previously defined as the period following a critical juncture where a n authoritarian country engaged in democratic reform in response to a crisis of the authoritarian status quo embarked upon different and distinct regime trajectories that influenced their likelihood of mai ntaining a democratic system of rule. 3 Utilizing Freedom House data to illustrate the point , s ome countries, such as Benin, Ghana, and South Africa, largely followed democratic trajectories that were characterized by democratic openings that subsequently c ontinued to 2 In particular, when engaging in theory building exercises (especial ly within nested analyses) it is often useful to target deviant cases, as they are more likely to provide insight into previously unconsidered factors and variables (Lieberman 2005; 445; see also Seawright and Gerring 2008). 3 In addition to the above crit eria defining democratic openings, it is important to note that the reforms occurring in these periods included the holding of an inaugural election.
172 democratically improve over time . 4 As Figure 6 1 shows, countries with democratic trajectories continued to expand political rights and civil liberties following their respective openings, all the while avoiding breakdown or any other significa nt setbacks . In contrast, other countries such as the Republic of Congo or Gabon instead followed clearly non democratic ( i.e. authoritarian) trajectories , where each country increasingly turned authoritarian following their democratic openings . In these i nstances, Figure 6 2 significant breakdown or degradation of democracy that never recovered, thus effectively making the initial democratic opening meaningless . Lastly, other countries , including Burkina Faso, Kenya and Zambia, followed what will be considered mixed trajectories . Figure 6 3 show s how these countries exhibited neither clearly democratic nor authoritarian tendencies following their respective openings. While this ambiguity s democratic opening (as in the case of Kenya) , it also could have been due to a maintained position with in the ambiguous zone between democracy and authoritarianism (suc h as Burkina Faso ) . 4 Figur es 6 1, 6 2, and 6 3 serve to illustrate the various regime trajectories that some African states pursued following their democratic openings, the Freedom House data does not define such trajectories, and as a result, the figures should instead be regarded as illustrative.
173 Source: Quality of Government 2013 . Figure 6 1 : Democratic Trajectories, 1990 2012 So urce: Quality of Government 2013. Figure 6 2: Authoritarian Trajectories, 1990 2012
174 So urce: Quality of Government 2013 . Figure 6 3: Mixed Traje ctories, 1990 2012 To be sure, the Freedom House data should not be taken as an exact associate d with conceptualization, measurement, and aggregation, as well as additional challenges associated with the replicability of its assessments (2009). Similarly, Coppedge et. al. have noted the problem of intercorrelation among associated indicators of poli tical rights and civil liberties, thus raising the concern as to whether such indicators are independently coded from one another (2011, 251). Nevertheless, in spite of these challenges the Freedom House data does at least provide researchers and scholars a general sense for where a country stands along the democracy authoritarianism continuum. More importantly, the data provides information regarding
175 where a country has been in the past, and how far it has come since its democratic opening in the early 199 0s. As a result, for the purposes of this dissertation it is reasonable to utilize Freedom House scores over time in order to obtain a visual democratic standing that matters, it is instead the assessment regarding where the country came from and the journey it took to arrive at its current position. Outcome of Interest: Regime Trajectory For the purposes of this dissertation , regime trajectory is defined as the overall direction that a regime traveled since it underwent a democratic opening. Regime trajectories do not necessarily have fixed time frame s, as they continu e until a subsequent critical juncture occurs that alters its direction . In addition, r egime trajectories are organized along a democratic authoritarian spectrum, and are subdivided into democratic, mixed, and authoritarian categories (defined below). To b e sure, while the classification of regime trajectories could have been executed in a number of alternative ways, I have chosen this tripartite typology as it provides a representative yet manageable range of outcomes that encompass the entire democratic a uthoritarian spectrum. In addition, these categories are discrete and mutually exclusive from one another (i.e. a country cannot have more than one type of regime trajectory) . Lastly, building from democratic, competiti ve authoritarian, and authoritarian regimes (2010, 365 9), the operationalization of these different types of regime trajectories have been fitted according to the scope conditions previously specified .
176 Democratic t rajectories First, democratic trajectori es are defined as cases that held regularly scheduled and uninterrupted elections since their initial opening, did not have any instance s of coups or democratic breakdown, had an overall continuous expansion of civil and political liberties with no signifi cant instances of retrenchment, and had met the basic requirements to be regarded as an electoral democracy by 201 2 (Freedom House 2014 a ) . Authoritarian t rajectories In contrast, authoritarian trajectories are defined as cases that experienced either a co nsistent degradation of civil and political rights since their democratic opening, a regime breakdown that never recovered to its previous status, or other instances where elections were banned or had become so manipulated over time that they became practi cally meaningless. Mixed t rajectories Lastly, mixed trajectories are defined as those cases that did not succeed in maintaining a democratic trajectory , yet still managed to avoid falling within the authoritarian category . Serving as more than a residual category, mixed trajectories instead highlight the challenges that proponents of democracy and authoritarianism experienced as a result of certain structural limitations (discussed below) that prevented these countries from realizing either outcome.
177 Condit ion Variables 5 Utilizing the insights derived from the Ghanaian experience, the observed v ariation in regime trajectories among countries experiencing democratic openings between 1990 and 1994 may largely be explained through the combined presence of three structural conditions, as well as executive agency : significant political opposition, international leverage, reliance on foreign aid repress democratic demands in their inaugural administration . Domestic poli tical opposition First, in a similar manner to idea concerning the balance of political forces as influencing the long term probability of success of democratic reform ( 1991, 79 88 ), the presence of a significant domestic political opposition is hypothesized to have affect ed such group s posing a formidable obstacle to any non democratic manipulation of the political system. In other words, in circumstances where the political oppo sition comprised a substantial segment of the population, it is reasonable to argue that the incumbent government would have been less likely to enact non democratic measures that would s previous political opening. As many opposition parties in the post democratic opening period campaigned on platforms of extending democracy in their respective countries (Riedl 2014), such actions by the government would have likely outraged the oppositi on (in addition to international donors), and, at the very least, would have significantly 5 based independence from one another (Rihoux and D e Meur 2009, 67, fn. 3; see also Ragin 1987).
178 diminished the odds of the incumbent being reelected in a subsequent election (barring overt manipulation and repression). 6 In order to empirically assess for the si gnificant presence of the domestic political opposition , this variable is operationalized according to three conditions: that the opposition was relatively unified, minimally competitive, and able to mobilize voters to turn out. Parties that were unified and minimally competitive obtained at least 30 per cent of the vote in the inaugural presidential election, or otherwise lost the election by 15 per cent or less. 7 In addition, opposition parties demonstrated their mobilization capacity by encouraging a sufficient number of their members and sympathizers to vote so as to obtain a minimum turnout rate of more than 50 per cent of registered voters. 8 Foreign aid dependence Secondly, the impact of foreign aid is al so hypothesized to have affected the long term trajectories of many African countries. In a period of increased conditionality and where aid was often attached with democratic demands, f or countries like Ghana whose share of foreign aid exceeded te n per ce nt of their annual GDP , there is significant reason to believe that there would have been a significant interest in 6 To be sure, while many opposition parties campaigned on platforms of extending democracy in their respective countries, it should not be misconstrued to assume that such opposition parties were committed demo of winning an election (Riedl 2014; see also Beaulieu 2014). 7 It is important to note that the latter condition losing with less than 30 per cent total but being wi thin 15 percentage points of the winning party inherently means that the winning party won with less than 50 per cent. 8 See Kaya and Bernhard (2013) for discussion on voter mobilization capacity. In addition, while the cutoff rate of 50 per cent turnout m ay appear low, this is to control for instances where voter rolls were inflated due to outdated voter registers, which were widely believed to have been a frequent occurrence throughout the continent during this period.
179 continued liberalization, not necessarily because such aid precipitated change in a kind of policy quid pro quo, but more likely because rec ipient governments wanted to continue receiving the benefits of foreign aid into the future (Wright 2001; Dunning 2004; Levitsky and Way 2010). In order to control for the impact of foreign aid on this process, as well as to control for yearly fluctuations period, this condition has been operationalized according to the average share of aid as constituting ten per cent or more of their GDP would re asonably be considered to have felt more pressure to democratize than other countries receiving less than ten per cent. While determining the threshold of aid dependency is an inherently problematic process, as establishing any level risks criticism of arb itrariness, I have chosen ten per cent as the threshold for aid dependency by using Ghana as the guiding model. In short, a s Ghana remained aid dependent and as its average share of aid was 11 per cent between 1990 and 1994, it was thus reasonable to calib rate aid dependency to ten per cent as a natural threshold. 9 International leverage However, while foreign aid may incentivize governments to democratize, it may also hinder such efforts by promoting corruption, or even by strengthening the capacity of th e government to repress (Bauer 2000; Brautigam and Knack 2004; Kono and 9 Interestingly, this threshold was further reinforced with preliminary analysis demonstrating Ghana to have the lowest share of aid (11 per cent) among all other countries with democratic trajectories, while the highest share of aid among authoritarian trajectories only went as high as nine per cent (Ivory Coast and Republic of Congo).
180 Montinola 2009) . 10 As aid inherently contains the ability to either help or hinder efforts at democratization, it is worth considering the role of international leverage and the manner in which leverage a governme (20 10, 40). Categorized in to three conditions, representing the presence of either high, medium, or low levels of international leverage, t his factor is composed of elements that largely reflect the level of a given relative economic and political power in the international system, in cluding GDP , level of oil production, possession of nuclear weapons, as well as whether the country retained any geostrategic or security interest among Western countries or whether the state was a particular benefactor from any non Western majo r powe r . 11 In other words, not to be misconstrued as experiencing actual pressure from pro democracy Western donors, countries with a high level of international leverage would be expected to be more s usceptible to democratic reform, as opposed to other cou ntries with less leverage. For this analysis, cases are assessed according to whether they were subject to significant or insignificant amounts of international leverage at the time of their respective democratic openings. Countries experiencing significa nt levels of criteria. As Western international aid was generally tied to democratic improvement in 10 See Resnick and van de Walle (2013) for a useful survey on the current debate concerning foreign 11 For additional details regarding this concept, including its co nstituent features, see Levitsky and Way 2010, 372 3.
