Domestic Politics and Drought Relief in Africa: Explaining Choices 9781626371026

Citation preview

munemo-fm.qxd:ff-fm

6/25/12

10:51 AM

Page 1

DOMESTIC POLITICS AND DROUGHT RELIEF IN AFRICA

munemo-fm.qxd:ff-fm

6/25/12

10:51 AM

Page 3

DOMESTIC POLITICS AND DROUGHT RELIEF IN AFRICA Explaining Choices

Ngonidzashe Munemo

munemo-fm.qxd:ff-fm

6/4/12

12:18 PM

Page 4

Published in the United States of America in 2012 by FirstForumPress A division of Lynne Rienner Publishers, Inc. 1800 30th Street, Boulder, Colorado 80301 www.firstforumpress.com and in the United Kingdom by FirstForumPress A division of Lynne Rienner Publishers, Inc. 3 Henrietta Street, Covent Garden, London WC2E 8LU © 2012 by FirstForumPress. All rights reserved Library of Congress Cataloging-in-Publication Data A Cataloging-in-Publication record for this book is available from the Library of Congress. ISBN: 978-1-935049-49-4 British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library. This book was produced from digital files prepared by the author using the FirstForumComposer. Printed and bound in the United States of America The paper used in this publication meets the requirements of the American National Standard for Permanence of Paper for Printed Library Materials Z39.48-1992. 5 4 3 2 1

In memory of Julius Mhondiwa Munemo 1934-2012

Contents

List of Tables and Figures Acknowledgments

ix xi

1

The Domestic Politics of Drought Relief in Africa

1

2

Drought in Africa and Programs for Relief

27

3

Insecurity, Autocracy, and Drought Relief in Kenya

51

4

The Pathologies of Drought Relief in Zimbabwe

87

5

The Evolution of Drought Relief Policy in Botswana

131

6

Conclusion

169

List of Acronyms Bibliography Index About the Book

183 185 209 217

vii

Tables and Figures

Tables

1.1 Variations in Drought Relief Programs for Able-Bodied Adults

11

2.1 Financial Efficiencies of Components in Botswana’s Drought Relief Program, 1982-1990

35

2.2 Changes in Kenyan Maize Prices by Province, 1984

43

2.3 Decisions Facing Responding Incumbents: Policy Challenges and Alternatives

46

4.1 Change in Electoral Support for ZANU by Province between 1980 and 1985

99

4.2 Average Maize, Wheat, and Sorghum Prices by Decade

106

5.1 District Discretionary Allocations for Relief, 1979-1980

149

5.2 Income from Crop Failure Replaced by Labor-based Relief Transfers, 1982-1985

153

5.3 Summary of ARAP Components and Types of Assistance, 1985-1989

155

5.4 Rate of Spending on Drought Relief, 1982-1989

157

Figures

1.1 First-Order Matching of Variables

17

6.1 The Relationship between Insecurity and Programs for Relief in Africa

174

6.2 Democratic and Autocratic Authority Trends in Africa, 1960-2009

180

ix

Acknowledgments

I accumulated many debts in the course of researching and writing this book. It is a pleasure to identify those institutions and individuals to whom I owe special gratitude. I am indebted to several mentors at Columbia University: Anthony (Tony) Marx, Lisa Anderson, and Alfred Stepan. Tony saw a question worth pursuing in state-society interaction during dearth, even when some thought it was not “political sciency” enough. I thank him for encouraging me to pursue this project. I owe special thanks to Lisa Anderson for her guidance. Her patience gave me time to let the story of relief develop in directions that proved to be far more rewarding than I had imagined. At the University of Zimbabwe, I owe a great debt to the late Professor Masipula Sithole for sparking my interest in politics. At Bard College, I thank Sanjib Baruah, Omar Encarnación, and especially David Kettler, my first intellectual mentor. The initial research for this book was conducted during a six-month trip to Zimbabwe in 2004, partially funded by a travel grant from the Institute of African Studies, Columbia University. This was followed by an eight-month trip to Botswana and then to Zimbabwe in 2005, under a fellowship from the IDRF Program of the Social Science Research Council, New York, with funds provided by the Andrew W. Mellon Foundation. In the field, the project benefited from the support and encouragement of countless individuals and organizations who prodded and quietly guided me in the right direction. I am grateful and indebted to old colleagues from my student years at the University of Zimbabwe for vouching for me, even when the questions I pursued were a bit indelicate. I benefited from institutional affiliation, access, and intellectual support from the Department of Political and Administrative Studies at the University of Zimbabwe, my first intellectual home; the Institute of Development Studies, University of Zimbabwe; the University of Zimbabwe library; the Department of Political and Administrative Studies at the University of Botswana; the University of Botswana library; the Institute of Development Studies, University of Nairobi; and the University of Nairobi library. I would like to thank the

xi

xii

Domestic Politics and Drought Relief in Africa

administrations and staffs of these institutions for their generosity. I appreciate the patience and courtesy of the staffs of the National Archives of Zimbabwe; the Grain Marketing Board of Zimbabwe; the Department of Social Welfare, Zimbabwe; the Botswana National Archives and Records Services; the library of the Ministry of Agriculture, Botswana; the library of the Southern African Development Community (SADC) Secretariat Office in Gaborone; and the Kenya National Archives and Documentation Service. After Columbia, the Department of Political Science at Williams College provided a collegial environment for me to continue my work on this project. Williams also provided funds to help with the preparation of the final manuscript for this book, for which I am grateful. With support from the Ursula Prescott Memorial Fund, the Department of Political Science supported additional fieldwork to Kenya in 2009, greatly enriching my understanding of that country. I am indebted to Cathy Johnson for making the trip possible. Michael MacDonald read draft chapters and shared his deep knowledge generously. You were right, Michael, revising is like sanding a wood floor; finishing one section reveals imperfections in others. I thank Paul MacDonald and Justin Crowe; each chapter in this book benefited from our research meetings at Tunnel City Café. I am also grateful to Kenda Mutongi, for reading portions of the book and offering strengthening questions and criticisms. The book benefited from the guidance and support of many. I owe a debt of gratitude to Adrienne LeBas, who not only loaned me her car during my first trip to Zimbabwe but is a friend and critic, prodding me to strengthen and extend my argument. Greg White and Linda Beck were a source of support and inspiration. At the invitation of David Hulme, I was provided with an opportunity to articulate a preliminary sketch of my argument for an edited volume on social protection. I thank the Social Science Research Council for supporting this project a second time by awarding me a Book Fellowship in 2008-2009. Through this fellowship I worked with a fantastic developmental editor, Mara Naselli, who helped shape the structure and arc of the book. I hope she sees her imprint in the pages that follow. A yearlong residence at the Oakley Center for the Humanities and Social Sciences, Williams College, where I was a Herbert H. Lehman Fellow in 2010-2011, provided time for revisions and support for new research. I thank Michael Brown, Francis Oakley, John Chandler, Gretchen Long, Karen Gover, Edward Gollin, Magnus Berhardsson, Michele Hannoosh, and Keith McPartland for reading and commenting on early drafts of this book.

Acknowledgments

xiii

It has been a delight to work with First Forum Press, a division of Lynne Rienner Publishers. My sincerest gratitude goes to Lynne Rienner for her unwavering interest and support for this book. The detailed and constructive comments from the two anonymous reviewers pushed me in productive directions. I would like to thank Milton Djuric, whose diligence and care in copyediting made the final stages pleasant. Over the near decade I worked on this project, I enjoyed the patience, understanding, and encouragement of my family, across two continents. To my family in Zimbabwe, Gogo, Farai, Nyara, Patience, and all my cousins, nephews, and nieces, thank you for accommodating my trips. To my family in the United States, Manga, Susannah, Josh, Lauren, and Herbie, nephews and nieces, thank you for your support. I am most grateful to my sons, Julius and George Munemo, for coping with my hours at the office, even on the weekends. And yes, I did write a book. Finally, the largest share of my gratitude goes to my wife, Julia Munemo, for her love, support, and encouragement. She traveled with me first to Zimbabwe and then to Botswana, endured my absences while writing, and edited countless drafts, all the while raising our two fabulous sons. It is inconceivable that I could have completed this book without your love, boundless patience, and encouragement. This book is dedicated to my grandfather, Julius Mhondiwa Munemo, who raised me as his own son and encouraged my quest for education, even when it meant leaving Zimbabwe. I regret that I was not able to finish the book before his passing. —Ngoni Munemo

1 The Domestic Politics of Drought Relief in Africa

From late 1981 through 1986, Botswana endured six continuous years of inadequate and erratic rainfall. The severity and length of the drought left most rural Batswana without sufficient food. In early 1982, President Q.K. Masire declared Botswana drought-stricken and initiated a national relief program.1 Between 1982 and 1990, drought relief and recovery measures in Botswana involved three key components. The first was a system of food distribution to pre-school children, primary school children, pregnant and lactating mothers, TB patients, and individuals defined as Destitute or Remote Area Dwellers.2 The government decided not to extend free food to able-bodied adults. Instead, to avoid waste and to compensate the rural population for lost incomes, the second component of the relief program was a labor-intensive public works program,3 which provided “temporary employment on local development projects (e.g., road maintenance, small dams and general construction)” for a subsistence wage of P7.50 per week (or US$14.42 in 2005 terms).4 Works projects were identified by standing Village Development Committees (VDC) and then administered and implemented by a Drought Relief Technical Officer assigned to the VDC. The duration of works projects ranged from three weeks to a maximum of eight months. As rain gradually began to return in 1986, the government added a third component, the Accelerated Rainfed Arable Programme (ARAP), to its ongoing relief program. ARAP contained eight packages (ploughing, row planting, weeding, destumping, field fencing, fertilizer provision, seed purchasing, and water provision), under which farmers could benefit from government support during the recovery period.5 The government also made funds available to local authorities to administer

1

2

Domestic Politics and Drought Relief in Africa

livestock relief, as well as provide temporary water supplies to affected segments of the population. By the time President Masire declared that the drought had broken in 19886, the government had provided free food aid to over 556,000 Batswana each year, mitigated lost incomes for roughly 20 percent of the rural population through labor-based relief, and ploughed, destumped, planted, weeded, or fenced a combined 500,000 hectares of land under ARAP. The breadth of the drought relief and recovery program in Botswana cost P236 million (roughly US$261 million in 2005 terms) in the first six years and an additional P166 million (US$138 million in 2005 terms) over the period of recovery. Spending by the government of Botswana accounted for over 90 percent of the total cost of relief. Thus, government-initiated and funded action protected Botswana from what could have become famine. Though it was effective, Botswana’s 1982-1990 drought relief program was not typical of responses initiated and funded by incumbents elsewhere in Africa in the early 1980s. During the same period, Zimbabwe and Kenya executed significantly different strategies. In the same year that Botswana faced the first year of its prolonged drought crisis, newly independent Zimbabwe confronted a similar droughtinduced threat of famine. Like Botswana, drought relief measures in Zimbabwe were adopted “early in 1982, and given a high political and financial priority.”7 Zimbabwe’s program contained a supplementary feeding scheme for children under five and lactating mothers, a water supply scheme, cattle protection, and input provision for peasant farmers, as in Botswana. However, unlike Botswana, Zimbabwe’s response did not rely on labor-based relief as the primary form of assistance for able-bodied adults. Instead, the main element of Zimbabwe’s 1982-1984 response was the distribution of take-home food rations to the rural adult population.8 The administrative structure used to implement relief in Zimbabwe was also different from Botswana’s. Whereas incumbents in Botswana delegated the implementation of drought relief to standing institutions, their counterparts in Zimbabwe used village and district members of the Zimbabwe African National Union (ZANU), the newly elected ruling party, to identify districts, villages, and beneficiaries for inclusion in the food aid program for adults.9 Beyond the identification of districts, villages, and beneficiaries, ZANU party cadres were also responsible for the day-to-day distribution of rations to households. As a consequence, the distribution of relief rations took place at public gatherings led by local ZANU cadres, and often attended by incumbent ZANU members of government—essentially at ZANU political meetings or rallies.10

The Domestic Politics of Drought Relief in Africa

3

Relief in Zimbabwe reached an average of 850,000 people a month in the first year and 1.46 million a month in the second and final year. At the peak of the program, in mid-1983, the government distributed free food to about 2.1 million able-bodied adults, or 37 percent of the rural population.11 To feed this many people, incumbents in Zimbabwe spent over Z$22 million (or US$39 million in 2005 terms) in the first year and Z$42 million (US$52 million in 2005 terms) in the last year of the relief program. Incumbents in Zimbabwe averted famine, but turned the relief program into a political tool while doing so. In Kenya, the government of President Daniel arap Moi responded to the 1983-1984 drought, at the time the worst in its history, with a program that combined food aid to rural Kenyans with market controls (in the form of price controls and food movement restrictions) and government acquisition of food for the urban market.12 Administratively, Moi took direct control of the food aid component of the program by placing it under the Office of the President and ordering the Provincial Administration, which reported to him, to identify beneficiaries and distribute relief. As Mugabe had done with ZANU cadres at the local level, Moi gave Provincial Administration staff complete discretion to determine who qualified for free food aid. He also created several ad hoc institutions, which all likewise reported to him, to manage the other components of the relief program.13 Food aid in Kenya reached over 930,000 rural Kenyans each month between August 1984 and October 1985. At its peak, in October 1984, the Office of the President distributed food aid to close to 10 percent of the rural population, or 1.6 million Kenyans. To feed this many people and provide food for the urban market, Moi’s government spent over US$150 million to import about 500,000 Mt of maize, about 77 percent of the maize imports during the relief effort.14 Despite enduring similar covariant shocks to food security in the mid-1980s, Botswana, Zimbabwe, and Kenya adopted very different drought relief programs to avert famine. Incumbents in Botswana choose labor-based relief, while those in Zimbabwe opted for free food aid to protect able-bodied adults from the production and income shocks produced by drought. In Kenya, the relief program combined food aid to rural Kenyans with market controls and government acquisition of food for the urban market. Why did incumbents in these three countries offer able-bodied adults different aid packages? What determined the administrative structures and personnel used by governments to administer and implement relief programs? Explaining why programs of relief offered by governments in Africa differ across contexts requires a serious exami-

4

Domestic Politics and Drought Relief in Africa

nation of the relationship between incumbency and forms of relief. When one pays attention to the strength or weakness of incumbency, the distinctive political interests that direct policy-making during food crises is revealed. It sheds light on why some African incumbents choose relief policies that stress free food aid to able-bodied adults, while others opt to generate employment through labor-based relief projects. The Question

Among scholars, policy analysts, and journalists writing on disasters in Africa, the tendency is to assume that African countries lack both the technical administrative capacity and political will to respond to crises produced by drought. In explaining government failure to protect citizens from drought-induced threats of famine, scholars, analysts, and journalists alike converge on three seemingly distinct, but overlapping, explanations: bad policies, weak political institutions, and inhospitable social conditions. First, countries in Africa are argued to be vulnerable to drought-induced famine because of poorly conceived agricultural policies that force its most well-endowed farmers (i.e., commercial farmers) to disinvest from domestic food crops in favor of cash crops.15 Because commercial farmers across much of Africa turned to cash-crop production for the world market, the burden of domestic food production has fallen to rain-dependent peasant farmers. Thus, the incidence of drought has generally produced large domestic food deficits, with which governments have not been prepared to deal. If some scholars have suggested that poorly conceived agricultural policies have structurally increased vulnerability to drought in Africa, others contend that bad politics turned this vulnerability into famine. Specifically, they zero in on weak political institutions in Africa as the major structural impediment to government responsiveness to disasters.16 According to these accounts, chronic state weakness together with few de jure or de facto constraints on executive power create an environment in which leaders are more interested in pursuing their own private goals than in providing for the citizens. Thus, the central claim is that governments in Africa are not likely to respond to threats of famine because politics and policy-making are personalized, not institutionalized. In such contexts, citizens survive or starve at the whim of the personal ruler. President Mobutu and the famine of 1977-1978 in Zaire (now the Democratic Republic of Congo), or the Ethiopia famine of 1984-1985 under Mengistu’s military regime, are two notable examples. The third explanation favored by scholars, policy analysts, and journalists situates the general problems of weak technical capacity (bad

The Domestic Politics of Drought Relief in Africa

5

policies and weak institutions) and the specific issue of poor governmental response to disasters in the overriding socio-political composition of post-colonial Africa; that is, its neopatrimonialism.17 The concept of neopatrimonialism draws heavily from Max Weber’s discussion of authority systems and their attendant bases of legitimacy.18 In Weber’s decidedly teleological typology, each authority system (traditional, transitional, or the modern state) is supported or associated with a justifying and legitimating order, which works to induce indivi-duals to obey those who claim authority over them. Weber identified and linked traditional society with personal patrimonial authority, tran–sitional society with charismatic authority, and the modern state with impersonal rational-legal or bureaucratic authority.19 In trying to make sense of the patterns of political order (and disorder) that developed across Africa, scholars noted that although African countries gained independence with the trappings of a modern state (armies, bureaucracies, legislatures, judiciaries, political parties, etc.), the legitimation and exercise of that authority seemed quite far from Weber’s impersonal, rational-legal, bureaucratic, and corporatist order.20 As noted by Bratton and van de Walle, “contemporary African regimes do not display the formal governing coalitions between organized state and social interests or the collective bargaining over core public policies that characterize corporatism.”21 Instead, leaders legitimated their hold on power and authority through personal patronage, rather than through ideology or law as Weber had imagined. Thus, highly personal relationships of loyalty and dependence, not professionalism and autonomy, are said to pervade the formal political and administrative system across Africa. It is this mismatch between the authority system (the post-colonial modern state) and its legitimating sub-structures and practices (personal patrimonial rule) that students of African politics sought to capture by the term neopatrimonialism. The mismatch notwithstanding, neopatrimonial practices served a function. They enabled Africa’s new leaders to fashion some semblance of political order in the post-colonial period. But the order obtained by generating compliance and obedience through personal networks came at the price of the pathologies of neopatrimonialism: patterns of politics and policies that created opportunities for mismanagement, corruption, and ethnic rivalries and conflict.22 As a consequence, scholars suggest that while post-independence leaders solved the immediate problem of political order in the modern political system by resorting to elements of patrimonial legitimation, this anachronistic practice created many problems that hindered the normal operation of public institutions— including the will and capacity of political leaders to respond to disasters

6

Domestic Politics and Drought Relief in Africa

such as drought. Due to the socio-political conditions of neopatrimonialism, the African state is seen, at its best, as too weak to respond to drought or, at its worst, as so corrupt that leaders willfully used starvation as a political tool.23 The three arguments differ in important respects. The first argument privileges policies in the agricultural sector, the second account points to political institutions, while the third stresses the composition of society and its negative effect on politics in Africa. But they draw the same conclusion: African institutions and leadership are not up to the task of famine relief. The bias is not without justification. Since 1960, Africa has witnessed twenty-two famines (including the Somali famine of 2011), some of them attributable to governmental failure to droughtproof agriculture, weak political and economic capacity to respond, or the political manipulation by competing elites of access to food. Thus, the overriding image produced by scholars, analysts, and journalists writing on drought and famine in Africa is that of emaciated children from famines in Somalia, the Sudan, Nigeria, and Ethiopia requiring aid from abroad. This narrative highlights external intervention intended to make up for the inadequacies of domestic governments. Yet this should not be generalized. On the ground, the situation is more complicated. The blanket assessment of “apathy, incompetence, and corruption of African governments in the context of famine prevention” that some scholars and journalists trumpet is not sufficient.24 Some governments have ignored drought-related threats of famine, including Angola, Ethiopia, Liberia, Mozambique, Nigeria, Sierra Leone, Somalia, Sudan, Uganda, and Zaire. But numerous others have responded to similar crises by adopting national relief programs of various forms, including Cape Verde, Tanzania, Botswana, Zimbabwe, and Kenya. Droughts in these countries have produced government relief programs, but with notable variation across countries and even across time within the same country.25 In this book, I offer a framework that accounts for the different programs governments adopted during times of dearth, explaining why forms of relief to the adult population vary across and within countries over time. By drought relief, I mean those domestic governmentinitiated and government-funded programs intended to provide segments of the population with temporary assistance for the duration of a drought. Drawing on sixteen months of fieldwork in Botswana, Zimbabwe, and Kenya, I develop my argument through a sustained examination of the relationship between the durability of political incumbency and the form of relief adopted. Examining the strength of political incumbency reveals the range of political and economic interests that

The Domestic Politics of Drought Relief in Africa

7

shape policy-making during dry times, while a focus on forms of relief highlights the way that each program is associated with a distinctive matrix of political and economic benefits and costs, which incumbents are keen to capture and avoid, respectively. Thus, this book reveals the political determinants of government responses to drought in Africa. The Argument

I argue that variation in the form of drought relief programs is a result of differences in the vulnerability of responding incumbents. The conditions that strengthen or weaken the security of incumbents—the susceptibility of the regime to possible coups and subversive movements, the strength of the government’s institutional power, the robustness of governing coalitions, the size of the president’s legislative majority, and the frequency of political and economic protests— determine executives’ overriding political concerns when making decisions about relief programs. In other words, political interests influence their decisions in distinctive ways. Specifically, governments that suffer drought when incumbents are secure tend to adopt laborbased relief; but droughts in countries in which incumbents are insecure favor food aid programs for adults because of the immense patronage it puts at the disposal of vulnerable executives. In responding to drought, incumbents must pick a relief program, determine access and eligibility, decide whether to intervene in the market for food, and establish administrative structures for implementing the program on the ground. Each choice has distributional consequences (which affect the immediate political support afforded to incumbents) and efficacy implications (which determine the long-term viability and sustainability of relief programs). Thus, examining the strength of incumbency reveals the political interests and motivations that guide incumbents in crafting drought-relief policy. Two kinds of interests demand our attention here. First, instrumental or technocratic considerations pull policy towards comprehensive but cost-effective programs that maximize protection for the most needy while limiting waste (labor-based relief). Second, political considerations push tenure-seeking incumbents to search for relief programs that spread the coverage net as broadly as possible and are also fungible enough to be targeted for building support (free food aid). I argue that incumbents’ expectations about their immediate and future prospects in power determine the hierarchy of interests that guides them when they must respond to drought. Specifically, incumbents’ expectations about holding on to power, and the discount rates

8

Domestic Politics and Drought Relief in Africa

associated with them, dictate the extent to which policy choice is motivated by the desire for immediate political gain or by technocratic concerns for efficiency. Scholars agree that micro-political conditions affect incumbents’ interests, which in turn affect the policies they adopt. If political environments are unstable and insecure, “rational politicians in office … concentrate on activities that lead to quick results and immediate rewards.”26 If, conversely, political conditions are stable and secure, incumbents are afforded the political space to consider the medium and long-term effects of their policies. Under these conditions, incumbents are more likely to be concerned with efficacy and technical efficiency, and thus find adopting measures with limited immediate political pay-offs more palatable. I extend this framework to government responses to drought in Africa to show that the policy choices adopted by governments reflects the rational political interests of the executives who control the state. The logic of behavior by politically (in)secure incumbents in other policy areas also operates in government-initiated and -funded droughtrelief programs across Sub-Saharan Africa. In fact, there are good reasons to expect that the logics of political (in)security will be magnified or amplified during dry times. As Schatzberg convincingly argues, maintaining food security under normal agricultural conditions has long been a central basis on which incumbent executives have been evaluated and gained legitimacy in African polities.27 When drought occurs across rain-dependent Africa, crops fail, households lose incomes, and the overall economy might slump. The consequences of drought amplify the basic survival, livelihood, and distributional effects of the decisions made by incumbents. Thus, droughts and the stresses they precipitate are “revelatory crises” that “disrupt conventional routine[s] sufficiently to allow actors (including government policy-makers as well as rural producers) to innovate with normative codes.”28 In other words, droughts and the threats of famine they produce are critical and highly visible political moments for incumbents. How they respond may bolster their claims to legitimate control of the state, or it may discredit them. I argue that the strength of political incumbency has a distinctive imprint on the form of relief, the amount of money appropriated, the level of access, and the administrative structures that incumbents establish to implement relief in Africa. This explanation differs from the one offered by scholars who have tended to impute irrationality or a lack of knowledge on the part of African governments that adopt free food aid. A familiar claim is that the reliance on food aid across Africa “partly reflect[s] a widespread failure to perceive droughts as a serious

The Domestic Politics of Drought Relief in Africa

9

and potentially long-term economic problem.”29 According to these scholars, too often drought-related crop failures are treated as though their impact were limited to food production, when in fact crop failure is also associated with significant income shocks for households.30 In their view, responding to drought by simply distributing free food might in fact not address all the needs of affected households. Scholars therefore conjecture that incumbents who continue to adopt free food aid do so because they are not fully aware of the diffuse effects of drought and the most appropriate way to respond to dearth. That is, they lack a key technical understanding of the situation they face. While Benson and Clay are correct to stress that drought-induced crop failure is also associated with large income shocks for households, they are on shakier ground in attributing the persistence of food aid to ignorance of the wider effects of drought. In all three countries covered in this study, I found that bureaucrats in key government ministries were keenly aware of the income and economic shocks of drought. This recognition led them, in most cases, to propose relief programs that smoothed income fluctuations through labor-based relief instead of providing free food aid to able-bodied adults. Knowledgeable though they were of the wider effects of drought and the most efficacious way to respond, under certain conditions bureaucrats lost the internal policy debate with incumbent politicians. This picture of the dynamics of drought relief policy-making suggests that what shapes the program of relief governments adopt is political not technical. In order to explain why food aid to adults persists, then, we need to understand better the circumstances that lead incumbents to accept or disregard recommendations from bureaucrats. I argue that prevailing political interests dictate when incumbents listen to bureaucrats (when they are secure) or disregard their policy proposals (when they are vulnerable). There is a rational explanation for the persistence of food aid to able-bodied adults: it is beneficial to insecure incumbents who use the immediate patronage and the mass mobilization that is characteristic of the direct delivery of food to consolidate their tenuous control over the state. To illustrate how prevailing levels of incumbent vulnerability condition responses to drought, this book draws on the post-colonial experience of three African countries, Kenya, Zimbabwe, and Botswana. Why these countries? Why have I not included Ethiopia, a country ever present in the minds of scholars, policy analysts, and journalists interested in drought and famine in Africa? Peter Gourevitch suggests that the incidence of covariant shocks provides unique opportunities for analysis across cases as such crises “provoke changes that reveal the

10

Domestic Politics and Drought Relief in Africa

connections between particularities and the general. If the comparativist can find countries subject to the same stresses, it then becomes possible to see how countries differ or converge and thereby to learn something about cause and effect.”31 Thus, first of all, I wished to examine a set of countries that experienced drought in roughly the same years. Kenya, Zimbabwe, and Botswana were all affected by the drought wave that swept across Africa from 1980 to 1985. Botswana and Zimbabwe experienced droughts again during the 1990 to 1993 and 1998 to 2005 waves. My choice of Kenya, Botswana, and Zimbabwe leverages the covariance in drought incidence as a way of controlling for worldhistorical time and the practices common in each period as an explanation of the variation in relief programs governments adopt. There was a strong argument in favor of including Ethiopia in my analysis based on this factor; the country suffered a major drought between 1982 and 1984, a period that overlaps with my three cases. However, a second consideration disqualified it. Each of the three countries included in this book confronted famine threats by adopting pre-emptive, domestically-initiated and -funded relief programs. Ethiopia in the 1980s, however, is the poster case for government failure to pre-emptively address the the threat of famine.32 When famine gripped Ethiopia between 1983 and 1985, international aid (popularized by Band Aid and Live Aid), rather than domestically-initiated action, gained prominence. Ethiopia is an interesting case in its own right, but not if you are interested in examining domestic programs of relief. The three states were also selected to maximize variation on some variables and minimize variation on others. In other words, these three countries provide significant variation on the dependent variable, that is, the form of relief offered to adults, level of access set by government, the extent of government intervention in the market, and the administrative structures used to implement relief. As summarized in Table 1.1, of the eleven cases of government drought relief programs I examine, four involved free food aid (Botswana 1979-1980, Kenya 1984-1985, and Zimbabwe 1982-1984 and 2002-2005), six involved labor-based relief (Botswana 1982-1990, 1992-1994, 1996-1998, and 2002-2005, and Zimbabwe 1986-1987, 1992-1993) and one involved a grain loan scheme (Zimbabwe 1995-1999).33 While looking for variation on some variables, I also chose these three states because of the lack of variation on others, most significantly on their colonial experience. All three countries were British colonies and their colonial administrations were strikingly similar in what they were willing to do: human relief was almost always limited to providing for the European population. Where it existed for Africans, colonial

The Domestic Politics of Drought Relief in Africa

11

governments administered relief more as a loan scheme than as a relief effort.34 Colonial officials also consulted closely with their counterparts in other colonies on drought relief strategies. For instance, colonial officials in Botswana were very keen to learn and copy drought relief policies adopted in colonial Zimbabwe. In this regard, it is clear that the post-independence programs of relief incumbents adopted were not a legacy of the colonial experience. Table 1.1: Variations in Drought Relief Programs for Able-Bodied Adults Country

Duration of Drought

Duration of Relief

Form of Drought Relief Program

Botswana

1978-79

1979-80

Free Food Aid

Botswana

1981-88

1982-90

Labor-Based Relief

Botswana

1992-93

1992-94

Labor-Based Relief

Botswana

1995-97

1996-98

Labor-Based Relief

Botswana

2001-04

2002-05

Labor-Based Relief

Kenya

1983-84

1984-85

Free Food Aid

Zimbabwe

1981-83

1982-84

Free Food Aid

Zimbabwe

1986-87

1987-88

Labor-Based Relief

Zimbabwe

1991-92

1992-93

Labor-Based Relief

Zimbabwe

1995-98

1996-98

Grain Loan Scheme

Zimbabwe

2001-04

2002-05

Free Food Aid

As illustrated in Table 1.1, Botswana and Zimbabwe have responded to five major national droughts, although the trajectory followed by these programs has been different.35 Botswana’s post-independence response to drought relief began with an expansive food aid program for able-bodied adults between 1979 and 1980. Two years later, Botswana shifted its drought relief program to one predicated on cash-paying

12

Domestic Politics and Drought Relief in Africa

works programs for all able-bodied adults. After the 1982-1990 relief program, the government took an unprecedented step by writing laborbased drought relief into its national development plans.36 During nondrought years, public works programs were to be used as a tool for generating employment in the rural areas. In drought years, the scale of these works programs would be increased to accommodate relief needs. Thus, when droughts struck again from 1991 to 1993, 1995 to 1997 and 2001 to 2004, Botswana’s responses were all aggressively predicated on labor-based relief for able-bodied adults.37 By writing labor-based relief into its national development plans, the government of Botswana signaled its standing commitment to respond to future droughts in the country. This administrative commitment both reassured the population by removing any uncertainty about what the government would do in response to future droughts and, significantly, took policy-making discretion away from future incumbents faced with a threat of famine. Botswana’s move toward a standing administrative commitment to respond to drought, and to do so primarily through labor-based relief for able-bodied adults, is quite different from the drought relief policy trajectory in Zimbabwe, where responses continue to be ad hoc and therefore subject to greater manipulation by incumbents. Unlike Botswana, independent Zimbabwe’s first relief program in 1982 provided able-bodied adults with free food aid. After this program, Zimbabwe went through nearly a decade (between 1986 and 1995) when it responded to drought with food-for-work programs, labor-based projects that paid in food. From 1995 to 1999, Zimbabwe adopted a regressive grain loan scheme, which offered government relief to drought-affected households as a short-term loan to be repaid in subsequent agricultural seasons. Following Zimbabwe’s descent into political and economic crisis in 2000, incumbents once again championed a change in drought relief policy, and moved back to free food aid for able-bodied adults. As these changes reveal, Zimbabwe has no standing administrative commitment to respond to drought through a predetermined program of relief. In Kenya, the government of President Moi responded to the 19831984 drought by adopting a relief program that was distinctive for combining food aid to the rural population with government takeover of the urban market. As in Zimbabwe, Kenya had no standing drought relief institutions or prior policy commitment. Nevertheless, the program of relief adopted between 1984 and 1985 was lauded for successfully addressing the threat of famine. Looking across countries with similarities in the proximate cause of threats of famine, why have both free food aid and labor-based relief