181 this period, it is reasonable to argue that aid recipient countries in Africa receiving a significant amount of international leverage would be more likely to enact democratic reform and utilize such aid as intended by donors , while others receiving less leverage would instead be less likely. Executive agency: tolerance or repression Lastly, while such structural conditions incentivized countries towards democracy, it is important to recall that during these periods of democratic opening, the environment remained highly contingent and open to multiple outcomes . As national e xecutives in these inaugural periods set the national agenda by the nature of their positions, they inherently retained a high degree of influence in determining the ultimate path their countries would follow. Despite possessing a wide range of policies to pursue, all national executives experienced pressure to continue with democratic reform in their respective countries , whether by opposition groups, international donors , or both. As a result, executive policies generally fell along one of two choices: wh ether to ignore such demands for reform, or whether to tolerate them instead. On the one hand , national executives could have chosen to ignore demands for reform and maintain their executive power and authority. However, to do so would have been costly : do nors would have likely suspended or withdrawn aid, and the domestic opposition would have likely protested such actions, resulting in increased levels of domestic instability. As a result, ignoring calls for reform would have resulted in the increased like lihood of a chief executive needing to rule through overt repression . In turn, this would have also . On the other hand, however, executives could have chosen to prolong their duration of rule by working
182 within the confines of a democratic system and by tolerating such demands for reform. In this instance, enacting democratic reform would serve to satiate donors and the political opposition , thus providing domestic stability and continued acces s to international assistance . In other words , by sacrificing some amount of executive authority in exchange for domestic political and economic stability, such actions would serve to extend the tenure of a chief executive beyond their current t erm in offi ce . To be sure, tolerating democratic demands does not necessarily mean that an executive was a converted democrat or otherwise willingly accepted further democratic reform. Tolerance of additional democratic reforms could include an executive being agains t reform, yet selectively enacting reforms or otherwise utilizing measures to control the process so as to minimize the loss of their power and authority in the process . Utilizing the reasonable assumption that executives in office wanted to stay in office as long as possible , these executive decisions were structured by the presence and absence of the three previous factors. Where executives recognized the presence of significant domestic political opposition, high international leverage, and significant a mounts of conditioned foreign aid, they were less likely to ignore calls for reform, as an attempt to consolidate their authority would have likely failed and would have furthermore jeopardized their tenure in office. In contrast, in instances where the op position was insignificant, international aid was relatively small, and where leverage was limited or inconsequential, executives possessed a higher probability of success in consolidating their rule without negatively affecting their tenure in office . In other words, the decision to repress or tolerate democratic reform largely depended on whether executives recognized the constraints to their rule and acted appropriately. To be sure,
183 executives could still have acted contrary to what the structural circum stances suggested, however to have done so would have likely resulted in an outcome below their desired expectations due to such structural constraints favoring the opposite trajectory . A Theory of Regime Trajectories in Africa In sum, these factors consti tute the model for explaining why different countries proceeded to follow different regime trajectories after their initial democratic openings. As levels of opposition presence, foreign assistance, and the degree of leverage exerted by external donors inf luenced the decision for national executives to pursue additional reform , such conditions laid the path term regime prospects . As Figure 6 4 shows, w here executive actions accorded with structural conditions favoring democratization, including significant political opposition, significant receipts of foreign aid, and high international leverage, countries and their regimes proceeded along democratic trajectories . W hen f ormally expressed within Boolean algebraic terms amenable to QCA, this argument resembles the following formula : H ypothesis 1 : F * L * O * T = D, where F equals Foreign Aid, L equals International Leverage, O equals Domestic Political Opposition, T equals Executive Tolerance, and D equals a Democratic Trajectory. In con trast, Figure 6 5 shows how in countries where the political opposition failed to establish a significant presence , where democratic leverage remained low, and where foreign aid was a relatively small percentage of state income , such structural conditions failed to constrain executive interests in consolidating power , thus providing the ideal circumstances for an authoritarian trajectory . When formalized, Hypothesis 2 becomes: H ypothesis 2 : f * l * o * t = d, where f equals the absence of Foreign Aid, l equ als the absence of International Leverage, o equals the absence of Domestic
184 Political Opposition, t equals the absence of Executive Tolerance, and d equals an Authoritarian Trajectory. Lastly, Figure 6 6 shows that for those countries where variations in s tructural to tolerate or repress democratic reform, such countries subsequently fo llowed inconsistent (mixed) trajectories exhibit ing neither clear democratic nor authoritarian tendencies. This hypothesis is represented by the following expression: H ypothesis 3 : (F + f) * (L + l) * (O + o) * (T + t) = M, where M equals a Mixed Trajector y precipitated by the combined presence of the abovementioned conditions, with at least one condition being absent and another being present . Figure 6 4: Democratic Trajector ies Figure 6 5: Authoritarian Trajector ies Democratic Trajectory Executive Tolerates International Pressure (Effective) Foreign Aid (HIGH) International Leverage (HIGH) Domestic Political Opposition (HIGH) Authoritarian Trajectory Executive Represses International Pressure (Ineffective) Foreign Aid (LOW) International Leverage Domestic Political Opposition (LOW)
185 Figure 6 6: Mixed Trajectories An alysis Table 6 1 present s a comparative analysis of all countries experiencing democratic openings in the immediate period following the end of the Cold War (1990 1994) . 12 The first four columns provide information concerning the significant presence of int ernational leverage, foreign aid dependence, domestic political opposition, and the 13 The fifth thi s theory of regime trajectories and subsequent hypotheses ), while the sixth column provides the actual observed regime trajectory. Regime trajector ies are assessed up through 2012 . In addition, conditions were coded independently from one an other in order to prevent one condition influencing the coding of others, including the actual trajectory of cases. 12 See Appendix A for summary statistics associated with each condition. Appendix B contains information regarding which countries were structurally predisposed to follow democratic, authoritarian, and m ixed regime trajectories. 13 Data for international leverage was derived from the World Bank (2014) and the USEIA (2014). Data for foreign aid dependence was derived from the World Bank (2014). Data for opposition presence was derived from Lindberg (2006). Mixed Trajectory Executive Agency (Uncertain) International Pressure (Uncertain) Foreign Aid (HIGH/LOW) International Leverage (HIGH/LOW) Domestic Political Opposition (HIGH/LOW)
186 While certain conditions were relatively straightforward with respect to coding, including leverage, aid dependence , and the presence of sig nificant domestic political opposition, other conditions contained inherent challenges. First, w ith regards to the coding of executive decisions and the outcome of interest ( i.e. actual trajectories), it was necessary to engage in careful content analysis from a variety of sources including the British Broadcasting Service, Freedom House, Africa Confidential , Africa South of the Sahara , and other published political histories so as to obtain a reasonable and satisfactory account for each case and its associ ated condition. Secondly , the assessment of certain cases proved difficult as well. For example, as eight countries in this analysis experienced inaugural elections that were widely regarded as not free and fair (Lindberg 2006), this feature consequently made assessing for the presence of the domestic political opposition in these countries a challenge . 14 manipulated by the incumbent government, and were furthermore successful insofar as the electoral results withstood opposition challenges, such circumstances suggest that the domestic political opposition was perhaps insignificant in these countries. On the other hand, however, the fact that the incumbent resorted to electoral manipulation i n the first place suggests that a sufficient level of domestic opposition existed that subsequently triggered the incumbent government fearing an electoral loss to take action . As making an assumption for either approach remained problematic, as doing so e ffectively required taking an intellectual leap of faith, in this analysis these cases were 14 Countries with inaugural elections that were considered to have been not free and fair include Burkina Faso, Cameroon, the Comoros, Djibouti, Equatorial Guinea, Kenya, Mauritania, and Togo.
187 assessed solely according to what the data indicated . In Cameroon, the Comoros, and Kenya , the political opposition demonstrated significance despite participating in unfair electio ns; in the remaining cases the data showed the opposition to be less than significant , all the while acknowledging that the data was likely skewed to the incumbent To be sure, even if some of these latter cases did, in fact, have the significant presence of a domestic political opposition, their predicted regime trajectories would have nevertheless remained the same, largely due to subsequent executive repression. Thirdly, a nother challenge concerned classifying the regime tr ajectory for SÃ£o TomÃ© and PrÃncipe . Since its democratic opening , the country of SÃ£o TomÃ© and PrÃncipe has been recognized as a free demo cracy by Freedom House (2014a). However , the country is classified as following a mixed trajectory due to a successful coup that briefly deposed then President Trovoada in 1995 . Moreover , as President Trovoada was reinstated as chief executive after only a few days after the coup, it begs the question as to whether such actions constituted a genuine coup, or whether this e vent was instead a failed attempted coup. 15 Such a distinction is important, as it not actual regime trajectory, it additionally ould predict for this country . Ultimately, because President Trovoada was, in effect, removed from office (if only for a short period), and because his reinstatement was unforeseen at the time of his removal , his ouster stands as a successful coup in this analysis . As a 15 Interestingly, Freedom House does not appear to recognize that a coup occurred in the country, as its Freedom House rating for 1995 remains unchanged from the previous year.
188 SÃ£o TomÃ© and PrÃncipe was deemed to have followed a mixed trajectory. Lastly, there was a challenge regarding whether to utilize data from presidential or parliamentary elections in ascertaining the level of political opposition in certain cases. In general, within this group of cases presidential elections either preceded parliamentary elections, or coincided with them. In six cases, however, parliamentary elections pre ceded presidential elections within a short time span (about one month ) , thus raising the question as to whether it would be more appropriate to assess the presence of the domestic political opposition in these countries by utilizing results from the inaug ural parliamentary elections. 16 At the same time, however, utilizing data from parliamentary elections presents further complications, most notably by the manner in which these types of elections had multiple parties competing for legislative representation . In one instance, for example, there were no less than forty political 1992 inaugural parliamentary election (Lindberg 2006, 189). In short, using these kinds of cases to determine the presence of a significant amount of domestic political opposition would be problematic, as the number of competing parties including both pro government and opposition parties would inescapably obscure the true level of government and opposition support. As a result, I chose to us e the results from the inaugural presidential elections in these cases, as presidential elections generally contained fewer competing parties, and thus served to narrow the choice of chief executive to a minimal number of candidates 16 These cases included Cape Verde, Guinea Bissau, Mali, Niger, the Republic of Congo, and SÃ£o TomÃ© and PrÃncipe .
189 (Lindberg 2006). The on ly exception to this decision, however, was the case of Cape Verde, as its inaugural parliamentary election contained only two competing parties, thus providing a clear choice between the incumbent government and the opposition. Table 6 1: Regime Trajecto ries in Afric a International Leverage Aid Dependence Domestic Opposition Executive Repress vs. Tolerate? Predicted Trajectory Actual Trajectory Benin (1991)* Yes Yes Yes Tolerate Democratic Democratic Burkina Faso (1991) Yes Yes No Tolerate Mixed Mix ed Cameroon (1992) No No Yes Repress Mixed Mixed Cape Verde * (1991) Yes Yes Yes Tolerate Democratic Democratic Central African Republic (1993) Yes Yes No Repress Mixed Mixed Comoros * (1990) Yes Yes Yes Repress Mixed Mixed Congo Rep. (1992) Yes No No Repress Authoritarian Authoritarian Djibouti (1993) * Yes Yes No Repress Mixed Mixed Equatorial Guinea** (1993) Yes Yes No Repress Mixed Authoritarian *** Gabon (1993) No No No Repress Authoritarian Authoritarian Ghana (1992) Yes Yes Yes Tolerate Democratic Democratic Guinea Bissau (1994) Yes Yes No Tolerate Mixed Mixed Ivory Coast (1990) No No No Repress Authoritarian Authoritarian Kenya (1992) Yes Yes Yes Repress Mixed Mixed
190 Table 6 1. Continued International Leverage Aid Dependence Do mestic Opposition Executive Repress vs. Tolerate? Predicted Trajectory Actual Trajectory Lesotho (1993) Yes Yes No Repress Mixed Mixed Madagascar (1993) Yes Yes No Repress Mixed Mixed Malawi (1994) Yes Yes Yes Tolerate Democratic Democratic Mali (1 992) Yes No No Tolerate Mixed Mixed Mauritania (1992 ) Yes Yes No Repress Mixed Mixed Mozambique (1994) Yes Yes Yes Tolerate Democratic Mixed *** Niger (1993) Yes Yes No Tolerate Mixed Mixed S Ã£ o Tom Ã© and Pr Ã ncipe* (1991) Yes n/d No Tolerate Mixed Mix ed Seychelles (1993) Yes No Yes Tolerate Mixed Democratic *** South Africa * (1994) No No No Tolerate Mixed Democratic *** Swaziland** (1993) Yes No N o Repress Authoritarian Authoritarian Togo (1993) Yes Yes No Repress Mixed Mixed Zambia (1991) Yes Yes No Repress Mixed Mixed * = data from parliamentary elections . ** = data from Bratton and Van de Walle 1997, 208 . *** = unpredicted regime trajectory. Parentheses indicate year of democratic opening. n/d = no data available for indicator. So urce: Quality of Government 2013 ; Bratton and van de Walle 1997, 208; Lindberg 2006. Based upon the results, there are a number of observations worth noting. First, the results from the analysis lend s
191 demonstrate a general relationship . Out of 2 7 total cases, the initial model correctly accounts for 2 3 . While the initial results are limited insofar as their inability to account for the regime trajectories of four countries, it is important to consider that some countrie resultant regime trajectories were influenced and subsequently altered by extraordinary events falling beyond the scope of this theory . For example , alth ough South Africa was predicted to follow a mixed trajectory , its divergence may be reasonably argued to have occurred due to the exceptional circumstances in which this country embarked upon its respective regime trajectory , including its prior histo ry of an Apartheid based government, as well as the extraordinary amount of external pressure and international attention it received before, during, and after its democratic opening (Cooper 2002; VonDoepp and VillalÃ³n 2005, 7; see also Nugent 2005, 430 43 3; Klotz 1999; Lyman 2007) . In addition, it is worth noting the role of executive agency in this country , including the manner in which Nelson Mandela exercised general self restraint in his initial administration , despite possessing a preponderance of dom estic support and executive authority at his general discretion. With regard to Equatorial Guinea, its divergence from initial expectations may be attributed to the discovery and subsequent production of oil in the period immediately following its democrat ic opening. Oil production in Equatorial Guinea climbed from just five thousand barrels per day in 1994 to over 80 thousand barrels by 1998, and continued to climb thereafter (USEIA 2014). As oil revenue continued to supplant other sources of financial sup port for the country , it is reasonable to argue that the effectiveness of both international aid and leverage had significantly diminished, thus enabling the government to consolidate power and reverse any democratic gains that
192 were made since the time of the 1993 inaugural elections. In other words, discovery of oil and subsequent revenues proved strong enough to alter the course of In sum, as South Afric a and Equatorial Guinea constitute what may be reasonably regarded as exceptional cases, and as the circumstances of their regime trajectories may be largely attributed to idiosyncratic forces that are inherently problematic to compare, it is thus reasonable to suggest that their circumstances fall outside the parameters of the overall model. In doing so, the initial model consequently accounts for 2 3 out of 25 cases, explain ing 92 per cent of the variation in regime trajectories. When c ompared against more recent explanations of democratization in Africa that subsequently account for 11 out o f 14 cases, or 79 per cent of variation (Levitsky and Way 2010, 306), the results from Table 6 1 thus provide a contribution by improving upon our c urrent understanding as to why some African countries maintained democratic systems of rule, while others did not. In regards to the other unexplaine d cases from the initial model , Mozambique and the Seychelles serve as borderline cases that demonstrated d eviance through their ambiguous behavior and associated challenges in coding. The chief executive in Mozambique (Joaquim Chissano) , for instance, acted in ways that were neither clearly tolerant of democratic demands nor clearly repressive . In brief, a ltho ugh discrepancies arose during the inaugural in holding constitutionally mandated municipal elections, as well as delays for the second presidential elections of
193 open repression of democratic demands deemed to have exercised general tolerance. As f or the Seychelles, this case prov ed problematic through the coding of its ultimate regime trajectory. Though the country generally proceeded in a steadfast democratic manner with no significant authoritarian setbacks, notable challenges have nevertheless persisted, including a relatively weak judiciary and legislature, government control over the media , as well as notable corruption within government (Baker 2008) . In addition, while Freedom House currently recognizes the Seychelles as an electoral democracy, it notes that the country has c ontinued to remain since its democratic opening in 1993 (Freedom House 2014a). Nevertheless, because the regime was recognized as an electoral democracy by 2012 , the country thus met the minimum criteria for a democratic trajectory and was coded as such. Nevertheless, after excluding Mozambique and the Seychelles based upon the abovementioned challenges , and whose resolution remains beyond the scope of this dissertation 2) support s all three hypothe ses. 17 I t is important to note that despite the relative low occurrence of democratic and authoritarian trajectories (compared to instances of mixed trajectories), such findings are not problematic as part of the purpose of QCA is to analyze for instances o f sufficiency , as opposed to assessing for the frequency of such instances . In other words, unlike standard statistical analyses where the frequency of a causal relationship 17 A truth table consists of a t able listing all possible combinations of conditions, with the purpose being to demonstrate logical consistency between combinations of conditions and outcomes of interest (Ragin 1987).
194 determines the strength of the overall results, QCA instead requires only one occu rrence to demonstrate that a combination of conditions will, indeed, produce the outcome of interest. Furthermore, despite the high number of instances of mixed trajectories (15 out of 23 cases), such findings are not extraordinary, particularly as recent scholarship finds the modal regime in Africa to reside between democracy and authoritarianism (Levitsky and Way 2010, 306). In addition, as Levitsky and Way find democracy to be the exception rather than the rule in Africa, these findings provide a contrib ution through explaining how democracy sustained itself, despite experiencing generally unfavorable conditions. 18 18 ianism in Africa is unable to account for any instance of democracy among their sample of 14 cases (2010, 306).
195 Table 6 2: Truth Table of Regime Trajectories in Africa Condition Outcome Countries L A O T D N 0 0 0 0 d 4 Congo Rep. Gabon , Ivory Coast, Swaziland 0 0 0 1 0 0 0 1 0 M 1 Cameroon 0 0 1 1 0 0 1 0 0 0 0 1 0 1 0 0 1 1 0 0 0 1 1 1 0 1 0 0 0 0 1 0 0 1 M 2 Mali , S Ã£ o Tom Ã© and Pr Ã ncipe 1 0 1 0 0 1 0 1 1 0 1 1 0 0 M 7 Cent ral African Republic, Dji bouti , Lesotho, Madagascar, Mauritania, Togo, Zambia 1 1 0 1 M 3 Burkina Faso, Guinea Bissau, Niger 1 1 1 0 M 2 Comoros , Kenya 1 1 1 1 D 4 Benin, Cape Verde, Ghana, Malawi , Sum 23 Residual Cases 1 1 1 1 M 1 Mozambique 1 1 0 0 d 1 Equatorial Guinea 1 0 1 1 D 1 Seychelles 0 0 0 1 D 1 South Africa Sum 4 Total 27 D = Democratic Trajectory d = Authoritarian Trajectory M = Mixed Trajectory N = Number of Cases Democratic Trajectories in Africa To further reinforc e the argument that international leverage, aid, domestic political opposition, and executive tolerance produced democratic trajectories in Africa and to control for the possibility that th e initial results provided spurious findings, Table 6 3 provides supplementary analysis utilizing fuzzy set QCA . The second, third, fourth, and fifth
196 columns provide membership details on each of the condition variables, while the sixth column provides the membership scores for the combined set of all four previous conditions. 19 The seventh column , in turn, membership within a democratic trajectory. In short, the results demonstrate the combination of hypothesize d factors to be sufficient in producing a democratic trajectory, as the membership scores for L*A*O*T are all equal to or less than membership scores for being in a democratic trajectory. What is further significant is that this test of sufficiency include s all residual cases, including Equatorial Guinea, Mozambique, the Seychelles, and South Africa. As in the previous analysis, all conditions were coded independently from one another to prevent the coding of one condition from influencing the coding of ano ther. First, international leverage was coded according to whether a country received high, medium, or low international leverage. C ases with high leverage were placed completely within the set, cases with low leverage were placed completely outside of th e set, and cases with medium leverage were classified as being more in the set than not, while still falling short of being considered a full member of the set. Secondly, foreign aid dependence was coded according to how much of a percentage international aid (official and development) constituted relative to each Using the case s of Ghana (a democratic trajectory) and Ivory Coast (an authoritarian trajectory) as aid s in calibrating membership scores for the set, t he inflection point (i.e. the point at which a country was determined to have been more aid 19 Membership scores refer to how much a country exemplified a given condition or characteristic. Membership in a given condition is based on a s core between 0.0 (not a member) and 1.0 (full member).
197 dependent than not) was adjusted to nine percent , and coding was clas sified along a six point scale ranging from 0 .0 to 1.0, in increments of 0.2 . Scores of aid dependence ranged from zero per cent up to a 12 per cent threshold. Countries where aid constituted 12 per cent or more were classified as residing fully within the set. All other cases were coded according to their position relative to the inflection point. Thirdly, membership in the do mestic opposition category was similarly adjusted along a six point scale, from 0.0 to 1.0. As this condition was a composite of the (unless otherwise specified), the coding of each country required careful content analysis to determine how significant the opposition was in each particular country. In cases where the opposition mobilized more than 50 per cent of the population to vot e, e share resided within 15 percentage points, these cases were classified as being fully within the set , and were coded as such . For countries whose domestic opposition lost the inaugural election by more than 15 per cent, yet still obtained more than 30 pe r cent of the vote, these cases were considered as being generally within the set, though classified just short of full membership. For instances where one of the criteria were missing (such as being electorally competitive but failing to mobilize at least 50 per cent of the electorate, and vice versa), these cases were listed as residing near the inflection point, though still below the 0.5 threshold, thus generally being outside of the membership set. All other instances were classified as generally not r esiding within the set. Fourthly, the classification of executive tolerance was relatively straightforward, as measures were derived from the same condition listed in Table 6 1. Instances where
198 the executive tolerated democratic demands were classified as being fully within the set, and instances where the executive clearly repressed such demands were listed as falling completely outside of the set. However, for the case of Mozambique and its repress or tolerate democratic demands, this case was listed as residing at the inflection point between tolerance and repression . Fifthly , membership in democratic trajectories was classified according to a of actual democratic trajectories were accorded full membership; countries with mixed trajectories were classified as residing at the inflection point; and authoritarian trajectories were listed as being completely outside of the membership set. The only e xception to this rule was the Seychelles, due to its previously noted ambiguities with its regime trajectory , which was ultimately determined to have b een democratic . Lastly, Membership scores for L*A*O*T were calculated by taking the lowest score from the set of conditions comprising Leverage (L), Aid (A), Opposition Presence (O), and Executive Tolerance (T).
199 Table 6 3 : Membership Within Democratic Trajectories in Africa Country Leverage Aid Opposition Tolerance L*A*O*T Democratic Trajectory Beni n 1 1 1 1 1 1 Burkina Faso 1 1 0 1 0 0.5 Cameroon 0.75 0.2 1 0 0 0.5 Cape Verde 1 1 0.8 1 0.8 1 Central African Republic 1 1 0.4 0 0 0.5 Comoros 1 1 1 0 0 0.5 Congo Rep ublic 1 0.4 0.2 0 0 0 Djibouti 1 1 0.2 0 0 0.5 Equatorial Guinea 1 1 0.2 0 0 0 Gabon 0.75 0.2 0.2 0 0 0 Ghana 1 0.8 0.8 1 0.8 1 Guinea Bissau 1 1 0.2 1 0.2 0.5 Ivory Coast 0.75 0.4 0.2 0 0 0 Kenya 1 1 1 0 0 0.5 Lesotho 1 1 0.2 0 0 0.5 Madagascar 1 1 0.2 0 0 0 .5 Malawi 1 1 1 1 1 1 Mali 1 1 0 1 0 0.5 Mauritania 1 1 0. 2 0 0 0.5 Mozambique 1 1 0.8 0.5 0.5 0.5 Niger 1 1 0.2 1 0.2 0.5 S Ã£ o Tom Ã© and Pr Ã ncipe 1 n/d 0.2 1 0.2 0.5 Seychelles 1 0.2 0.8 1 0.2 0.7 South Africa 0 0 0.2 1 0 1 Swaziland 1 0.2 0.2 0 0 0 Togo 1 1 0 0 0 0.5 Zambia 1 1 0 0 0 0.5 n/d = no data available f or indicator. . When visually represented, Figure 6 7 shows how the combined presence of international leverage, international aid, domestic political opposition, and executive tolerance constitute a sufficient condition that is fully consistent for producing democratic trajectories. In spite of the previously highlighted cases whose regime
200 trajectories were not predicted by the model, as all cases reside within the left upper triangular quadrant and subsequently demonstra te their membership scores to be at or below their outcome scores the data shows that these cases nevertheless accord with an explanation that emphasizes the occurrence of democratic trajectories as occurring through the combined presence of all four hypot hesized conditions. To be sure, this is not to suggest that this combination of conditions is the only path to a democratic trajectory , as the cases of South Africa and the Seychelles demonstrate that there exist alternative routes. However, the data does suggest that this path still remains as the modal route. A n additional analysi s of authoritarian trajectories (not shown) also demonstrate s the combined absence of all four conditions to be a fully consistent sufficient condition for producing an authorita rian trajectory . Figure 6 7: Plot of Degree of Membership Within a Democratic Trajectory Against the Degree of Membership in L * A * O * T.
201 Conclusion contributes to the wider study of contemporary democratization and regime change in Africa. Previously considered an idiosyncratic case of democracy that was essentially dependent up on the executive it to develop (Levitsky and Way 2010), the case of Ghana shows how dome stic and international pressure manifested to, in fact, influence the decision of many African chief executives to tolerate or repress democratization in their respective countries . Throug h the comparative analysis of 27 African countries, this chapter dem onstrated the utility that a theory of democratic trajectories in Africa provides for obtaining a greater understanding regarding the question why some countries in a period of democratic optimism following the end of the Cold War and the subsequent wave o f democracy on the African continent succeeded in maintai ning democratic systems of rule whi le many others did not. Operationalized into democratic, authoritarian, and mixed regime trajectories, this chapter demonstrated how the domestic political oppositi on, international aid, and international leverage from Western donors influenced the decision of African executives to either tolerate or repress opening. Where domestic and international pressure suggested that leaders would compromise their tenure as chief executive were they to engage in overt repression, such executives exercised restraint. In contrast, where domestic and international conditions suggested a general absen ce of consequences for undemocratic behavior,
202 such executives accordingly often engaged in repression in order to maximize their power and authority. in its ability to account for 23 out of 27 cases, the deviance of these cases may be attributed to challenges derived from exceptional and uniq ue circumstances ( Equatorial Guinea and South Africa ), as well as from borderline cases whose difficulties were primarily associated with ambiguity (Mozambique and the Seychelles). T nevertheless account for the vast majority of variation in regime trajectories, and in doing so, provide a framework for advancing our current understanding of regime change in Africa s ince the end of the Cold War. Furthermore, as subsequent analysis among all 27 countries finds all four hypothesized conditions to be sufficient and fully consistent with producing democratic trajectories, such findings lend additional support for this cha . To be sure, an explanation emphasizing the interaction of the domestic and international environment as influencing the decision of the executive to d emocratize is rather intuitive, perhaps even banal. However, despite the common understanding that domestic and international politics shaped democracy in Africa, there have been few works that have attempted to formalize this idea into a theory, including operationalizing such abstract concepts into empirically derived in dicators. As this chapter does provide an account as to how the domestic and international environment shaped regime change on the African continent, and does so by emphasizing regime trajectories rather than regime transitions, this chapter thus illuminat es an under explored area within African politics, and contributes to the scholarly study of democracy in Africa.