The Domestic Politics of Drought Relief in Africa

13

programs (some paying in cash and others with food rations) been used to protect able-bodied adults from starvation? Specifically, why did incumbents in Botswana opt for labor-based relief in response to drought in the early 1980s, while incumbents in Zimbabwe and Kenya decided to rely on food-aid programs? Focusing within countries over time, why did Botswana adopt a food aid program in 1979 and then switch to labor-based relief in 1982? Why, a decade later, did incumbents in Botswana take the unprecedented step of institu-tionalizing labor-based relief? Finally, what explains Zimbabwe’s shifts from food aid to foodfor-work to the grain loan scheme and then back to food aid after 2000? By placing the interests of incumbents at the center of the explanation of policy choice, this book also offers an alternative account to that suggested by Bates, who argues that “public policies are not explained as the choices made by some reified single actor, called a government. … Rather, policy is the product of the interested actions of private parties who bring their resources to bear upon politically ambitious politicians and the political process.”38 Although Bates’s approach is useful insofar as it directs our attention to a broader and more complex explanation of policy outcomes, his excessive focus on ‘lobbying’ by organized interests fails to capture the nature of policymaking across much of Africa. In many of these governments, the policy-making process tends to be top down, with little space or opportunity for organized groups outside the state to exert pressure on incumbents. Thus, policy-making is closed and centered on the executive. Most citizens have their first encounter with policy when it is announced, rather than through involvement in its formulation.39 I show that during periods of dearth, government relief programs are not the product of lobbying by external agents; they are reducible to the interests and goals of incumbents who direct policy-making. Because organized groups have a structurally weak position in the policy-making process in Africa, I focus my attention on the internal agents of the state (incumbent politicians and bureaucrats) in my explanation of drought relief policy formulation, and assume that incumbents have a great deal of autonomy in making policy. Along with other scholars, I understand state autonomy to describe contexts in which incumbent government officials are free or insulated from societal or external pressure in their decision-making.40 By assuming incumbents to be structurally insulated from society I do not imply that their actions lack coherence, but that the coherence comes from their desire to maximize or prolong their hold on power. That is, other things being equal, incumbents will be motivated to use government resources to solidify their hold on power by building governing coalitions through

14

Domestic Politics and Drought Relief in Africa

the distribution of patronage, or otherwise using their institutional powers to weaken potential challengers.41 In this approach to explaining policy outcomes, “policy elites [are] less reactive to interest group pressures and more active in attempting to maximize their chances of staying in power by putting together supportive coalitions and using public resources to ‘buy’ support” even when (or perhaps, especially if) that policy is drought relief.42 The form of relief governments in Africa offer is a product of a rational decision-making process guided by incumbents’ political interests. Institutional and Structural Explanations

In this book, I argue that variation in the vulnerability of responding incumbents explains why governments offer free food to able-bodied adults in some instances and ask the same adults to work for relief in other contexts. My argument builds on an idea well established in the political economy of policy choice in developing countries: that the interests that animate policy-making are determined by the non-trivial trade-offs between the political and instrumental goals held by incumbents. However, I challenge conventional views about the determinants of famine relief by insisting that the type of relief provided to ablebodied adults is a direct function of the strength or weakness of incumbency when governments are called upon to respond to dearth. I argue that incumbents provide free food to able-bodied adults not because it is the easiest form of relief to administer or because they are ignorant of the wider effects of drought, but because it is strategically beneficial for insecure incumbents to leverage free food to reward supporters and punish challengers. Explaining why the common covariant shocks of drought have elicited different relief programs across countries or within countries over time in this way separates the form of relief from common institutional and structural accounts. The institutional argument, which draws heavily from Amartya Sen’s work on famines and responsiveness, contends that the form of relief is a function of each government’s regime type43, and I label it the regime argument. To account for the form of drought relief, the regime argument focuses on the formal rules structuring the selection of political elites and the overall relationship this engenders between government and society. Substantively, the regime argument rests on two key institutional mechanisms to explain drought relief policy adoption: the credible threat of punishment provided by regular elections and the whistle-blowing function of a free press. Scholars contend that these two key features of

The Domestic Politics of Drought Relief in Africa

15

a democracy induce incumbents to adopt sound and efficacious drought relief programs. That is, the knowledge that their choice of policy will be reviewed and debated by the public and, more importantly, by potential challengers, prevents incumbents from picking wasteful and otherwise less effective responses. According to some versions of the regime argument, incumbents need not actually face public agitation, a vigilant press, or feisty challengers; the mere rational anticipation of potential electoral concerns leads them to the one program that is capable of forestalling many of these problems: labor-based relief.44 The structural argument, what I call the agricultural sector position, attributes variation in drought relief programs to agricultural performance in normal years. Significantly, this explanation jettisons the regime argument of policy adoption and contends that, even in democracies, the form of drought relief is conditioned by whether the country in question is a surplus or a deficit food producer. Along these lines, de Waal contends that countries that are normally surplus producers of food opt to remove the variability in food supply through free food aid, while countries that are typically net importers of food prefer to maintain rural incomes through labor-based relief programs that pay in cash.45 In Figure 1.1, I use the generally accepted POLITY score to measure the extent of political democracy in my cases. Cases on the left of Figure 1.1, with a POLITY score of -10 to 0, can be considered to be deficient in most of the key features of democracy: free or fair elections to select leaders who then decide on policy; institutionalized constraints on the powers of the executive; and guarantees of civil liberties, including a free and open press.46 Five of the eleven drought relief programs covered in this book were adopted by governments judged to lack one or more of these institutional features of a political democracy: Kenya in 1984, with a POLITY score of -7; and Zimbabwe in 1986, 1992, 1996 and since 2000, with POLITY scores ranging from -4 to -6. Cases on the right, with a POLITY score ranging from 0 to +10, exhibit the institutional structures regime theorists insist lead democracies to adopt labor-based relief in response to threats of famine. They hold free and fair elections, have institutional constraints on the powers of the executive, and guarantee civil liberties. The six cases in which governments were judged to exhibit elements of a democratic system when they adopted a drought relief program include all of Botswana’s interventions between 1979 and 2002, with a POLITY score ranging from +6 to +8; and Zimbabwe in 1982, when it had a POLITY score of + 4.47 The vertical axis of Figure 1.1 represents the structural argument by measuring the per capita cereal surplus or deficit held by the government

16

Domestic Politics and Drought Relief in Africa

when incumbents had to respond to a threat of famine produced by drought. As with the institutional measure, my cases contain significant variation on this variable. At the top of the vertical axis, at a surplus of 120 kgs per capita of grain and above, are cases that possessed enough reserves to distribute as food aid to affected groups until the next harvest. This was the case for Zimbabwe in 1982, 1986, and 1992. A second set of cases, with positive stocks below 120kgs per capita, did not have reserves large enough to distribute to all drought-affected groups until the next harvest. Zimbabwe in 1992 and 1996 and Kenya in 1984 are examples of these cases. Finally, a third set of cases had no grain stocks to call upon when governments picked a drought relief program. In these cases, Zimbabwe after 2000 and Botswana in all of its relief programs, incumbents had to import all of the grain they used in their relief program. As the first order matching in Figure 1.1 illustrates, neither the regime-institutional account, which focuses on the mechanisms associated with a democratic political system, nor the institutional story, which is rooted in the structure of the agricultural sector, offers a compelling explanation of the variation in drought relief programs across Africa. The regime argument cannot explain why sometimes democracies adopt food aid for adults, as Botswana did in 1979. Also, because the regime argument is not explicit about what we should expect from non-democracies (countries on the left end of the political openness continuum in Figure 1.1), it cannot explain why nondemocracies (e.g., Zimbabwe between 1986 and 1992) were just as likely to adopt labor-based relief. Democracy, it seems, is neither necessary nor sufficient for the adoption of labor-based relief as the primary form of protection for able-bodied adults. Similarly, the structural account of variation in relief programs falls well short of explaining why some food-deficit countries choose the expensive option of importing food for food aid programs (Botswana in 1979, Kenya in 1984, and Zimbabwe after 2000) or why some foodsurplus countries choose non-food aid relief programs (Zimbabwe between 1986 and 1999). While thinking about the structure of the agricultural sector forces us to understand normal agricultural performance in drought-affected countries, it is not the only or the decisive factor in the selection of drought relief programs. This book demonstrates that the structure of the agricultural sector is only one of the factors that incumbents weigh when picking a program. I further illustrate that the salience of food stocks or deficits is a product of the larger political milieu.

The Domestic Politics of Drought Relief in Africa

17

Figure 1.1. First-Order Matching of Variables

Structure of the Agricultural Sector - Cereal Balance Per Capita (kgs)

Zim'86 - food-for200 work

150 Zim'96 - grain loan scheme

Zim'82 - food aid

Zim'92 - food-for100 work

50 Ken'84 - food aid

0 -8

-6

-4

-2

0

2

4

6

8

10

-50

Zim'02 food aid

-100

Bots'82 - laborbased relief

Bots'92, '96 & '02 - labor-based relief

Bots'79 - food aid

-150 Political System - POLITY Score

Source: I used the POLITY score from the Polity IV dataset to measure regime type in these cases. The POLITY score ranges from +10 (strongly democratic) to -10 (strongly autocratic). See Marshall and Jaggers, Polity IV Project: Political Regime Characteristics and Transitions, 1800-2007. To access the structure of the agricultural sector at the time of drought I used the Cereal Balance, which is calculated using the WFP and UNICEF adult-equivalent cereal requirement of approximately 150 kgs of grain per capita per year. Countries with a negative balance have to import food, while countries with a positive balance are potentially netexporters of grain.

18

Domestic Politics and Drought Relief in Africa

Although the regime argument provides a key building block from which I develop my argument—that is, the understanding that the policies adopted by incumbents are intended to be tenure-securing or tenure-prolonging—for a more comprehensive explanation of the crosscountry and within-country variation in drought relief programs, we must look beneath regime type. By the same token, while attention should be paid to de Waal’s claim that the food resources available to governments matter when they have to decide on a program of relief— for instance, relying on domestic food stocks is certainly cheaper than importing the same amount of food—to understand drought relief policy selection we must look beyond the agricultural sector. To that end, I consider the strength or weakness of political incumbency as the main factor in shaping the form of relief a government chooses, not regime type or the agricultural sector. I will demonstrate in this book that the strength of political incumbency accounts for democratic Botswana’s adoption of food aid in 1979, its switch to labor-based relief in 1982, and its subsequent consolidation of labor-based relief in 1992. I also leverage incumbency to explain Zimbabwe’s particular drought relief trajectory, which saw the country start with food aid in 1982, switch to labor-based relief in 1986 and 1992, and move to a grain loan scheme in 1996, before returning to food aid after the collapse of secure incumbency in 2000. I demonstrate the uniform political determinants of drought relief programs in Africa across different regime types, economic conditions, and agricultural contexts. Research Methods

In developing my argument, I simplify the dynamics of a much more complicated set of phenomena and decision-making processes in order to call out the most salient factors. I make no apology for this. My intention is not to offer the most complete account, but rather to highlight the primary connections in the politics of domestic government responses to disasters in Africa. As a result of this, I pay much less attention to external actors and offer an endogenous account of drought relief policy-making. While international aid agencies were present in the cases I examine, their role and activities were secondary to those of the domestic government. I employ a number of qualitative research methods to investigate why governments facing similar drought-induced threats of famine adopt sometimes convergent and at other times divergent programs of relief. In describing the research methods used in this book, it is useful to high-light the contribution of each of these approaches to the deve-

The Domestic Politics of Drought Relief in Africa

19

lopment and testing of the argument of the book. My theory-generating analysis of Kenya most approximates what is commonly thought of as process tracing.48 I test the argument developed from Kenya using a wider set of cases (10 drought relief programs) drawn from Zimbabwe and Botswana. Zimbabwe and Botswana provide broadly different contexts in which to test whether the argument linking forms of relief to levels of political vulnerability is informative. The two countries suffered several roughly covariant droughts. Their regime type, economic standing, and agricultural capacity are quite different; yet each country responded to drought. Zimbabwe is most similar to the theory-building case, Kenya, and I was reasonably confident that the basic incumbency model would explain its drought relief programs. Like Kenya, Zimbabwe was a British settler colony prior to independence. In both Kenya and Zimbabwe, settler colonialism had a distinctive imprint on political and economic developments. Although separated by a decade and a half in achieving independence, the victorious nationalist party in both Kenya and Zimbabwe had a weak commitment to securing incumbency under multi-politics. Kenya became a de jure one-party state and Zimbabwe established a de facto one-party state. Like Kenya, Zimbabwe possessed, for a period, the agricultural capacity to produce food surpluses in nondrought years. To provide a more rigorous test of the argument, I wanted to include a country in which one might not expect the causal relationship between incumbency and drought relief to exist. Botswana differs from Kenya and Zimbabwe in important respects. Unlike them, Botswana was a protectorate of Britain and became independent with the effects of benign neglect: weak modern state institutions. After independence in 1966, incumbents in Botswana managed to secure their position under regime conditions quite different from Kenya and Zimbabwe, in a multi-party democracy. Botswana is also different from Kenya and Zimbabwe in that its arid and semi-arid conditions limited the capacity of its agricultural sector. Structurally, Botswana is a net importer of food, even in non-drought years. My analysis suggests that programs of relief adopted in all three countries were determined by prevailing micro-political conditions. I show that governments that endured drought when incumbents were insecure (Botswana in 1979, Kenya in 1984, and Zimbabwe in 1982 and again in 2000) favored relief programs associated with immediate mobilizational and tenure-enhancing benefits. I contend that political interests in favor of immediate mobilization and fungible patronage led insecure incumbents, even in democratic Botswana, to adopt food aid

20

Domestic Politics and Drought Relief in Africa

programs for able-bodied adults. In these cases, insecure incumbents offered universal food aid (Botswana in 1979) or otherwise sought to politically target free food rations (Kenya in 1984 and Zimbabwe in 1982 and after 2000). This book also demonstrates that secure incumbents (such as Botswana in 1982 and 1992, and Zimbabwe in 1986 and 1992) all expressed the desire to adopt programs that targeted relief to the most needy, avoided waste, and were sustainable. These considerations, I argue, led secure incumbents to adopt labor-based relief (paying in cash in Botswana and with food rations in Zimbabwe). Organization of the Book

The remainder of the book develops the incumbency and drought relief model fully. Chapter 2 provides a description of the incidence of drought across Africa, examines in greater detail the core conceptual components of the book, such as the effects of drought, the policy options available to incumbents, and the challenges they face in responding to dearth. In analyzing the policy challenges facing droughtaffected incumbents, the chapter situates the form of relief adopted between programs that stress short-term political benefits with those that value cost-effectiveness and sustainability. In responding to drought, incumbents must pick a relief program, determine access and eligibility, decide whether to intervene in the market for food, and establish administrative structures for implementing the program on the ground. Each choice is associated with a distinctive matrix of political and economic benefits and costs, which incumbents are keen to capture and avoid, respectively. Thus, the chapter advances a broader conception of the policy-making challenges during drought than that presented in the existing famine studies literature. How do political vulnerability and the interests it generates affect drought relief policy-making? To answer this question, Chapter 3 examines the conditions that led to insecurity and autocracy in Kenya between 1978 and 1983, the year the country suffered its worst drought. The chapter suggests that the strength or weakness of political incumbency is subject to a number of factors, namely the susceptibility of the regime to possible coups and subversive movements, the strength of the government’s institutional power, the robustness of governing coalitions, the size of the president’s legislative majority, and the frequency of political and economic protests. With a clearer understanding of the conditions that made President Moi insecure, I turn to how political vulnerability affected his decision making in resolving the four droughtrelief policy challenges discussed in Chapter 2. I show that prevailing

The Domestic Politics of Drought Relief in Africa

21

insecurity led a rationally self-interested Moi to favor relief programs that could assist him in addressing his immediate political problems. The chapter demonstrates that the interests produced by prevailing insecurity dictated which general relief policy Moi favored, they shaped the level of access to aid he established, they compelled him to aggressively intervene in the urban food market, and they determined the administrative institutions and personnel Moi trusted to implement his food aid program. Furthermore, the chapter shows how each element of the relief program contributed to his consolidation of power. Chapters 4 and 5 consider whether my argument linking forms of relief to levels of political vulnerability explains choices about the form of relief in other countries, testing the argument in two quite different cases: Zimbabwe in Chapter 4 and Botswana in Chapter 5. Like Kenya, both Zimbabwe and Botswana demonstrate that the conditions that strengthen or weaken the security of incumbents determine programs of relief offered to able-bodied adults more so than regime type or agricultural performance. Chapter 4 shows that since attaining independence in 1980, Zimbabwe has gone through roughly four distinct periods of political security, each afflicted with at least one drought, and each with distinct agricultural and economic circumstances under which incumbents had to respond to dearth. Although a number of important agricultural and economic changes occurred alongside changes in the political vulnerability of incumbents in Zimbabwe, this chapter demonstrates that these have, thus far, not affected drought relief policy selection. In Zimbabwe, as in Kenya, programs of relief offered to adults shifted in accordance with the interests generated by changes in President Robert Mugabe’s political vulnerability at the time of each drought. Chapter 5 suggests that even under democracy, as in Botswana, the interests generated by political vulnerability dictate how incumbents responded to drought. The chapters also show that the way incumbents build security is consequential for the form and durability of subsequent programs of relief. Chapter 6 summarizes the contribution of the book to political science, policy choice in Africa, and famine studies, an issue very much on the minds of scholars, policy analysts, and politicians in Africa today. The book argues that variation in drought-relief interventions, and other disasters more generally, cannot be understood with reference to the technical capacity of the state alone. Instead, I show that we need to pay closer attention to micro-political conditions and how they structure the interests that guide incumbents during policy selection. In short, politics matters. The political status of incumbents in charge of the state affects

22

Domestic Politics and Drought Relief in Africa

how governments respond to the threat of famine and shapes the programs of relief they provide to citizens. The incumbency-led model of disaster response gives rise to clear hypotheses about the form, efficacy, and political manipulation of relief offered to the public across Africa. Notes 1

Botswana Daily News (April 2, 1982). Republic of Botswana, The Drought Situation in Botswana. 3 Boers, Botswana Drought Relief. 4 Amis, Financial Efficiencies in Drought Relief, p. 6. 5 Montshiwa and Mhlanga, Accelerated Rainfed Arable Programme [ARAP], Annual Report 1987-88. 6 “Drought Broken – Masire Declares,” Botswana Daily News (March 2, 1988). 7 Drèze, “Famine Prevention in Africa,” p. 147. 8 Leys, “Drought and Drought Relief in Southern Zimbabwe.” 9 Drèze, “Famine Prevention in Africa,” pp. 123-72. 10 Interviews with communal farmers in Buhera, Sanyati, and Mt. Darwin, April-May 2004. 11 See Government of Zimbabwe, Zimbabwe’s Experience in Dealing with Drought, 1982 to 1984; Bratton, “Drought, Food and the Social Organization of Small Farmers in Zimbabwe”; Weiner, “Land and Agricultural Development.” 12 Borton, “Overview of the 1984/85 National Drought Relief Programme,” pp. 24-64. 13 Cohen and Lewis, “Role of Government in Combating Food Shortages,” pp. 269-98. 14 Borton, “Overview of the 1984/85 National Drought Relief Programme.” Food aid donations made-up the remaining 23 percent. 15 Key aspects of the argument linking poor agricultural performance to bad policy are evident in Bates, “Pressure Groups, Public Policy, and Agricultural Development”; Lofchie and Commins, “Food Deficits and Agricultural Policies in Tropical Africa,” pp. 1-25; Lofchie, “Political and Economic Origins of African Hunger,” pp. 551-67; Bates, Markets and States in Tropical Africa; and van de Walle, African Economies and the Politics of Permanent Crisis, 19791999. 16 According to de Waal, Famine Crimes, p. 2, responsiveness is dependent on the establishment of an anti-famine political contract involving “political commitment by government, recognition of famine as a political scandal by the people, and lines of accountability from government to people that enable this commitment to be enforced.” In his view, neither the nationalist movements of the 1960s and 1970s nor the recent wave of political liberalization beginning in the 1990s was able to forge this political contract in Africa. 17 The literature on neopatrimonialism as a critical or distinctive feature of post-independence African politics is quite expansive and rich, but also contested. For a theoretical treatment of neopatrimonialism and its likely effect on 2

The Domestic Politics of Drought Relief in Africa

23

the dynamics of politics in Africa, see Médard, “Patrimonialism, Neopatrimonialism and the Study of the Post-colonial State in Sub-Saharan Africa,” pp. 76-97; Médard, “The Underdeveloped State in Tropical Africa,” pp. 162-92; Crook, “Patrimonialism, Administrative Effectiveness and Eco- nomic Development in Cote d'Ivoire,” pp. 205-28; Zolberg, “The Structure of Political Conflict in the New States of Tropical Africa,” pp. 70-87; Jackson and Rosberg, “Personal Rule,” pp. 421-42. On how neopatrimonialism might affect political reform, see Bratton and van de Walle, Democratic Experiments in Africa; and Bratton and van de Walle, “Neopatrimonial Regimes and Political Transitions in Africa,” pp. 453-89. Other scholars also link neopatrimonialism with the prevalence of corruption in Africa. See Harsch, “Corruption and State Reform in Africa,” pp. 65-87; de Sardan, “A Moral Economy of Corruption in Africa?,” pp. 25-52. For a critical reflection on what scholars diagnose as neopatrimonialism, see Mamdani, Citizen and Subject; and deGrassi, “‘Neopatrimonialism’ and Agricultural Development in Africa,” pp. 107-33, an engaging discussion of the methodological and substantive contributions and weaknesses of the concept in studies of Africa. 18 For a broad reading of Max Weber, see Weber, Economy and Society. 19 See Max Weber, “Politics as a Vocation,” in Weber, From Max Weber, pp. 77-128; Weber, “Bureaucracy,” in ibid., pp. 196-244. 20 The now classic statements on this were offered by Zolberg, Creating Political Order; Zolberg, “The Structure of Political Conflict in the New States of Tropical Africa,” pp. 70-87; Jackson and Rosberg, Personal Rule in Black Africa; Jackson and Rosberg, “Personal Rule,” pp. 421-42; Callaghy, The StateSociety Struggle; Young and Turner, The Rise and Decline of the Zairian State; Joseph, Democracy and Prebendal Politics in Nigeria. 21 Bratton and van de Walle, “Neopatrimonial Regimes and Political Transitions in Africa,” p. 458. 22 Szeftel, “Clientelism, Corruption & Catastrophe,” pp. 427-41; Tarr, “Undermining the Political Logic of African Governments’ Poor Economic Policies,” pp. 9-34. 23 See Omaar and de Waal, “Land Tenure, the Creation of Famine and the Prospects for Peace in Somalia”; de Waal, Evil Days; de Waal, Famine That Kills; Pankhurst, The History of Famine and Epidemics in Ethiopia Prior to the Twentieth Century; Sen, Poverty and Famines, pp. 86-112; and Shepherd, The Politics of Starvation. See also Keen, The Benefits of Famine, on the political uses of famine. 24 Drèze, “Famine Prevention in Africa,” p. 127. 25 I noted this variation in Munemo, “Political Incumbency and Drought Relief in Africa,” pp. 264-81, and sketched the framework for explaining it that I develop fully in this book. 26 Geddes, Politician’s Dilemma, p. 13. 27 See Schatzberg, Political Legitimacy in Middle Africa. 28 Solway, “Drought as a ‘Revelatory Crisis,’” p. 471. 29 Benson and Clay, The Impact of Drought on Sub-Saharan African Economies, p. 2. 30 Notably, the 1991-1992 drought in Zimbabwe resulted in a 9.5 percent decline in the manufacturing sector. See Benson and Clay, Understanding the Economic and Financial Impacts of Natural Disasters; Benson, “Drought and

24

Domestic Politics and Drought Relief in Africa

the Zimbabwe Economy, 1980-1993,” pp. 241-73; Benson and Clay, The Impact of Drought on Sub-Saharan African Economies. 31 Gourevitch, Politics in Hard Times, p. 221. 32 See Gill, Famine and Foreigners; Henze, Ethiopia in Mengistu’s Final Years; Sen, Development as Freedom; de Waal, Famine Crimes; Pankhurst, Resettlement and Famine in Ethiopia; de Waal, Evil Days. 33 See Chapter 2 for a fuller description of variation in programs for relief in these three countries. 34 See Government of Bechuanaland, Drought Relief Legislation, 1962; Government of Bechuanaland, Drought Relief Legislation, 1963-64; Government of Kenya, GO/3/2/31 Report on Famine Relief in Kenya, 1962 [Room 3, Shelf 2718, Box 4]; Government of Southern Rhodesia, Drought Conditions and Food Shortages, 1942-1947 [S2998/1]. 35 Both countries received significant praise for their responses to drought in the 1980s. See, for instance, Sen, Development as Freedom, pp. 178-79, 18384; and Drèze, “Famine Prevention in Africa,” pp. 123-72. 36 See Republic of Botswana, National Development Plan 7, 1991/97; and Republic of Botswana, National Development Plan 6, 1985/91. 37 Letter from the Permanent Secretary, Ministry of Agriculture (April 21, 2005). A34/6XVII(32)PS. 38 Robert Bates, Beyond the Miracle of the Market, p. 5. 39 Grindle and Thomas, “Policy Makers, Policy Choices, and Policy Outcomes,” pp. 213-48. 40 State autonomy as an analytical construct regained prominence with the publication of Evans et al., eds., Bringing the State Back In, and the debate the book subsequently generated. See Nordlinger et al., “The Return to the State: Critiques,” pp. 875-901; Cammack, “Bringing the State Back In?,” pp. 261-90; Almond, “The Return to the State,” pp. 853-74. 41 For a more comprehensive discussion, see Ames, Political Survival. 42 Grindle, “The New Political Economy,” p. 49. Similar arguments are also prominent in Grindle and Thomas, “Policy Makers, Policy Choices, and Policy Outcomes,” pp. 213-48. 43 For Sen’s explanation of famine causation, see “Famines as Failures of Exchange Entitlements”; “Starvation and Exchange Entitlements,” pp. 33-59; Poverty and Famines. Sen alludes to the determinants of government responsiveness in Poverty and Famines, and then develops this more fully in Development as Freedom. 44 See de Waal, Famine Crimes, pp. 43-48; Thompson, Drought Management Strategies in Southern Africa, ch. 2; and Drèze, “Famine Prevention in Africa,” pp. 150-53. 45 de Waal, Famine Crimes, p. 58. 46 Marshall and Jaggers, Polity IV Project, pp. 13-14. 47 I show in Chapter 5 that this score for Zimbabwe following the independence elections in 1980 was overly generous. In addition to an election marred by clashes between the main contending parties, there were numerous assassination attempts on various candidates. Zimbabwe’s founding elections were neither free nor fair. 48 As described by George and Bennett, Case Studies and Theory Development in the Social Sciences, process-tracing involves careful

The Domestic Politics of Drought Relief in Africa

25

examination of the micro-foundations of the causal chain between variables of interest.

2 Drought in Africa and Programs for Relief

This chapter frames the subsequent analysis of the form of drought relief programs across Africa. It begins by summarizing the frequency of drought in Africa. I then examine the alternative relief programs governments might adopt to prevent famine. Although deciding between food aid and labor-based relief is the primary policy challenge facing responding governments, it is not the only one. Incumbents must also determine access to relief, decide how they will intervene in the market for food to augment their general relief program, and establish a system of administering government relief. The chapter shows that each of these decisions involves significant political and economic trade-offs for incumbents. I conclude the chapter by summarizing the variations in drought relief decision-making in the three countries under study in this book. Drought in Africa

Drought is a sustained absence of, or reduction in, the availability of water. There are several different kinds of drought. If the reduction in water is due to a fall in seasonal rainfall significantly below expected averages, scholars talk of meteorological drought.1 If, on the other hand, sustained water scarcity is produced by the depletion of reservoirs, lakes, or aquifers, hydrological drought is in effect. A severe meteorological drought may lead to an agricultural drought affecting crop production, impairing rural farmers from providing for their families’ daily needs and livestock. In response to shortages of water and crop failure, farmers might decide to relocate families to areas they hope will support them in the short term. A decline in precipitation, therefore, may produce a social drought: a forced migration of people dependent on

27

28

Domestic Politics and Drought Relief in Africa

subsistence agriculture in search of relief or livelihoods elsewhere. Finally, there is what some scholars refer to as economic drought, which encompasses the repercussions of meteorological, agricultural, and social drought on the economy as a whole.2 Thus, one kind of drought may lead to another. Compared to other natural disasters (e.g., floods, earthquakes, epidemics, extreme temperatures, insect infestations, storms, wildfires, and mass migrations), the 382 meteorological droughts recorded between 1960 and 2010 make up about a fifth of the hazards of nature that most African countries suffered over that 50-year period. This frequency of drought incidence in Africa is about twice that of Asia over the same decades.3 Thus, the high frequency in the variability of rainfall in the arid and semi-arid regions of the continent, combined with the poor capacity of most soils in Africa to retain moisture, has made most of the continent particularly food insecure.4 Available data suggests that droughts across Africa have affected an estimated 326 million people and led to the death of close to a million.5 Two of the countries examined in this book, Botswana and Zimbabwe, have a drought incidence frequency characteristic of arid and semi-arid Africa. For instance, much of Botswana lies within the Kalahari Desert or in its arid and semi-arid periphery. As a consequence, rainfall tends to be sparse and unreliable and since independence Botswana has endured several drought periods. The country attained independence in 1966 just as the drought of the 1960s was ending. Another drought followed in 1978-1979, and, after a partial recovery during the 1979 and 1980 growing seasons, the onset of the long drought of the 1980s was marked by a precipitous 38 percent fall in annual rainfall (from 577 mm in 1980 to 359 mm in 1981). Seven continuous years of below average and erratic rainfall6 were then followed by droughts on fairly regular three-year intervals: from 1991 to 1993, 1995 to 1997, and 2001 to 2004. Like Botswana, Zimbabwe has endured fairly frequent deviations from its normal annual rainfall of about 660 mm.7 The first postindependence drought in Zimbabwe started in 1981 and lasted for two agricultural seasons, with average annual rainfall about 218 mm below the mean. That drought was followed by another between 1986 and 1987, with average rainfall about 220 mm below the mean. After about three good seasons, Zimbabwe endured its worst recorded drought over the 1991 to 1992 rainy season. Annual rainfall was a staggering 320 mm below the mean, about half of the normal rainfall. This drought period was then followed by two more droughts: from 1995 to 1997 and 2000 to 2004.