203 In observing an existing gap in the literature and the unresolved question concerning the legacies of democratic experiments that were experie nced more than two decades ago including the questions as to why some ostensibly successful democratic transitions resulted in contemporary authoritarian regimes, while other previously considered flawed transitions instead produced widely recognized democ ratic systems this chapter previously considered deviant and unlikely case of democracy, and leveraged i ts experience against that of 26 other countries attempting similar proce sses at the same time as Ghana. In sum, th ough t contain some cases whose outcomes are dependent on factors beyond the scope of this study , the analysis nevertheless provide s a scholarly contribution through its analysis of a wider rang e of cases of regime change in Africa, and its ability to account for a higher proportion of such cases than what the current literature provides.
204 CHAPTER 7 CONCLUSION What began as a question concerning one puzzling trajectory of democracy has since produced a broader understanding concerning the development of democracy and authoritarianism in Africa after the Cold War. At the time of its democratic opening, Ghana appeared predisposed towards autho ritarianism: its democratic opening was largely managed by the incumbent, its inaugural election failed to provide democratic legitimacy to the victors (resulting in a boycott of the opposition from government ) , and the executive continued to possess a hig h amount of discretionary authority, often at the expense of the Parliament. In addition, the media and private press con tinued to be restricted by the NDC led government, thus government (Morrison 2004; cited in Levitsky and Way 2010, 302). 1 In sum , as none of these issues had been reconciled prior to the inauguration of the Fourth Republic , Levitsky and Way argue that instead of producing a democracy, suppo sed democratic transition instead However , despite these inauspicious origins, Ghana nevertheless continued to democratically improve in a steadfast fashion. By the time of the 2000 general electi ons , electoral conditions had improved so much so that the opposition could freely compete and win, resulting in t he peaceful transition of power from one party to another , the first time that such an event had occurred in th . Previous cited the roles of civil society (Gyekye Jandoh 2006) , the Electoral Commission and formation of the Interparty 1 Private television and radio outlets would not be legalized until 1995 (Morrison 2004, 433).
205 Advisory Committee (Frempong 2007) , and even the influence of repeated elections (Lindberg 20 06) as facilitating democratic development in the country . W hile such explanations rightly account for some of the forces that have served to deepen democracy in the country, they do not fully explain what initially brought the country to follow a democrat ic trajectory in the first place. In short , such explanations take the the country overcame its initial obstacles (noted above) so as to eventually allow other factors s uch as IPAC, civil society, and repeated elections to promote democratic reform. In doing so , such explanations thus take democratic trajectory for granted, subsuming it as a product of its alleged democratic transition in 1992 . Furthermore , remain limited by the other countries that attempted the same process in the same time period . For example, while IPAC certai nly facilitated interparty dialogue between the NDC, NPP, the EC, and major international donors, it is important to recall that IPAC was created in a context election environment. As a result, it is inherently difficult to compare IPAC with any other kind of informal institutional arrangement as there will likely be differences between the constituting parties, the reasons for creating the institution, as well as the manner in which any recommendations from such an informal body wo uld result in any legal and/or policy changes. Moreover , comparing civil society across cases remains an inherently problematic task for researchers and scholars. Not only do multiple definitions of civil society abound (Harbeso n, Rothchild, and Chazan 199 4), in addition, civil society may not always be a force for democratic improvement
206 (Way 2014). As ivil society is neither homogeneous, nor wholly emancipatory; in fact it is contradictory, exhibiting both democrat ic and despotic tendencies (1995, 93). In sum, as the scope, boundaries, and purpose of civil society remain contested amongst scholars, it thus becomes problematic to compare the actions of civil society across cases without first investigating the reaso ns that initially precipitated its rise. Through the previous chapters that have condensed the most salient arguments to their common denominator, this dissertation has shown how the continued presence of the political opposition primarily (though not exclusively) through the largest opposition party, the NPP remained instrumental in producing As explained in Chapter 4 , d espite boycotting the inaugural parliamentary elections and aban doning any formal role within the inaugural government of the Fourth Republic, the NPP remained active within Ghanaian society: they maintained a high public profile, and led what may be reasonably considered a successful legal campaign against the NDC gov ernment. To handedly coerce Rawlings and the NDC to democratically reform, when combined with pressure from international donors (as previously noted in Chapter 3), their public pr esence and efforts nevertheless incentivized the government to do so. continued political presence and subsequent legal victories could be regarded as a remember that
207 with the opposition, and often at the expense of Rawlings (Frempong 2007) . In sum, a s the NPP demonstrated that it could pose a credible th reat to the NDC by its competitiveness in the inaugural elections, as well as through its ability to weaken threat by removing their most salient argument from the equation. In fact, the historical record demonstrates that this was likely the case, as Rawlings allowed for continued democratic reform in his first term as President. Perhaps the most important ref included the rapprochement with the opposition, ultimately resulting in the creation of IPAC and the implicit acknowledgement that the opposition would have a say in , while the creation of IPAC in 1994 creation stand as more salient than what the current literature suggests (Frempong 2007; Asante 2013), a s a fundamental elite consensus re determining th emerged prior to the other kind of elite consensus that appeared To be sure, such an explanation does not marginalize the importance of the inter national environment, as well as the actions of international donors during this period. As explained in Chapter 3, Ghana , as well as many other Africa n countries, was dependent on foreign aid to support its ongoing economic recovery. In addition, due to t relatively weak economy, its diminished strategic importance in a post Cold War environment, and its inability to rely on non Western support as an alternative
208 to Western aid, the country thus had little leverage in negotiating the terms of as sistance with donors (Levitsky and Way 2010) When oppositi on parties boycotted the inaugural p arliamentary elections under the percep tion that the presidential elections had been rigged in favor of the NDC, Rawlings and his government subsequently faced a crisis of legitimacy, as donors were unwilling to provide assistance to a government regarded as illegitimate by a significant propor tion of the electorate. As a result, Rawlings was strongly encouraged to find a means for resol ving the impasse and reintegrating the opposition into the electoral system . Otherwise, the country faced the prospect of international aid being suspended, or e ven withdrawn (Green 1998, 199; Boafo Arthur 1998, 174; Devarajan, Dollar, and Holmgren 2001, 77) . Together, Chapters 3, 4, and 5 show that in order to resolve this crisis , Rawlings reengaged with the opposition. Operating under the Electoral Commission an d under the recommendations of donors, Rawlings al lowed for the creation of IPAC and an environment where the opposition had an influential role in shaping the rules and norms of the system . In addition, Rawlings and his government allo wed for additional democratic reforms, including loosening restrictions on the press and the expansion of private me dia outlets. In short, Rawlings succeeded in reintegrating the opposition by both acceding to a certain number of their demands, as well as by providing them a stake in the political system that they had previously decried. Thus , it is reasonable to argue that without the sustained and significant presence of the domestic political opposition, Rawlings and the NDC would have had
209 little reason to create IPAC, as there would not have been a viable domestic political elections. In addition, even if international pressure succeeded in pressing the government to c reate such an institution in the absence of a minimally competitive political opposition, it is likely that there would have been little incentive to enact any recommended reforms, as the opposition would have likely represented less than a sizeable portio n of the population. 2 As a result, both the political and electoral environment would likely have remained unchanged, thus limiting the ability of independent , pro democracy civil society organizations to operate and facilitate the return of a vibrant and active civil society , as argued by previous scholars (Gyekye Jandoh 2006; Abdulai and Crawford 2010; Arthur 2010) . In sum , though factors such as civil society, IPAC, and the continued holding of regularly scheduled elections have certainly been influentia l in promoting democratic reform in Ghana, this dissertation and the chapters therein have shown that these forces would likely not have occurred without the presence of four factors: (1) the country being in significant need of international assis tance; ( democracy orientation, as well as the significant amount of leverage donors possessed over the terms of their aid; (3) a resilient and minimally competitive opposition that additionally possessed the inherent ability to undermi s continued access to foreign aid ; and (4 ) the pragmatic leadership of Rawlings, who ultimately chose to follow the path of least . 2 To be sure, while Beaulieu (2014, 50) argues that when faced with strong international pressure non demo cratic governments across the developing world may implement democratic reforms despite the absence of strong opposition, the results of this dissertation demonstrate such circumstances are nevertheless likely to produce non democratic trajectories in the long term.
210 As Chapters 3 thro ugh 5 have demonstrated, a bsent any of these factors, Ghana democratic opening would have likely produced a non democratic regime trajectory. As a previously deviant case of democracy in Africa that remained unexplained by general theories of democratiza tion (Bratton and van de Walle 1997; Levitsky and Way 2010) , I utilized the insights from the Ghanaian example towards building a n alternative theory of regime trajectories in Africa. In short, in a period of significant change following the end of the Col d War, domestic and international pressure informed the consolidation or extension of their rule, including the decision whether to tolerate democratic demands from the political opposition. The results from t his exchange, in turn, produced distinct paths of democratic, mixed, and authoritarian trajectories that would prove difficult to change due to path dependent constraints and self reinforcing behavior. Operationalized into three factors, including domestic political opposition, reliance on foreign aid, and international leverage, Chapter 6 demonstrated how such domestic and international pressure constrained the range of choices available to newly elected executives regarding the issue of repressing or tole rating continued democratic reform . Where executive actions accorded with structural conditions forecasting democracy or authoritarianism , such countries proceeded along democratic or authoritarian trajectories , respectively ; where such decisions differed, or where structural considerations provided ambiguous information to executives regarding the ir , countries instead followed mixed trajectories.
211 Excluding what may be reasonably considered extraordinary case s, the results from this analysis account for 23 out of 25 cases, providing insights as to why some countries succeeded in maintaining democratic syste ms of rule while others did not. Figure 7 1 provides a diagrammatic summary of the model and its findings . Though the model does not account for the regime trajectories of Mozambique and the Seychelles , the fact that these were borderline cases whose divergence was largely attributable to ambiguity in executive decisions (Mozambique) and ultimate regime traje ctory (Seychelles) demonstrates that their deviance was a matter of degree and not of kind . As a result, this model and its framework for explaining regime trajectories in Africa after the Cold War provides the most accurate assessment of democracy and aut horitarianism in Africa to date. 3 To be sure, while some may argue that the contributions of international aid and leverage are epiphenomenal to the production of democratic trajectories, as such countries were likely to democratize on their own accord (D utta , Leeson, and Williamson 2011; cited in Resnick and Van de Walle 2013, 3) , the results from this dissertation instead emphasize how international pressure and aid influenced democratic development in Africa after their democratic openings . In other wor ds , this model of regime trajectories does not investigate international aid in regards to how it initiated democratic openings in Africa , but rather investigates how such aid produced and maintained democratic trajectories after the ir inaugural elections. 3 The most recent assessment of democracy in Africa accounts for 11 out of 14 cases, with every instance of democracy in Africa, including Ghana, remaining unexplained (Levitsky and Way 2010, 306).
212 Figure 7 1: Diagrammatic Summary of Regime Trajectories in Africa 4 In sum, by investigating the process by which a country with a previously flawed democratic transition nevertheless became the only one of its kind to be recognized as a contemporary d emocracy (Bratton and van de Walle 1997, 120), and to have 4 Note: Equatorial Guinea, Mozambique, Seychell es, and South Africa not shown.
213 furthermore succeeded as being recognized as a liberal democracy in Africa (Diamond 2010, xxvi), alternative model of democ ratization in Africa. The results from this model building exercise provided an updated explanation for highlighting the significance of the domestic political opposition as being a sustained and resilient presence durin domestic and international politics interacted towards pressuring the government to tolerate additional democratic reform, and consequently placing the country on a democratic trajectory in the pro cess . Furthermore , the results of this exercise provided additional insights regarding the experiences of 26 other African countries attempting the same process that Ghana started in the period immediately following the end of the Cold War. In doing so, th e results from this analysis aid in explaining the reasons why some countries in a period of democratic change succeeded in developing democratic systems of rule, while others instead reverted back to authoritarianism or otherwise adopted ambiguous and /or inconsistent regimes. path to democracy serves to demonstrate that democratization is not an event that occurs at a single moment in time; rather, it is a process that develops over time. At in 1992, Ghana appeared predisposed to a non democratic, competitive authoritarian regime (Levitsky and Way 2010). Scholars additionally ci rcumspect about its democratic future (Nugent 1995; Oquaye 1995; Bratton and van de Walle 1997, 120; Green 1998, 205 6; Gyimah Boadi 1999, 171; Lyons 1999, 163). Nevertheless, the country proceeded to follow a steadfast path of
214 democratic improvement. Only when the main opposition party, the NPP, won the 2000 elections did researchers generally regard the country as a democracy (Freedom House 2015). As a result, the Ghanaian experience demonstrates that at times, an emphasis on democratic transitions may mi slead researchers to overlook the events and consolidation. As noted in Chapter 1, Ghana and van de Walle 1997), yet nevertheless dev eloped into a contemporary democracy. the Central African Republic, and the Republic of Congo) subsequently became contemporary non democratic regimes. If the kind of transition a country experiences explains little with regard to its ultimate regime type, then it is worth reevaluating the utility that such an emphasis on transitions provides, and it is additionally worth considering what other approaches may provid e instead. T his dissertation shows how a reconceptualization of regime change that emphasizes the kind of regime trajectory a country has followed over a course of time, rather than the political system a country possesses at a given moment in time, can pr ovide additional its standing in the future. status omits the political turbulence the country has experienced since its initial democra tic democratic status forgets the democratic optimism that surrounded the countr lections
215 in 1992 when the political opposition candidate, Pascal Lissouba, defeated the then incumbent presi dent, Denis Sassou Nguesso . In addition, such a perspective omits the a s well as Sassou In sum, a perspective that utilizes regime trajectories as its unit of analysis (produced after democratic openings that are precipitated by critical junctures), as opposed to an emphasis on tr ansitions, provides new information as to why some African countries either succeeded or failed to maintain democratic systems of rule , in spite of each attempting democratic reform within the same time period . To be sure, this does not suggest a n outright rejection of the transitions paradigm or the obsolescence of the transitions framework . At the same time, however, the Ghanaian experience and the results from this analysis suggest the existence of alternative s to the transitions perspective. The use and operationalization of regime trajectories can serve as an additional tool for students of democratization. While an emphasis on democratic transitions has rightly aided in the study of political change both within and outside of the African continent, the results of this dissertation remind us that the ways in which political phenomena are conceptualized affect the results of (Goertz 2006) . Simply assuming that political change occurs solely through transitions risks both under and over estimating the ir impact: it may underestimate the gravity that such political changes produce, such as mistaking a democratic opening (with a resultant regime trajectory) for a transition ; and it may also overestimate the impact that alleged democratic tra nsitions may have with regards to producing robust democratic regimes. The point is that rather than assuming that
216 democratization can only occur through one mechanism, researchers can scrutinize cases of political change in order to ascertain whether they are, in fact, observing a democratic transition, or whether they are instead viewing something else entirely. In addition, the results of this analysis remind us that a ny cross sectional study purporting to demonstrate a cause of democracy or even its imp rovement thereof ought to be treated with an extra degree of circumspection. This is because such studies must inherently acknowledging the journey that country has taken to get to that particul ar moment in time. Moreover, because this study has shown that some countries can and do move between democracy and authoritarianism , sometimes in a span of less than one decade , such studies cannot account for the likelihood that one of their cases will r emain within a democratic or non democratic regime. In other words, by treating all cases of democracy (or authoritarianism) the same without differentiating between the kinds of trajectories that these cases may follow, any findings derived will be limite d insofar as a number of cases will subsequently and inescapably experience change in their regime status . In sum, t he lessons of the Ghanaian experience show that sometimes democratization can neither be regarded as occurring at particular points in time during at other moment s in time. As the Ghanaian case illustrates, the coun try embarked on a slow and steadfast process of democratic change that took years to develop before the country was widely considered a democracy. However, to regard its success as a product of a democratic transition remains problematic, as the event that led it to be widely recognized as a
217 democracy occurred approximately eight years after its alleged transition, whil e at the same time, all other countries experiencing similar types of transitions subsequently developed into non democratic regimes. 5 As a result, the Ghanaian experience demonstrates that periods of political change may put in motion a set of forces whos e consequences are not observed until much later. Challenges to tic Trajectory Since its de democratic trajectory has experienced two peaceful turnovers of power in 2000 and 2008. In addition, the death of Jo that facilitated a smooth transition of power to the then Vice President John Dramani 6 Most recently, the 2012 elections demonstrat Supreme Court. 7 However, while democratic trajectory has demonstrated resilience whe n faced with adverse circumstances , it nevertheless faces other First , the issue of corruption has been a longstanding concern within Ghanaian politics. Recent reports of misappropria tion of funds within government agencies not 5 The main exception to this pattern being Kenya, which may be reasonably considered a contemporary democracy, though its history demonstrates a mixed regime trajectory. 6 BBC News , 25 July, 2012. Online. http://www.bbc.com/news/world africa 18980041 (April 8, 2015). 7 The Economist , 7 September, 2013.
218 (Freedom House 2014b) . 8 In addition, a 2013 report from Transpa rency International noted that in a survey regarding perceptions of corruption , respondents perceived the police (92 per cent), the corrupt institutions in the cou ntry. Furthermore, 79 per cent of respondents in the same survey reported having paid bribes or other kinds of illegal payments to the police services, while 54 per cent reported corruption to have worsened over the previous two years. 9 In addition , concer ns of increased narcotics smuggling facilitated by elites and government officials create the fear that democracy will suffer legal and security apparatus decay due to the corrupting influence of monies used to smuggle such drugs i nto the country . 10 More recently, the arrest of a Ghanaian substantial controversy in Ghana after it was found that the smuggler was associated with several government minis ters, and was additionally granted access to the private lounge, thus being exempt from inspection prior to boarding a plane to London. 11 The previous year , 8 The Daily Graphic , 3 March 2014. 9 The Economist , 11 July, 2013. Online. http://country.eiu.com/article.aspx?articleid=310709615&Country=Ghana&topic=Politics# (October 29, 2013). 10 The Guardian , 14 December, 2010. Onlin e. http://www.theguardian.com/world/2010/dec/14/us embassy cables ghana airport drugs (April 8, 2015). 11 BBC News , 26 November, 2014. Online. http://www.bbc.com/news/world africa 30212820 (April 8, 2015).
219 international airport was charged with conspiring to smuggle heroin into the U.S., while in 2005 a minister of parliament was similarly arrested for drug trafficking. 12 Secondly , like many other African countries, neopatrimonialism has been a persistent feature in Ghana (Bratton and van de Walle 1997 ). Cli entelism remains a strong feature within domestic politics , as politicians and elites are often required to dispense favors (including money, jobs, or other good s and services) in exchange for political support (Lindberg 2003; 2010 ). As Lindberg notes, in addition to the formal duties of parliamentarians which traditionally include providing legislation, executive oversight, constituency representation, and constituency service parliamentarians in Ghana are additionally expected to provide private goods in the form of favors, personal assistance, cash handouts and Ghana often oblige persons with wealth or power to provide assistance to their community or kin group, ministers of parliament in Ghana often become fr ustrated as their most common expectation is to provide private goods to constituents, as opposed to their four other formal duties. As a consequence, resources that can be used for public development projects, as well as other kinds of public goods, dimin ish over time. In sum, the effects of corruption and clientelism continue to challenge the the state to enforce its laws . This is important, a s an essential prerequisit e of democracy is a sovereign state (Linz and Stepan 1996 , 17). 13 As corrosive forces of 12 BBC News , 5 June 2013. Online. http://www.bbc.com/news/world africa 22780761 (April 8, 2015). 13 occupying a definite t 1975, 70; cited in Linz and Stepan 1996, 17).
220 corruption undermine the ability of the state to act needs and demands, in a non discriminatory ma nner, it is reasonable to argue that corruption stands as one of the greatest challenge s to democratic direction. Thirdly, the discove ry of oil in 2007 and the potential income that its extraction will produce for the country has become an area of concern , primarily in regards to how (Gyimah Boadi and Prempeh 2012). While there is optimism that the c will help ensure ts will lead to corruption and a consequent decrease in democracy ), and that oil revenues will be channeled towards improving infrastructure and overall development, it is not guaranteed. As Gyimah Boadi and Prempeh note, the inclusion of oil rents into the Ghanaian political system threaten to exacerbate both political and ethnic rivalries ( 2012, 101 2). A s politics in the country is already largely considered a zero sum affair (due to, in part, by a first past the post electoral system), and as the executive generally retains a large amount of discretion in managing and allocating government revenues, the fear is that the inclusion of oil money into Ghanaian politics will only serve to amplify political competition in the country towards a or will be so great, while the costs of losing will be too much to bear . Thus, it remains to b e seen whether the country will successfully manage this resource without sacrificing its democratic progress in the process.
221 effectively decided the 2012 presidential election in favor of the incumbent, John Mahama (NDC). Addo, nevertheless calculated a 14 In addition, personal observations at a symposium sponsored by the Danquah Institute (a generally pro NPP Ghanaian NGO) . In attendance at the symposium were many elites within the NPP leadership, including the party chairman, and various stated purpose was to analyze the result of the Supreme Court judgment , the executi on of the event instead turned into a general rebuke of the Court and its decision , as speakers generally argued that the Court had grossly erred in its judgment and that the NPP should have won the election. 15 One of the speakers at the Symposium remarked that rationale for deciding in favor of the NDC was so flawed that it paradoxically supported the NPP position , thus demonstrating that the Court ought to have ruled in favor for the NPP. 16 Another remark ed t the co and went as far to ask those in attendance take a firm position against any system of impunity that has plagued this nation over the years , implying that either the Court was tol erating illegality, or worse, was acting in an 14 The Economist , 7 September, 2013. 15 Danquah Institute Symposium, National Theater, Accra, Ghana, 25 September, 2013. 16 e Court Judgment Was 5 The Daily Graphic , 26 September, 2013.
222 illegal manner when forming its decision. 17 That these remarks openly challenged the received by those in attendance a t the event, including prominent leaders of the NPP, suggests that future elections may be problematic as both citizens and elites may be less likely to accept electoral defeat, and may furthermore be less likely to resolve electoral disputes through forma l channels of resolution , especially if they doubt the credibility, integrity, and legitimacy of the process . At the very least, the execution and subsequent analysis of the 2012 election s demonstrate that additional electoral reforms particularly i n area s of accountability and the training of election officers will likely be necessary ahead of the 2016 elections. 18 concerns of corruption in the country, it is not unr easonable to suggest that such factors, if left untreated, could eventually become strong enough so as to alter the returning the country to the kinds of conditions that i nitially brought about the period of military rule approximately 34 years ago . Ghana at a Crossroad: The 2016 Elections approaching a crossroad with regard to its democracy. Frustrat ions with rising levels of inflation and corresponding increases in basic commodities have resulted in widespread 17 The Daily Graphic , 11 October, 2013. 18 Nearly all respondents interviewed for this dissertation acknowledg ed that electoral reform would be necessary ahead of the 2016 elections, which would likely include improving the training and preparation of election officers, as well as establishing better accountability mechanisms in order to accurately and consistentl
223 protests around the country, including the capital city, Accra. 19 In addition to these concerns, grievances also include rising electricity and water fees, fuel shortages, and insufficient energy production resulting in rolling power outages throughout the country. More recently, despite receiving debt relief in 2002, government overspending has brought total public debt to record levels, constit ut ing more than 67 per cent of the and requiring IMF intervention . 20 Looking forward to the 2016 elections, it appears as if the country has two directions in which it can proceed . On the one hand, the country can maintain the status quo and continue to the effects of corruption and entrenched clientelism (Lindberg 2003; 2010). Many re spondents shared this concern, and additionally noted their fears that the 2016 elections cont ain the potential for provoking unrest and po litical violence in the country . 21 On the other hand, however, the country and the population can respond to such challenges by performing the necessary electoral and economic reforms, and subsequently ensuring t future. To be sure, t here are promising indications that the government is genuinely interested in reform. With consultation from citizens and civil society organizations, the 19 The Daily Graphic , 2 July 2014. 20 H Â¢ 34.6 Billion The Daily Graphic , 19 February 2015. 21 Interv iew with Senior Lecturer, University of Ghana, Legon Center for International Affairs and Development, Legon, Ghana (September 11, 2013); interview with professor at the University of Ghana Institute of Statistical, Social, and Economic Research, Legon, Gh ana (October 14, 2013); interview with senior NPP Minister of Parliament, State Parliament House, Accra, Ghana (November 21, 2013); interview with senior official at the Ghana Institute for Democratic Governance, Accra, Ghana (November 25, 2013); interview with official at the Ghana Bar Association, GBA Headquarters, Accra, Ghana (March 18, 2014).