Drought in Africa and Programs for Relief

29

However, meteorological droughts in Africa have not been limited to the arid and semi-arid regions. Droughts have also afflicted countries like Kenya, whose location on the Equator and between two major bodies of water (the Indian Ocean to the east and Lake Victoria to the west) typically assures it of two rainy seasons: the long and heavier rains from March to May and the short and lighter rains from October to December. Compared to Southern Africa’s single rainy season from November to March, the short rains in Kenya act as a natural insurance for rain-dependent farmers. However, over the 1983 to 1984 agricultural season, nature’s insurance failed Kenya. Poor short rains in late 1983 were then followed by a complete failure of the long rains in early 1984. Typically, rainfall during the long rains ranges from 100 mm to as much as 600 mm for the Western and Coastal regions of the country.8 During the long rains of 1984, two-thirds of the country received less than 100 mm of rain. This was well below the 300 mm needed for a successful maize crop. Total rainfall in the key agricultural regions of the country (Eastern Province, Central Province, and the Rift Valley) was more than 60 percent below normal. The drought that gripped Kenya in 1983-1984 was its worst in 100 years.9 In addition to frequency, drought across Africa also tends to be covariant, generally affecting several countries in any given year. Since 1960, there have been roughly four major peaks in the continental incidence of drought: between 1971 and 1975, when no fewer than eight countries experienced a drought each year; from 1980 to 1985, when as many as 32 countries were hit by drought in 1983; a relatively short El Niño–related drought peak between 1990 and 1993; and finally a rather prolonged period from 2001 to about 2005, when no fewer than five countries experienced below normal rainfall. By looking at how different countries dealt with the same drought periods, we can more closely examine and compare the conditions under which each executive directed policy. Programs for Relief

Drought shocks both food production and household income. As a consequence, most discussions of drought relief suggest that the primary decision governments have to make in responding to the threat of famine is to choose between two approaches.10 The first, which reflects the manner in which drought relief is intuitively imagined, asserts that the primary task of public agencies during threats of famine is to acquire and distribute food aid to those people whose food production entitlements (i.e., the ability to meet food requirements from your own

30

Domestic Politics and Drought Relief in Africa

production) have collapsed.11 The second approach, with roots in traditional laissez-faire economic theory, draws different conclusions about the most appropriate general program of relief. Importantly, this approach stresses that drought has significant income effects which food aid leaves unaddressed. As such, scholars and experts contend that it is more appropriate and efficacious to respond to drought-induced threats of famine by addressing lost incomes through labor-based relief programs.12 Beneficiaries then have the freedom to decide for themselves whether what they most need is food, which they can buy on the market, or other requirements like health care, education, etc. These two approaches inform the discussion of relief programs in this book, yet neither strategy captures the full range of choices, or the trade-offs facing incumbents who decide to respond to drought-induced threats of famine. In the following section, I describe what each policy looks like in theory and practice, after which I discuss three other policy challenges incumbents need to resolve in order to respond to droughtinduced threats of famine. Smoothing Consumption: Food Aid in Africa

The incidence of meteorological drought across Africa has resulted in shocks to agriculture. For instance, the 1983-1984 drought in Kenya cut the aggregate national production of maize, the primary staple, by close to 38 percent from the 1983 harvest of 2.3 million metric tonnes (Mt). Other key food crops were also severely impaired by this drought. The production of wheat, Kenya’s second most important grain, fell by close to 43 percent compared with 1983. To keep the people of the cities and rural areas fed and meet the suggested adult-equivalent yearly cereal requirement of 150 kgs of grain per person, Kenya had to make up a cereal deficit of over 1.1 million Mt. The 1983-to-1984 drought also severely affected Kenya’s potato and bean production. The potato crop was reduced by about 50 percent and bean production fell by 75 percent, from 270,000 Mt in 1983 to only 68,000 Mt in 1984.13 The collapse of bean production had noticeable immediate negative nutritional effects because many rural Kenyans, the urban working class and poor, relied on beans as a relatively cheap source of protein.14 As in Kenya, droughts in Botswana and Zimbabwe also severely impaired domestic food production. In Botswana, the failure of rain during the 1981 agricultural season resulted in a 68 percent fall in crop production in 1982 (from 54,285 Mt in 1981 to 17,220 Mt in 1982). In 1984, the third consecutive year of the drought, crop production fell to just 7,275 Mt.15 Measured by the adult-equivalent yearly cereal require-

Drought in Africa and Programs for Relief

31

ment of about 150 kilograms, Botswana’s harvest in 1982 was only enough to feed the country of 973,000 people for a month, while the 1984 harvest was less than half a month’s requirement. In Zimbabwe, too, drought decimated rain-fed domestic food production. In 1982, the first year of drought, total cereal production fell from 3.3 Mt in 1981 to 2.2m Mt and then was halved to 1.1 million Mt in 1983, the worst year of the drought, before rising slightly to 1.4 million Mt in 1984. Although the aggregate cereal produced in 1982 seemed respectable, the drought affected the high-population rural areas, where about 77 percent of Zimbabwe’s population lived. More importantly, the presence of food in some parts of the country did not guarantee its availability in affected regions. In the province of Matabeleland, local women risked arrest by illegally crossing into apartheid South Africa in search of food and water.16 Officials in the region estimated that close to 1 million people were in need of government assistance if starvation was to be averted.17 Because drought shocks to food production are well documented,18 smoothing consumption through food aid seems to be the most intuitive program of relief. Direct delivery, or food aid, principally involves the government in acquiring, transporting, and then distributing food to the population at no cost to the beneficiaries. Four of the cases of government relief covered in this book used food aid as the primary form of support for everyone, including able-bodied adults. Food aid programs were adopted by Botswana in 1979-1980, Kenya in 19841985, and Zimbabwe between 1982 and 1984 and then again between 2000 and 2004. Thus, food aid programs were adopted in countries with different regime types (in democratic Botswana, de jure one-party autocratic Kenya, and dominant-party authoritarian Zimbabwe) and quite different domestic food production capacities (with Botswana a net-importer of food, Kenya sometimes a surplus producer, and Zimbabwe a breadbasket in the 1980s). Drought-afflicted countries adopting free food have three potential sources of food for free distribution. First, countries that are typically surplus food producers in non-drought years may hold sufficient stocks when drought strikes (for instance, Zimbabwe in 1982 and 1986). Governments may also acquire food for direct delivery on the world market, a strategy typical of famine-threatened countries that do not have sufficient stocks when drought strikes (for instance, Kenya in 1984 and Zimbabwe in 1992 and after 2000), or of countries that are structurally net importers of food even in non-drought years (for instance, Botswana). Finally, food aid relief might also be adopted using food donations from foreign countries or international relief agencies such as the World Food Programme, Oxfam, and USAID, among others.

32

Domestic Politics and Drought Relief in Africa

Because I am primarily interested in government-initiated and funded drought relief programs, I limit my analysis in this book to countries that relied on the first two sources of food—domestic stocks and government imports—for the bulk of food used in their relief programs.19 These two sources of food for direct delivery are very costly for responding governments, and thus imply a significant opportunity cost for incumbents. Aid programs that rely on imported food are likely to be more financially burdensome than those that draw on domestic stocks. The first reason for this difference in cost, as noted by students of agricultural economics and political economy decades ago, is the tendency across Africa for governments to keep the price of food crops on the domestic market below the price of the same crops on the world market.20 According to Bates’s now classic argument, keeping domestic food prices below the world-market equivalent advanced two goals of Africa’s rulers. Low food prices served the populist tenure-enhancing interests of incumbents who searched for cheap ways to keep cities fed and therefore less politically volatile. As is well demonstrated by Schartzberg, food security, won in part through cheap food prices for the cities, was a common basis on which incumbents across Africa were evaluated.21 Ironically, what was good for the stability of cities and the legitimacy of incumbents in the short-term structurally weakened food production in the medium to long run.22 Bates and others also argue that incumbents kept domestic crop prices low in order to generate significant revenues by using the monopsony powers of agricultural marketing boards to buy crops at the domestic price and then export the same crops at the higher world market price. For instance, in January of 1984, just five months before President Moi made an appeal to international donors for assistance in the country’s drought relief effort, Kenya’s National Cereals and Produce Board (NCPB) exported 47,434 Mt of white maize and earned close to US$8.3 million in foreign currency.23 Using the 1983 domestic price of maize of about US$121.05 per Mt, the government of Kenya had at most spent US$5.72 million to purchase this grain. Revenues generated from this indirect tax on food producers could be used by incumbents to fund development projects or, in the more egregious cases, to fund incumbents’ networks of patronage and corruption. A comparison of the per-Mt cost of Kenya’s food aid program in 1984-1985 with Zimbabwe’s in 1982-1984 confirms this disparity in the cost of relief efforts that rely on imported food versus local stocks. Between 1984 and 1985, for its relief program the Kenyan government imported close to 500,000 Mt of maize at a cost of about US$87

Drought in Africa and Programs for Relief

33

million,24 about US$174 per Mt. However, over the 1983 marketing season, the year before relief started, the NCPB had offered farmers a domestic price of Ksh 2,300.00, or about US$121 per Mt of maize.25 Thus, had the NCPB had a healthy strategic maize reserve leading up to the drought in 1983-1984, it would have saved roughly US$53 for each Mt of maize devoted to the relief effort. In contrast, Zimbabwe drew on its very large maize stocks (about 1.2 million Mt) from bumper harvests in 1980 and 1981 to use in its food aid program between 1982 and 1984. The price of maize in these two years was US$141 per Mt in 1980 and US$166 per Mt in 1981. Thus, even in the unlikely event that all the maize used in Zimbabwe’s relief program was purchased in 1981, the cost of food for direct delivery in Zimbabwe was at least US$8 Mt cheaper than the US$174 Kenya paid for maize imports. Once food has been acquired for direct delivery, responding governments then have to establish and manage a system of transporting it from ports or line-of-rail storehouses or national depots to distribution centers around the country. Systems of food distribution might involve the government’s fleet of trucks, the military, private truckers, or combinations of these. Each option is associated with some potential consequences. For instance, using the government’s civilian fleet of trucks, while cheap and expeditious, might compromise existing development projects. The use of government vehicles is also vulnerable to logistical problems. Second, relying on the military often minimizes public disorder during food distribution but poses security risks if the army is not professional. Finally, enlisting private truckers sometimes overcomes the logistical difficulties associated with public transportation, but it raises the cost of relief; opening food transportation to private companies also creates opportunities for corruption. All three systems of food transportation are represented in my cases. In 1979-1980, Botswana combined private transportation with the use of government vehicles to transport food from line-of-rail points to distribution centers across the country, generally schools and clinics. Zimbabwe between 1982 and 1984 started by relying exclusively on the government-operated Central Motor Engine Department to transport maize from government depots to distribution centers. However, logistical problems with this system forced the government to sub-contract the transportation of food to private operators. Finally, Kenya in 1984 relied on a system of food transportation that included significant military involvement in ferrying maize shipments from Mombasa to provincial and district distribution centers. Analysts of drought and famine relief concede that direct delivery through food aid seems the most intuitive and straightforward response

34

Domestic Politics and Drought Relief in Africa

to dearth.26 But food aid relief requires significant logistical and administrative competence on the part of government, above all, an efficient system of food acquisition, transportation, and distribution that can transfer aid reliably to beneficiaries.27 All four cases where government adopted food aid exhibited many of the problems scholars cite. An internal government review concluded Botswana’s food aid program in 1979-1980 was short-staffed and the personnel were poorly trained; the government was slow in dealing with these problems.28 According to the review, another shortcoming of the relief program was its organization. Critically, by entrusting the central government’s management of relief to three separate committees, the organizational structure resulted in a duplication of functions and a poor definition of responsibilities. Finally, at the local level, district officials had the attitude that drought relief work should not interfere with their ‘normal’ responsibilities, an attitude that affected the commitment of District Drought Committee staff to the relief program. Kenya and Zimbabwe experienced similar problems with the administration of their food aid programs. Finally, in addition to raising the concern that food aid generates waste if it is not targeted, scholars note that the high cost of acquiring and transporting food and staffing distribution centers means that only a small proportion of the cost of relief is transferred to affected people in the form of aid.29 A comparison of the financial efficiencies between the food aid and the labor-based components in Botswana’s 1982-1990 drought relief program provides striking evidence for this claim. As indicated in Table 2.1, only about 10 percent of the yearly cost of the food aid program was transferred to beneficiaries in the form of a food ration, while close to 90 percent of its budget went toward administrative costs; by contrast, 62 percent of the cost of the labor-based relief component went to its beneficiaries. It is also notable that the income transfer for the labor-based relief program, US$155.00, was almost four times greater than the US$39.20 achieved under the food aid program. This suggests that by adopting labor-based relief instead of free food aid governments can actually increase their famine protection transfers per beneficiary. However, as I argue in subsequent chapters, adopting laborbased relief limits the ability of incumbents to take advantage of some of the patronage opportunities created by systems of food aid.

Drought in Africa and Programs for Relief

35

Table 2.1: Financial Efficiencies of Components in Botswana’s Drought Relief Program, 1982-1990 Food Aid Component

Laborbased Relief Component

Average annual total cost (in 2005 US$)

$9.19 million

$12.6 million

Recurrent cost as percent of total

48%

20%

Salaries of administrators as percent of total

16%

5%

Transport costs as percent of total

26%

13%

Percent of total cost transferred to beneficiaries as relief

10%

62%

Average income transfer per beneficiary (in 2005 US$)

$39.20

$155.00

Source: Amis, Financial Efficiencies in Drought Relief.

Smoothing Incomes: Labor-Based Relief in Africa

Another criticism of food aid relief is that it fails to address the income shocks produced by drought. Quite often, meteorological drought progresses into economic drought. A short account of the wider effects of drought in Kenya and Zimbabwe illustrates this. In the pastoral north of Kenya, drought started much earlier with the failure of long rains in early 1983 and persisted until late September of 1984, leaving Northern Kenya without normal rainfall for close to 21 months. As drought progressed, livestock were the first causalities of scarce food and water. As many as 95 percent of cattle, 60 percent of goats, 40 percent of sheep, and 5 percent of camels perished in Northern Kenya.30 These high losses were devastating for a region of the country in which livestock served several important functions in household livelihoods, in society, and in the market economy. They provided farmers with food, either directly in the form of meat or milk or indirectly through income generated from their sale. Animals were also a store of wealth and a marker of social standing. 31 Outside of the pastoral north, the 1983-1984 drought had noticeable income shocks in other regions of the country, such as Eastern and Central Kenya, which in normal years offered farmers the insurance of a

36

Domestic Politics and Drought Relief in Africa

wider set of livelihood strategies. In addition to keeping livestock, farmers in Eastern and Central Kenya grew a variety of food crops (maize, beans, sorghum, and millet) and cash crops (coffee and tea in the higher range districts, and cotton and tobacco in the lower range districts). From food crops, farmers covered their own household food requirements and in good years sold surpluses or cash crops on the market to cover school fees, input purchases (seeds, fertilizers, etc.), medical expenses, and other market-based requirements.32 As in the north of Kenya, the income effects of the drought in Eastern and Central Kenya were first evident in the livestock sector and then became evident in rain-fed food production as the agricultural season progressed. In Eastern and Central Kenya, an estimated 30 to 40 percent of cattle died during the drought.33 As in Northern Kenya, this was the major mechanism by which drought affected the economic and social standing of households. The death of animals meant an immediate loss of income, savings, and security from future shocks to household livelihoods.34 Then, in what became a common coping strategy during the drought,35 farmers tried to limit losses by bringing their cattle to market at Kenya Meat Commission (KMC) depots. The KMC was established in the 1950s by the colonial administration as a governmentrun marketing board. In exchange for market monopoly over meat purchases domestically and meat exports externally, the commission was required by statute to provide marketing and processing facilities across the country.36 After independence, the new government retained and expanded the marketing and control structure the KMC provided.37 However, while decisions by households to bring their animals to market were individually rational, the collective market outcome from these decisions was sub-optimal. Generally, livestock prices were low and KMC depots lacked the processing capacity to handle all the animals brought to them in a timely manner. The government tried to deal with the processing problems by increasing the permitted slaughtering rate at “two of its plants from 7,000 head per month to 23,000 head by August 1983.”38 However, this 229 percent jump in processing quotas proved to be insufficient for the large numbers of cattle being brought to market during the drought. As a consequence, by late 1983, some households were losing their animals while they waited in the long lines at KMC depots.39 As it did in Zimbabwe in 1992, drought may also produce income shocks by impairing macro-economic performance. Between 1991 and 1992, Zimbabwe endured its most severe drought in a century.40 The more than 300 mm drop in annual rainfall had a severe impact on the availability of water for domestic and industrial use, as dams, wells, and

Drought in Africa and Programs for Relief

37

boreholes across the country ran dry. To conserve water in the dams, city councils adopted rationing in hopes of stretching the available water until the next rainy season. One unintended consequence of rationing was that it severely impaired the operation of the high-employment light-industrial sector. Without water, and often also without electricity, firms across the country were forced to close and send their employees home. Benson estimates that the 1991-1992 drought in Zimbabwe was associated with “year-on-year declines of 5.8 percent in GDP at factor cost … and 8.5 percent in industrial GDP.”41 Thus, insufficient rainfall often means households suffer multiple shocks to their income-earning potential. Mindful of these major income effects of drought and also weary of the many problems associated with food aid, scholars contend that an alternative to food aid is to smooth for lost incomes.42 At heart, laborbased programs typically involve the government in organizing large public works programs that provide temporary employment to affected groups. Governments adopting this strategy of relief have to make decisions on appropriate projects, and then acquire and distribute materials to participants. The ease with which governments can set up works projects on a national scale is somewhat dependent on the complexity of the projects and the overall capacity of the state. For instance, projects that ask participants to work on rural road maintenance are relatively simple and much easier to set up than projects that involve general construction or the building of small dams. The latter require more complex materials and equipment, constant supervision of the work, and, sometimes, technical assistance. Four of the cases covered in this book involved labor-based relief as the primary source of aid for able-bodied adults. Botswana adopted this form of relief between 1982 and 1990 and then went on to institutionalize this strategy in its 1991 National Development Plan. Since then, labor-based relief paying in cash has been the norm in Botswana. But my cases reveal that labor-based relief is not the preserve of democracies. Zimbabwe also successfully used this form of relief twice: in 1986-1988 and again in 1992-1993. Governments adopting labor-based relief have to decide how they are going to pay participants. Labor-based programs, like Botswana’s between 1982 and 1990, tend to smooth incomes by paying works participants a nominal relief wage. As I noted in Chapter 1, at the beginning of Botswana’s relief program in 1982, a drought subsistence wage of US$14.40 a week was paid to each willing worker and US$19.20 to supervisors.43 Work-based relief programs may also ‘pay’ participants in

38

Domestic Politics and Drought Relief in Africa

food rations. For instance, Zimbabwe’s two labor-based relief programs in 1986-1988 and in 1992-1993 paid works participants in food rations. The consensus among some scholars is that, for poor developing countries, labor-based relief is the most effective way for governments to respond to threats of famine.44 That is, labor-based relief is shown to be efficacious, cost-effective, and sustainable. Scholars stress that the minimal requirement of asking relief beneficiaries to work eliminates the indiscriminate distribution of government aid typical of food programs. The principal argument here is that “public works as a mechanism for transferring income have the major advantage that they are ‘self-targeting’; that is, that only those without income and work (or with very low incomes) would seek such employment.”45 If labor-based relief pays in cash, as was the case in Botswana, scholars contend that smoothing for lost incomes in this way enables affected groups to compete effectively for food on the domestic market, while at the same time it encourages trade and commerce. As Sen put it: The employment route also happens to encourage the processes of trade and commerce, and does not disrupt economic, social and family lives. The people helped can mostly stay in their own homes, close to their economic activities (like farming), so that these economic operations are not disrupted. The family life too can continue in a normal way, rather than people being herded into emergency camps.46

However, it should be noted here that cash support is predicated on the assumption that food prices will remain stable on the market. Should food prices rise dramatically over the course of the relief effort, beneficiaries risk being stuck with a nominal relief wage they cannot turn into food on the market.47 More recently, Subbarao suggests that another advantage of laborbased relief programs is in the construction of much-needed infrastructure, if programs are well designed.48 According to Subbarao, the construction of infrastructure minimizes the trade-off between public spending on relief and public spending on development. It should be noted that the infrastructural advantages that Subbarao has in mind assume that governments decide on the more complex set of works projects like dam construction or the building of schools and local clinics. All these advantages of labor-based relief programs have led some scholars to propose rural public works projects as standing instruments of development beyond their use for relief purposes.49 In summary, it should be stressed that these two general relief policies (smoothing consumption and smoothing income) need not be

Drought in Africa and Programs for Relief

39

mutually exclusive. The most comprehensive and efficacious drought relief programs combine the strengths of each. A number of relief experts contend that governments might start their drought relief program by offering food aid for a period of time and then progressively switch to labor-based relief once they have set up appropriate works projects around the country.50 In addition to this sequencing of forms of relief, experts also recommend that governments adopt food aid for some categories of the population (e.g., children, pregnant and lactating mothers, the elderly, households without a prime-age adult, etc.) and labor-based relief for the remaining able-bodied adults. All four of the cases in which governments adopted labor-based relief for adults included provisions for food aid to children and, in some cases, to other vulnerable groups. Because food aid may be adopted as a transitional program or targeted to categories of the population that are not capable of working, the key decision I explain in this book is why governments sometimes decide to provide food aid to able-bodied adults when they could do better (administratively and in terms of the per capita support they provide to drought-affected people) by adopting labor-based relief. Access to Relief: Universalism or Non-Exclusion

Yet these general policies alone do not fully capture the full range of decisions and challenges that incumbents have to resolve in responding to drought-induced threats of famine. Following from this first-order decision about a general program of relief are secondary questions concerning who qualifies for government support during a food crisis. Decisions about access and qualification reflect the degree to which incumbents are willing to bear the political cost of limiting public aid to provide adequate support to a narrower, but most needy, segment of the population. The literature suggests that access to drought and famine relief programs may be guided by two broad principles: universalism, on the one hand, and non-exclusion, on the other.51 Relief programs guided by universality aim to provide “unconditional support to everyone without distinction.”52 For many scholars, activists, and journalists, programs of relief informed by universal protection are ethically appealing because they affirm every citizen’s right to government support. On a more practical implementation level, universal access removes the administrative burden and, as I will show in the case studies, the political cost to incumbents for targeting relief. This aspect of universal relief or welfare programs has led some scholars to contend that they might be

40

Domestic Politics and Drought Relief in Africa

better because the broad beneficiary pool they generate garners the widest political support, which might lead to more secure budgetary allocations.53 In terms of stated central government policy, none of the drought interventions examined in this book adopted a program of relief that was committed to universal access. However, in practice, the political cost associated with following the administrative criteria for selection into relief under Botswana’s 1979-1980 food aid program led officials to make food rations available to as many residents as they could. Instead of waiting until there was evidence that 40 percent of children under five in their district were undernourished, many districts declared Stage II relief without regard to administrative criteria. Three out of nine districts distributed food aid to all adults. In a fourth district, officials used their own funds to purchase and distribute aid to everyone, even though only a small area qualified for Stage II relief.54 Thus, access to food aid among adults in 1979-1980 in Botswana was practically universal. Critics of universal support note that its advantages are also its biggest weakness. Fundamentally, scholars suggest that providing assistance on a blanket basis offers no guarantee that the most needy segments of the population have adequate support. Quite often, to spread aid to as many people as possible, governments might have to lower the per capita protection level. But while smaller aid levels might not be of consequence to more-well-off households, poorer households face the prospect of persistent undernourishment. Alternatively, if governments commit to providing protection to everyone at near the adult-equivalent monthly cereal requirement (about 12.5 kgs per month), universal coverage makes relief programs unduly expensive for poor governments. Consequently, some scholars suggest that a simple solution is for government to imbed targeting or selection mechanisms into its relief program. Scholars contend that when time is short and resources are limited, the cost of failing to come to the rescue of the most vulnerable by targeting relief to them can be very high.55 From this perspective, scholars and experts agree that relief programs should have mechanisms which privilege access to aid to the most vulnerable instead of casting the protection net widely with the hope of capturing the most needy. This approach describes the principle of non-exclusion. Under a non-exclusive program, government concentrates on protecting segments or groups of people considered to be most vulnerable to starvation. Non-exclusive programs of relief call for governments to establish clear mechanisms and benchmarks for identifying the most vulnerable groups and then providing them with sufficient support.

Drought in Africa and Programs for Relief

41

Common mechanisms for identifying beneficiaries include: top-down centralized administrative criteria (such as nutritional measures, demographic characteristics, or geographical location), at one extreme, and bottom-up self-acting tests by potential beneficiaries (like choosing to work on public works projects in exchange for government aid), at the other. Four of the drought relief programs examined in this book adopted a self-selection mechanism for targeting government protection. These were Botswana’s cash-paying works programs in 1982 and 1992 and Zimbabwe’s food-for-work programs in 1986 and 1992. Under each of these, to receive relief, able-bodied adults had to opt into works programs organized by the government. Households with able-bodied adults who did not sign up for works projects did not receive any protection. Only one case, Botswana in 1979-1980, saw government attempt to establish a top-down administrative benchmark for determining access to relief for able-bodied adults. But , as I showed earlier, Botswana’s decision to provide food aid to able-bodied adults only when 40 percent of children under five in a district were determined to be nutritionally at risk proved to be too politically costly for incumbents.56 . As a consequence, food aid to adults was provided on a more universal basis. A third, though much less transparent, mechanism for targeting drought relief involves incumbents delegating the responsibility of deciding who gets aid to trusted functionaries at the local level. Three of the cases of drought relief covered in this book had a discretionary system for targeting government protection. In Zimbabwe’s 1982-1984 food aid program, incumbents gave local ZANU party cadres the authority to draw up adult beneficiary lists. After switching to selfselection between 1986 and 1996, Zimbabwe then returned to a system of discretionary inclusion into drought relief programs between 2000 and 2004. In a similar fashion, in 1984, Moi delegated the responsibility of drawing up Kenya’s beneficiary lists to district commissioners, traditional chiefs, and sub-chiefs at the local level. One case, Zimbabwe’s grain loan scheme from 1996 to 1999, combined top-down administrative criteria with self-selection and the discretion of local agents. As a consequence, the process for winning government aid was long and complex. A district first had to be declared drought stricken by the Agricultural Extension Department of the Ministry of Agriculture. Potential beneficiaries were then required to self-select by forming grain loan groups, which collectively applied for the aid loan from government. The government gave village level kraal heads and district level chiefs, members of Zimbabwe’s traditional

42

Domestic Politics and Drought Relief in Africa

system of government, the authority to certify loan groups;57 without their support, grain loan groups had no hope of receiving aid. Chapter 4 offers an explanation of why Mugabe’s government adopted this decidedly regressive drought relief program. In brief, I contend that excessive political security produced pathologies in how government responded to crises by the mid-1990s. Government Intervention in the Market for Food

Droughts and the threat of famine they produce often lead to hoarding, food speculation, and price increases.58 Kenya in early 1984 illustrates some of these market responses to dearth. The drought that struck Kenya from late 1983 through 1984 decimated domestic food production. As a consequence of the collapse in seasonal food production and the inadequacy of household stocks, peasant farmers, who in past years had supplied food to domestic markets, now were turning to the market to purchase food for their families. A common practice among rural farmers was selling their surplus to the state-run marketing board, the NCPB. In normal years, selling surpluses right away was a prudent strategy: it generated income and allowed farmers to avoid the cost of storing maize. However, what was a smart strategy in normal years proved to be glaringly exposed in 1983-1984 when both rains failed. Consistent with the literature on the functioning of food markets during famine-threat periods, the above-normal demand for food produced by the collapse of subsistence agriculture led to food price increases, particularly in the most hard-hit regions of the country. Table 2.2 below compares the changes in the price of maize, Kenya’s staple cereal, between drought-hit and drought-free districts. In drought-hit provinces, the price went up by over 130 percent between January and November 1984, compared to an increase of about 23 percent in nondrought provinces over the same period. Households responded to price increases by hoarding, and this in turn produced a vicious cycle of rising prices and hoarded food. These rational household (hoarding) and market responses (price increases) to uncertain food availability and demand pressure exacerbated Kenya’s food crisis and produced a rather strong response from the government. In early June 1984, President Moi directed the police to arrest and prosecute anyone caught hoarding food. When the prospect of arrest seemed to have little impact, Moi threatened hoarders with losing their stock of food if caught.59 Retailers were also threatened with stiff penalties if they were found to be selling food above the governmentmandated levels.60

Drought in Africa and Programs for Relief

43

Table 2.2: Changes in Kenyan Maize Prices by Province, 1984

Market

Average price of maize in January 1984 (KShs/kg)

Average price of maize in November 1984 (KShs/kg)

Average percentage increase in the price of maize

Drought-hit provinces

2.53

5.88

132%

Non-drought provinces

2.27

2.79

23%

Source: Government of Kenya, Central Bureau of Statistics, Ministry of Finance and Planning, Market Information Bulletin (January-June 1984 and July-December 1984) cited in Drèze, “Famine Prevention in Africa.”

Hoarding, price increases, and speculation in food present such major challenges for incumbents that relief policy of any type is often augmented by at least some government intervention in the food market. Commenting on these challenges, some scholars contend that governments have two basic options for dealing with them: market-excluding or market-complementary interventions.61 According to Drèze and Sen, a market-excluding intervention during a food crisis is one that leads to the replacement of the market with state controls (i.e., the state takes over the role of the market by setting food prices and/or controlling the private movement of food). In theory, exclusionary interventions are intended to address the market failures common during famine threat situations, such as trafficking in food and the failure of the market to direct food to affected areas.62 Five of the drought relief programs analyzed in this book involved some form of exclusionary market interventions by governments: Zimbabwe from 1982 to 1984, from 1986 to 1988, from 1992 to 1993, and again from 2000 to 2004, and Kenya from 1984 to 1985. Intent notwithstanding, scholars have shown that often when governments take over the market for food during crisis situations their interventions generate negative externalities, which may make the problems they are intended to resolve much worse.63 Exclusionary interventions are linked with inequitable distribution of existing food. They have also been shown to freeze the movement of food from surplus areas to deficit areas, which generates opportunities for rent-seeking and corruption. Finally, state controls of the market for food might result in large differentials in food prices across regions.

44

Domestic Politics and Drought Relief in Africa

Because of these potential problems with exclusionary interventions and the immense state capacity needed for them to be minimally effective, Drèze and Sen contend that complementary interventions might be a better strategy for governments facing food scarcity. The interventions that Drèze and Sen have in mind limit the government’s involvement in the market for food to supporting the market in acquiring it (typically, importing it from abroad). After food has been acquired, governments then rely on the existing market structure to distribute it to the general population. Four of my cases saw incumbents adopt complementary interventions. This included all three of Botswana’s drought relief programs and Zimbabwe’s 1996-1999 program. In theory, this intervention assumes that the existing market structure is largely robust and functional and that all it needs is help acquiring food. In practice, the functionality and fairness of markets for food varies, as I will show in discussing Kenya, Zimbabwe, and Botswana. Administering Relief

As with the other components of relief discussed above, responding incumbents have important choices to make concerning the administration of relief programs. Two general approaches to relief administration are debated in the literature. First, there are scholars and relief experts who argue that drought-prone countries should establish and maintain permanent standing institutional and administrative structures.64 The key issues that these scholars and experts raise include the need for government to have developed a coherent national food security strategy and a contingency plan in case of a threat of famine and to have standing government agencies with the responsibility of administering relief. In other words, these scholars contend that the most effective administration of drought relief programs is predicated on institutions that are sustained even during non-famine years. These experts stress that prior planning and meetings build the necessary hierarchy and system of communication required to administer a relief program successfully. Other scholars, however, are not convinced that poor governments ought to establish standing relief institutions.65 Citing the difficulty and cost of setting them up, these scholars contend that governments might be better off relying on ad hoc institutions supported by line staff in key service ministries to administer relief. In recommending the use of ad hoc institutions, Cohen and Lewis insist that “many of the functional capabilities required to mount and manage an effective food-shortage response program are generically similar in character to those required

Drought in Africa and Programs for Relief

45

for normal administration and development activities.”66 Thus, a functional standby strategy is seen as an alternative to the common recommendation of standing drought relief institutions. The cases examined in this book include both types of administrative structures and their combinations. Four cases— Zimbabwe from 1982 to 1984, from 1996 to 1999, and from 2000 to 2004, and Kenya from 1984 to 1985—used substantively ad hoc institutions to administer relief. These institutions all included personnel appointed by incumbents. In three cases (Botswana in 1979-1980 and Zimbabwe in 1986-1988 and 1992-1993), drought relief was administered using a combination of standing bureaucratic institutions and ad hoc committees established by incumbents. Only Botswana, after 1981, established standing drought relief institutions and relied on them to administer relief from 1982 to 1990 and again from 1992 to 1994. Summary

The foregoing discussion underscored the frequency and effects of drought in Africa. It also highlighted that incumbents intending to respond to the threat of famine produced by drought have four major challenges to resolve. They must choose between smoothing consumption and smoothing incomes, decide who qualifies for relief, determine the type of intervention in the market for food, and consider the institutions and personnel they will rely on to administer relief on the ground. Table 2.3 summarizes the range of challenges responding incumbents have. Each decision, as I will show in subsequent chapters, involves important trade-offs for responding incumbents, trade-offs whose resolution is determined by the interests generated by their political vulnerability. With this understanding of the full range of choices that constitute the form of a country’s drought relief program, it is possible to offer a more comprehensive summary of the variation in relief programs I explain in this book. The drought relief programs examined here vary significantly on all four dimensions. My task for the remainder of the book is to account for this variation in responses to drought-induced threats of famine. I develop my argument by examining Kenya’s intervention in 1984 (Chapter 3), where incumbent vulnerability was outwardly apparent and its imprint on how Moi resolved the four policy challenges less concealed. To check and extend this argument, I then turn to comparisons with Zimbabwe (Chapter 4) and Botswana (Chapter 5), two countries that have endured multiple droughts under changing political conditions for incumbents.