224 government rece ntly undertook a constitutional review exercise that sought to remedy removing sitting m embers of Parliament, that the number of Supreme Court justices be limited to 15, and that the country institute a National Development Plan (CRC 2011). The government has been rather supportive, though notable exceptions include its disapproval over establishing a National Development Plan (GoG 2012). In addition, the Government of Ghana has recently entered into an agreement with the IMF in order 22 Though complete details are unclear, it is expected that the G overnment of Ghana will receive approximately $940 million from the IMF in exchange for implementing economic reforms, including a likely decrease in government spending, a reduction in fuel and energy subsidies, as well as other undisclosed measures. In s not without their challenges. As issues of corruption, drug smuggling, oil revenues, and lingering tensions from the 2012 elections threaten to undermine the future of the democratic trajectory, leaders and citizens nevertheless recognize these concerns and are engaging in ways to combat their pernicious effects and to maintain challe the demand for democracy has not waned. For example, the most recent Afrobarometer (2012) survey reports that 82 per cent of Ghanaian respondents still prefer democracy to any other kind of system, 22 BBC News , 26 February, 2015. Online. http://www .bbc.com/news/business 31652406 (April 8, 2015).
225 while 86 per cent of respondents agreed with the statement that politic al violence is satisfaction with democracy remains wanting, as 71 per cent of respondents reported their count ry to be a democracy with problems, while 25 per cent of respondents not ed dissatisfaction with the way democracy operates in the country. In sum, much like elsewhere in Africa, there exists a gap between Ghanaians expectations of democracy and their sati sfaction with it (Bratton 2010). As Ghana looks previously noted challenges, and it is similarly important that the citizenry select the leaders that are most likely to s ucceed in doing so. Fortunately, the history of Ghanaian politics in the Fourth Republic demonstrates voters to be increasingly retrospective when choosing to retain or remove their political leaders ( Lindberg and Morrison 2008 ) . S hould the current governm it is likely that it will be replaced in the subsequent election. T hus , while there continues to exist a viable alternative to the political status quo in Ghana, and as both major parties continue to require the support of the electorate in order to govern, there remains the expectation that democratic trajectory will continue into the future. Prospects For Democracy In Africa This dissertation has shown post Cold War democratization in Africa t o be a process that occurs over a period of time, as opposed to an event occurring at a moment in time. In addition, it has shown that regime trajectories do not emerge by themselves, but are instead a function of the democratic openings from which they or iginated. As a result, these findings suggest that it is often necessary to adopt a long term perspective when commenting on the spread of democracy in Africa. In addition,
226 this dissertation shows that it is necessary to evaluate the kinds of democratic ch ange that occur on the continent in order to determine if such political changes are sufficient Furthermore, while the results of this dissertation explain why many regimes in Africa ha ve remained in a state somewhere between democracy and authoritarianism, this does not suggest that stable democracy remains outside the reach of millions of those residing within countries with these types of trajectories . Indeed, democracy remains a high priority for a majority of Africans (Bratton 2010). In stead, what the results show is that for those countries continuing along both mixed and authoritarian trajectories, the pace at which such desired reforms are likely to arrive will be less than those countries following democratic traj ectories. In addition, developing democracy in such conditions will likely require additional action perhaps even confrontation with the incumbent governments from the citizenry and civil society organizations representin g their democratic interests (LeBas 2013) . In other words, changing non democratic trajectories towards democracy in such countries will likely require bringing about new democratic opening s in order to o gain by working within a democratic system than without it. As recent events in Burkina Faso, Mali, and Kenya offer hope that these countries will shed their mixed regime trajectories and restart the democratic process, it is important to consider the co ntexts in which such changes have occurred, and it is additionally important to consider whether these events indicate a change from the regime democratic opening. Are these new att empts at democracy the signal of new
227 democratic opening s , or are they instead a n effect of the same political forces that have kept these countries on their respective mixed regime trajectories for more than twenty years ? The answer to this question remain s beyond the scope of this dissertation, however scholars and researchers interested in these countries have the opportunity to determine whether the forces that caused breakdown in these countries are an effect of a critical juncture that has yet to be id entified. Should this , in fact, be the case, such countries thus have the opportunity to experience another democratic opening, and a second chance at establishing a democratic trajectory. In addition, while the scope of this dissertation only covers those African countries that experienced democratic openings in the wave of democratization that occurred follo wing the end of the Cold War (27 countries), this is not to suggest that other currently non democratic countries on the continent are foregone to aut horitarianism. Rather, the results of this project only demonstrate one way of democratization. While there are multiple ways at obtaining democracy in Africa, the results of this dissertation show that for those c ountries experiencing future democratic op enings, they can reinforce their gains through strong international support, effective international leverage, a sustained domestic opposition, and a pragmatic leadership willing to tolerate democratic reform . Implications T he implications of these findin gs contain policy relevance for those interested in the promotion of democracy on the continent. First, by ability to foster democratic reform is a function of the international leverage behind such aid, as well as the domes tic context in which it is provided, future democracy promotion policies may be augmented by focusing aid allocations towards countries with a viable
228 and sustainable domestic political opposition . For other countries, the findings of this dissertation show that current strategies for promoting democracy through civil society and representative organizations have the potential to not only strengthen civil society , but may also help facilitate the growth of a political environment where opposition parties can thrive and compete. Secondly, the implications of this analysis may also be used towards identifying future democratic openings in Africa and assessing the likelihood that such openings will produce democratic or non democratic trajectories . For countries with propitious democratic openings and whose structural conditions suggest a high likelihood for producing a democratic trajectory, policymakers may act to convince either incumbent or newly elected executives to advance democratic reform in their respec tive countries . For other countries whose openings suggest a non democratic trajectory, both policymakers and practitioners may nevertheless identify areas requiring improvement, and may thus act to improve such conditions before a non democratic trajector y becomes fully institutionalized. To be sure, the scope of a theory of regime trajectories is not infinite. I ts applicability ought to be contained to situations involving democratic reform brought about by periods of significant change sufficient for pro ducing critical junctures . It is incumbent on researchers to critically analyze their cases and determine whether the political change they perceive occurs as a result of isolated circumstances, or whether it occurs as an effect of something much larger. In addition, t his model and its findings are not represen tative of all African countries: only those that experienced democratic openings in response to crises of the
229 status quo. At the same time, however, this model may nevertheless be relevant to other s tudies of democratization following critical junctures , both within Africa and abroad. In addition, this theory of regime trajectories is not an argument emphasizing the rigidity and inflexibility of regime trajectories. To be sure, embarking on a particul ar trajectory does not indicate that a particular country is either secured with in a democratic trajectory or a foregone conclusion in others. As the cases of Equatorial Guinea and South Africa demonstrate, subsequent domestic and international forces may over time grow strong enough to alter authoritarianism to democracy, and vice versa. Changing from one regime trajectory to another remains possible, though to be sure, such change will likely remain difficult due to sunk costs involving path dependence and positive feedback (Pierson 2004). Lastly, while the results of this dissertation assist in explaining the variety of regimes that emerged following the democratic wave of the early 1990s, additional avenues of research remain. First, as this investigation has documented executive decisions as they were given, there still remains the question as to why some national executives choose to follow what international and domestic conditions suggest, while others instead disreg ard such indications and follow their own preferences. Secondly , additional research is needed with regards to exploring mixed regime trajectories. As the results of this investigation demonstrate that most democratic openings in Africa resulted in mixed t rajectories, research in this area may further investigate differ ences among these kinds of trajectories, as well as inquire into the ways in which some mixed trajectories proceed along turbulent political route s while others instead remain stable with in a n ambiguous zone somewhere between democracy and authoritarianism . In
230 addition, research in this vein may investigate where, if anywhere, these different types of regime trajectories are headed, including whether some types of mixed trajectories are more o r less likely to be democratic than others. Lastly, research concerning the limits of regime trajectories would do well to investigate how regime trajectories have changed, if at all, to a changing global environment. As concerns of terrorism, the Arab Spr ing, and alternative sources of foreign finance challenge the post Cold War status quo, it will be necessary to inquire into whether such factors are strong enough to alter the course that such countries have followed for the past two decades. In sum, in a nswering the question as to how one country succeeded in developing a democratic regime despite experiencing inauspicious initial conditions, the results of this investigation have served to assist in answering a much larger question concerning why some co untries in a period of democratic change succeeded in developing democratic regimes while others did not. As the case of Ghana illustrates, democratization does not occur at a certain moment in time; rather, it occurs as a process over time. By continuing to investigate democrat ization as a long term process rather than as an event , as well as a function of the interplay between the interna tional and domestic environment during such periods of change, we may continue to increase our understanding with rega rd to explaining why some countries have remained democratic since their initial democratic openings , and we may additionally understand how others may be likely to remain so in the future.
231 APPENDIX A SUMMARY STATISTICS OF DEMOCRATIC OPENINGS IN AFRICA, 19 90 1994 Table A 1: Measuring Domestic Political Opposition, Relative Aid Dependence, and International Leverage Country Year of Opening Election Type Free and Fair? Voter Turnout Vote Share Aid /GDP , 1990 1994 (%) * Leverage Benin 1991 Pres. Yes 64% 36/27 1 5% High Burkina Faso 1991 Pres. No 27% 99/1 17% High Cameroon 1992 Pres. No 72% 40/36 5% Medium Cape Verde 1991 Parl . Yes 75 % 62/32 30% High Central African Republic 1993 Pres. Yes 69% 37/22 15% High Comoros 1990 Pres. No 64% 23/24 20% High Congo Rep ublic 1992 Pres. Yes 60% 36/20 9% High Djibouti 1993 Pres. No 51% 61/22 29% High Equatorial Guinea 1993 Parl. No 68% 85/8 38% High Gabon 1993 Pres. Yes 88% 51/27 3% Medium Ghana 1992 Pres. Yes 50% 59/30 11% High Guinea Bissau 1994 Pres. Yes 89% 46/22 51% High Ivory Coast 1990 Pres. Yes 69% 82/18 9% Medium Kenya 1992 Pres. No 66% 36/26 12% High Lesotho 1993 Parl. Yes 72% 99/1 19% High Madagascar 1993 Pres. Yes 74% 45/29 12% High Malawi 1994 Pres. Yes 80% 47/34 29% High Mali 1992 Pres. Yes 23% 45/1 4 18% High Mauritania 1992 Pres. No 47% 63/33 20% High Mozambique 1994 Pres. Yes 88% 53/34 54% High Niger 1993 Pres. Yes 33% 27/34 18% High S Ã£ o Tom Ã© and Pr Ã ncipe * 1991 Pres. Yes 60 % 82/0 n/d High Seychelles 1993 Pres. Yes 86% 60/37 5% High South Afri ca 1994 Parl. Yes 86% 63/20 0% Low Swaziland 1993 Parl. Yes 61% n/d 4% High Togo 1993 Pres. No 36% 96/3 12% High Zambia 1991 Pres. Yes 46% 76/24 24% High * = includes development and official aid n/d = no data available for indicator. Source: Quality o f Government 2012; Diamond 2010: xxvi; Lindberg 2006; Levitsky & Way 2010.