46

Domestic Politics and Drought Relief in Africa

Table 2.3: Decisions Facing Responding Incumbents: Policy Challenges and Alternatives General Relief Policy

Access to Relief

Interventions in the Market

Administration of Relief

i.

i.

i.

i.

Smoothing consumption (food aid)

Universal access

Complementary interventions

Standing relief institutions

ii.

ii.

ii.

ii.

Smoothing incomes (labor-based relief)

Administratively targeted

Exclusionary interventions

Ad hoc relief institutions

iii. Self-targeting iv. Discretion to administrative personnel

Notes 1

For a more comprehensive discussion of types of drought, see Glantz, “Drought and Economic Development in Sub-Saharan Africa,” pp. 37-58. 2 Clay, “Aid and Drought,” pp. 199-220. 3 Gautam, “Managing Drought in Sub-Saharan Africa.” 4 Clay, “Aid and Drought,” pp. 199-220, links the frequency of drought to the arid and semi-arid climate of Africa, while Bloom and Sachs, “Geography, Demography, and Economic Growth in Africa,” pp. 207-72; and Gallup et al., “Geography and Economic Development,” tie the effects of drought on food security to poor soil quality. 5 EM-DAT. www.emdat.be. 6 Bhalotra, The Drought of 1981-87 in Botswana. See also Republic of Botswana, Ministry of Agriculture, Botswana Agricultural Statistics, Yearly Reports, 1978-1990. 7 Drought incidence and rainfall data from Government of Zimbabwe, The Drought Relief and Recovery Programme, 1992/93. 8 Macodras et al., “Synoptic Features Associated with the Failure of the 1984 Long Rains in Kenya,” pp. 69-81. For an analysis of the variability of rainfall in Kenya leading up to the 1984 drought, see Farmer, “Rainfall Variability in Tropical Africa and Kenya,” pp. 82-93. 9 “Downpour After Prayers,” Daily Nation (June 4, 1984).

Drought in Africa and Programs for Relief

47

10 The historical roots of the ideas, attitudes, and ideologies informing famine relief programs are discussed in some detail by Hall-Matthews, “Historical Roots of Relief Paradigms,” pp. 216-30. 11 See USAID, Office of Foreign Disasters Assistance, OFDA Annual Report, FY 1985. 12 See Passmore, “Famine Relief,” pp. 92-94; Loveday, The History and Economics of Indian Famines; Mallory, China; Government of India, The Imperial Gazetteer of India, Vol. 3, pp. 475-99. 13 Downing et al., “Drought in Kenya,” p. 23. 14 Nuemann et al., “Impact of the 1984 Drought on Food Intake, Nutritional Status, and Household Response in Embu District,” pp. 211-30. 15 Republic of Botswana, Ministry of Agriculture, Botswana Agricultural Statistics, Yearly Reports, 1978-1990. 16 “Drought Forces Women Into Water Smuggling,” Herald (April 28, 1982). 17 “900 000 in Urgent Plea for Food Aid,” ibid. (April 27, 1982). 18 Southern African Development Community Secretariat, Measures to Address Food Security in the SADC Region; Gautam, “Managing Drought in Sub-Saharan Africa”; World Bank, Managing Agricultural Production Risk. 19 For a discussion of what drought/famine relief looks like when it relies on food donations from outside, see Gill, Famine and Foreigners; and de Waal, Famine Crimes. 20 Meilink, Food Consumption and Food Prices in Kenya; Bates, Markets and States in Tropical Africa; Bates and Lofchie, eds., Agricultural Development in Africa; de Wilde, “Price Incentives and African Agricultural Development,” pp. 46-66. Cf. Berry, No Condition is Permanent; and Bratton, “The Comrades and the Countryside,” pp. 174-202. 21 Schartzberg, Political Legitimacy in Middle Africa. 22 See Lofchie and Commins, “Food Deficits and Agricultural Policies in Tropical Africa,” pp. 1-25, for an explanation of this form. 23 Bates, Beyond the Miracle of the Market, also discusses the decision by the government of Kenya to continue to export maize in early 1984. 24 Maize import data from Downing et al., Coping With Drought in Kenya. Maize price data from UN Food and Agriculture Organization, FAOStat. 25 Meilink, Food Consumption and Food Prices in Kenya, p. 22. 26 See Benson and Clay, “The Impact of Drought on Sub-Saharan African Economies”; and Drèze and Sen, Hunger and Public Action, who note how common food aid is as a response to drought. 27 For a discussion of the administrative and logistical challenges associated with direct food delivery, see Drèze and Sen, Hunger and Public Action; Drèze, “Famine Prevention in Africa,” pp. 123-72; Leys, “Drought and Drought Relief in Southern Zimbabwe,” pp. 258-74; and Gooch and MacDonald, Evaluation of 1979/80 Drought Relief Programme. 28 Gooch and MacDonald, Evaluation of 1979/80 Drought Relief Programme. 29 See, for instance, von Braun et al., Famine in Africa. 30 Robinson, “Reconstructing Gabbra History and Chronology,” pp. 15168.

48

Domestic Politics and Drought Relief in Africa

31 Berry et al., Eastern Africa Country Profiles: Kenya, offer a more comprehensive description of livelihoods across the different agro-climatic regions in Kenya. 32 Anyango et al., “Drought Vulnerability in Central and Eastern Kenya,” pp. 169-210. 33 Nkanata, Workshop on Provision of Livestock Production Research Information for Use by Extension Workers in the Training and Visit Programme. 34 Devereux, “Goats Before Ploughs,” pp. 52-59, stresses that this type of asset loss has significant long-term structural effects as it impairs recovery. 35 The idea of coping strategy, understood as a temporary response to shortterm shocks to food access, has been central in drought and famine studies. See Devereux, “Goats Before Ploughs,” pp. 52-59; Davies, “Are Coping Strategies a Cop Out?,” pp. 60-72; Campbell, “Strategies for Coping with Severe Food Deficits in Rural Africa,” pp. 143-62; de Waal, Famine That Kills; Moris, “Indigenous versus Introduced Solutions to Food Stress in Africa”; Swift, “Why Are Poor People Vulnerable to Famine?,” pp. 8-15; Corbett, “Famine and Household Coping Strategies”; Longhurst, “Household Food Strategies in Response to Seasonality and Famine,” pp. 27-35. 36 Government of Kenya, Report of Enquiry into the Kenyan Meat Industry; Abbott, “Agricultural Marketing Boards in the Developing Countries,” pp. 70522. 37 Bates, Markets and States in Tropical Africa, offers a compelling analysis of the uses of marketing boards by post-independence governments in Africa. 38 Downing et al., “Drought in Kenya,” p. 8. 39 See “Cattle Die of Starvation at KMC Depot,” Daily Nation (December 27, 1983); “Cattle Death Toll Rises,” ibid. (December 28, 1983). These reports forced the government to order an investigation of the operations at KMC depots during the drought. See “KMC Told to Explain Cattle Deaths,” ibid. (December 29, 1983). 40 Benson, “Drought and the Zimbabwe Economy, 1980-1993,” pp. 241-73; Government of Zimbabwe, The Drought Relief and Recovery Programme 1992/93; Thompson, Drought Management Strategies in Southern Africa. 41 Benson, “Drought and the Zimbabwe Economy, 1980-1993,” p. 248. 42 As noted by Drèze and Sen, Hunger and Public Action, the idea of responding to threats of famine using cash support is not new. In the Bible, cash support is mentioned in Acts 11: 27-30. The use of labor-based relief in response to drought-induced threats of famine in developing countries is also articulated in the Indian Famine Codes. See Drèze, “Famine Prevention in India,” pp. 13-122; Government of India, The Imperial Gazetteer of India, Vol. 3, pp. 475-99; Madras Presidency, Famine Code; Government of India, Report of the Indian Famine Commission, 1901; Government of India, Report of the Indian Famine Commission, 1880. Labor-based relief programs have also been adopted in Botswana (see Hay et al., A Socio-Economic Assessment of Drought Relief in Botswana; Tabor, Drought Relief and Information Management; Amis, Financial Efficiencies in Drought Relief; Boers, Botswana, Drought Relief; Rockliffe-King, Drought, Agriculture and Rural Development) and Cape Verde (see Drèze, “Famine Prevention in Africa,” pp. 131-36; Economist Intelligence

Drought in Africa and Programs for Relief

49

Unit, Quarterly Economic Review of Angola, Guinea Bissau, Cape Verde, Sao Tome, Principe”; USAID, Cape Verde: Food for Development Program.) 43 Amis, Financial Efficiencies in Drought Relief; Boers, Botswana, Drought Relief; Rockliffe-King, Drought, Agriculture and Rural Development; Tabor, Drought Relief and Information Management. 44 See Drèze and Sen, Hunger and Public Action; Drèze and Sen, eds., The Political Economy of Hunger, Vol. 2; Subbarao, “Systemic Shocks and Social Protection”; von Braun et al., “Labour-Intensive Public Works for Food Security in Africa,” pp. 19-33; and Sen, Development as Freedom, for a more general defense of labor-based relief during droughts. 45 Stewart, “Adjustment with a Human Face,” p. 22. 46 Sen, Development as Freedom, pp. 177-78. 47 See Gill, Famine and Foreigners, p. 111, for a discussion of what happened to beneficiaries of Ethiopia’s cash relief programs when global food prices rose quickly in 2008. 48 Subbarao, “Systemic Shocks and Social Protection.” Stewart, “Adjustment with a Human Face,” pp. 22-23, makes similar claims about the potential for labor-based relief to be used to build beneficial infrastructure. 49 Terhal and Hirway, “Rural Public Works and Food Entitlement Protection,” pp. 373-403. 50 von Braun et al., Famine in Africa, pp. 123-88. 51 See the discussion of this in ch. 7 of Drèze and Sen, Hunger and Public Action, pp. 104-21. 52 Ibid., p. 104. 53 Gelbach and Pritchett, “More for the Poor Is Less for the Poor”; Devereux, “Social Pensions in Namibia and South Africa”; Moene and Wallerstein, “Targeting and Political Support for Welfare Spending,” pp. 3-24. 54 Gooch and MacDonald, Evaluation of 1979/80 Drought Relief Programme. 55 Drèze and Sen, Hunger and Public Action, p. 105. 56 MacDonald and Austin, A Human Drought Relief Programme for Botswana. See also Gooch and MacDonald, Evaluation of 1979/80 Drought Relief Programme. 57 Munro, Poverty and Social Safety Nets in Zimbabwe, 1990-99. 58 Drèze and Sen, eds., The Political Economy of Hunger, Vol. 2; Drèze and Sen, Hunger and Public Action; and Ravallion, Markets and Famines, offer useful discussions of the functioning of food markets during famine threats. 59 “Arrest Hoarders, Orders President,” Daily Nation (June 8, 1984). See also “Punish Hoarders, Moi Directs A-G – Leaders Told not to Spread Alarm Over Food,” ibid. (July 2, 1984); and “Hoarders to Lose Stock, says Moi,” ibid. (July 18, 1984). 60 “Do Not Charge Extra – Moi,” ibid. (July 16, 1984); and “Hoarders to Lose Stock, says Moi,” ibid. (July 18, 1984). 61 For a more comprehensive discussion of the difference between marketexcluding and market-complementary interventions, see Drèze and Sen, India, pp. 21-26. 62 A more comprehensive discussion of market responses to dearth is offered by Ravallion, Markets and Famines. 63 See, for instance, Drèze and Sen, Hunger and Public Action; Drèze and Sen, India; von Braun et al., Famine in Africa; Sen, Development as Freedom.

50

Domestic Politics and Drought Relief in Africa

64 See, among them, Hay, “The Concept of Food Supply System with Special Reference to the Management of Famine,” pp. 65-72; Jodha, “Famine and Famine Policies,” pp. 1609-23; Aykroyd, The Conquest of Famine; Blix et al., eds., Famine. In Botswana’s case, see Gooch and MacDonald, Evaluation of 1979/80 Drought Relief Programme; McGowan and Associates, A Study of Drought Relief and Contingency Measures Relating to the Livestock Sector of Botswana; Hinchey, ed., Proceedings of the Symposium on Drought in Botswana; Sandford, Dealing With Drought and Livestock in Botswana. 65 See Cohen and Lewis, “Role of Government in Combating Food Shortages,” pp. 269-96. 66 Ibid., p. 293.

3 Insecurity, Autocracy, and Drought Relief in Kenya

How do political vulnerability and the interests it generates affect drought relief policy-making? To answer the question, this theorybuilding chapter examines Kenya in the early 1980s, when President Daniel arap Moi’s chronic political vulnerability became acute as the food crisis threatened to undermine the legitimacy of his government. The chapter demonstrates that the interests produced by prevailing insecurity dictated the general relief policy Moi favored, they shaped the level of access to aid he established, they compelled him to intervene aggressively in the urban food market, and they determined the administrative institutions and personnel Moi trusted to implement his food aid program. Kenya in 1984 and 1985 illustrates how political insecurity determines the resolution of the four challenges facing incumbents during dry times. To check this argument, I then turn in subsequent chapters to comparisons with Zimbabwe and Botswana. The Origin of Autocracy and Insecurity in Kenya

On the eve of Kenya’s independence in 1963, Daniel Toroitich arap Moi would have appeared an unlikely candidate to eventually succeed Mzee Jomo Kenyatta as president of Kenya. Ahead of him stood prominent nationalists and powerful members of the Kikuyu ethnic elite. Moi’s selection to take Kenyatta’s place in 1978 and the weakness of his incumbency frame the political milieu in which Kenya’s drought relief response was conceived and implemented, highlighting the underlying considerations that led Moi to marshal his government to respond to the threat of famine. Daniel arap Moi was born in Baringo District of the Rift Valley in 1924. He was a Kalenjin, a member of a small Rift ethnic group. From a

51

52

Domestic Politics and Drought Relief in Africa

relatively non-political start as a teacher at the Government African School at Tambach, Moi rose to become headmaster of the Government African School at Kapsabet in 1954,1 which afforded him some political standing in his community. He became the Rift Valley District Council’s African Representative to the colonial assembly, the Legislative Council, in 1955, which in turn thrust him to the center of Kenya’s political stage at a time when the nationalist movement was starting to win significant concessions from the colonial administration.2 From this position, Moi was nominated to serve as one of the African delegates at the independence talks held at Lancaster House beginning in 1960. Competing visions of how an independent Kenya was to be organized and governed led to the formation of two broad nationalist parties: the Kenya African National Union (KANU) and the Kenya African Democratic Union (KADU).3 KANU was a coalition comprised of the two largest, better-educated, and wealthier ethnic groups, the Kikuyu and Luo, and it supported the creation of a strong, unitary, centralized state. In contrast, KADU was a wider coalition of smaller ethnic groups seeking a constitution that would grant ethno-regional constituencies significant powers and autonomy.4 For KADU, federalism, or Majimboism, appeared to be the best way to protect the interests of the regionally dispersed smaller ethnic groups and guard against Kikuyu and Luo hegemony.5 Initially, Moi was aligned with KANU. However, disquiet over the growing dominance of the KikuyuLuo coalition led him to defect to KADU, where he was elected treasurer. In the transitional elections of 1961, KANU won strong backing for its unitary state agenda with 67.4 percent of the vote to KADU’s 16.4 percent.6 General elections in May 1963, the first for independent Kenya, saw KANU consolidate its political advantage: the party won a total of 83 seats in parliament, while KADU managed to win only 32.7 Jomo Kenyatta, the KANU leader, a Kikuyu who became the country’s first prime minister, swiftly followed this resounding victory with a tenacious push to consolidate his position by creating a strong Executive Presidency, an office he assumed in 1964, and by establishing a voluntary one-party state.8 For Kenyatta, unity between KANU and KADU was the starting point of true Uhuru, or independence. He also sought to weaken the regional agenda by co-opting senior members of KADU into his cabinet. Helpless against the onslaught of constitutional amendments that empowered Kenyatta, battered by rhetoric that framed KADU as anti-Uhuru for maintaining its opposition in parliament, and undermined by the co-option of its leaders into Kenyatta’s cabinet, KADU dissolved in November 1964 and its members in parliament joined KANU.9

Insecurity, Autocracy, and Drought Relief in Kenya

53

Moi benefited from Kenyatta’s strategy of co-opting KADU leaders. During the transitional period, he served as Minister of Education and then Minister of Local Government. After the KANU/KADU merger in 1964, Kenyatta appointed Moi as Minister of Home Affairs.10 The literature on political developments in Kenya after 1963 suggests that Moi’s inclusion in Kenyatta’s cabinet was owed to neopatrimonial ethnic arithmetic (including the Kalenjin Moi in his Cabinet would help Kenyatta to co-opt the Rift Valley into a governing coalition rooted in the Kikuyu-dominated Central Province)11 and ideological affinity (both men were described as conservative, “pragmatic, reformist, procapitalist, [and] anti-communist.”12). The power struggle between Kenyatta and his left-leaning Luo vicepresident, Oginga Odinga, also helped Moi advance, with Kenyatta viewing him as an ally in that fight.13 As Kenyatta’s campaign to isolate and weaken his immediate rival intensified between 1964 and 1966, Odinga was constantly on the defensive against accusations that he wanted to turn Kenya in a pro-Soviet or -Chinese direction, or, worse, that he was using his links with these two communist countries to plan a coup against Kenyatta.14 Kenyatta proposed to re-organize KANU as another way of undermining Odinga’s national position. At the party’s conference at Limuru in 1966, Kenyatta sought three substantive changes to KANU and then tasked Moi and Tom Mboya, a Luo cofounder of KANU, with the responsibility of ensuring their passage: abolishing the post of party national vice-president and replacing it with eight vice-presidents to represent the seven provinces and Nairobi; freeing the party’s secretary general and treasurer from being full-time employees of the party; and establishing a closer working relationship between party members in parliament and local party organs.15 Moi and Mboya proved capable: delegates at Limuru overwhelmingly voted in favor of the three reforms.16 In turn, Moi was elected vice-president for the Rift Valley, and Mboya was returned as secretary general.17 These changes to KANU stripped the party of effective organizational leadership and subordinated it to parliamentary control. Moreover, by creating eight vice-presidents, each with narrow base of support, Kenyatta reasoned that no one could use the party to establish a national platform strong enough to challenge him. Having tamed and subordinated the party,18 Kenyatta exerted and consolidated his presidential power in part through formal institutions over which only he had control, such as the coercive Provincial Administration (PA). Kenya’s PA system contained eight provinces subdivided into 41 districts, which largely coincided with Kenya’s major ethnic groups. Each district was split into divisions, and each division

54

Domestic Politics and Drought Relief in Africa

into a number of “locations,” until reaching finally the village, or sublocation, level. In 1965, the PA system was removed from the Ministry of Home Affairs and placed under direct control of the Office of the President. The reforms also gave the president authority to appoint provincial commissioners, district commissioners, district officers, chiefs, and assistant chiefs, and in this way, the whole PA system came to represent the interests of the executive at the local level. 19 But Kenyatta also relied on informal institutions of his invention, such as the patronage-based system of harambee, or self-help development funds.20 As Kenyatta conceived it, harambee involved local-level societal fundraising for development projects. Each community was expected to establish a harambee group and use proceeds from fundraisers to construct a local social-service infrastructure.21 At harambee fundraisers, MPs and other political figures were expected to make considerable donations if they entertained hope of receiving Kenyatta’s support or winning (re)election. Kenyatta used his stature and influence over the largest and wealthiest faction within KANU (the Gikuyu, Embu, and Meru Association [GEMA]) to induce and direct harambee contributions to loyal politicians. Defections and disloyalty could be punished by withdrawing harambee contributions and, in so doing, limiting the chances of reelection. Kenyatta had built, and kept obedient, his governing coalition using this system. After the vote at Limuru in 1966, Odinga resigned from KANU and as national vice-president and formed the Kenya People’s Union (KPU) with 30 or so other left-leaning MPs and trade union leaders who also resigned from KANU.22 Odinga’s formation of a new opposition party angered Kenyatta, who wished to maintain the voluntary one-party state created by the dissolution of KADU in 1964. Describing the KANU dissidents as “puny and misguided men,” Kenyatta called a special session of parliament to pass a new law requiring members who resign from the party under which they were elected to seek a new mandate via a by-election. Forced to stand for re-election by this retroactive constitutional amendment, the KPU challenge faltered; twenty of its twentynine candidates lost to KANU in by-elections held in May 1966.23 To replace Odinga as national vice-president, Tom Mboya appeared a natural candidate. First, he was a conservative capitalist like Kenyatta. Second, Mboya was now the de facto leader of the Luo community and in a purely neopatrimonial calculus, his elevation would enable Kenyatta to maintain the Kikuyu/Luo governing coalition and forestall any backlash from Odinga’s ouster. Finally, Tom Mboya had proved to be a loyal and a skillful ally at Limuru in presenting the party reforms on Kenyatta’s behalf. However, Mboya’s strong domestic political base

Insecurity, Autocracy, and Drought Relief in Kenya

55

proved to be a major handicap for a president keen on finding a vicepresident with little independent standing.24 Kenyatta settled on Joseph Murumbi instead, one of the weaker members of his cabinet. Murumbi had spent the decade before independence in London and had not played the critical role in the internal nationalist movement that had provided people like Odinga and Mboya the opportunity to build popular bases of support. Murumbi also had no obvious ethnic base—he was Maasai and Goan, groups that made up less than 2 percent of Kenya’s population. Yet for all of Murumbi’s attractiveness from Kenyatta’s point of view, their political partnership was short-lived. In September of 1966, just six months after becoming Kenya’s second vice-president, Murumbi abruptly resigned his position and retired from politics. In the coming months, Kenyatta resumed his search for a nationally weak and therefore dependent and loyal non-Kikuyu as his vicepresident. Eventually, this led him to Moi. At the time, Moi had no real political support outside of the Kalenjin-dominated Rift Valley. He too had proved his loyalty to Kenyatta in the backroom politics that won the vote at the Limuru Conference. Thus, on January 5, 1967, four months after Murumbi’s resignation, Kenyatta appointed Daniel arap Moi as vice-president of Kenya. Moi held this position until he succeeded Kenyatta in August 1978. As Kenyatta had intended, Moi’s survival in the vice-presidency for over a decade relied on the president’s constant patronage, protection, and support. Moi’s enduring weakness was apparent as late as 1976, when a coalition of Kikuyu MPs attempted to amend the constitution so that “a triumvirate of the Speaker of the House, the Chief Justice, and the Head of the Civil Service would organize presidential elections” in the event of Kenyatta’s death.25 This amendment would change the succession rules established in the 1968 constitutional amendment (No. 10, Act 48) that granted the national vice-president an institutional advantage over other aspirants to the presidency: as vice-president, Moi would automatically become interim president upon Kenyatta’s death, with the requirement that he organize a presidential by-election within 90 days. When amendment No. 10 passed in 1968, Kenyatta’s health was not in question and the Kikuyu elites considered Moi “as ‘a passing cloud’, someone who, lacking true leadership qualities, would eventually drift into the political distance.”26 However, by the mid-1970s, Kenyatta’s deteriorating health made the vice-presidency the most valuable position in the intensifying succession battles.27 Although Moi remained a nationally weak vice-president with little support outside of the Rift Valley, the Kikuyu elites feared that his institutional advantage would be

56

Domestic Politics and Drought Relief in Africa

hard to overcome. Paul Ngei, a cabinet minister and member of the ‘change-the-constitution’ movement, openly acknowledged this concern, saying that during the three months the vice-president served as president, “a lot can happen. If you give me that period I can really teach you a lesson and I can assure you it would not be a pleasant one.”28 Moi could not fight off this constitutional challenge alone. He needed support from two senior Kikuyu politicians, Charles Njonjo, the attorney general, and Mwai Kibaki, the finance minister, with Njonjo warning that “it [was] a criminal offense for any person to encompass, imagine, devise, or intend the death or disposition of the President.”29 Njonjo and Kibaki’s backing for Moi helped to sway the debate, but it was Kenyatta’s intervention in support of Moi—and his public endorsement of Njonjo’s statement—that finally halted discussion of the constitutional amendment in parliament. Moi survived the change-theconstitution movement, but not out of the strength of his own political coalition. Throup suggests there was a split among Kikuyu elites on the question of Moi’s possible ascension.30 In his assessment, the faction that produced the change-the-constitution movement sought to keep the presidency in Kikuyu hands. Kenyatta was concerned by the transparent power-grab by the clique around him, commenting: “these people are mistaken. You do not show a cow the thong you will use for tying its neck and killing.”31 Others feared that a Kikuyu monopoly of the presidency would intensify ethnic resentment and conflict to the detriment of the group’s interests. Thus, Throup argues, Kibaki and Njonjo supported Moi because they thought that Kikuyu hegemony would be more secure if it were, for the short term, concealed in a weak and dependent non-Kikuyu president. What Throup ignores is Njonjo and Kibaki’s individual political motivation in supporting Moi at this stage in the Kenyatta succession battles. By blocking constitutional change, they could blunt the political rise of Kenyatta’s brother-in-law, Mbiya Kionage; his nephew, Dr. Njoroge Mungai; Kilika Kimani; Paul Ngei; and other members of the change-the-constitution movement. Once the clique around Kenyatta had been prevented from capturing the presidency, they saw their own opportunity in a fight against a weak President Moi. After the fiasco of the change-the-constitution movement, Kenyatta called in April 1977 for a KANU party congress, the first since 1966. On the agenda was the election of a new national executive and a proposal to bring back the position of national party vice-president, which offered an ailing Kenyatta the opportunity to resolve escalating succession battles. But the conference also provided Moi’s detractors an oppor-

Insecurity, Autocracy, and Drought Relief in Kenya

57

tunity to prevent him from succeeding Kenyatta. If they could defeat him here in the race for KANU’s vice-presidency, they could force Kenyatta to replace him as the national vice-president thus preventing his automatic ascencion to the presidency. To challenge Moi, the Kikuyu coalition chose fellow Kalenjin and former KADU member Taita arap Towett, the Minister of Education. The choice of Towett was intended to give their challenge cover from the anti-Kikuyu sentiments that had emerged during the change-the-constitution push. However, at the last minute, the party conference was cancelled and the vote for a new party vice-president was never held.32 In the early hours of August 22, 1978, President Kenyatta died of a heart attack at State House in Mombasa and at 3:00 P.M., Moi was sworn in as interim president.33 Paul Ngei’s earlier fears about the advantages of automatic succession were now realized. Moi used the trappings, powers, and resources of the Presidential Office to ensure that he stood unopposed for election on October 10, 1978. But Moi’s position remained vulnerable. In what follows, I show that he inherited weak institutions, had fewer resources at his disposal for securing his position through the distribution of patronage, and encountered economic crises in the early 1980s that produced instability in the urban areas, all of which limited Moi’s ability to establish a governing coalition or legitimate his rule and so deepened his political vulnerability. In increasingly desperate attempts to retain power, Moi’s heavy-handed responses to insecurity culminated in the adoption of a de jure one-party state in 1982. But security remained elusive, and instead of order, there was an attempted coup in late 1982 as the country realized the transition to an autocratic Kenya. The coup attempt forced Moi to look for other instruments and strategies to strengthen his position, and his prevailing anxieties about maintaining and securing tenure dominated his drought relief policy-making in 1984 and 1985. Insecurity and Autocracy in Kenya, 1978 to 1984

Following independence, Kenyatta had used his “charisma and reputation as the ‘Grand Old Man’ of African nationalism”34 to build political order and consolidate his position at the top, relying on institutional powers and the distribution of patronage to maintain control of the state and keep elites happy and loyal. The manner in which Kenyatta secured his own incumbency had three constraining effects on his successor. First, a weak party left Moi with no obvious instrument for political mobilization and recruitment. Moi had the misfortune of inheriting a political party that, after years of being systematically gutted,

58

Domestic Politics and Drought Relief in Africa

had weak representational and aggregative institutions and lacked a decision-making hierarchy and any real programmatic content to draw widespread popular support. Widner rightly contends that the KANU that Moi inherited was little more than a “loosely organized ‘debating society’” essentially composed of several competing corporate groups tolerated and ultimately kept together by Kenyatta’s stature.35 KANU was therefore of little utility in Moi’s attempts at consolidation and legitimation. Second, Kenyatta’s reliance on the coercive powers of the PA (police, internal security, and the registration and sanctioning of all public gatherings) to control political activity limited its utility as a legitimacy-generating institution for Moi. Finally, as Widner stresses, Moi was also not able to deploy the institution of harambee, the informal extra-party system Kenyatta leveraged to maintain political control and shore up his own incumbency. Raising sufficient self-help development funds was too closely fused to Kenyatta’s standing and the support accorded him by wealthy, and especially Kikuyu, elites. As a second-generation leader, Moi did not have the stature to induce contributions, he did not yet control resources, and, most importantly, he did not belong to the largest and wealthiest faction, GEMA. 36 Other structural factors also constrained Moi. He found it difficult to rely on the distribution of patronage to build a governing coalition and consolidate his incumbency. Kenyatta had benefited from the ability to use access to government jobs and to the expanding coffee and tea sector to create intricate patron-client networks that undergirded his 15 years in power. For instance, the Brazilian frost of 1975 benefitted Kenya’s coffee sector, by damaging Brazil’s coffee crop and bringing an increase in world prices. This produced a minor coffee boom in Kenya, with some “farmers, including smallholders, [said to have] uprooted their food crops and planted coffee instead.”37 Access to the fruits of a booming coffee export market enabled an aging Kenyatta to forestall growing discontent. While Kenyatta had benefited from the post-colonial Kenyanization of state positions and the state’s expansion in its size and responsibilities during his administration, Moi’s position in the late 1970s and early 1980s was more constrained. By late 1978, when he assumed the presidency, land was scarce and public sector jobs in short supply. Moi inherited a country in which there were frequent ethnic clashes over land, such as between the Nandi and Luhya near Kapsabet in the Rift Valley Province,38 and between squatters and Arabs in the Coast Province.