232 APPENDIX B DEMOCRA TIC OPENINGS IN AFRICA AND THEIR PREDISPOSED TRAJECTORIES Table B 1: Structural Conditions and Their Influence on Regime Trajectories in Africa High International Leverage Aid Dependent Domestic Political Opposition Predisposed Regime Trajectory Benin Yes Yes Yes Democratic Burkina Faso Yes Yes No Mixed Cameroon No No Yes Mixed Cape Verde Yes Yes Yes Democratic Central African Republic Yes Yes No Mixed Comoros Yes Yes Yes Democratic Congo Rep. Yes No No Authoritarian Djibouti Yes * Yes No Mixed Equatorial Guinea Yes Yes No Mixed Gabon No No No Authoritarian Ghana Yes Yes Yes Democratic Guinea Bissau Yes Yes No Mixed Ivory Coast No No No Auth oritarian Kenya Yes Yes Yes Democratic Lesotho Yes Yes No Mixed Madagascar Yes Yes No Mixed Malawi Yes Yes Yes Democratic Mali Yes No No Authoritarian Mauritania Yes Yes No Mixed Mozambique Yes Yes Yes Democratic Niger Yes Yes No Mixed S Ã£ o Tom Ã© an d Pr Ã ncipe Yes n/d No Mixed Seychelles Yes No Yes Mixed South Africa No No No Authoritarian Swaziland Yes No No Authoritarian Togo Yes Yes No Mixed Zambia Yes Yes No Mixed n/d = no data available for indicator . Source: Quality of Gover nment 2012: Bratton and van de Walle 1997 ; Lindberg 2006.
233 LIST OF REFERENCES Abdulai, Abdul Democratization 17(1): 26 67. Afari Gyan. Kwadwo. 1995. The Mak ing of the Fourth Republican Constitution of Ghana . Accra: Friedrich Ebert Foundation. Afrobarometer. 2012. Afrobarometer Data, Ghana, Round 5, 2010 2012, available at http://www.afrobarometer.org ( Accessed Apri l 8, 2015). Agyeman Ghana: Governance in the Fourth Republic , ed. Baffour Agyeman Duah. Legon: Ghana Center for Democratic Development, 155 94. Agyeman Duah, Baffour. 2012. My Ghanaian Odyssey . Tema: Digibooks Publishers. Ake, Claude. 2000. The Feasibility of Democracy in Africa . Dakar: CODESRIA. Amenumey, D.E.K. 2011. Ghana: A Concise History From Pre Colonial Times to the 20 th Century . Accra: Woeli Publishing Services. Armstrong, Rober t P. 1996. Ghana Country Assistance Review: A Study in Development Effectiveness. Washington, D.C.: The World Bank. Africa Today 56(2): 45 73. tic Consolidation in Ghana: The Role and Contribution of Commonwealth & Comparative Politics 48(2): 203 26. Public Entities and the Drama of Elections i Journal of African Elections 12(2): 56 74. Electoral Studies 16(3): 416 27. Demo cratization 9(2): 148 174. Ghana: One Decade of the Liberal State , ed. Kwame Boafo Arthur. Dakar: CODESRIA: 165 87. Journal of Contemporary African Studies 26(3): 279 93.
234 Economic and Political Liberalization in Sub Saharan Africa , ed. Jennifer A. Widner. Baltimore: Johns Hopkins University Press , 13 28 . Bauer, P.T. 2000. From Subsistence to Exchange and Other Essays . Princeton: Princeton University Press. Beaulieu, Emily. 2014. Electoral Protest and Democracy in the Developing World . New York: Cambridge University Press. Bernard. H. Russell. 2006 . Research Methods in Anthropology: Qualitative and Quantitative Approaches , Fourth Edition. New York: Alta Mira. Boafo Ghana: Transition to Democracy , ed. Kwame A. Ninsin. Dakar: CODESRIA, 167 86. Boahen, Adu. 1992. The Ghanaian Sphinx: The Lectures that Broke the Culture of Silence. Accra: Sankofa Educational Publishers. Bob of Power Pol Journal of Modern African Studies 50(4): 573 601. Brady, Henry E., and David Collier (eds.) 2010. Rethinking Social Inquiry: Diverse Tools, Shared Standards , Second Edition. Boulder: Rowman and Littlefield . Bratt Cicil Society and the State in Africa , eds. John W. Harbeson, Donald Rothchild, and Naomi Chazan. Boulder: Lynne Rienner Publishers, 51 81. Democratization in Africa: Progress and Retreat , Second Edition, eds. Larry Diamond and Marc F. Plattner. Baltimore: Johns Hopkins University Press, 103 17. Bratton, Michael, and Nicolas Van de Walle. 1997. Democratic Exp eriments in Africa: Regime Transitions in Comparative Perspective . Cambridge: Cambridge University Press. Economic Development and Cultural Change 52: 255 85. Journal of Modern African Studies 50(4): 603 24. International Organization 63: 309 40.
235 Comparative Political Studies 43 : 931 68 . Journa l of Democracy 13(1): 5 21. Journal of Democracy 13(3): 33 8. Political Science Quarterly 97(3): 461 85. COG ( Commonwealth Observer Group ) . 1992. The Presidential Election in Ghana: 3 November, 1992: The Report of the Commonwealth Observer Group . London: Commonwealth Secretariat. Cooper, Frederick. 2002. Africa Since 1940: The Past of the Present . New York: Cambridge University Press. Collier, David, a Annual Review of Political Science 2 : 537 65. Collier, Ruth Beri ns, and David Collier. 1991. Shaping the Political Arena . Princeton: Princeton University P ress. Collier, Ruth Berins. 1999. Paths Towards Democracy: The Working Class and Elites in Western Europe and South America . New York: Cambridge University Press. CRC (Constitution Review Commission). 2011. Report of the Constitution Review Commission: Fro m a Political to A Developmental Constitution , Accra: Republic of Ghana. Coppedge, Michael, John Gerring, David Altman, Michael Bernhard, Steven Fish, Allen Hicken, Matthew Kroenig et al. 2011. "Conceptualiz ing and measuring democracy: A New A pproach." Per spectives on Politics 9( 2 ) : 247 267. Journal of Contemporary African Studies 27(1): 57 83. Democratization 6(4): 114 38. Dadzie, Nana Ato, and Kwamena Ahwoi. 2010. Justice Daniel Francis Annan: In the Service of Democracy . Accra: Institute for Democr atic Governance.
236 Dahl . Robert. 1971. Polyarchy: Participation and Opposition , New Haven: Yale University Press. New York Times September 19, 2002. http://www.nytimes.com/2002/09/19/world/in quietly courting africa white house likes dowry.html?src=pm ( Accessed 18 April, 2012). cess and the 2000 General Elections in Liberal Democracy and Its Critics in Africa , ed. Tukumbi Lumumba Kasongo. Dakar: CODESRIA, 125 48. Devarajan, Shantayanan, David R. Dollar, and Torgny Holmgren. 2001. Aid and Reform in Africa . Washington, D .C.: World Bank Publications. Democracy: A Reader , eds. Larry Diamond and Marc. F. Plattner . Baltimore: Johns Hopkins University Press, 229 43. Democratizati on in Africa: Progress and Retreat , Second Edition, eds. Larry Diamond and Marc F. Plattner. Baltimore: Johns Hopki ns University Press, ix xxviii. Diamond, Larry, Francis Fukuyama, Donald L. Horo witz, and Marc F. Plattner. 2014 . Journal of Democracy 25(1): 86 100. Diamond, Larry, and Marc F. Plattner (eds.). 1999. Democratization in Africa . Baltimore: Johns Hopkins University Press. Diamond, Larry, a nd Marc F. Plattner (eds.). 2010 . Democratization in Africa: Progres s and Retreat , Second Edition. Baltimore: Johns Hopkins University Press. Democratic Trajectories in Africa: Unraveling the Impact of Foreign Aid , ed s . Daniel le Resnick and Nicolas van de Walle. Oxford: Oxford University Press, 56 86. International Organization 58: 409 23. Dutta, N., P. Leeson, Institute Working Paper 821. EIU (Economist Intelligence Unit). 1990. Country Report: Ghana, Sierra Leone, Liber ia . No. 4. London: Economist Intelligence Unit.
237 ECG ( Electoral Commission of Ghana ) http://ec.gov.gh/page.php?page=402§ion=51&typ=1 ( Accessed Ju ne 24, 2014). ECG ( Electoral Commission of Ghana ) http://www.ec.gov.gh/assets/file/1996%20Parliamentary%20Election%20 Results. pdf ( Accessed January 18, 2015). Elischer, Sebastian. 2013. Political Parties in Africa: Ethnicity and Party Formation . New York: Cambridge University Press. l African Studies Review 38(2): 67 99. Fordwor, Kantinka K. Donkoh, 2010. The Danquah Busia Tradition in the Politics of Ghana: The Origins, Mission and Achievements of the New Patriotic Party , Accra: Unimax Mac millan Ltd. Fraser, Alastair. 2009. The Politics of Aid: African Strategies for Dealing With Donors , ed. Lindsay Whitfield. Oxford: Oxford University Press, 22 44. Freedom House. 2014 a Online. http://www.freedomhouse.org/sites/default/files/Country%20Ratings%20and%20 Status%2C%201973 2014%20%28FINAL%29.xls ( Acc essed November 3, 2014). Freedom in the World 2014 . Online. https://freedomhouse.org/report/freedom world/2014/ghana#.VSU6Q5TF 4I Acce ssed (April 8, 2015). Freedom in the World 1998 1999 . Online. https://freedomhouse.org/report/freedom world/1999/ghana#.VSQDGpTF 4I ( Ac cessed April 7, 2015). Ghana: One Decade of the Liberal State , ed. Kwame Boaf o Arthur. Dakar: CODESRIA, 128 64. Frempong , Alex Kaakyire Duku. 2012. Electoral P the Context of Post Cold War Africa . Accra: Yamens Press Limited. African Affairs 106(423): 281 305. Annual Review of Political Science 2: 115 44.
238 Gerring, John. 2001. Social Science Methodology: A Criterial Framework . Cambridge: Cambridge University Press . GSS ( Gh ana Statistical Service ) . 2002. 2000 Housing and Population Census, Summary Report of Final Results , Accra: Ghana Statistical Service. Goertz, Gary. 2006. . Princeton: Princeton University Press. Go G (Government of Gh ana). 2012. White Paper on the Report of the Constitution Review Commiss ion Presented to the President , Accra: Republic of Ghana. International Organization 55(1): 123 48. Green. Daniel. 199 The African State at a Critical Juncture: Between Disintegration and Reconfiguration , ed s . Leonardo A. VillalÃ³n and Phillip Huxtable. Boulder: Lynne Rienner, 185 212. Gyekye Jandoh, Maame Adwoa Gyimah Journal of Democracy 5(2): 75 86. Gyimah ety, and Democratization in Cicil Society and the State in Africa , eds. John W. Harbeson, Donald Rothchild, and Naomi Chazan. Boulder: Lynne Rienner Publishers. Gyimah Democratization in Africa , ed s . Larry Diamond and Marc F. Plattner. Baltimore: Johns Hopkins University Press, 170 83. Gyimah Journal of Democracy 12(2): 103 17. Gyimah Democratization in Africa: Progress and Retreat , ed s . Larry Diamond and Marc F. Plattner. Baltimore: Johns Hopkins University Press, 137 51. Gyimah Journal of Democracy 23( 3): 94 108. Gyimah Democratic Trajectories in Africa: Unraveling the Impact of Foreign Aid , ed s . Danielle Resnick and Nicolas van de Walle. Oxford : Oxford University Press, 256 80.
239 Harbeson, John W., Donald Rothchild, and Naomi Chazan (eds.). 1994. Civl Society and the State in Africa . Boulder: Lynne Rienner Publishers. Commonwe alth & Comparative Politics 41(1): 48 76. Herbst, Jeffrey. 1993. The Politics of Reform in Ghana, 1982 1991 . Berkeley: University of California Press. Horowitz, Donald. 2000. Ethnic Groups in Conflict . Berkeley: University of California. Huntington, Samuel P. 1991. The Third Wave: Democratization in the Twentieth Centrury . Norman: University of Oklahoma Press. Hyden, Goran. 2013 . African Politics in Comparative Perspective , Second Edition. New York: Cambridge University Press. Jebuni, Charles D., and Abena Ghana: Transition to Democracy , ed. by Kwame A. Ninsin. Dakar: CODESRIA, 21 48. Urban Popular Attitudes towards the Economic Recovery Programme and the PNDC Government in Ghana African Affairs 91(363): 207 26. African Affairs 97(387): 189 208. African Affairs 92 (368): 331 66. Making , ed s . Kwame A. Ninsin and Francis K. Drah. Accra: Gha na Universities Press, 77 83. Party Politics in Ghana: Transition to Democracy , ed. Kwame A. Ninsin. Dakar: CODESRIA, 83 108. to Multi Party Democracy in in Ghana: Transition to Democracy , ed. Kwame A. Ninsin. Dakar: CODESRIA, 187 210. Authoritarian Stability or Democratization? Evidence from Postcom munist Perspectives on Politics 11(3): 734 52.