Insecurity, Autocracy, and Drought Relief in Kenya

59

Moi turned to populism in his early attempts to secure his position. He promised to fight corruption among bureaucrats and the political elite. He suspended “the allocation of residential and commercial plots on grounds that some big men were grabbing everything.”39 The president’s fight against corruption was then ostensibly extended to bureaucracy when he appointed a board in early 1979 to review and evaluate the operations of over 60 statutory boards and parastatals and six cooperatives.40 Moi declared that he would work to transfer tangible economic and political benefits to ordinary Kenyans. The new president announced that primary school fees would be abolished and that his government would provide free milk to schoolchildren.41 Finally, on December 12, 1978, the date of Kenya’s independence, Moi released Kenyatta’s political detainees and “promised that his government would only use detention without trial as a last resort.”42 The president also vowed to hold the long-overdue council elections in an attempt to gain popular legitimacy by reviving democracy, at least at the local level. In the short term, these decisions won Moi some internal support and political developments in 1979 seemed to go in his favor. He used the campaign against corruption to purge the government, the party, the police, and army of individuals thought to be disloyal.43 Even GEMA, the powerful organization that represented Kikuyu interests, was put on notice by the new president when its whole board of directors was summoned to court on charges that the association had failed to post annual returns and accounts.44 Although the board got away with only a K.Sh. 205,000 (roughly US$61,600 in 2005 terms) fine for the violation, Moi appeared strong enough to challenge its members. As Kenyatta had done in previous general elections, Moi used his control of the Provincial Administration to secure victories for his loyalists in the 1979 election. Kenya’s system of government and elections gave the PA, through the provincial commissioners and the district commissioners beneath them, powers to sanction and control political gatherings at the local level.45 Moreover, district commissioners also ran elections and acted as returning officers. Throup also stresses that Moi deployed troops from the paramilitary General Service Unit (GSU) to compel voters to the polls and ensure they voted for the ‘correct’ candidate.46 When a minor drought in 1979 threatened urban food security, Moi responded with “Operation Maize,” a credit-claiming operation in which the commander-in-chief ordered the Army and the National Youth Service to transport maize from growing districts to the cities.47 Operation Maize was then repeated again in November of 1980, with more explicit attempts by Moi to take credit for his efforts in ensuring

60

Domestic Politics and Drought Relief in Africa

that the people were fed. Speaking to local reporters, the Nairobi Area Provincial Administrator made the purpose plain: “We in the Government want to show our people we are capable and willing to feed them under the wise leadership of President Moi.”48 With this battery of interventions, the results of the 1979 general elections went heavily in Moi’s favor. First, 51 percent of Kenya’s incumbent MPs lost their bids for reelection and were replaced by candidates that Moi backed. Second, Moi’s manipulation of the electoral process also ensured that his two rivals among the Kalenjin, J.M. Seroney and arap Towett (the politician Kikuyu elites had picked to run against Moi at the cancelled party congress of 1977), lost their bids for election to parliament. Finally, and just as significantly, seven ministers, including three members of the Kikuyu elite opposed to Moi, lost their seats.49 Moi was thus able to drop many politicians associated with the Kenyatta years from his reformed and expanded cabinet of 1979.50 The president seemed to revel in the outcome of the election. In his speech after the results were in, Moi thanked Kenyans for being politically mature, for understanding their interests and knowing those people most able to assist them in achieving their interests, and for being “determined to uphold the Nyayo philosophy of peace, love, and unity.”51 However, Moi found it hard to consolidate these early gains. Progressively, the gap between Moi’s populist pronouncements and his ability to deliver change on the ground posed new challenges. Poor planning was evident in Moi’s decrees on free education and the provision of free milk to school children. On the ground, weak fiscal health by late 1978 meant that Moi’s government could not follow through on the promise of free education to Kenyans. Moi’s shortsighted free-milk policy led to shortages in the market which, in turn, created huge incentives for corruption among the school headmasters and government officials who controlled the distribution of milk to schools.52 This burgeoning informal market for fresh milk weakened Moi’s fight against corruption. Furthermore, Moi came to power at a time when the price of Kenya’s main import, oil, was doubling as OPEC attempted to maintain profits in the face of a global economic recession. By 1980, the Kenyan government was spending close to half of its foreign currency earnings on crude oil imports.53 Just a few months after assuming the presidency, Moi tried to deal with the growing economic crisis by calling on the public to ‘tighten their belts’ and prepare for increases in the domestic price of fuel.54 To limit the use of oil for electricity generation, the government introduced power rationing, which had negative effects on

Insecurity, Autocracy, and Drought Relief in Kenya

61

the economy, as factories across Kenya were forced to open on a parttime basis or to lay off workers until power supply improved.55 The world recession of the early 1980s also cut demand for Kenya’s primary export commodities, coffee and tea, further deepening the country’s trade deficit. At the end of 1978, that deficit was about US$176 million (in 2005 terms); within two years it had nearly tripled to over US$522 million (in 2005 terms).56 At the behest of the IMF and the World Bank, the government devalued the currency twice over the course of 1981, moves intended to correct the overvalued Kenya’s currency and make it more expensive for wealthy Kenyans to transfer their assets into foreign currency or import consumer goods from abroad. The Central Bank of Kenya then imposed strict import restrictions on all Kenyans, but this measure only created a ‘market’ for import licenses. Kenya’s economic crisis also limited Moi’s opportunities to use appointments to the bureaucracy to bolster his position. In 1981, his government was forced to suspend recruitment and promotion into the civil service. This in turn exacerbated a situation in which only about 7 percent of the nearly 300,000 school-leavers each year could find jobs.57 Increasing conflict over access to land and declining opportunities for public employment limited Moi’s positive sum use of patronage to construct support and weaken his challengers. To consolidate his grip on the bureaucracy and the army, Moi resorted to replacing established Kikuyu bureaucrats and officers with co-ethnic Kalenjins or members of smaller loyal ethnic groups, a Kalenjiziation of the state that generated elite and officer resentment and in the short term only weakened his position further. Persistent economic hardships eventually matured into social and political unrest. The early 1980s were characterized by growing ethnic conflict over access to land in the rural areas and industrial action for better wages in the cities. For instance, clashes between landless peasants and the landed elites in the Central Province started with squatters moving onto estates and then escalated into open conflicts that destroyed cash crops.58 In Western Kenya, restive rural populations burned sugar plantations, while armed bandits were terrorizing nomads and villagers in the Eastern Province.59 The situation in the urban areas was equally contentious, with high school students who faced terrible job prospects staging strikes, followed quickly by demonstrations at the universities. In May 1981, government doctors went on strike deman-ding pay increases, better conditions of service, and the lifting of rules that barred them from private practice. Within days of the doctors’ strike, bank workers threatened similar action.

62

Domestic Politics and Drought Relief in Africa

Moi blamed the troubles at the universities on “Marxist lecturers” with support from jealous left-wing agitators outside of Kenya.60 He closed the universities and required all male students to report to their chiefs on a weekly basis, effecting a parole-like system that forced male students out of the city and back to the rural areas, where they could be monitored and controlled. Moi then proscribed the University Academic Staff Union and the largest union in the country, the Kenya Civil Servants Union, for failing to follow the true nyayo line. As Widner observes, Moi developed nyayo (or footsteps) as the guiding ideology of his administration.61 Nyayo was intended to enable Moi to draw on Kenyatta’s charisma and reputation to legitimate his position by claiming that his administration followed in the founding president’s footsteps. Nyayo also insinuated that the population ought to follow in Moi’s footsteps. But the efficacy of nyayo in transmitting Kenyatta’s stature to Moi was limited. The charisma and decolonization credentials on which founding presidents could draw to fashion stable political incumbency were not easily transferred.62 Moi’s footsteps after Kenyatta were insecure and his hold on power was tenuous and increasingly reliant on coercion. Thus, striking doctors were arrested and thrown in jail, while bank workers were threatened with dismissal. Moi’s heavy-handed steps did little to address the economic crisis or to build his legitimacy; rather the president intensified social conflict and spread economic agitation, forcing challenges to his administration to become subversive. Moi’s eroding legitimacy led to the formation of an underground opposition group, the December Twelve Movement.63 Rhetorically Marxist, this movement launched its own newspaper, Pambana, or struggle. In May1982, Pambana declared: Kenyans have been massively betrayed. The revolution we launched with blood has been arrested and derailed. Today, more than twenty-two years after KANU was formed and almost twenty years after a fake independence was negotiated, the broad masses of Kenya are materially and politically worse off than ever before. The criminally-corrupt ruling clique, sanctioned by KANU, has isolated itself from the concerns of our daily life and has committed crimes, among many others, more brutal than any British colonialism ever did: they have silenced all opposition and deprived us, forcibly and otherwise, of the very right to participate in Kenya’s national affairs. The sacred rights of expression and association have been cast aside.64

Insecurity, Autocracy, and Drought Relief in Kenya

63

The movement’s grievances also included Kenya’s dependence and the restrictions on free political activity, with most of the anger directed at Moi’s coercive Provincial Administration: “It is also our natural right to express ourselves, to disseminate ideas and to associate. These rights do not any longer have to be begged from PCs [provincial commissioners] and DCs [district commissioners].”65 The movement called on Kenyans to be prepared for the class struggle required to recapture the promises of independence. As challenges to Moi became directly political and subversive, the president’s responses became more hard-line, autocratic, and repressive. In his Madaraka Day address on June 1, 1982, he declared that “My back is to the wall” and warned: “I want to make it clear that we shall not allow a few individuals who regard themselves as revolutionaries promoting foreign ideologies to disrupt our education and training programme.”66 After this speech, detentions without trial resumed, and in the coming months, scores of university lecturers were arrested for possessing seditious material, plotting to acquire arms to distribute to students, or being Marxists bent on bringing about anarchy and totalitarianism.67 An uncompromising Moi threatened that the “detention of anti-government persons [would] continue until disgruntled elements and unpatriotic people [were] rooted out of Kenya.”68 In addition to growing intolerance of citizen expressions of discontent, Moi also closed access to formal and legitimate political expression. As political conditions deteriorated further in Kenya, Odinga, his son Raila Odinga, and others took steps to formally register a new political party, the Kenya Africa Socialist Alliance, to challenge Moi and KANU. News of this plan produced an overreaction by the government that signaled the anxieties Moi had about his incumbency. On June 9, 1982, as the chaos surrounding the detention of university lecturers unfolded, the government rushed through a bill amending Section 2A of the constitution, which permitted the formation of opposition parties. To get the votes he needed, Moi threatened members of parliament with detention. After just 40 minutes of debate in parliament, Kenya became a de jure one-party state.69 Moi’s inability to resolve his vulnerability resulted in autocracy. The president’s legitimacy had deteriorated to such an extent that Kenyans openly talked of the possibility of a military coup. Most notably, Goerge Githii, the editor-in-chief of The Standard newspaper, published an editorial arguing that the “government should take steps to put an end to the prevailing fear and insecurity in the body politic,”70 and warning that failure to free political detainees, abolish detentions

64

Domestic Politics and Drought Relief in Africa

without trial, and revive Kenya’s dying democracy would lead to violence. Githii’s editorial elicited an aggressive response from incumbents in Kenya. He lost his job and the owners of The Standard were forced to print an apology. At an emergency session of parliament, MPs called for Githii’s immediate detention, and Vice President Mwai Kibaki warned that “no individual in the country has the right to question the legitimacy of the Government, and should any person do so, he shall face the consequences.”71 Currie and Ray rightly conclude that Githii’s editorial revealed that conditions in Kenya were “in many respects…a classic situation of instability.”72 Social conflict was prevalent, economic crisis was deepening, political repression was growing, and the illegitimacy of Moi’s government was openly expressed. In his work on military intervention in Brazil, Stepan suggested that coups are more likely during periods when significant civilian elites implicitly sanction them by articulating justifications for such action.73 Just eleven days after Githii’s editorial was published, gunfire broke the unease across Nairobi. Junior and noncommissioned officers from the Kenyan Air Force attempted to wrest power from Moi. Airmen at the three main bases in Kenya mutinied around 2:00 A.M. on August 1 and “rebel soldiers took the GPO, the International Airport, Wilson Airport and by 5:00 A.M. they had seized the Central Bank of Kenya, the Voice of Kenya radio station, and other communications installations.”74 At approximately 6:00 A.M. an announcer on the Voice of Kenya read a statement informing the public that Moi’s government had been overthrown. Claiming to represent the People’s Redemption Council (PRC), the announcer accused the government of being corrupt, dictatorial, and having failed to manage the economy. I announce to you today the overthrow of the corrupt regime of Daniel Toroitich arap Moi by the patriotic forces of our country. As I speak to you now, our country is fully and firmly under the control of our armed forces. Every care has been taken to make the revolution as bloodless as possible. Fellow Kenyans, over the past few years this country had been heading from an open to a closed inhuman and dictatorial society. … Over the last six months we have witnessed with disgust the imposition of a de jure one-party state system without the people’s consent, arbitrary arrest and detention of innocent citizens, censorship of the press, intimidation of individuals, and general violation of fundamental human rights. …

Insecurity, Autocracy, and Drought Relief in Kenya

65

This revolution is entirely an internal affair and our friends have nothing to fear. … As for now, the Constitution has been suspended and a national liberation council has been set up to preside over the affairs of the government and state. All the detainees and political prisoners are released forthwith, with immediate effect. Long live Kenya. Long live the People’s Redemption Council!75

PRC rebels then went to the University of Nairobi residence halls adjacent to the Voice of Kenya station and called on “students from the University of Nairobi and Kenyatta University College … [to] take to the streets and demonstrate their support for the coup.”76 After months of state action against college lecturers and students, “hundreds of students and urban poor…rioted, looted, and took arms from the insurgents.”77 The extent of public support for the coup contradicted Moi’s insistence that the rebellion was only the result of the actions of a “few hooligans and misguided youth.”78 Like the social and economic crises and political agitation, “the attempted coup served to point out the shortcomings in the personal status of the new President.”79 Describing the coup as simply a “disturbance by air force hooligans and their age mates at the University which had lasted less than an hour,”80 Moi deployed the 7th Battalion of the army under General Mohamed. At the same time, units at the Kahawa Garrison were sent to restore order at the Eastleigh Air Force base outside Nairobi. In addition to deploying loyal forces from the army and GSU, the president also imposed a dusk-to-dawn curfew on Nairobi and Nanyuki,81 which would last until September 4. Under the curfew, pro-Moi troops were given unlimited powers of search, seizure, arrest, and the use of deadly force. With these soldiers on the ground, the Moi government, four and a half hours after the PRC made its initial announcement, insisted that “everything was under control and the legally constituted Nyayo government was in power.”82 However, it took another eleven days of fighting before the last of the rebels were defeated. When forces loyal to Moi’s government finally subdued rebels in and around Nairobi and Nanyuki, as many as 1,500 people were dead,83 the police had over 1,000 looters in custody, and property estimated at close to US$223 million (in 2005 terms) had been destroyed.84 Following the coup attempt, there was a notable increase in political repression. Kenya’s descent into autocracy accelerated. The army was ordered to surround and seal Kibera slum to ensure that the throngs of

66

Domestic Politics and Drought Relief in Africa

urban residents who had come out in support of the coup would remain under constant surveillance. Hundreds of people were rounded up, placed in detention centers, and charged with supporting or celebrating the coup.85 Students were arrested and the University of Nairobi and Kenyatta University College were closed indefinitely; they re-opened the following year, after their administrative structure had been changed to give the president more direct control over their operations.86 As part of Moi’s hard-line position, Police Commissioner Ben Githi, Peter Mbuthia, head of the General Services Unit, and Major General P.N. Kariuki, commander of the Kenyan Air Force, were dismissed and later court-martialed and convicted of having taken “insufficient action to prevent the coup.”87 Most airmen were detained and close to 600 of them were court-martialed between October and November 1982.88 Twelve leaders of the coup were executed, and Moi dissolved the entire air force and put General Mohamed in charge of reconstituting it. Beyond repression, Moi tried to buy back the loyalty of the military. Shortly after the coup attempt, the president announced a 15-percent pay increase for the general forces and a 30-percent increase for officers, and started to appoint military officers to head state agencies.89 Among ordinary Kenyans, Moi sought to rebuild support and legitimacy through a retooled populist strategy.90 On August 20, he announced a 15-percent wage increase for rural and urban workers. The president then embarked on a wave of nationwide rallies at which he appealed directly to Kenyans for support, consistently telling them that the problem was not with ordinary people, but with the big men in the government.91 In exchange for loyalty, he promised a government more open to the needs of the people and less corrupt. In this political rebirth, the president deliberately tried to identify himself with the people against the elites in the civil service and government—the people who had not warned of the coup plot. His self-appointed attack dog, Minister for Local Government Stanley Oloitipitip, joined the fray by suggesting that big men had been sabotaging development projects in the country.92 In addition to bypassing the old guard with direct appeals to Kenyans, Moi’s populist turn included a recruitment drive for the moribund KANU.93 The president set a target of 4 million new members, hoping to build a strong party to support his incumbency. However, his attempt at grassroots recruitment and mobilization failed. The drive, which was concluded in April 1983, a few months before the start of drought, resulted in 1.2 million people (22 percent of Kenyans over 18 years) registering as members of KANU in a system where it was the only legal political party. Registration patterns also revealed great ethnoregional variation in the public’s support for Moi. For instance, in Luo-

Insecurity, Autocracy, and Drought Relief in Kenya

67

dominated Nyanza Province and among the Luhya in Western Province, registration rates were low: 10.2 percent and 19 percent respectively. They were much higher in Moi’s home district of Nandi (89.27 percent), but rather low for the entire Rift Valley, at about 35 percent.94 Despite his attempt at populism, the president had not been able to extend his support base beyond his home district. The final step of Moi’s “new contract” with Kenya was an early general election in September 1983, just as the drought was starting. Citing the unusual political developments, Moi dissolved parliament a year early and framed the elections as an opportunity to have “the people decide who were true nyayo followers fit to serve the president.”95 The early poll also permitted Moi to take advantage of post-coup restrictions to ensure the election of loyalists. Although Moi and Vice President Kabaki stood unopposed and 40 percent of incumbent MPs lost their seats to candidates Moi identified as true nyayo followers, the overall turnout in the 1983 elections indicated that the president still had not established himself among ordinary Kenyans. Of the close to 7.3 million Kenyans registered to vote in 1983, only 45.9 percent turned out at the polls.96 In sum, Moi’s early experience in office between 1978 and 1983 offers a clear picture of the conditions characteristic of vulnerable political incumbency. Moi inherited weak legitimacy-generating institutions from Kenyatta and had no governing coalition of his own. He was both institutionally and personally weak when he became president in 1978, and although he attempted to strengthen his institutional position by resurrecting KANU, sought personal security by transferring political and economic power from the Kikuyu to his co-ethnic Kalenjin core or to other loyal smaller ethnic groups, and embarked on a populist agenda, no strategy produced secure incumbency. The spiraling instability of Moi’s government, which led to the coup attempt and its aftermath of populist ploys, serve as the political backdrop to the drought that gripped Kenya a month after the 1983 elections. I show below that insecurity led Moi to respond to dearth with food aid rather than laborbased relief; to grant relief administrators the discretion to determine access, rather than adopt universal or administrative standards for qualification for food aid; to combine exclusionary and complementary interventions in the market; and to rely on the Provincial Administration to implement relief at the local level.

68

Domestic Politics and Drought Relief in Africa

How Vulnerability Shaped an Autocrat’s Response to Drought

Several months after the failed coup attempt, President Moi faced a different type of crisis – a major drought. Kenya typically has two rainy seasons: the long rains and the short rains. In normal years, total annual rainfall from these two seasons ranges from about 100 mm in the semiarid Eastern Plateau to over 2,000 mm in the Mount Kenya region. Twothirds of Kenya’s staple crop, maize, grows during the long rains. Though lighter, the second rains enable about a third of the national maize crop to grow. If the long rains fail, farmers can usually count on rain to yield some measure of maize in the short season. However, in the 1983-to-1984 agricultural season, neither period saw enough rainfall for adequate agricultural production. As the drought progressed into late May of 1984, usually the peak of the long rains, the country was consumed by the implications of parchedness. An editorial in the local newspaper, the Daily Nation, considered whether the continued lack of rain was a sign of changing climate patterns. The Minister of Agriculture and Livestock Development warned that the lack of rain was bound to affect Kenya’s harvest, while the Governor of the Central Bank of Kenya expressed his concern about the impact of the drought on the nation’s employment levels.97 President Moi was among the many government officials who confirmed the severity of drought. Adopting what I will show evolved into a clear strategy of using the drought and the high profile the food crisis afforded him to try to legitimate his incumbency, Moi declared Sunday, June 3, 1984, a National Day of Prayer for Rain.98 On this day, Moi joined ordinary Kenyans at a church in Nakuru to pray for rain. Although the day of prayer was declared a success—a downpour was reported in parts of Kenya—the drought that continued to grip Kenya until March 1985 was the worst in 100 years.99 The drought exposed the tenuousness of provisioning in Kenya. Many of the farmers who once fed Kenya could now not even feed themselves. Kenyan cities, the focal point of most political and commercial activity, also risked running out of food. The 38 percent fall in maize production and 43 percent decline in wheat output in 1984 resulted in a cereal deficit of 1.1 million Mt, which could only be addressed by the government. Government, however, was unprepared. Moi’s Minister of Agriculture insisted that there was enough food in the country.100 Despite these attempts to maintain calm, backbench MPs were unconvinced. Their unease turned into public questioning of Moi’s management of the food stocks. In a direct challenge, agitated MPs demanded the Deputy

Insecurity, Autocracy, and Drought Relief in Kenya

69

Minister of Agriculture and Livestock Development give an account of what had happened to the country’s stockpile of beans.101 The government of Kenya, through its marketing board, the NCPB, typically sold crop surpluses, marking them for export. Despite the failure of rains in 1983, the government did not suspend its aggressive export policy, which was not limited to maize. Beans too brought a good price on the world market. But the failure of the government to maintain an adequate strategic reserve added to prevailing negative assessments of Moi’s administration and further undermined its legitimacy. Moi sent the more senior Minister of State, Justus ole Tipis, to respond to the panicked parliament, but his explanation, that the government had sold one million bags of the 1983 bean harvest rather than risk that it would it rot in storage,102 failed to calm parliament’s nerves, and so Moi resorted to bullying MPs by characterizing them as alarmists.103 It was now quite clear that neither farmers nor the government could count on saved surpluses to make up the shortfall. The situation intensified, because, as the drought destroyed crops, both peasants and the government turned to the market to buy food. This resulted in above-normal demand for food, hoarding, and price increases in the market.104 In early June 1984, Moi directed the police to arrest anyone caught hoarding food, and when that had little effect, he threatened hoarders with losing their stockpiles if caught.105 I will address the specifics of the government’s subsequent drought relief policy below, but it is important to note here that government intervention in the market proved to be less than effective at arresting the negative externalities of individual and market responses to droughtinduced food scarcity. Market controls did, however, prove to be potent instruments through which President Moi could wrest control of provisioning away from key constituencies in the country. In sum, a disastrous harvest from the poor long rains and uncertain second rains produced a desperate situation in Kenya in 1984. Less than a year after the coup attempt, a still-insecure Moi faced a different type of political crisis, with MPs openly questioning the government’s management of a major drought. What effect did Moi’s failure to establish a governing coalition, combined with restiveness in the urban areas, social conflict in the rural areas, and a recent coup attempt, have on the president’s resolution of the four policy challenges set out in Chapter 2? I show below that by placing himself at the center of his government’s efforts, Moi could take full credit for any success. Thus, his response to dearth reveals the effects of political insecurity on relief programs and underscores the ways in which incumbents attempt to address prevailing vulnerabilities by their answers to drought relief policy challenges.

70

Domestic Politics and Drought Relief in Africa

Taking the Lead: Provisioning for a Nation

In line with Bates’s observation that Moi “consolidated his power in the course of the political struggles precipitated by dearth,”106 I show that political insecurity induced Moi to take the lead in the government’s response to drought. In April 1984, right in the middle of the poor long rains, President Moi pre-empted the existing early-warning system by requesting papers analyzing the potential impact of the drought from three government ministries: the Ministries of Agriculture and of Finance and Planning, and the Office of the President.107 The normal famine warning system in Kenya drew on meteorological data but relied most heavily on agents of the Provincial Administration. Declarations of drought and requests for aid went up the administrative hierarchy from sub-chiefs to chiefs to district commissioners to provincial commissioners. It was then the responsibility of provincial commissioners to make a request to the central government for famine relief funds for their provinces.108 This system was time-consuming and often uneven. Moi’s pre-emption of the system confirmed how potentially devastating famine would be for his already-weak presidency. With information from the three reports, Moi moved quickly to respond to the growing crisis. On June 13, 1984, the president announced the establishment of the National Famine Relief Fund (NFRF). Under the control of his office,109 the NFRF would be the main institution through which private contributions were mobilized and channeled to affected communities. Then, on June 19, Moi made a formal appeal for assistance to all international donor agencies represented in the country and announced that his government would embark on a relief program. President Moi faced four major challenges to resolve if his administration was going to prevent the worst drought in a century from progressing into a widespread famine. First, Moi’s government had to make a decision regarding the general policy of relief it would adopt and pay for (smoothing consumption versus smoothing income). After that, the government had to decide who would have access to (or qualify for) the assistance it was providing. Here the choice was between universal access and non-exclusion. The third policy challenge arose from the failures in the market for food. Would the government intervene only to support the market (complementary intervention) or would it temporarily take over all market activities (exclusionary market intervention)? Finally, Moi had to establish an administrative system to implement and oversee each of these components of the government’s relief effort. Since becoming president, Moi had struggled to garner wide support in rural areas outside of the Rift Valley, relying on coercive powers of

Insecurity, Autocracy, and Drought Relief in Kenya

71

the Provincial Administration to maintain order. Prior to 1984, Moi had had few opportunities to build his governing coalition through the distribution of tangible goods. The drought provided him with the political opportunity to use the most valuable resource, food, as patronage to resolve some of his political problems. As a result, Moi took great care to ensure that the program of relief that he favored— food aid to the rural populace, mainly peasant farmers—became the centerpiece of the government’s drought program, not the alternatives favored by Kenya’s technocrats, such as labor-based relief. Technocrats called for “job-creation schemes that would create temporary employment opportunities on public works, such as dam construction, soil conservation projects, and road works.”110 In the early stages, the option of food-for-work was considered but that approach was rejected in favor of “paying workers in cash, and then letting them have the discretion to spend it on food or other necessities.”111 This plan proposed an expansion of the number of labor-intensive local development projects, funded under the Rural Development Fund of the Ministry of Finance and Planning. Experts estimated that the government would have to create about 250,000 jobs per month (2.25 million jobs total) in order to offer adequate protection to those affected by the drought. Technocrats envisioned that the works programs would compensate households for lost incomes, as well as give beneficiaries the discretion to spend the income based on their own priorities. Moi avoided a direct and public debate over the merits of his foodaid program and its labor-based rival; instead he leveraged his control of the public purse. Through the Ministry of Finance and Planning, the president set spending limits on the labor-based relief program, permitting no more than 25,000 jobs nationwide for the duration of the relief effort, compared to the 2.25 million jobs experts estimated were needed to counter the severity of the drought. The spending limits effectively killed labor-based relief without Moi ever having to come out against the program in public. All the political and financial backing of the government was thrown behind the food aid program the president preferred. Thus, a week after establishing the NFRF, Moi announced that his office would adopt direct support to protect rural Kenyans from starvation.112 The food aid program ran from July 1984 to November 1985 and reached twenty-five of forty-one districts. The number of recipients of free food rations reached its height in October 1984, at 1.575 million.113 Of those receiving food aid, 75 percent were in Eastern Province, 15 percent in the Rift Valley, 5 percent in Coast Province, 3 percent in Central Province, and 2 percent in Northern Province. Three districts in the

72

Domestic Politics and Drought Relief in Africa

Eastern Province accounted for close to 65 percent of the national total in food aid recipients. Leading up to the drought, Moi drew most of his popular support from the Rift Valley, his home province. However, as I illustrated earlier, support there was not uniformly high. Public backing of the coup attempt in the Western Province and cities made it clear that, in order to survive, Moi needed to expand his coalition, and one such opportunity came with the distribution of free food to the three million people in Eastern Province. The three main districts in Eastern Province were kept well fed, with robust average monthly ration sizes of between 10 and 12 kgs of food per adult per month. This was very close to the adultequivalent cereal requirement of approximately 150 kgs of grain per capita per year (12.5 kgs of grain per month). Moi also placed great value in maintaining and consolidating support in his core region of support, his home province, which also received much higher levels of aid. In some Rift Valley districts, adults were provided with rations in excess of 20 kgs per month, underlining the relative political value of each district to the new coalition that Moi was fashioning. Dealing With Cities and Challengers Using Market Interventions

The political crises that Moi faced since succeeding Kenyatta in 1978 revealed that cities, Nairobi and Mombasa in particular, were the center of student discontent, economic agitation, and political challenges, restiveness that further deepened the illegitimacy of Moi’s administration. Thus, Moi’s second concern in responding to drought in 1984 was to deal with the highly volatile Central Province and the city of Nairobi; his solution was to ensure that urban residents, too, were fed. With the cities fully supplied with food, challengers would find it harder to mobilize support against Moi’s emerging coalition. Moi’s desire to keep the cities fed and quiet led to the adoption of market-complementary interventions, which meant committing a significant amount of resources to food imports. Moi’s decision to import yellow maize and wheat for the urban market assured millers, retailers, and households of a steady and adequate supply. The hope was that this could prevent shortages, queues, price increases, and, most importantly to Moi, social unrest. The first tenders for maize and wheat were made in July of 1984; the first shipments arrived in September. Between September 1984 and June 1985, the government of Kenya spent over US$113 million (in 2005 terms) to import 498,705 Mt of maize and 73,585 Mt of wheat. The Kenyan government also received 331,438 Mt of maize and wheat as relief donations from a number of countries and

Insecurity, Autocracy, and Drought Relief in Kenya

73

international aid organizations, which amounted to about 37 percent of the grain imported during the relief program. Over 60 percent of these supplies went directly to the cities. As I described earlier, the government’s interventions at both rural and urban levels when drought threatened in 1979 and 1980 offered Moi the opportunity to claim all credit for ensuring that Kenyans were well provisioned. The same politics of dry times was evident in the government’s intervention in 1984. Officials made sure that people were informed about what Moi’s administration was doing to keep them fed. Thus, every single movement in grain—from the initial government tenders, its shipment and estimated arrival date, its arrival at the port of Mombasa and subsequent delivery to markets—received front-page coverage in the government-run Daily Nation.114 Moi also held several public briefings, at which he declared that he did not want to hear of starving Kenyans because his government had ensured that food was available.115 The threat of famine offered Moi, as it had in 1980, an opportunity for political benefit by leveling the blame for the food shortage on political challengers whom he called “unscrupulous businessmen…bent on getting rich overnight.”116 In keeping with his populist rhetoric, Moi accused the big men of hoarding food from the cities in order to generate political discord. For Moi, the food crisis was also an opportunity to attack and weaken the major maize traders in the country who had captured the NCPB.117 In addition to the complementary interventions discussed above, the autocratic Moi added several market-excluding policies to his relief program. First, the government imposed a regime of price controls on maize and wheat in mid-1984. Second, the president announced that, as a matter of national security during the drought, the Provincial Administrative system would take over the functions of the NCPB at the regional level, forcing previously independent large farmers associations into a subordinate position under Moi’s office. Moi then appointed the permanent secretary in the Office of the President to chair the NCPB at the center in Nairobi.118 Having asserted his control of the formal marketing institutions, Moi then strengthened his position by curtailing informal internal movements of food. All traders, merchants, or private individuals were required to obtain a “permit” from the PA and the NCPB before they could transport food around the country.119 Provincial commissioners, who commanded the police and internal security forces at the local level, were ordered to deploy these agents and “stop and examine all lorries and to search for unauthorized movements of grain.”120 To further

74

Domestic Politics and Drought Relief in Africa

centralize Moi’s control of the market for food, all permits had to be approved by officials in Nairobi. Although these exclusionary interventions in Kenya were less than efficacious in dealing with food prices (see Table 2.2), the centralization of provisioning was of great political value. Price controls and restrictions on the movement of food worked to further Moi’s position vis-à-vis the established commercial food producers in the country. Market-excluding interventions were used to weaken and dismantle the power of the established and entrenched commercial producers who had benefited from the old regime. The 100-year drought of 1983-1984 provided Moi with a legitimate reason to challenge established groups and the farmer organizations they had created, as he did with the dismantling of the Kenya Farmers’ Association (KFA), which was dominated by commercial farmers from the Rift Valley. KFA members supplied the NCPB with about 20 percent of its yearly intake of maize, and were increasingly using their wealth to enter politics. For instance, in the 1983 elections, a director of the KFA ran against a candidate Moi supported.121 Although the KFA director lost his bid for election, his wealth, and that of the other KFA directors, became a threat to Moi in his own home province. Thus we see Moi’s motivations in taking over the reins of the entire food market through exclusionary interventions. In August of 1984, Moi replaced the KFA management team,122 and tried to conceal his overt political attack by blaming its members for the food crisis. In October, the president launched a new grain union, the Kenya Grain Growers Co-operative Union (KGGCU),123 speaking at the ceremony of how he looked forward to greater food security and the transformation of Kenya into a netexporter of food. In championing the formation of the KGGCU, Moi was attempting to transfer dominance in the domestic food sector from commercial farmers rooted in the KFA to smaller-scale peasant farmers, the core of his new governing coalition. Without the resources to provide services, the new union on its own had little chance of wresting control of food production from the 61-year-old KFA. But Moi used his executive powers to ensure that the it had the necessary resources, directing all agricultural co-operatives and farmer unions in the country to buy shares in the KGGCU.124 In effect, Moi forced his enemies to fund the association he was forming to replace them. Such an intervention was unprecedented and signaled Moi’s determination to consolidate his position in the Rift Valley. Faced with the combination of implicit blame for Kenya’s food insecurity and a state-supported competitor farmer’s union, the KFA had little hope of survival, and ultimately it

Insecurity, Autocracy, and Drought Relief in Kenya

75

disappeared into the KGGCU,125 which now would represent over 10,000 farmers. Thus, the drought and relief program provided Moi the opportunity to use Kenya’s food insecurity as cover for the destruction of a farmer’s union whose economic power was a potential challenge to the president. Transforming the PA: Access and the Administration of Relief

In 1984, President Moi’s government had to determine access to its food aid program, deciding whether food aid would be available to everyone or only to particular categories of Kenyans. As with the other policy challenges that he faced, Moi adopted a selection system that enabled him to extract the most political advantage from the relief program. Consequently, access to food aid in 1984-1985 was neither openly universal nor narrowly targeted. Instead, Moi chose to give discretion to local functionaries to determine who qualified for government relief, thus authorizing officials in the PA system to make free food available to those in need. District commissioners were to rely on “famine relief committees and local chiefs to identify those in need of support.”126 Lists of those selected for free food relief at the village and district levels were forwarded through district and provincial commissioners to the Office of the President, which oversaw the program. In the rural areas, food aid provided the Provincial Administration with tangible resources to transfer to Kenyans. This was quite different from Moi’s earlier attempt to woo the people (wananchi) by leading the fight against corruption. Through food aid, the president successfully leveraged control over food distribution to consolidate his power. After settling on this program of relief and determining access to it, Moi’s government then had to appropriate sufficient funds to pay for it and decide on an administrative structure to implement it. Moi went outside Kenya’s existing administrative apparatus and established new ad hoc organs to implement relief and committees that reported directly to his office. A senior-level inter-ministerial Steering Committee at the top advised the cabinet on key policy issues and provide overall coordination for the program. The chief secretary to the cabinet chaired the steering committee, whose members included a representative of the Office of the President, which was responsible for the provincial and district administration, and permanent secretaries from the Ministry of Finance and Planning and the Ministry of Agriculture. Below the steering committee was a Task Force on Food Supply and Distribution, charged with managing the government’s complementary food import program. The task force chair and secretary were

76

Domestic Politics and Drought Relief in Africa

drawn from the Office of the President, with members from the Ministry of Finance and Planning, the National Cereals and Produce Board, the Ministry of Transport, the Kenya Railways, Kenya Ports Authority, and other key institutions. A subcommittee of the task force addressed logistics issues, particularly with respect to the port at Mombasa and road and rail transport inland.127 The task of coordinating NGOs involved in the relief program was given to the pre-existing Kenyan National Council of Social Services, a small unit under the Ministry of Culture and Social Services. The council had been established a number of years before the 1984-1985 relief program to act as a coordinating organ, and, less benignly, to monitor and regulate the many NGOs active in Kenya. Moi placed the administrative responsibility for the food aid program with the Provincial Administrative system Kenyatta had resurrected. Provincial commissioners were responsible for communicating with all the district officials in their province regarding the government’s relief program, for submitting district requests for aid to the Office of the President, and, subsequently, for monitoring the distribution of rations at the local level. Provincial commissioners were assisted by District Famine Relief Committees, which were headed by district commissioners and also included chiefs and district representatives of relevant ministries. A similar structure was established at the division, location, and sub-location levels, with district officers, chiefs, and assistant chiefs as the chairs at each level. Under this system, lists of beneficiaries for the food aid program made their way up from the sublocation until they reached the provincial commissioners, who forwarded these names to the Office of the President, whose approval was needed for food rations to be released to the provinces for distribution. The placement of ad hoc organs reporting directly to the president offered Moi incredible autonomy in directing and massaging the relief effort in his favor and worked to ensure that credit for the success of the relief effort was directed to him alone. For example, over the course of the 1984-1985 relief program, President Moi took advantage of the administrative system in place to have a number of photo-op meetings with his provincial administrative heads, at which he urged them to ensure that the distribution of food was fair and orderly.128 The administrative structure of the program also permitted Moi to bypass a state bureaucratic system that was still largely ambivalent towards its nonKikuyu leader.

Insecurity, Autocracy, and Drought Relief in Kenya

77

Conclusion and Implications

When drought threatened Kenya with famine in 1984, President Moi was actively involved in the government’s relief program, using the presidential pulpit to marshal the government’s early responses to hoarding, price increases, and food scarcity. His office was the focal point of decision-making on a general relief policy and oversaw the implementation of the food aid component of the relief program. Just north of Kenya, in Ethiopia, Mengistu Haile Mariam’s military regime (the Derg) provides a contrasting picture. Incumbents in Ethiopia ignored the threat of famine produced by droughts throughout the early 1980s,129 and by 1983, the result was a severe famine in which 7.75 million people were affected and over 300,000 starved to death.130 Images of the suffering in Ethiopia caught the world’s attention through Bob Geldof’s Band Aid campaign of the mid-1980s, and the famine subsequently contributed to the civil war that cost Mengistu power in 1991. In this chapter, I have provided evidence that Moi’s political standing in 1983, when the drought began, determined how he chose to respond to the forthcoming dearth. The analysis suggests that political vulnerability dictated which general relief policy Moi favored, it shaped the level of access to aid he established, it nudged his government to intervene in the market for food in distinctive ways, and it determined the administrative institutions and personnel Moi trusted to implement his food aid program. First, insecurity compelled Moi to respond to the drought as the food crisis undermined the legitimacy of his government. The desire to find measures that could help mitigate the immediate political problems his vulnerability generated led Moi to push for food aid over labor-based relief in the rural areas and to complement the market by importing food for the urban sector, while simultaneously imposing exclusionary government controls on food pricing, distribution, and movements. Each element of the relief program contributed to President Moi’s consolidation of power. First, Kenya’s rural districts could be recruited and bound into a new governing coalition by the distribution of free food rations. Building an expansive rural governing coalition outside of his home district of Nandi, in the Rift Valley, was something Moi had tried and failed to do through his populist agenda between 1978 and 1983. The widespread crisis of provisioning caused by the drought offered Moi the opportunity to do what he had failed to do in nondrought years: transfer tangible benefits to rural Kenyans and thereby win their support. Second, I showed that the volatile cities could be kept relatively calm and quiet if a steady supply of food for the market was

78

Domestic Politics and Drought Relief in Africa

maintained. Finally, I revealed that under the guise of responding to a national crisis, Moi used exclusionary government interventions in the food market to dismantle the economic power base of potential challengers, especially those in the Rift Valley. He thereby reconfigured Kenya’s political economy, including food production, more decisively than he had been able to do through anti-corruption campaigns between 1978 and 1983. This element of Kenya’s intervention reveals that drought relief is not just a political opportunity to build support through the distribution of patronage, as Moi was able to do by distributing food aid to rural areas. Drought relief also affords incumbents a chance to strengthen their position by weakening the economic base of known challengers. Thus, by responding to the drought and placing himself at the center of his government’s efforts, Moi used dearth to his advantage. The threat of famine in 1984 reveals what Bates calls the “agrarian foundations for politics” in Africa: it shows “how anxieties about food supplies can be used to win friends and to discredit enemies and thereby to consolidate political power.”131 Moi’s intervention from 1984 to 1985 confirms that weak incumbents can use drought relief to consolidate their position in office and illustrates how political insecurity determines the resolution of the challenges that face incumbents during dry times. Notes 1

See Morton, Moi; Widner, The Rise of a Party-State in Kenya. The growth and maturation of African nationalism in Kenya is thoroughly covered by Kyle, The Politics of the Independence of Kenya; Gordon, Decolonization and the State in Kenya; Alam, Rethinking the Mau Mau in Colonial Kenya; Spencer, The Kenya African Union; and Clough, Mau Mau Memoirs. 3 Cleavages in the nationalist movement in Kenya are covered by Anderson, “‘Yours in Struggle for Majimbo,’” pp. 547-64; Kyle, The Politics of the Independence of Kenya; Bennett and Rosberg, The Kenyatta Election; Widner, The Rise of a Party-State in Kenya; Special Correspondent, “The Changing Face of Kenya Politics,” pp. 44-50. 4 This coalition included the Kenya African People’s Party led by Muliro; the Kalenjin Political Alliance led by arap Towett and arap Moi; the Maasai United Front; the Coast African People’s Union led by Ngala; and the Somali National Association. For a discussion of the interests and factions behind the formation of KADU, see Bennett and Rosberg, The Kenyatta Election. 5 Okoth-Ogendo, “The Politics of Constitutional Change in Kenya,” p. 13; Anderson, “‘Yours in Struggle for Majimbo,’” pp. 547-64. 6 For the most comprehensive discussion of the 1961 transitional elections, see Bennett and Rosberg, The Kenyatta Election. 2

Insecurity, Autocracy, and Drought Relief in Kenya

79

7 See Sanger and Nottingham, “The Kenya General Election of 1963,” pp. 1-40. KANU won 54 percent of the vote and 72 of the 112 openly-contested seats as well as 11 of the 12 National Member seats selected by the directlyelected members of the House, while KADU won just one. The remaining 8 popularly-elected House seats were won by Paul Ngei’s African People’s Party (with 7 percent of the vote). Ngei defected from KANU in the run-up to the election and managed to leverage his strong support among the Kamba to win seats in parliament. 8 For a discussion of the early constitutional amendments, which entrenched the Executive Presidency and a strong central government over regional alternatives, see Okoth-Ogendo, “The Politics of Constitutional Change in Kenya,” pp. 18-21. Kenyatta’s push for a voluntary one-party state is discussed by a number of scholars including Ogot and Ochieng’, eds., Decolonization and Independence in Kenya, 1940-93; Throup, “Elections and Political Legitimacy in Kenya,” pp. 371-96; Tamarkin, “The Roots of Political Stability in Kenya,” pp. 297-320; and Gertzel, The Politics of Independent Kenya, 1963-8, among others. 9 Throup, “Elections and Political Legitimacy in Kenya,” pp. 371-96, contends that KADU’s dissolution must be understood as a product of two considerations by its MPs as their ranks thinned: 1) they were enticed by the offers of co-option that Kenyatta made, and 2) they were threatened by what would happen to members of a disloyal opposition as Kenyatta continued to be given more and more powers through the process of constitutional amendment. 10 Morton, Moi, ch. 6. The other members of KADU’s executive (Ngala, Muliro, Shikuku, and arap Toweett) did not get cabinet posts. 11 Khapoya, “Kenya under Moi,” pp. 17-32, discusses the ethnic balancing in all of Kenyatta’s cabinet appointments. See also Karega-Mûnene, “Polarisation of Politics in Kenya Along Ethnic Lines.” 12 Morton, Moi, p. 86. This description of the ideological cleavages within KANU is consistent with that offered by Hyden and Leys, “Elections and Politics in Single-Party Systems,” p. 393. 13 The struggle between Jomo Kenyatta and Oginga Odinga was rooted in competition for power, substantive ideological differences, and Kikuyu/Luo jostling for political and economic dominance in Kenya. See Orvis, “Bringing Institutions Back Into the Study of Kenya and Africa,” pp. 95-110; Anderson, “‘Yours in Struggle for Majimbo,’” pp. 547-64; Special Correspondent, “The Changing Face of Kenya Politics,” pp. 44-50. 14 See “Kenya: Arms & Odinga”; and Fellows, “Soviet Guns Due for Kenya.” Odinga also writes about this period in his autobiography, Not Yet Uhuru. 15 Gertzel, The Politics of Independent Kenya, 1963-8, pp. 71-72; Special Correspondent, “The Changing Face of Kenya Politics,” pp. 44-50. See also “The Limuru Conference Paves the Way for the Emergence of a New Opposition Party,” East African Standard (March 14, 1966). 16 Scholars question the openness of the vote at Limuru. For instance, Throup, “Elections and Political Legitimacy in Kenya,” p. 373, contends that the elections were rigged by Tom Mboya. This assessment is consistent with early analyses of the delegates’ conference, which noted that the distribution of votes was weighted in favor of two provinces, the Rift Valley and Coast, both controlled by Kenyatta allies (Moi and Ronald Ngala). Together these two

80

Domestic Politics and Drought Relief in Africa

provinces accounted for 40 percent of the votes at Limuru. See Special Correspondent, “The Changing Face of Kenya Politics,” pp. 44-50. 17 The other provincial vice-presidents elected were: James Gichuru, Central; Mohamed Jubat, North Eastern; Eric Khasakhala, Nyanza; Mwai Kibaki, Nairobi; Ronald Ngala, Coast; Jeremiah Nyagah, Eastern; and Lawrence Sagini, Nyanza. 18 According to Tamarkin, “The Roots of Political Stability in Kenya,” pp. 297-98, although Kenyatta is generally assumed to have enjoyed high “standing, prestige and reverence” among fellow nationalists as the father figure of Kenyan nationalism, his position was far less secure. Of note for Tamarkin is the fact that before independence Kenyatta was largely absent from Kenya: between 1931 and 1946, when he lived abroad; and between 1952 and 1961, when he was imprisoned for his involvement with Mau Mau. These absences, Tamarkin rightly contends, meant Kenyatta was out of touch with political realities of KANU on the ground. 19 For an in-depth analysis of Kenya’s Provincial Administration system see Ndemo, Epitome of State Power; Branch and Cheeseman, “The Politics of Control in Kenya,” pp. 11-31; Orvis, “Bringing Institutions Back into the Study of Kenya and Africa,” pp. 95-110; Swainson, “State and Economy in PostColonial Kenya,” pp. 357-81. 20 Kenyatta’s weakening of KANU and subsequent reliance on the Provincial Administration and harambee to build political control in Kenya is discussed in detail by Cheeseman, “Political Linkage and Political Space in the Era of Decolonization,” pp. 3-24; Orvis, “Bringing Institutions Back Into the Study of Kenya and Africa,” pp. 95-110; Branch and Cheeseman, “The Politics of Control in Kenya,” pp. 11-31; Widner, The Rise of a Party-State in Kenya; Gertzel, “The Provincial Administration in Kenya,” pp. 201-15. 21 Barkan and Holmquist, “Peasant-State Relations and the Social Base of Self-Help in Kenya,” p. 360. The system of harambee and how Kenyatta used it to construct political stability in Kenya is also discussed in detail by Cheeseman, “Political Linkage and Political Space in the Era of Decolonization,” pp. 3-24; and Orvis, “Bringing Institutions Back into the Study of Kenya and Africa,” pp. 95-110. 22 “Odinga Resigns from KANU,” East African Standard (April 15, 1966). A discussion of the political goals of the KPU is offered by Ogot and Ochieng, eds., Decolonization and Independence in Kenya; Gertzel, The Politics of Independent Kenya, 1963-8; Gertzel et al., eds., Government and Politics in Kenya. 23 Ogot and Ochieng’, eds., Decolonization and Independence in Kenya.. Odinga also discusses this moment in his autobiography, Not Yet Uhuru, pp. 300-4. 24 Mboya rose to prominence in the early 1950s as a vocal trade unionist behind the Mombasa Dock Strike of 1955. He was the founding secretary general of the Kenya Federation of Registered Trade Unions (KFRTU) in 1953. In 1955 he transformed the KFRTU into the Kenya Federation of Labour (KFL). Following this early left-leaning activism grounded in the labor movement, Mboya went through a rather remarkable transformation into a key member of the conservative capitalist faction within KANU. In 1957, he was elected to the Legislative Council, and from that position was instrumental in the negotiations at Lancaster that eventually led to Kenya’s independence in

Insecurity, Autocracy, and Drought Relief in Kenya

81

1963. When KANU was formed in 1960, Mboya was elected secretary general. On March 7, 1960, his face adorned the cover of Time magazine. After independence, Mboya held a number of cabinet positions in Kenyatta’s government. See Goldsworthy, Tom Mboya.. 25 Currie and Ray, “State and Class in Kenya,” p. 568. Widner, The Rise of a Party-State in Kenya, also discusses the change-the-constitution movement in some detail. Key members of the movement included prominent Kikuyu politicians such as: Mbiya Kionage, Kenyatta’s brother-in-law; Dr. Njoroge Mungai, Kenyatta’s nephew; Kilika Kimani; Paul Ngei; James Gichuru; Njenga Karume; Jackson Angaine; and senior civil servants and army officers. 26 Morton, Moi, p. 91. 27 The politics surrounding the Kenyatta succession have been comprehensively dealt with elsewhere. See, for instance, Widner, The Rise of a PartyState in Kenya; Currie and Ray, “State and Class in Kenya,” pp. 559-93; Karimi and Ochieng, The Kenyatta Succession. 28 Quoted in Karimi and Ochieng, The Kenyatta Succession, p. 21. 29 Ibid., p. 22. 30 Throup, “Elections and Political Legitimacy in Kenya,” pp. 380-82. 31 Quoted in Morton, Moi, p. 110. 32 For a discussion of the conference and its subsequent cancellation, see Throup, “Elections and Political Legitimacy in Kenya, pp. 371-96; Tamarkin, “From Kenyatta to Moi,” pp. 26-29. 33 Tamarkin, “From Kenyatta to Moi,” pp. 26-29. 34 “Kenya: The End of an Illusion,” p. 223. 35 Widner, The Rise of a Party-State in Kenya, p. 2. For a contrasting picture of what Kenya might have looked like had Kenyatta invested in building a strong, political KANU, see Hyden and Leys, “Elections and Politics in Single-Party Systems,” pp. 389-420. 36 Widner, The Rise of a Party-State in Kenya. 37 “Kenya: The End of an Illusion,” p. 224. 38 Ibid., pp. 221-44. 39 Ibid., p. 225. 40 “Statutory Boards Review Starts,” Daily Nation (February 9, 1979). 41 “Kenya: The End of an Illusion,” p. 226. In many parts of the country, free milk was provided twice a week to schools. 42 Ibid. 43 See Daily Nation (November-December 1978). 44 “GEMA Bosses Summoned to Court,” ibid. (January 18, 1979). The GEMA board consisted of James Kurume, Duncan Ndegwa, Jacob Mwango, Dixon Kimani, Karigu Stephen, Wilson Macharia, and George Waruhiu. 45 The powers that Moi had at his disposal under Kenya’s system of provincial administration are discussed by Ndemo, Epitome of State Power; Cheeseman, “Political Linkage and Political Space in the Era of Decolonization,” pp. 3-24; Branch and Cheeseman, “The Politics of Control in Kenya,” pp. 11-31; Gertzel, “The Provincial Administration in Kenya,” pp. 201-15. 46 Throup, “Elections and Political Legitimacy in Kenya,” p. 384. 47 East African Standard (February 9, 1980). 48 Sunday Standard (November 30, 1980). 49 Throup, “Elections and Political Legitimacy in Kenya,” pp. 371-96; Hornsby and Throup, “Elections and Political Change in Kenya,” pp. 179-99.

82

Domestic Politics and Drought Relief in Africa

50 Moi’s reforms and additions to the cabinet included appointing three ministers of state in the President’s Office; establishing new Ministries of Energy, Transport and Communications, and Industry; adding responsibilities for the environment to the Ministry of Natural Resources and for urban development to the Ministry of Local Government; and, finally, splitting education into two ministries (Basic Education and Higher Education) and agriculture into Ministries of Agriculture and of Livestock Development. 51 Moi, “Speech Announcing the ‘First’ Cabinet.” 52 “Kenya: The End of an Illusion.” 53 Maxon and Ndege, “The Economics of Structural Adjustment,” pp. 15186; Throup, “Elections and Political Legitimacy in Kenya,” p. 383. 54 See “Tighten Your Belts – Moi,” Daily Nation (March 7, 1979); and “Petrol Price Rise on the Way,” ibid. (March 9, 1979). 55 “Kenya: The End of an Illusion,” pp. 221-44. 56 Ibid., p. 226. 57 Throup, “Elections and Political Legitimacy in Kenya,” pp. 383-84. 58 Currie and Ray, “The Pambana of August 1,” pp. 47-59; “Kenya: The End of an Illusion,” pp. 221-44. 59 “Kenya: The End of an Illusion,” p. 230. 60 , “Moi Slams Marxists,” Daily Nation (June 14, 1982). 61 Widner, The Rise of a Party-State in Kenya. 62 See Bienen and van de Walle, Of Time and Power; and Bienen and van de Walle, “Time and Power in Africa,” pp. 19-34, for a comparison of leadership survival rates between first-generation and second-generation rulers in Africa. Bienen and van de Walle contend that anti-colonialist credentials and founding-father stature generally afforded first-generation leaders in Africa wider support than their successors. As a consequence, in their first year in office, first-generation leaders across Africa had only a 0.08 probability of losing power compared with a 0.18 probability for subsequent leaders. 63 December 12 is the date of Kenya’s independence. 64 “Pambana: Organ of the December Twelve Movement,” p. 322. 65 Ibid., p. 323. 66 Morton, Moi, p. 131. Madaraka Day commemorates Kenya’s attainment of internal self-rule on June 1, 1963, before full independence on December 12, 1963. 67 See “Two Lecturers Picked-up,” Daily Nation (June 9, 1982); “University Crackdown – Police Search Campus,” ibid. (June 11, 1982); “16 College Students are Expelled,” ibid. (June 11, 1982); “Another Lecturer Arrested,” ibid. (June 17, 1982); “Lecturer Refused Bail,” ibid. (June 19, 1982); “Makaru, Mazrui Detained,” ibid. (June 25, 1982); “Moi Reveals Weapons Plot – Lecturers ‘were to arm dissidents,’” ibid. (June 7, 1982); “Moi Slams Marxists,” ibid. (June 14, 1982). 68 Quoted in “Detentions will Continue, Moi – Security Must be Defended,” ibid. (June 19, 1982). 69 “One-Party State: It’s Now Official – House in Full Attendance,” ibid. (June 10, 1982). 70 “Editorial,” Standard (July 20, 1982). 71 Nairobi Times (July 22, 1982). 72 Currie and Ray, “The Pambana of August 1,” p. 52. 73 See Stepan, The Military in Politics, pp. 85-121.

Insecurity, Autocracy, and Drought Relief in Kenya

74

83

Currie and Ray, “The Pambana of August 1,” p. 52. “The Coup Broadcast,” pp. 325-26. 76 ViVa, Hours of Chaos and the Aftermath, p. 4. 77 Currie and Ray, “The Pambana of August 1,” pp. 47-59; Currie and Ray, “State and Class in Kenya,” p. 570. 78 Quoted in Standard (August 6, 1982). 79 Waruhiu, From Autocracy to Democracy in Kenya, p. 81. 80 “Kenya: The End of an Illusion,” p. 241. 81 “Curfew is On,” Daily Nation (August 3, 1982). 82 ViVa, Hours of Chaos and the Aftermath, p. 4. 83 Currie and Ray, “The Pambana of August 1, ” pp. 47-59. 84 “Hundreds Throng City Mortuary – Missing Relatives Presumed Dead,” Daily Nation (August 7, 1982); “200 Confirmed Killed in Coup Bid,” ibid. (August 9, 1982); “1,000 Looters Held by Police,” ibid. (August 4, 1982); “Moi Puts Coup Loss at Sh. 1.2bn,” ibid. (August 11, 1982). 85 “Coup Revelry 70 Set Free,” ibid. (September 13, 1982). 86 “Varsity to Open Next Year – Moi,” ibid. (November 12, 1982). 87 Currie and Ray, “State and Class in Kenya,” p. 571; “Kenya: The End of an Illusion,” pp. 325-26. 88 See Currie and Ray, “The Pambana of August 1,” p. 54. 89 N’Diaye, “How Not to Institutionalize Civilian Control,” pp. 627-30. 90 A number of scholars have used the term populism to describe Moi’s approach to leadership. See, among these, Ogot, “The Politics of Populism”; and Widner, The Rise of a Party-State in Kenya, esp. ch. 5. 91 “Troublemakers are ‘Big Men’, says President,” Daily Nation (August 25, 1982). 92 See “Minister Hits Out at ‘Big Men,’” ibid. (September 18, 1982). 93 See “KANU Drafts Election Plan – Members Drive to Start Soon,” ibid. (November 12, 1982); “All Set for KANU Members Drive,” ibid. (December 13, 1982). 94 Currie and Ray, “State and Class in Kenya,” pp. 559-93. 95 “Thugs in Polls Race, says Moi – Voters Urged to Pick a Nyayo Man,” Daily Nation (November 13, 1982). 96 Hornsby and Throup, “Elections and Political Change in Kenya,” pp. 179-99. In de jure one-party states the absence of any real political space or substantive choice over candidates tends to mean that exit from elections is the primary instrument through which citizens can express their dissatisfaction with the political system. 97 See “Is Rainfall Pattern Changing in Kenya?” and “Drought Threatens Bumper Harvest,” Daily Nation (May 28, 1984); “Drought Hits Jobs – Ndegwa,” ibid. (July 21, 1984). 98 “Prayer Day for Rain Set,” ibid. (May 30, 1984). 99 “Downpour After Prayers,” ibid. (June 4, 1984); Downing et al., “Drought in Kenya,” p. 22. 100 See “Drought Threatens Bumper Harvest,” Daily Nation (May 28, 1984); and “Food SOS Report Denied,” ibid. (June 27, 1984). 101 “MPs in Uproar Over Beans – Mwicigi Under Pressure in Parliament,” ibid. (July 5, 1984). Schatzberg, Political Legitimacy in Middle Africa, contends that presidents and governments in Africa very often win or lose legitimacy based on how well they maintain food security or respond to food insecurity. 75

84

Domestic Politics and Drought Relief in Africa

102

“We Sold Beans Outside – Tipis,” Daily Nation (July 6, 1984). “Punish Hoarders, Moi Directs A-G – Leaders Told not to Spread Alarm Over Food,” ibid. (July 2, 1984). 104 See Drèze and Sen, Hunger and Public Action; Drèze and Sen, eds., The Political Economy of Hunger; Ravallion, Markets and Famines. 105 See “Arrest Hoarders, Orders President,” Daily Nation (June 8, 1984). See also “Punish Hoarders, Moi Directs A-G – Leaders Told not to Spread Alarm Over Food,” ibid. (July 2, 1984); and “Hoarders to Lose Stock, says Moi,” ibid. (July 18, 1984). 106 Bates, Beyond the Miracle of the Market, p. 135. 107 Borton and Stephenson, Disaster Preparedness in Kenya. 108 See documents and correspondence in Government of Kenya, BB/1/144 Famine and Famine Relief, 1965-1973. (Room 4, Shelf 3491, Box 1), Kenya National Archives. 109 “Moi Launches Famine Relief Fund, Asks for Aid,” Daily Nation (June 14, 1984). 110 Borton, “Overview of the 1984/85 National Drought Relief Programme,” p. 27. 111 Cohen and Lewis, “Role of Government in Combating Food Shortages,” p. 281. 112 “Moi Launches Famine Relief Fund, Asks for Aid,” Daily Nation (June 14, 1984). 113 Beneficiary data is from Deloitte, Haskins and Sells Management Consultants 1985 in Borton, “Overview of the 1984/85 National Drought Relief Programme,” p. 54. 114 “Yellow Maize Expected in 90 Days,” Daily Nation (July 21, 1984); “Yellow Maize Due at Port,” ibid. (August 10, 1984); “Food Coming – Envoy,” ibid. (September 6, 1984); “Yellow Maize Arrives at Port,” ibid. (September 22, 1984); “Maize Arrives,” ibid. (September 25, 1984). 115 “Moi’s Directive on Food Distribution,” ibid.(October 10, 1984). 116 Ibid. (March 12, 1980). 117 Ibid. (July 1, 1980). 118 Bates, Beyond the Miracle of Market. 119 See Drèze, “Famine Prevention in Africa.” 120 Bates, Beyond the Miracle of Market, p. 124. 121 For an extended discussion of the KFA, see ibid., pp. 128-31. 122 “KFA Management Replaced,” Daily Nation (August 17, 1984). 123 “President Praised for Launching Grain Union,” ibid. (October 11, 1984). 124 “Co-ops Ordered to Buy Union Shares,” ibid. (December 19, 1984). 125 See “KFA, Grain Union are Set for Merger,” ibid. (December 5, 1984); and “Moi Hails Merger of KFA, Grain Union,” ibid. (December 6, 1984). 126 Drèze, “Famine Prevention in Africa,” p. 139. See also Cohen and Lewis, “Role of Government in Combating Food Shortages,” p. 281. 127 Borton, “Overview of the 1984/85 National Drought Relief Programme.” 128 “Moi Meets Provincial Heads,” Daily Nation (July 6, 1984). See also “Moi’s Directive on Food Distribution,” ibid. (October 10, 1984). 103

Insecurity, Autocracy, and Drought Relief in Kenya

85

129 Sen, Development as Freedom; de Waal, Evil Days. The behavior of Mengistu’s regime in the 1980s was similar to that of the Emperor before him; see Sen, Poverty and Famines, pp. 86-112. 130 Estimates of the number of people affected and killed during the 1983 Ethiopian famine are from EM-DAT: The OFDA/CRED International Disaster Database – www.emdat.be. 131 Bates, Beyond the Miracle of the Market, p. 115.

4 The Pathologies of Drought Relief in Zimbabwe

If you know of any points that are quite desperate let us know and something will be done to rush to those areas.1 – President Robert Mugabe addressing the media, July 2, 1992 Why foist this food upon us? We don’t want to be choked. We have enough.2 – President Robert Mugabe refusing international food aid, May 24, 2004.