240 The Association For Diplomatic Studies and Training Foreign Affairs Oral History Project . Frontline Diplomacy, Manuscript Division, L ibrary of Congress, Washington, D.C. , September 8, 1999. Online. http://www.adst.org/OH%20TOCs/Fritts,%20Robert%20E.toc.pdf ( Accessed February 25, 2015). Kennedy, Charles Stuart. 20 The Association For Diplomatic Studies and Training Foreign Affairs Oral History Project . Frontline Diplomacy, Manuscript Division, Library of Congress, Washington, D.C. , September 8, 1999. Online. http://www.adst.org/OH%20TOCs/Robertson,%20Nicolas.toc.pdf ( Accessed February 25, 2015). King, Gary, Robert O. Keohane, and Sidney Verba. 1994. Designing Social Inquiry: Scientific Inference in Quali tative Research . Princeton: Princeton University Press . Klotz, Audie. 1999. Norms in International Relations: The Struggle against Apartheid . Ithaca: Cornell University Press. Democ Journal of Politics 71(2): 704 18. Lancaster, Carol. 2007. Foreign Aid: Diplomacy, Development, Domestic Politics . Chicago: University of Chicago Press. LeBas, Adrienne. 2013. From Protest to Parties: Party Building & Democratization in Afr ica . Oxford: Oxford University Press. The Journal of Democracy 13(2): 51 65. Levitsky, Steven, and Lucan A. Way. 2010. Competitive Authoritarianism : Hybrid Regimes After the Cold War , Cambridge: Cambridge University Press. Method Strategy for American Political Science Review 99 (3): 435 52. American Political Science Review 65: 682 698 . Patrimonialism Rather Than Counter Democratization 10(2): 121 40. Lindberg, Staffan. 2006. Democra cy and Elections in Africa , Baltimore: Johns Hopkins University Press.
241 Lindberg, Staffan (ed.). 2009. Democratization by Elections: A New Mode of Transition , Baltimore: Johns Hopkins University Press. es Do MPs in Africa Face and Journal of Modern African Studies 48(1): 117 42. Lindberg, Staffan I. and Minion K. C. Morrison. 2008. Political Science Quarterly 123 (1): 95 122. Handbook of Political Science , eds. F.I. Greenstein and N.W. Polsby. Reading: Addison Wesley, 175 353. Linz, Juan J., and Alfred Stepan. 1996. Pr oblems of Democratic Consolidation: Southern Europe, South America, and Post Communist Europe . Baltimore: John Hopkins University Press. Beyond Humanitarianism: What You Need to Know About Africa and Why It Matters , eds. Princeton N. Lyman and Patricia Dorff. New York: Council on Foreign Relations, 45 68. Democratization in Africa , ed. by Larry Diamond and Marc F. Plattner. Baltim ore: Johns Hopkins University Press, 157 69. Mahama, John Dramani. 2013. Lost Decades of Africa . New York: Bloomsbury. A Tale of Two Cultures: Contrasting Quantit Political Analysis 14: 227 249 . Moore, Barrington. 1966. Social Origins of Dictatorship and Democracy . Boston: Beacon Press. to Democrat Comparative Politics 36(4): 421 42. Munck, Gerardo L. 2009. Measuring Democracy: A Bridge Between Scholarship and Politics . Baltimore: Johns Hopkins University Press. Mustapha, Abdul Raufu, and Lindsay Whitfield. 2009 a isation: The Turning Points in African Democracy , eds. Abdul Raufu Mustapha and Lindsay Whitfield. Rochester: James Currey . Mustapha, Abdul Raufu, and Lindsay Whitfield (eds.). 2009 b . Turning Points in African Democracy. Rochester: Jame s Currey.
242 NCD ( National Commission for Democracy ) , and Provisional National Defense Council (Ghana). 1991. the Establishment of a New Democratic Order: Report Submitted to the PNDC . Accra: National C ommission for Democracy. NPP ( The New Patriotic Party ) . 1993. The Stolen Verdict: Ghana November 1992 Presidential Election: Report of the New Patriotic Party . Accra: New Patriotic Party. Ghana: Transition to Democracy , ed. By Kwame A . Ninsin. Dakar: CODESRIA, 211 29. Ninsin, Kwame A. 2008. Executive Parliament Interface in the Legislative Process (1993 2006): A Synergy of Powers? Accra: Institute for Democratic Governance. No rth, Douglas C., 1990. Institutions, Institutional Change, and Economic Performance . Cambridge: Cambridge University Press. Nugent, Paul. 1995. Big Men, Small Boys, and Politics in Ghana . Accra: Asempa Publishers. African Issues 29(1/2): 2 7. Nugent, Paul. 2004. Africa Since Independence . New York: Palgrave Macmillan. Journal of Democracy 7(2): 34 51. Journal of Democracy 13(3): 6 12. 1986. Transitions From Authoritarian Rule: Tentative Conclusions about Uncertain Democracies . Baltimore: Jo hns Hopkins University Press. Commonwealth & Comparative Politics 42(2): 242 73. A Dissentin African Affairs 94(375): 259 75. Oquaye, Mike. 2004. Politics in Ghana: 1982 1992 . Accra: Tornado Publications. Oquaye, Mike. 2010. The Hybrid Constitution and Its Attendant Difficulties . Accra: Institute of Economic Affairs.
243 Pierson, Paul. 2000. The American Political Science Review 94 (2): 251 67. Pierson, Paul. 2004. Politics in Time: History, Institutions, and Social Analysis . Princeton: Princeton University Press. Przeworski, Ad am. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America . Cambridge: Cambridge University Press. Przeworski, Adam, Michael E. Alvarez, JosÃ© Antonio Cheibub, and Fernando Limongi. 2000. Democracy and Development . Cambridge: Cambridge University Press. Ragin, Charles. 1987. The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies . Berkeley: University of California Press. in A Revolutionary J ourney: Selected Speeches of Flt Lt. Jerry John Rawlings, Chairman of the PNDC , Volume 1 . Accra: Information Services Department, 1 3. Rawlings, Jerry John. 1983 b A Revolutionary Journey: Selected Speeches of Flt Lt. Jerry John Rawlings, Chairman of the PNDC , Volume 1 . Accra: Information Services Department, 38 9. Rawlings, Jerry John. 1990. Constitution Must Be Borne By Truth and Integrity . Accra: Information Services Department. Freedom, Justice and Accountability Essential for Nation Building : Selected Speeches of Flt Lt. Jerry John Rawlings, Chairman of the PNDC , Volume 9 . Accra: Information Services Department, 64 9. RoG ( Republic of Ghana ) . 1992. C onstitution of the Republic of Ghana . Accra: Ghana Publishing Corporation. Democratic Trajectories in Africa: Unraveling the Impact of Foreign Aid , ed s . Danielle Resnick and Nicolas van de Walle. Oxford: Oxford University Press, 1 27. Riedl, Rachel Beatty. 2014. Authoritarian Origins of Democratic Party Systems in Africa . New York: Cambridge University Press. Rihoux, Beno Ã® t , and Gis Ã¨ l e De Meur. 2009. isp Set Qualitative Comparative Analysis Configurational Comparative Methods , ed s . Beno Ã® t Rihoux and Charles Ragin. Los Angeles: Sage, 33 68. Rooney, David. 2010. Kwame Nkrumah: Vision and Tragedy . Accra: Sub Saharan Publishers.
244 Rueschemeyer, Dietrich, Evelyne Huber Stephens, and John D. Stephens. 1992. Capitalist Development and Democracy . Chicago: University of Chicago Press. The Oxford Handbook of Political Institutions , ed s . R.A.W. Rhodes, Sarah A. Binder, and Bert A. Rockman. New York: Oxford University Press, 39 55. American Political Science Review 64(4): 1033 53. Sartori Giovanni. 1987. The Theory of Democr acy Revisited . Chatham: Chatham House. Journal of Democracy 9(2): 91 107. Sche ring Democratic Consolidation, Studies in Comparative International Development 36(1): 66 92. Schumpeter, Joseph A.  1962 . Capitalism, Socialism, and Democracy , Third Edition. New York: Harper and Brothers Publishers. Schraeder, Peter J. 1994. United States Foreign Policy Toward Africa: Incrementalism, Crisis, and Change . Cambridge: Camb ridge University Press. Seawright, Jason Political Research Quarterly 61(2): 294 308. Shillington, Kevin. 1992. Ghana and th e Rawli ngs Factor Electoral Studies 21: 519 26. Summers, R., and A. Heston. n.d. University of Pennsylvania. Compute r disk. Teorell, Jan, Nicholas Charron, Stefan Dahlberg, S Ã¶ r en Holmberg, Bo Rothstein, Petrus Sundin, and Richard Svensson. 2013. The Quality of Government Dataset, version 20Dec13. The University of Gothernburg: The Quality of Government Institute. Online . http://qog.pol.gu.se ( Accessed June 25, 2014). The Formation of National States in Western Europe , ed. Charles Tilly. Princeton: Princeton University Press.
245 Global Corruption Barometer . Online. http://www.transparency.org/gcb2013/country/?country=ghana ( Accessed April 8, 2015 ). Aid and Reform in Africa , ed. by S. Devarajan, D. Dollar, and T. Holmgren. Washington, D.C.: The World Bank, 45 100. USEIA ( U.S. Energy Information Administration ) . 2014. International Energy Statistics . Online. http://www.eia.gov/countries/data.cfm ( Accessed June 29, 2014). v African Affairs 109(434): 1 21. VillalÃ³n, Leon The African State at a Critical Juncture: Between Disintegration and Reconfiguration , ed s . Leonardo A. VillalÃ³n and Phillip Huxtable. Boulder: Ly nne Rienner, 3 26. VillalÃ³n, Leonardo A., and Phillip Huxtable (eds.). 1998. The African State at a Critical Juncture: Between Disintegration and Reconfiguration , Boulder: Lynne Rienner. VillalÃ³n , Leonardo A., and Peter VonDoepp (eds.). 2005. The Fate of A Democratic Experiments: Elites and Institutions. Bloomington: Indiana University Press. VonDoepp, Peter, and VillalÃ³n Democratic Experiments: Elites and Institutions , eds. Leonardo A. Vill alÃ³ n and Peter VonDoepp. Bloomington: Indiana University Press, 1 26. Journal of Democracy 25(3): 35 43. Wh itfield, Lindsay (ed.). 2009a. The Politics of Aid: African Strategies for Dealing With Donors . Oxford: Oxford University Press. Whitfield, Lindsay. 2009 b in Turning Points in African Democracy , eds. Abdul Raufu Mustapha and Lindsay Whitfield . Rochester: James Currey, 50 70 . African Affairs 108 (433): 621 41. Whi The Politics of Aid: African Strategies for Dealing With Donors , ed. by Lindsay Whitfield. Oxford: Oxford University Press, 18 5 216.
246 World Bank 1989. Sub Saharan Africa: From Crisis to Sustainable Growth . Washington, D.C.: The World Bank. World Bank. 2014. World Development Indicators . Online. http ://data.worldbank.org/data catalog/world development indicators ( Accessed June 24, 2014). American Journal of Political Science 53 (3): 552 71. Wrong. Michela. 20 02. In the Footsteps of Mr. Kurtz: Living on the Brink of Disaster in . New York: HarperCollins Publishers.
247 BIOGRAPHICAL SKETCH Nicholas Knowlton is an instructor in the Department of Political Science at the University of Florida. He prev iously earned a Bachelor of Arts and Master of Arts in Political Science from the University of Utah. His research includes the study of regime trajectories in Africa, as well as the politics of health and disease in Africa. He was awarded a National Secur ity Education Program Boren Fellow ship in 2013 and was granted the opportunity to conduct his dissertation fieldwork for seven months in Ghana. His previous work has been published in the Routledge Handbook of Global Health Security , as well as the African Journal of Political Science and International Relations . He received his Ph.D. from the University of Florida in the winter of 2015.