In 2004, Robert Mugabe expelled the Crop and Food Supply Assessment Mission of the United Nations’ Food and Agricultural Organization and World Food Programme from Zimbabwe. Joseph Made, Mugabe’s Minister of Agriculture, declared the 3-year drought broken and suggested that the country would double its cereal production to over 2.5 million Mt.3 On the basis of this fictional good harvest, Paul Mangwana, head of the ministry responsible for administering drought relief programs, concluded that the country did not require food aid or imports to meet domestic needs.4 Citing the same report, Mugabe ordered all foreign aid organizations to leave the country because his government had enough food and did not need their help. The consensus outside government, especially among NGOs and aid agencies, was that the country was still afflicted by drought and that the 2004 harvest would be inadequate. In fact, preliminary food security assessments concluded that the threat of famine remained: the Zimbabwe Vulnerability Assessment Committee estimated that close to 2.3 million Zimbabweans would need food assistance between 2004 and 2005.5

87

88

Domestic Politics and Drought Relief in Africa

Mugabe’s brazen denial of a food crisis was surprising considering that throughout much of the 1980s and 1990s he had been instrumental in marshaling his government to preemptively declare drought and respond to it through expansive relief programs. In fact, in 1988, the USbased Hunger Project awarded Robert Mugabe the Africa Prize for Leadership for the Sustainable End of Hunger, stressing that his agricultural programs "pointed the way not only for Zimbabwe but for the entire African continent.” But in August 2001, the Hunger Project amended its position on Mugabe.6 Amartya Sen also praised Mugabe and his government for preventing famine in the face of drought by adopting broad relief policies, including employment-creation programs.7 Yet by the mid-1990s, Mugabe progressively moved away from offering the broad programs of relief that had won him international recognition and defined his relationship with the rural population, a retreat that culminated in his denial of drought in 2004 and decision to expel foreign food aid providers. This chapter explains why the relief programs Mugabe offered Zimbabweans over the course of his thirty-plus years in power shifted, sometimes offering the rural population food aid (between 1982 and 1984), other times providing food-for-work relief programs (in 1987 and in 1992), or offering aid in the form of a loan (between 1995 and 1999). Just as political insecurity led Moi to resolve the four policy challenges discussed in Chapter 2 in ways that contributed to his consolidation of power, Mugabe’s adoption of drought relief programs reflected his own political strength or weakness, shifting as his standing changed. I will show that when Mugabe and his party, the Zimbabwe African National Union (ZANU), faced an insecure political environment, as in the early 1980s, they responded to droughts by adopting food aid programs for adults. When incumbents were more secure, as occurred between 1985 and 1995, drought-relief programs for adults shifted away from free food aid to cost-effective programs that avoided dependency and limited waste, such as food-for-work. But the commitment to expansive programs of relief faltered between 1995 and 1999 as personalized incumbency, authoritarianism, and excessive political insulation weakened Mugabe’s interest in providing adequate relief, while the political crisis of the first decade of the 21st century led to the revival of food aid programs and the politicization of relief. This chapter therefore demonstrates that, as in Kenya, the logics of secure and insecure incumbency determined government responses to drought in Zimbabwe.

The Pathologies of Drought Relief in Zimbabwe

89

Instability and Insecurity in the Early Years of Independence, 1980-1985

After years of war and numerous failed constitutional negotiations, a breakthrough at the Lancaster House Conference of 1979 set the stage for independence elections in Zimbabwe.8 The new constitution provided for a majority-rule government with a bicameral legislature. The House was to have 100 seats, 80 to be elected by an African Common Roll and the remaining 20 to be reserved for the white population. The Lancaster House agreement also established a Senate, the majority of whose 40 members were elected by the House, while 10 seats were reserved for the white population. The country was divided into eight provincial electoral districts, with seats to be apportioned among parties roughly according to votes they received in each region. As in a parliamentary structure, the party or coalition of parties with a majority of seats in the House would get the position of prime minister and be asked to form the country’s first majority-rule government. The legislature would also choose a ceremonial president.9 The transitional period in Zimbabwe produced an election marred by violence and intimidation.10 The conduct of all the parties entrenched a level of suspicion and animosity among elites and their followers that came to color the anxieties of the new incumbents after 1980. Significantly, the two major nationalist parties that had fought white rule and joined to negotiate at Lancaster House as the unified Patriotic Front (PF), split and contested the elections as independent parties. During the fight for independence, the strategy of the Zimbabwe African National Union, led by Robert Mugabe, had stressed guerilla– style combat. It embedded its forces in the rural population, capturing territory that ZANU then nominally governed as a liberated zone.11 Its rival, the Zimbabwe African People’s Union (ZAPU), had by contrast aimed to recruit and build a conventional force capable of confronting the army of the white Rhodesian government head-on. Thus, while ZAPU, led by Joshua Nkomo, built this capacity at its bases in Zambia, ZANU gained a foothold in key provinces (Manicaland, Masvingo and all three provinces of Mashonaland, Central, West, and East) through its guerilla strategy. In the 1980 elections, ZANU took advantage of its control of a significant portion of the countryside to deny other political parties campaign entry. According to Edison Zvobgo, ZANU’s campaign director in 1980, the party’s position was quite simple: “Why should any party go where it is not wanted? Why should any party wish to go and reap where it did not sow?”12 ZAPU attempted the same tactic, but its broad strategy of building a conventional army meant it had only

90

Domestic Politics and Drought Relief in Africa

established itself in Matabeleland North and South and parts of Midlands, only three of the eight provinces.13 These strategies led to violent clashes between ZANU and ZAPU, including attempts on Robert Mugabe’s life. The first came on February 6, 1980, when a grenade was thrown into Mugabe’s home in the Mount Pleasant suburb of Salisbury (now Harare).14 Although the damage was limited, the vulnerability of the ZANU candidate was again shown a few days later, on February 10, 1980, when Mugabe survived the detonation of a roadside bomb as his convoy left a campaign rally in Fort Victoria (now Masvingo).15 ZANU blamed the white Rhodesian forces that continued to resist majority rule.16 Despite the violence, majority rule elections were held on schedule,17 and in results announced on March 5, 1980, Mugabe and ZANU were declared the victors, with 57 of the 80 Common Roll seats.18 Nkomo’s ZAPU managed to win only 20 seats, while Bishop Abel Muzorewa’s United African National Congress won the 3 remaining Common Roll seats. White candidates from the politically and racially conservative Rhodesian Front Party led by Ian Smith, the former prime minister of the minority Rhodesian government, won all 20 of the seats reserved for whites, standing unopposed in 14 races. ZAPU’s poor showing came as a surprise, since the expectation was that the two major nationalist parties would split the African vote and neither would emerge with a clear parliamentary majority. That situation would have opened the door for a coalition government in which whites could leverage their 20 seats to help decide the winner, thus moderating the contentious transition to majority black rule. ZANU’s decisive majority foreclosed the possibility of such coalition maneuvers. Victory for ZANU, perceived to be most uncompromising in its nationalist and Marxist-Leninist ideology, raised questions about whether other key veto players (the Rhodesian forces, ZAPU, and the white community) would accept the outcome.19 In fact, the white community was unsettled by the election (over the next three years roughly half of the white population would leave the country20), and it emerged that the Rhodesian forces were ready to execute a coup,21 while reports from the Zambian border suggested that ZAPU’s considerable Soviet-trained and -armed conventional force was mobilizing.22 Thus, as Weitzer stresses, the new ruling party was rightly “suspicious of subversion by disaffected segments of the white community and its black rivals.”23 Mugabe moved quickly to try to resolve these immediate threats to his incumbency, appointing Lieutenant General Peter Walls, commander of the white Rhodesian forces, to oversee the integration of his army

The Pathologies of Drought Relief in Zimbabwe

91

with ZANU and ZAPU fighters into a unified Zimbabwean army. Mugabe also invited Nkomo to join a coalition government24 and then turned his attention to the white community, extending an olive branch of reconciliation on the eve of the official handover of power: If yesterday I fought you as an enemy, today you have become a friend and ally with the same national interest, loyalty, rights and duties as myself. If yesterday you hated me, today you cannot avoid the love that binds you to me and me to you. The wrongs of the past must now stand forgiven and forgotten.25

Months later, Mugabe would reiterate his wish for racial comity: “I have drawn a line through the past. From now on, we must trust each other if we are to work together for the benefit of the majority. I want people to believe in my policy of reconciliation and to respond accordingly.”26 These early decisions and statements were indicative of a weak incumbent who needed to begin to build support while forestalling direct challenges from key actors. Nkomo’s credentials as the patriarch of the nationalist movement and ZAPU’s conventional military capacity presented Mugabe with an immediate political threat. Though the nearly 6,000 white farmers supported Ian Smith and the Rhodesia Front and were one of the most privileged groups in the country, they owned close to 70 percent of the best agricultural land and so were critical to the early economic performance of Mugabe’s Zimbabwe. Meredith estimates that white farmers accounted for nearly 75 percent of the agricultural sector, grew 90 percent of the staple maize crop, dominated cotton production, in addition to other cash crops, accounted for 33 percent of export earnings from agriculture, and employed 271,000 people, about a third of the labor force in 1980.27 Thus, their economic veto made white farmers key to Mugabe and ZANU’s attempt to consolidate their position in office. The accommodations seemed to work: Mugabe and Nkomo agreed on the terms of a coalition government; the policy of reconciliation assuaged white fears of recriminations and stemmed the tide of white flight; and significant patronage to white farmers in the form of robust prices and technical support for their crops quickly transformed their anxieties over the outcome of the election to a guarded optimism for life under a government led by “Good Old Bob!”28 However, these decisions were not a permanent solution to Mugabe’s insecure incumbency. Early interactions between coalition members were not encouraging. In July 1980, Senator Enos Nkala, Mugabe’s Minister of Finance, pu-

92

Domestic Politics and Drought Relief in Africa

blicly expressed ZANU’s preference for the creation of a one-party state in Zimbabwe.29 A few days later, Nkala came to the ZAPU stronghold of Bulawayo in Matabeleland and provocatively declared to a rally that his task “from now on [was] to crush Joshua Nkomo and forget about him.”30 ZAPU responded to Nkala’s outburst by organizing a demonstration outside parliament in Harare and demanding that Mugabe explain his minister’s statements.31 The acrimony between members of the coalition government quickly escalated into violent clashes between their supporters, deepening political instability.32 In early November 1980, an open battle erupted at Entumbane housing estate in Bulawayo.33 The fighting forced many of the civilian residents in and around Entumbane to flee into the surrounding bush and left over 50 dead and 400 injured.34 Nkala urged ZANU supporters in Bulawayo to form vigilante committees to defend themselves, and a few months later, new battles at Entumbane, as well as at Ntabazinduna and Connemara military encampments, left over 300 people dead.35 The persistent fighting between troops loyal to the two nationalist parties revealed the tenuousness of the peace in Zimbabwe and complicated the new government’s task of integrating the over60,000 armed veterans of the fight for majority rule and close to 20,000 former Rhodesian forces into a single army.36 In response to the speading violence, Mugabe imposed a dusk-todawn curfew in Entumbane and deployed a loyal ZANU battalion of the army to Bulawayo. Fearful that this battalion would not be even-handed in disarming the fighting troops, battle-ready ZAPU units from all around the country left their camps and headed for Entumbane. Joshua Nkomo was eventually able to persuade his troops to return to their camps, but the incident revealed the fragility of the political arrangement as the situation in Zimbabwe was deteriorating into open conflict with Mugabe and his ZANU government struggling to establish monopoly over violence. The open engagement between soldiers in Matabeleland exacerbated a situation that was already tense following the discovery of caches of weapons at ZAPU-owned farms. In early February 1980, joint police and army patrols in Matabeleland discovered arms and munitions sufficient to outfit a brigade of 5,000 men.37 When another cache was discovered in early 1982, the strained political accommodation between ZANU and ZAPU finally collapsed. Mugabe fired Nkomo and other ZAPU members of government and imprisoned several of the party’s military commanders, declaring that: “Zapu and its leader, Dr Joshua Nkomo, are like a cobra in a house. The only way to deal effectively with a snake is to strike and destroy its head.”38

The Pathologies of Drought Relief in Zimbabwe

93

The coalition’s collapse fed an insurgency that had developed across the Midlands and Matebeleland. Jocelyn Alexander argues that there were two separate groups rebelling against Mugabe’s government: the first drew support from disaffected former ZAPU forces who questioned Nkomo’s decision to enter a coalition government with ZANU; while the other was supported by the apartheid government in South Africa.39 Between 1980 and 1983, there were over 220 reported insurgent attacks on security forces, the police, and members of the public.40 Mugabe’s response was disproportionately violent. Initially, he ordered a four-day cordon of Bulawayo, which resulted in over 1,000 detentions.41 He then deployed troops on two fronts: First, the 4th and 6th Brigades, the Paratroopers, the Central Intelligence Organization, and the Police Support Unit were sent to engage the insurgents directly. Second, in a violent and indiscriminate deployment called Gukurahundi, “The Rains That Clear Out the Chaff,” Mugabe unleashed the North Korean-trained 5th Brigade, the Central Intelligence Organization, the Police Internal Security Intelligence Unit, and ZANU Youth Brigade on the civilian population in rural and urban Matabeleland and parts of the Midlands. A report published in 1997 estimates that this deployment of troops led to the death of over 1,500 civilians,42 fulfilling Mugabe’s intention to exact a high civilian price for support of ZAPU. Widespread strikes in major urban centers and mining towns compounded the insecurity created by the insurgency; two weeks after Mugabe’s victory, “16,000 workers in 46 firms went on strike.”43 In a subsequent study, insecure incumbents acknowledged that they had to cope with over 200 strikes between 1980 and 1981.44 Appeals by Mugabe’s labor minister, Kumbirai Kangai, for patience and calm went ignored, and the prime minister’s own belated call for restraint and order had little effect.45 The government resorted to mass dismissals and mobilized the police to stem the tide of labor unrest.46 The rapid spread of strikes and other protests prompted one senator to call for the army to support the police in maintaining order in the country.47 This early confrontation with labor shaped how the ZANU government subsequently reformed industrial relations in the country. Although incumbents moved quickly to remove racial categorization and discrimination and to adopt minimum-wage legislation and price controls, they were reticent to enshrine any blanket rights to industrial action. The government instead facilitated the merger of close to 60 unions into a single entity, the Zimbabwe Congress of Trade Union (ZCTU),48 thus keeping unions legal while establishing a quasi-corporatist organization through which it was able to monitor, anticipate, and control labor activism. This strategy worked until the mid-1990s, when economic

94

Domestic Politics and Drought Relief in Africa

reforms damaged and eventually shattered the corporatist relationship between ZANU and labor. Early Responses to Instability and Insecurity

To curb instability, incumbents responded aggressively to industrial action in the early 1980s. As noted above, the government responded to workers who had created urban discord by adopting a robust minimumwage regime for domestic workers, agricultural workers, industrial workers, and mine workers.49 The government also put in place a system of price controls that kept the price of essential foods (maize meal, cooking oil, bread, etc.) relatively low. However, unlike other African governments that achieved low domestic urban food prices by depressing crop prices, the strategy in Zimbabwe kept crop prices fairly attractive through regular increases and subsidies to food producers so that the urban consumer would not have to bear the full cost of food. This approach benefited the urban population through inexpensive food, peasant and commercial farmers through good returns on their crops, millers and other food producers who received hefty government subsidies and, of course, incumbents who could use this patronage to build immediate political support across very different segments of the population. Yet these policy fixes, for all the short-term gain they bought incumbents, also generated, as I will show below, longer-term costs, which contributed to the adoption of unpopular economic reforms in the 1990s that in turn mobilized a determined challenge to Mugabe’s grip on power. In addition to the minimum-wage legislation and price controls, the government extended benefits by providing free health care to those earning less than Z$150 per month (US$564 in 2005 terms) and their families.50 Through the National Nutrition Unit, Mugabe’s government initiated a Child Supplementary Feeding Programme, by which 250,000 undernourished children were provided with food rations through a network of over 8,000 rural-area feeding points.51 Incumbents also aggressively reformed access to education; the government constructed new schools and abolished fees for primary education (grades 1 through 7).52 In the first five years of this policy, primary school enrollment grew by over 200 percent, from 892,668 in 1979 to 2.7 million by 1985.53 The new government’s most extensive patronage transfers were seen in the agricultural sector. In 1980, Mugabe and ZANU inherited an agricultural sector dominated by white commercial farmers in a highly skewed land-ownership pattern. To secure their position in power, they needed to attract the support of the rural population most affected by this

The Pathologies of Drought Relief in Zimbabwe

95

unequal distribution of land, while at the same time ensuring that the white farmers remained productive. Challenged by these seemingly irreconcilable goals, the new rulers of Zimbabwe used the inherited Agricultural Marketing Authority and its organs, the Grain Marketing Board (GMB) and the Cotton Marketing Board (CMB), to transfer patronage to close to 5 million peasants while appeasing the 6,000 or so white farmers.54 In 1980-1981, the government offered farmers $Z89 (US$335 in 2005 terms) per Mt for maize. The following year, over the 1981-1982 season, Mugabe’s cabinet increased the price of maize to US$358 per Mt. Producer prices for maize increased five more times before the end of the decade. Generous maize prices did not just benefit Mugabe and ZANU’s emerging rural governing coalition, but were also useful in allaying the apprehensions of white commercial farmers who, until 1986, when maize production in Zimbabwe came to be dominated by peasants, produced the bulk of the domestic staple.55 In the first six years of independence, the government transferred close to US$1.86 billion to the 6,000 or so commercial farmers in the country compared to US$1.48 billion it spent to buy maize from 5 million peasant farmers over the same period. As Bratton notes in his study of agricultural pricing in Zimbabwe, the desire by incumbents to transform their coercion-based presence in the rural areas into patronage-based support also encouraged Mugabe and ZANU to offer generous prices for many other of the crops grown by peasants (including mhunga, rapoku, sorghum, groundnuts, and cotton).56 However, this strategy of creating political order was very costly for the government and produced fiscal burdens that forced incumbents to adopt economic reforms a decade later. Herbst shows that, to effect these clientelism-building transfers, the government-run GMB was required by statute to buy all maize peasants delivered.57 As a consequence, Herbst estimates that the GMB had a trading account deficit of US$88 million by the end of the 1987-1988 fiscal year. Drought Relief During Unstable Political Conditions

Despite winning 71 percent of the seats and close to 63 percent of the African vote in the Common Roll elections of 1980, Robert Mugabe and his party assumed office as very insecure incumbents, as I have demonstrated. Like Moi in Kenya, Mugabe and ZANU had a strong interest in finding ways to secure their positions. This section demonstrates that a drought from 1981 to 1983 provided the prime minister with a highly visible national crisis around which relief measures could be fashioned

96

Domestic Politics and Drought Relief in Africa

to both prevent famine and consolidate ZANU’s position. As Moi did in Kenya, insecure Mugabe used drought relief to redress his own political vulnerability. After the first two years of independence produced agricultural surpluses, Mugabe and ZANU exploited an extended period of dearth between 1981 and 1984 to resolve their vulnerability.58 As in Kenya, there was an intense debate in government over how best to respond to the threat of famine. Insecure ZANU cadres urged the government to respond immediately through a food aid program,59 insisting that the identification of beneficiaries and the distribution of relief rations be given to ZANU party heads at the village and district level. Instead of food aid, technocrats favored a form of labor-based relief, either food-for-work or work-for-cash, arguing that food aid would be wasteful and could lead to a culture of dependence.60 They further pointed out that a relief program involving works projects would resolve questions about access through self-selection; families unaffected by the drought or with sufficient resources might opt out of working for relief wages or food. They argued that the marginal drop in the village-byvillage number of households seeking assistance would aggregate into significant savings for the government. On the question of administration, technocrats argued it should become the responsibility of district level officials from the Ministry of Labour and Social Services and other relevant line ministries, not ZANU party cadres.61 They feared that a reliance on ZANU to identify beneficiaries and distribute food rations would introduce political considerations into the relief effort especially in in provinces such as Matabeleland, Midlands, and parts of Manicaland where ZANU faced serious challenges from ZAPU. Yet political considerations prevailed and the direct-food aid program championed by ZANU incumbents won out in 1982. As in Kenya in 1984, the main element of Zimababwe’s 1982-1984 drought relief program was the distribution of free take-home food rations to able-bodied adults in the rural areas, but it also included a supplementary feeding program for children under five and lactating mothers, a water-supply scheme, and cattle protection and seed and fertilizer provision for peasant farmers.62 Mugabe and ZANU used their existing control of the food market, through the marketing board system and price controls, to ensure that the urban areas that had been the center of industrial action in 1980 remained well supplied during the drought. While Kenya imported a significant portion of the grain used in its program, incumbents in Zimbabwe drew on maize stores held by the GMB. In 1982, when the government’s relief program started, the GMB

The Pathologies of Drought Relief in Zimbabwe

97

held over 2.7 million Mt of maize from the previous two years’ good harvests.63 Mugabe’s government provided food aid to approximately 850,000 people each month in 1982, a number that increased to 1.46 million the next year, and which cost Mugabe and his ZANU government over Z$64 million (roughly $141 million in 2005 USD). Concerned about paying for future relief programs, Zimbabwe’s finance minister introduced a temporary Drought Relief Levy on income and corporate taxes.64 In an early review of the relief program, Roger Leys concludes that “for those without access to cash and other entitlements . . . [free food rations were] their only food intake.”65 Supporting Leys, Weiner estimates that “during the 1982-84 period the government drought relief programme became the primary means of survival for about 2.5 million people.” 66 But these high beneficiary figures concealed the political manipulation of relief to secure Mugabe’s incumbency. ZANU party cadres at the village and district level assumed leading roles in determining access to relief and administering the 1982-1984 program.67 This offered incumbents the opportunity to drive home the worth of supporting ZANU. For example, cadres often asked rural residents to present their ZANU membership cards before adding their names to beneficiary lists. Beyond the identification of districts and beneficiaries, ZANU party cadres were also responsible for the day-to-day distribution of rations, which took place at public gatherings at the village level and were led by the local chairmen of the party. Party cadres, often clad in ZANU T-shirts, explained that free food during the drought was the party’s way of thanking villagers for their support during the liberation war. Slogans praising Mugabe (Pamberi naComrade Robert Gabriel Mugabe! Pamberi neZANU! – Forward with Comrade Robert Gabriel Mugabe, Forward with ZANU) would often be shouted before food was distributed.68 In provinces that voted for ZANU in the 1980 elections, incumbents used food aid to consolidate their political dominance and leverage their control of food to transform their violence- and intimidation-based acquiescence in the rural areas into a patronage-based support. Roger Leys confirms that relief was distributed fairly smoothly in provinces like Masvingo that were essentially one-party regions dominated by ZANU.69 ZANU won 84 percent of the vote in Masvingo in 1980 and with that secured all 11 parliamanetary seats assigned to the province. However, in Matabeleland and Midlands, where ZANU managed received 10 percent and 59 perent of the vote respectively in 1980 (see Table 4.1 below), the strategy was different. Mugabe’s 5th Brigade

98

Domestic Politics and Drought Relief in Africa

exploited the drought and the government’s relief operation and used food as a weapon, imposing a general blockade on commercial and private movement of food and assuming control over government food depots.70 In their quest for security, incumbents intentionally withheld food in Matabeleland while distributing free food through ZANU structures in other parts of the country. Creating Stable and Secure Personal Incumbency in Zimbabwe, 1985-1995

A year after the 1982-1984 drought relief program, Zimbabwe held its first post-independence elections. As in 1980, the electorate was split into a Common Roll with 80 seats and a White Roll of 20 seats. However, incumbents had replaced the province-based proportional representation system of 1980 with a single-member district structure; election to parliament was to be by simple plurality within each constituency rather than being determined by the proportion of votes received by each party.71 Changes to the electoral system gave incumbents an advantage by effectively raising the threshold for winning a seat to 51 percent, thus making it harder for smaller parties, and even ZAPU outside of its Matabeleland stronghold, to win against ZANU. Another advantage for ZANU was that delays in the government’s preparation of the Common Roll voters list and delimitation of constituencies resulted in the elections being postponed from March to June-July 1985. This delay left the parties with only 19 days to campaign before the poll was to be held. And, in fact, ZANU won 64 of the 80 Common Roll seats to ZAPU’s 15, and Mugabe was returned as prime minister with an even larger legislative majority.72 In the White Roll, Ian Smith’s Conservative Alliance of Zimbabwe (the new name for the old Rhodesian Front) won 15 of the 20 seats. Sithole argues that a ZANU victory was never in doubt.73 Agreeing, Kriger stresses the shortened election campaign period in 1985 as a ZANU advantage, marked by targeted intimidation of political challengers, particularly ZAPU followers and former members of its military wing.74 Violence and intimidation were also present in the townships surrounding Harare; ZANU cadres went from house to house recruiting supporters, and sometimes beat and burned the houses of those who refused to join. Although compelling, this explanation fails to acknowledge that Mugabe and ZANU also used their control of the most valuable resource during the drought, food, to continue to build and solidify their control of the state. It is notable that, as their share of the national vote grew by

The Pathologies of Drought Relief in Zimbabwe

99

14 percent, Mugabe and ZANU made their largest gains in regions that were at the center of the relief effort: 13 percent in Mashonaland East; 15 percent in Masvingo; 16 percent in Mashonaland Central; and 24 percent in Mashonaland West and the Midlands (see Table 4.1 below). Table 4.1: Change in Electoral Support for ZANU by Province between 1980 and 1985 Percent of Vote for ZANU in Province

Province

Vote Share Gain

1980 Elections

1985 Elections

Manicaland

82

87

5

Mashonaland Central

82

98

16

Mashonaland East

79

92

13

Mashonaland West

70

94

24

Matabeleland North

10

14

4

Matabeleland South

7

14

7

Midlands

59

83

24

Victoria (Masvingo)

84

97

15

Source: Government of Zimbabwe, 1985 General Election Report; and Sithole, “The General Elections, 1979-1985.”

Soon after the 1985 poll, Mugabe moved to firm up his gains by intensifying unity negotiations with ZAPU.75 Towards the end of 1987, ZANU and ZAPU signed the Unity Accord, which ended open hostilities and competition between the two nationalist parties.76 As part of the agreement, the two parties merged into ZANU-PF, with the PF (for Patriotic Front) added to mark ZAPU’s inclusion in the party. A number of senior ZAPU members were given prominent cabinet posts, including Joshua Nkomo, who became a senior minister in 1989, and after 1990 one of two national vice presidents. Moreover, a general amnesty was declared for all involved in the violence of the insurgency.77 This unification removed the only political party with nationalist credentials that posed a threat to ZANU incumbents.78 Similar to the voluntary dis-

100

Domestic Politics and Drought Relief in Africa

solution of KADU in Kenya, the unification of ZANU and ZAPU into ZANU-PF was a crucial step in Mugabe’s march toward his goal of assured incumbency by creating a de jure one-party state. In August of 1987, Mugabe moved to abolish the constitutionallymandated hold on 20 seats by white Zimbabweans. He proposed amending the constitution for the sixth time in seven years and Amendment No. 6 was duly passed in the lower house and ratified by the Senate in September, and 20 ZANU-nominated candidates (11 of whom were white) were indirectly elected by the legislature, giving Mugabe and ZANU-PF control of 99 of the 100 seats.79 Mugabe, however, did not rest on this parliamentary dominance. Quickly he demanded further changes to the constitution that would secure his own position and power. The constitutional system agreed to at Lancaster House in 1979 split executive power between a president, the nominal head of state and commander-and-chief of the army, and a prime minister, who was the substantive center of political power in the country. In October of 1987, Mugabe’s government introduced Amendment No. 7 (Act 23), which fused the ceremonial authority of the president with the governmental powers of the prime minister in an executive presidency, whose independence from the legislature was guaranteed by direct popular election to the new position.80 Mugabe’s quest to secure his own position by centralizing power through the creation of a presidential system is in line with how many post-independence incumbents attempted to create stable political order.81 The executive presidency effected three changes for Mugabe: he would cease to be merely the first among equals and be placed above the day-to-day parliamentary business and questioning he had endured with growing impatience as prime minister; his personal powers, including a number of presidential prerogatives that could not be legally challenged, were reinforced; and, most importantly for his own security in power, he now had the opportunity to win direct popular mandates and legitimacy. However, when the amendment was passed, the executive president was to be indirectly elected by the House. Mugabe was nominated as the sole candidate for the presidency and duly elected by parliament; on December 31, 1987, Robert Gabriel Mugabe was sworn in as Executive President of Zimbabwe. At the next general election, in 1990, he would have to run for popular election to a six-year term. Mugabe had two years to use the trappings of his new post to prepare for his first national contest at the polls. Elections in 1990 would be held under a new structure designed to further entrench ZANU-PF’s political hegemony. In 1989, incumbents passed another constitutional amendment—the 9th—abolishing the 40-member Senate and expanding

The Pathologies of Drought Relief in Zimbabwe

101

the size of the lower house from 100 to 150; 120 seats were to be elected by single-member district elections under universal suffrage, while the remaining 30 members would hold non-constituency seats as appointees of the president (10 traditional chiefs, 12 non-partisan members, and 8 provincial governors).82 The amendment also increased the power of ZANU-PF to enforce discipline in parliament by requiring members of the House who defected or were expelled from their party to vacate their seat. A new political party, the Zimbabwe Unity Movement (ZUM) led by Edgar Tekere, contested elections in 1990. Though a founding member of ZANU and close ally of Mugabe, Tekere was fired from his cabinet post, removed from the party’s central committee, and stripped of his chairmanship of ZANU in Manicaland Province for persistently speaking publicly about corruption and the progressive embourgeoisment of the ZANU-PF brass. Tekere also questioned Mugabe’s failure to adopt a more aggressive land reform program, thus leaving the issue of land distribution unresolved. In forming ZUM, Tekere called on his supporters to “get out there and bury the corpse of this rotten Government.”83 Jonathan Moyo contends that a great deal was riding on the outcome of the 1990 elections.84 General elections would serve as a referendum on the Unity Accord. Could Nkomo and the other former ZAPU members now running under the ZANU-PF banner deliver Matabeleland and add it to Mugabe’s governing coalition? Incumbents worried that ZUM might capture the votes of disaffected former ZAPU supporters opposed to the Unity Accord, which would have made the agreement meaningless, since ZANU-PF would again be obliged to compete with another party, and Mugabe would be matched against Tekere, a former ally with strong nationalist credentials of his own. Further deepening incumbent apprehensions was a scandal that came to be known as “Willowgate.” In 1988 and 1989, senior members of the government were implicated in a scheme to purchase cars at subsidized prices from the state-owned Willowvale Motors Industries and then sell them to private individuals at marked-up prices. The scandal forced Mugabe to appoint an independent panel, the Sandura Commission, to investigate. The commission’s report led to the resignation of Maurice Nyagumbo, senior minister of state for political affairs, in 1989. “Willowgate” gave ZUM a superb opportunity to capture the votes of those disillusioned by corruption in the government. However, the challenge ZUM mounted proved to be a shadow of the build-up; ZANU-PF candidates stood unopposed in 11 constituencies, and won 117 of the 120 parliamentary seats, tallying close to 80 percent

102

Domestic Politics and Drought Relief in Africa

of the vote.85 In Matabeleland North and South provinces, former ZAPU members helped ZANU-PF capture all of the 24 seats. Mugabe too won his presidential contest handily; Edgar Tekere only managed to gain 17 percent of the vote, while Mugabe swept the rest. The victory gave Mugabe the power to appoint the 30 non-constituency members to parliament as provided for by the 9th amendment to the constitution. These appointments gave ZANU-PF control of 147 of the 150 seats in the legislature. The resounding victories marked the beginning of a decade of complete political dominance in Zimbabwe by President Mugabe and his ZANU-PF. Relief During Periods of Political Security

What effect did the consolidation of incumbency have on Zimbabwe’s drought relief programs? The government’s response to poor rains over the 1986-1987 growing season, a year after Mugabe and ZANU had won re-election and brought stability to political contestation, offers evidence of the effects of secure incumbency on the interests that animated drought relief decision-making in Zimbabwe. In 1986-1987, the average annual rainfall across Zimbabwe was again about 220mm below the mean, resulting in a total cereal harvest of 1.5 million Mt in 1987, half of what the country had produced the previous year. As ever, rain-dependent peasant farmers suffered the most severe losses of food production.86 For instance, an estimated 100,000 head of cattle were dying as a result of lack of water and pasture. As hypothesized in Chapter 3, secure incumbents in Zimbabwe switched their relief program to labor-based relief in 1987. The reasons given for the switch by the prime minister’s government confirm the extent to which growing security in office enabled incumbents to take into consideration the technocratic arguments they had dismissed when they were more vulnerable. Dr. Frederick Shava, the Minister of Labour, Manpower Planning and Social Welfare, signaled the change in interests guiding drought relief policy-making when he suggested that people who received relief would soon be required to work on community development projects under a new food-for-work scheme. According to Dr. Shava, food-for-work was in accordance with the government’s new approach to social welfare, whereby activities would be oriented towards self-reliance, co-operative action, and citizen participation in development.87 Government was concerned, he maintained, with avoiding dependency and waste. Other members of Mugabe’s government echoed Shava. Addressing people in Kwekwe, Senator Tranos Makombe, the Midlands provincial governor, stated that government

The Pathologies of Drought Relief in Zimbabwe

103

was considering introducing drought relief work as a permanent feature of delivering development to people, instead of giving them handouts.88 Shava’s ministry then wrote two internal reports that paved the way for a policy change by concluding that the 1982-1984 drought relief program had been wasteful and had engendered a dependency syndrome among the rural population.89 Mugabe and his cabinet subsequently endorsed the reports and resolved that future drought relief would be laborbased, in the form of food-for-work for able-bodied adults. After 1986, incumbents in Zimbabwe were much less interested in using relief to thank the rural population for its support during the war and more concerned with cost-effectiveness and avoiding dependency. Thus, the new program of relief had two components: food-for-work for households with an able-bodied adult, and free food distribution for households lacking an able-bodied adult. The food-for-work component of the government’s program required that one able-bodied adult per household work on a local construction or maintenance project. Existing projects under provincial and district development plans became food-for-work projects, and some new projects were created. At this early stage in the government’s use of labor-based relief, however, undertakings capable of employing large numbers of people (unskilled and semi-skilled) proved to be difficult to create in the short time frame of the relief. Despite these difficulties, Munro notes that the food-for-work program created valuable rural infrastructure, such as school classrooms, clinics, and rural feeder roads even though “the quality of work, especially in dam building, was often poor.”90 In addition to changing the general program of relief and the criteria for inclusion, secure incumbents in Zimbabwe also transformed the administrative structures in line with their desire that the program be sustainable and avoid waste. For the free food distribution component of relief, bureaucrats took over from party officials. ZANU cadres who had been responsible for the identification of beneficiaries and distribution of food aid during the 1982-1984 relief campaign were replaced by line staff in the Department of Social Welfare.91 By shifting the responsibilities to the bureaucracy, incumbents sought to rationalize and distinguish their drought relief program from what it had been between 1982 and 1984. Secure incumbents also set clear administrative criteria for qualification to the free food distrubution component, removing the discretion that had allowed ZANU cadres to use food aid to reward supporters and punish ZAPU supporters between 1982 and 1984. The transition to a relief program in which able-bodied adults had to work in order to receive aid did not seem to compromise protection for

104

Domestic Politics and Drought Relief in Africa

those most in need of government support. The 1987-1988 drought relief program provided aid to 1.08 million people each month and transferred about 9.77kg of maize per month per beneficiary.92 The government spent US$36 million on the relief program and managed to reach about 20 percent of the population in the rural areas. 93 Incumbents drew on the Drought Fund to pay for it.94 The change of policy in 1986, from food aid to labor-based relief, challenges explanations that link the form of relief to agricultural performance.95 Alex de Waal, for instance, argues that “because Zimbabwe is structurally a grain-surplus country, the drought response was radically different from that of neighboring Botswana: it was primarily aimed at ironing out variability in the food supply, rather than maintaining rural incomes.”96 In de Waal’s view, countries that in non-drought years are typically surplus food producers will simply draw upon existing food stocks and adopt food aid, while deficit food countries will likely opt for a labor-based program. But Zimbabwe’s adoption of labor-based relief in 1986 reveals the problem with explanations that link relief programs with agricultural performance. After the 1981-1983 drought, the government raised the price of maize twice: from Z$120/Mt in 1983 to Z$140/Mt in 1984 and then again in the following year to Z$180/Mt. Maize sales to the GMB increased sharply from a low of 616,000 Mt in 1983, the worst year of the drought, to a record high of 1.8 million Mt in 1985. The stellar harvest of 1985 was followed by another in 1986, with maize deliveries reaching over 1.6 million Mt.97 When drought next afflicted Zimbabwe over the 1986-1987 season, the GMB had 1.8 million Mt of maize in stock, enough to distribute free food rations to everyone in the country for a whole year. Yet despite adequate food stocks, the interests produced by improved political conditions led incumbents to adopt labor-based relief. As Mugabe’s incumbency grew stronger, Zimbabweans saw changes in the relief programs he offered. In 1986, incumbents scrapped the food aid program and replaced it with food-for-work, and delegated the administration of relief to bureaucrats, instead of relying on ZANU cadres. The structure of the agricultural sector did not shape the form of relief in Zimbabwe. Instead, changed political conditions by 1986 led the same incumbents who had championed food aid four years earlier to switch to labor-based relief because political conditions now permitted the consideration of the instrumental factors that technocrats had cited and incumbents had dismissed in 1982. Thus, progressive security for Mugabe and ZANU produced a decision-making environment in which incumbents were not just concerned with the immediate effects of their intervention, but also with long-term outcomes.

The Pathologies of Drought Relief in Zimbabwe

105

Consolidating the Food-for-Work Program

A number of crises towards the end of the 1980s led to the adoption of a neo-liberal economic reform program, the Economic Structural Adjustment Programme (ESAP). First, Mugabe’s use of the two marketing boards to transfer patronage to key political constituents throughout the 1980s had saddled them with very large deficits. Jeffrey Herbst estimates that by the end of the 1980s, government-run agricultural marketing agencies had trading account deficits in excess of US$159 million.98 In addition, between 1980 and 1989 the GMB’s operating costs rose sharply, to nearly 45 percent of producer prices, while across all crops, marketing subsidies consumed about 50 percent of the agricultural budget.99 Second, the size of the civil service grew from 62,000 in 1980 to over 181,000 in 1989.100 By 1991 this figure stood at a staggering 191,000. Third, the stagnating economy was creating only about 70,000 jobs each year, an inadequate figure for the more than 200,000 school graduates or dropouts each year.101 World Bank- and IMF-supported ESAP reforms were announced in 1990. The primary objective was to reduce budget deficits and cut government expenditure. The government removed price controls on essential foods, reduced spending on subsidies, and relaxed import licensing and foreign exchange controls.102 Mugabe and ZANU-PF discontinued the corporatist minimum-wage regime they had used to recruit the working class throughout the 1980s. In the agricultural sector, reforms brought cuts in government “support for research and development” of new crop varieties, funding that had contributed to the agricultural burst of Zimbabwe’s first decade.103 The government also dramatically scaled back its generous pricing policy. Thus, as illustrated in Table 4.2, the prices offered for most crops, including the staple crop maize, were less in the 1990s than they had been in the 1980s. For the urban sector, the introduction of cost recovery in education and health and the removal of food subsidies and wage controls promptly produced severe hardships for the working class.104 In response to the immediate social costs of reform, unions organized a series of strikes in the early 1990s105 that led eventually to the collapse of the quasi-corporatist regime of the 1980s and the souring in the late 1990s of relations between the government and the ZCTU. In the next period of political insecurity, at the end of the 1990s, members of the labor movement would become a significant bloc in the opposition Movement for Democratic Change (MDC).

106

Domestic Politics and Drought Relief in Africa

Table 4.2: Average Maize, Wheat, and Sorghum Prices by Decade Decade

Average Price of Maize US$/Mt

Average Price of Wheat US$/Mt

Average Price of Sorghum US$/Mt

1980-89

$224

$379

$224

1990-99

$125

$221

$96

Source: Herbst, State Politics in Zimbabwe; FAOStat; and World Development Indicators.

To add to the hardships these economic reforms produced, Zimbabwe suffered its most severe drought in a century.106 Poor rains over the 1991-1992 agricultural season led to a rural harvest 90 percent below normal and left an estimated 5.6 million people, nearly 52 percent of the population, at risk of suffering famine.107 As the drought’s severity became apparent under conditions of deepening economic austerity, some analysts expected the government to switch back to food aid on the presumption that organizing food-for-work programs for half the population would overextend the capacity of the state.108 However, in a response announced in early 1992, food-for-work was retained, expanded, and reinforced to cover the entire country.109 The commitment by secure incumbents to respond to dearth using laborbased relief for able-bodied adults and the bureaucratic administration of relief persisted even when the country had to deal simultaneously with the immediate costs of economic reform and a devastating drought.110 Food-for-work projects included road and borehole construction, brick making, fencing, weir constructions, gully reclamation and other conservation measures, pipelines, and nutrition gardens.111 Over the course of the relief program, the number of applicants approved rose from a low of 873,118 in January of 1992 to 5.05 million by November 1992.112 This increase in beneficiaries led to a progressive decline in the payment ration from an initial official target of 15kg of maize per person per month (20 percent above the standard adult equivalent cereal requirement), to 10kg per person per month, and then to about 5.6kg per person per month (close to 40 percent below the adequate level of support) at the peak of the program. Secure incumbents decided to support the urban food market by importing grain on behalf of domestic millers. Between January 1992 and April 1993, the government imported over 2.4 million Mt of food at a cost of about US$355 million (in 2005 terms).113 Despite growing

The Pathologies of Drought Relief in Zimbabwe

107

security, Mugabe and ZANU-PF also decided on some exclusionary market interventions; they re-introduced maximum prices for essential foods (maize meal, bread, cooking, etc.) in the urban market.114 The government’s decision to adopt exclusionary interventions followed its post-reform recognition of the social costs of economic reform.115 Thus, incumbents hoped that price controls would keep food cheap for the urban population bearing the brunt of economic reform and forestall food strikes. Attempts were also made to prevent hoarding by limiting the number of essential foods each person could buy at a time. The third component of the 1992-1993 relief effort was a Child Supplementary Feeding Programme, through which children under 5 and primary school children were eligible for free food aid.116 The government decided early that blanket supplementary feeding would be provided to any village where over 15 percent of children under 5 were found to be malnourished, using an upper arm measurement. As in 1986, this set clear administrative criteria for the distribution of food aid in Zimbabwe. Children under 5 were provided with rations amounting to 40 percent of their daily caloric needs (60 grams of maize meal and 20 grams of beans per day per child) using a network of over 18,819 feeding points across the country.117 In addition, school lunches were provided for grades 1 through 3 in designated schools during the lean period of January to March 1993. An estimated 350,000 children per month were given school lunches under this program. The administration of the feeding program for children was delegated to local committees of four women, typically mothers of children in the program, who rotated cooking shifts. This administration by stakeholder volunteers proved to be successful. Overall, the feeding program’s coverage was extensive and few problems of accountability were reported at the village level.118 The relative efficacy of this locallyadministered component of the 1992-1993 relief program parallels the success of the locally-managed health provision system discussed by Judith Tendler.119 When drought struck Zimbabwe in 1991-1992, Mugabe had been comfortably elected executive president and ZANU-PF held 98 percent of the seats in parliament. Yet the scale of the task the government faced, with over half of the population needing assistance, had led many to presume that it would switch back to food aid. Instead, the prevailing political security reinforced the earlier transition to food-for-work and produced one of the most expansive labor-based relief programs on the continent. Even though the 1991-1992 drought presented the government with its most severe food crisis, protection continued to be through food-for-work for all able-bodied Zimbabweans.

108

Domestic Politics and Drought Relief in Africa

Political Insulation, 1995-1999

Political developments in Zimbabwe in the mid-1990s point to the perverse effects of political insulation and security on programs of relief. ZANU-PF and Mugabe’s romp to victory in the 1990 elections marked the beginning of a political environment which, for about a decade, saw no obvious challengers to their incumbency. In fact, in the run-up to the Unity Accord in 1989, Mugabe had stressed that he and ZANU were “anxious to prove that the one-party parliamentary democracy is the best form of democracy for Zimbabwe.”120 Thus, following the 1990 elections, Mugabe and his party took steps to establish a de jure oneparty state.121 However, unlike what we saw in Kenya, this initiative failed, in part because Mugabe’s well-known intentions led to mobilization by segments of domestic civil society and pressure from the international community. At a time when the government was embarking on a World Bank-funded economic reform program, Mugabe and ZANU-PF abandoned the legislated one-party state agenda. The failure of that project led some scholars to foresee an imminent crack in ZANU-PF’s hegemony. For instance, in a 1997 article in the Journal of Democracy, Masipula Sithole boldly asserted that, “Authoritarianism in Zimbabwe is eroding. This erosion began to be visible in 1991, ironically quite soon after authoritarianism had reached a peak of sorts in the general and presidential elections of 1990.”122 Sithole and other scholars were impressed and charmed by the broad coalition opposing a one-party state in Zimbabwe. The coalition included ZANUPF backbenchers, members of civil society, university lecturers and students, human rights groups, international groups, and what remained of opposition parties. Ever optimistic, Sithole suggested that: “Although opposition parties are weak, a strong civil society is emerging and should eventually be able to provide a basis for viable and democratic political parties.”123 Yet Mugabe’s decision to abandon the one-party state project did not portend the collapse of ZANU-PF’s hegemony, nor did civil society immediately mature into a viable opposition by the next election. Instead, for roughly a decade after Sithole predicted the party would weaken, Mugabe and ZANU-PF enjoyed a level of political dominance that eluded Moi and KANU with their de jure one-party state system. Although multi-party politics remained enshrined in the constitution, in practice ZANU-PF incumbents controlled and dominated the political landscape and tightly guarded political opportunities against challengers. Civil society remained weak and exclusively urban in its reach, while opposition parties were too narrow, fragmented, and poor to worry

The Pathologies of Drought Relief in Zimbabwe

109

ZANU-PF under a first-past-the-post system. As a result, challenges to ZANU-PF and Mugabe became more and more futile. Although Mugabe’s pursuit of a de jure one-party state failed, what he achieved was a de facto one-party system in which incumbents enjoyed unparalleled security for about a decade and a half. The 1995 parliamentary elections revealed the stability of that oneparty state, when ZANU-PF won 118 out of 120 seats. Opposition parties were intimidated and confused, fielding candidates in only 65 of the 120 constituencies. In effect, the only major contest in close to half of the openly contested constituencies was the ZANU-PF primary. With 55 unopposed seats plus the 30 seats Mugabe as president appointed, ZANU-PF had a majority in parliament (85 of 150 seats) even before the polls opened on April 8, 1995. When they closed the next day, the results showed ZANU-PF candidates winning 63 of the 65 contested seats. Christine Sylvester was quite justified to tease that “on 8 and 9 April 1995, Zimbabweans turned out for an election that mostly was not.”124 The mirage of a challenge to ZANU-PF in the 1995 parliamentary elections was topped by the triviality of the presidential election the following year. Ndabaningi Sithole of ZANU-Ndonga and Abel Muzorewa of United Parties challenged Mugabe for the presidency. What little chance they had to threaten the incumbent was foreclosed when Sithole was arrested and charged with an alleged assassination attempt on Mugabe in 1980. Throughout the campaign period, Sithole was under house arrest, and a few weeks before the presidential poll, both he and Muzorewa withdrew from the contest. Their names were left on the ballot, to preserve the illusion of an election. Mugabe was declared the winner with 93 percent of the votes to 5 percent for Muzorewa and 2 percent for Sithole.125 Thus, although the legislated one-party state project was abandoned, ZANU-PF and Mugabe became even more secure between 1995 and 1999. The Pathologies of Political Insulation

Complete insulation from political vulnerability negatively affected the relief programs the government offered. With ZANU-PF assured of victory in the coming election, President Mugabe announced in March of 1995 that all previous drought relief programs (food-for-work and free food distribution) were to be replaced by a grain loan program,126 an idea with origins in a campaign stop Mugabe had made in Matabeleland. There the president entertained a delegation of chiefs who complained about the relief programs the government had offered since indepen-

110

Domestic Politics and Drought Relief in Africa

dence. They suggested that, even when beneficiaries were asked to work, drought relief was creating a dependency syndrome among rural households, while qualification guidelines were engendering individualism and undermining traditional community responsibility.127 To address these problems, the chiefs proposed a return to what they framed as a more traditional drought relief response: community-based loans administered by traditional leaders. After this consultation in Matabeleland, a region that ZANU-PF had long had trouble folding into its ruling coalition, Mugabe took up the issue with his cabinet. The cabinet, in turn, instructed the Ministry of Public Services, Labour and Social Welfare (in charge of previous drought relief programs) and the Ministry of Local Government, Rural and Urban Development (in charge of chiefs and kraal heads) to come up with a program that addressed the chiefs’ concerns. The result was the Grain Loan Scheme,128 which would be adopted three times between 1996 and 1999. The Grain Loan Scheme (GLS) adopted by insulated incumbents changed the general relief program, altered qualification criteria, and inserted a new layer of personnel into the administration of relief. First, the new policy imposed strict administrative criteria for aid eligibility. After 1995, only areas the Department of Agricultural Extension declared drought-affected were eligible for relief. Second, whereas foodfor-work between 1986 and 1992 had established a system of access to relief predicated on individual/household need, the GLS introduced group qualification and collective responsibility into the government’s program as a clear attempt to address chiefs’ concerns that individualism was undermining traditional community values.129 Households in need of relief were required to form grain loan groups and then apply for aid from chiefs and kraal heads. Approved groups would receive about 10 kg of maize per person per month. At the end of a drought period, borrowers were required to repay the government in kind after the following year’s harvest. In the event that the same area was again declared drought-affected, a further grain loan could be granted and repayment of the initial loan deferred for a year. If an area was droughtaffected for three consecutive years, the initial grain loan would be forgiven.130 In 1996-1997, the government loaned close to 435,000 Mt of maize at a cost to the government of about US$43 million (in 2005 terms). The following year, the persistence of drought in some districts resulted in the distribution of 224,000 Mt of maize at a cost of US$39 million.131 In the last year, 1998-1999, the government loaned out about 166,000 Mt of maize, valued at US$16 million.

The Pathologies of Drought Relief in Zimbabwe

111

An insulated Mugabe also used the GLS and its administrative structure to recruit chiefs into his governing coalition. Chiefs and kraal heads were given the responsibility of identifying and approving loan groups, and often used this newfound responsibility to increase their power.132 Once traditional leaders approved a loan group, bureaucrats in the Department of Social Welfare (DSW) in the Ministry of Public Service, Labour and Social Welfare were then required to purchase maize from the GMB and deliver it to the chiefs and kraal heads for distribution to beneficiaries. Under this arrangement, those in charge of spending the Department of Social Welfare’s budget allocation were not under their authority; chiefs at the local level were effectively given power over a portion of that budget. Problems with the GLS were soon apparent. Government relief in the form of loans discriminated against poorer rural households,133 the same households who were core ZANU-PF supporters since independence. Better-placed to repay and with much stronger political influence over the kraal heads and chiefs at the local level, wealthy rural households had significant advantages under the grain loan program. In this sense, the program was highly regressive. In fact, “evidence … soon emerged that many of the better off in rural areas were getting Grain Loans, selling the grain, and using this interest-free loan as seed capital for other commercial ventures, thus by-passing the high real interest rates on bank loans at the time. In effect, the grain loan program was subsidizing speculation in various futures markets.”134 Another problem with the scheme concerned the rules and mechanisms for the repayment of the loans. The GLS required farmers to repay them in kind, but did not specify the type or quality of the grain to be used. Well-positioned farmers took advantage of this loophole. Some made their repayments in lower quality maize, grade C or D, even though all government grains were grade A or B. Other farmers took this a step further and chose to repay their loans in less desirable grains like millet and sorghum. In some circumstances, chiefs and kraal heads collected such poor quality grain that the GMB refused to accept it.135 In April 1999, just as the political fortunes of Mugabe and ZANU began to change, the grain loan scheme was canceled. In their first joint announcement on the relief program, the Minister of Local Government and the Minister of Public Service indicated that it was being eliminated because few had been able to repay the loans even as many who did not deserve grain had been receiving it.136 As this section shows, the complete thinning of any real political challenge produced a political context in which incumbents could pander to narrow sectional interests, nega-

112

Domestic Politics and Drought Relief in Africa

tively affecting the relief program without immediately risking their position. Secure Incumbency Crumbles, 2000 and Beyond

Not long after the GLS was canceled, the assured incumbency that Mugabe and ZANU-PF built throughout the 1990s collapsed dramatically. At the turn of the century, ZANU-PF’s dominance started to erode in the face of new challengers, and incumbents’ defensive efforts to keep themselves in office sent the country into deeper crisis. The electoral threat posed by the Movement for Democratic Change (MDC) produced a series of responses from Mugabe and ZANU-PF—farm invasions, economic mismanagement, and political repression—that left the country polarized and stalled any hope of successful political reform.137 Patrick Bond and Masimba Manyanya suggest that cracks in the fortress that Mugabe and ZANU-PF built were evident much earlier as a result of the contradictions produced by exhausted nationalism and the social cost of ESAP, the unsuccessful neoliberal reform.138 ESAP was officially abandoned in 1998 and nominally replaced by the homegrown Zimbabwe Programme for Economic and Social Transformation. Although launched with a lot of pomp and ceremony, the homegrown reform program nevertheless relied heavily on support from international donors. In the end, the program was never implemented due to differences between the government and donors. According to Bond and Manyanya, by the late 1990s, ZANU-PF’s exclusive claim to power on the basis of the liberation struggle held little sway for an increasingly young and urban electorate that had benefited little from the failed reforms of the early 1990s. This explanation touches on one of the long-term recruitment challenges Mugabe and ZANU-PF faced, but does not adequately address the proximate causes of the return of political vulnerability. I argue that the decline of ZANU-PF and the shredding of Mugabe’s political insulation were produced by three successive developments, which revealed deep divisions within ZANU-PF’s governing coalition and exposed Mugabe’s regime to new, coordinated challenges. The first development was civil society’s push for a new constitution in the late 1990s. This middle-class agitation culminated in the formation of the National Constitutional Assembly (NCA) in December 1997,139 which called for three fundamental changes to the constitution: limiting the powers of the president, and especially the temporary presidential powers, which covered all policy areas, were effective for six months, and could not be

The Pathologies of Drought Relief in Zimbabwe

113

challenged in court; adopting presidential term limits; and creating an independent electoral commission.140 As it was initially composed, this coalition was of no real political threat to ZANU-PF incumbents. The NCA energized civic groups based in the capital, a thin network of human rights groups, university intellectuals, and college students.141 Because it lacked both the support of a real political constituency and the prospect of obtaining one, the NCA was initially dismissed by critics like Edson Zvobgo, a ZANU founder and former cabinet minister, who described it as nothing but “a small grouping of individuals sitting under a tree thinking they can come up with a Constitution for Zimbabwe.”142 However, policy choices in other areas of contention unwittingly transformed the NCA into a real problem for the government. By the late-1990s, the hardships caused by failed economic reform were so widespread and deep that even core ZANU-PF coalition members were suffering. As the formal economy continued to shrink throughout the 1990s, veterans of the liberation struggle, especially those who had not been absorbed into the army, police, security forces, or other departments of government, used their privileged position in the ZANU-PF coalition, large numbers (over 60,000 war veterans), and strong organization (the Zimbabwe National Liberation War Veterans Association) to press Mugabe and his government for patronage,143 thus igniting the second development leading towards political vulnerability in Zimbabwe. The war veterans demanded compensation and other social benefits for the sacrifices they had made during the struggle for independence and, in 1997, this agitation came to a head when they decided to flex their muscles by staging public demonstrations outside ZANU-PF headquarters in Harare. As Kriger notes, the government had a hard time dealing with the crisis because of the support the protestors enjoyed from fellow veterans in the police and army, who stood to gain from any concessions the civilian vets won. Fearful, perhaps, of the damage unhappy war veterans could cause to ZANU-PF unity and therefore to the government’s stability, Mugabe unilaterally decided to pay vets a lump-sum compensation of Z$50,000 (US$3,270 in 2005 terms) plus a monthly pension of Z$5,000 (US$327 in 2005 terms).144 The decision sent a shock wave through the economy. It was not clear how the government would meet its immediate obligation of roughly US$210 million, or the subsequent monthly commitment of roughly US$21 million. Fearing that the only way Mugabe could do so would be by printing money, businesses and wealthy individuals dumped the Zimbabwean dollar and sought security in foreign currencies. In what some scholars have come to call “Black Friday,” the value of the local local dollar fell by over 200 percent (from

114

Domestic Politics and Drought Relief in Africa

about Z$10 to over Z$35 to one US dollar) in a few hours of trading on November 14, 1997.145 Faced with a run on its currency, the government announced that it would pay for the war veteran compensation by introducing a War Veterans Levy on wages and salaries and by raising taxes on fuel, goods, services, and electricity.146 The economic crisis caused by Mugabe’s decision to compensate all registered war veterans pushed the increasingly assertive Zimbabwe Congress of Trade Unions (ZCTU), led by Morgan Tsvangirai, into the streets in protest,147 signaling the third domino to fall in Mugabe and ZANU-PF’s previous stronghold on political security. Noting that nearly three-quarters of the working population earned less than the veterans’ monthly pension award, the ZCTU organized a national stay-away on December 9, 1997. That led to more strikes throughout 1998 and placed the war veterans and the ZCTU in direct opposition with each other. In the process, Mugabe and ZANU-PF lost what was left of their labor support as the imposition of the war veteran levies placed the burden squarely on workers’ shoulders. Significantly, the breakdown of relations between government and the ZCTU propelled Tsvangirai and the labor movement into an alliance with the NCA, whose proposed constitutional reforms seemed to be part of the solution to workers’ bread and butter issues. ZCTU officials noted that if the country had sufficient checks on the powers of the president, Robert Mugabe would not have been able to award compensation and pensions to war veterans without broader debate and consultation. As links between the NCA and the ZCTU grew, dismissing the calls for a new constitution was becoming less and less palatable for Mugabe’s government. In response to this new, electorally-viable coalition, the government switched course and decided to try to take control of the reform process.148 The fear for Mugabe was that a civil-societydriven constitutional discussion might spin out of control and turn into something akin to the national conferences that had cost some incumbents their positions in Francophone Africa.149 On April 26, 1999, Mugabe attempted to regain the initiative by appointing “a 400-member Constitutional Commission charged with setting in motion a process that would produce a draft constitution to be submitted to a national referendum.”150 The strategy had unintended consequences. The appointment of a commission legitimized civil society’s claim that the country needed a new constitution, but the composition of Mugabe’s commission delegitimized its product. Presented with a draft constitution they had had little input in composing and about which they had strong reservations, the NCA and Tsvangirai’s MDC campaigned against its adoption. The

The Pathologies of Drought Relief in Zimbabwe

115

referendum was held on February 12-13, 2000, and although turnout was low, with only about 30 percent of the electorate participating, 54 percent of voters rejected Mugabe’s constitution.151 For the first time since independence, he and his party had lost an election. Masipula Sithole correctly concluded that the “defeat of the February referendum on the draft constitution marked a turning point in Zimbabwean electoral politics.”152 The referendum result caught ZANU-PF by surprise: “We are still trying to recover from the shock,” conceded one senior member of the party,153 and it was a solemn Mugabe who addressed the nation shortly after the results were announced: “Let us all, winners and losers, accept the referendum verdict and start planning our way forward.”154 But the stakes were high for ZANU-PF incumbents, as parliamentary elections were scheduled to be held two months after the referendum, in April 2000. The party knew it needed to find a response to the mobilization and coordination that the MDC had demonstrated in the referendum, and so it was within a week of its defeat, on February 18, 2000, that ZANUPF held an emergency meeting of its Central Committee. Opening the proceedings, Mugabe began the search for an answer: “It has been suggested and justifiably so, that part of the “No” vote [in the referendum] was in protest against economic hardships people are currently facing. This is a voice I believe we should listen to.”155 After the emergency meeting, and in a decision that exposed ZANU-PF’s new sense of political vulnerability, Mugabe postponed the date of the 2000 elections from early April to late June. The postponement was, transparently, intended to give ZANU-PF time to strategize, but it also revealed the seriousness with which ZANU-PF and Mugabe now were taking the MDC challenge. In the weeks and months following the referendum defeat, the two distinctive pillars of ZANU-PF’s strategy to retain power took shape. First, with increasing regularity, violence, and coordination, whiteowned farms were occupied by war veterans, causing chaos and confusion amid which land was haphazardly carved up and offered to potential political supporters in what incumbents described as the jambanja (or violent overthrow). The argument Mugabe and ZANU-PF mounted was that this was action long overdue, based in the call for radical land reform after independence.156 Incumbents’ turn to the emotive, nationalist, and polarizing question of land reform as a campaign issue was intended to breathe life into the old rural electoral base they had let wilt at the height of their political dominance. It was also, as suggested by a number of scholars, an attempt by Mugabe and ZANUPF to use “patriotic history” to resuscitate their exclusive claims to

116

Domestic Politics and Drought Relief in Africa

power.157 According to Tendi, “Patriotic History asserts the centrality of Zimbabwe’s radical revolutionary tradition and it is premised on four themes: land; race; a dichotomy between sell-outs and patriots; and the rejection of Western interference based on what are perceived as Western ideals such as human rights.”158 Mugabe and ZANU-PF increasingly drew on their role in the liberation struggle to legitimate their opposition to the prospects of an MDC government. The second strategy may have received less press than radical land reform: Mugabe’s government launched an aggressive campaign of intimidation against the opposition, a campaign of terror.159 Masipula Sithole estimates that this campaign against the MDC and its supporters “left 32 people dead and thousands of villagers homeless.”160 The violence was intended to force opposition party candidates, activists, and potential (or just suspected) supporters out of a given constituency, thus increasing the likelihood that ZANU-PF would be successful in retaining its seat. In this regard, the two strategies reinforced each other as the occupation of white-owned farms and displacement of farm laborers was also intended to rob the MDC of a significant portion of the votes that had helped them win the referendum. Despite the escalation in incumbent-led violence and intimidation, the MDC did three things previous opposition parties had been unable to do: they put up candidates in all 120 constituencies; they took advantage of their strong urban labor support to challenge ZANU-PF in constituencies in which it had historically been weak;161 and they campaigned heavily in Matabeleland, a part of the country where ZANU-PF had a complicated history and had struggled to gather support even during its most secure times. The results of the parliamentary elections held in July 2000 confirmed ZANU-PF’s changed political standing: The MDC won 57 of the 120 contested seats, ZANU-PF retained 62 seats, and one went to an independent candidate. The previously invulnerable ruling party lost 55 seats, and overall, the newly-formed MDC received 1,171,051 votes compared to ZANU-PF’s 1,212,302. In sum, Mugabe’s failed strategy to co-opt and diffuse the civic push for a new constitution, followed by ZANU-PF’s defeat in the referendum in February 2000, marked the erosion of their political insulation. Five months after the referendum, the gains made by the MDC in the general election confirmed the change in the country’s micropolitical conditions. The genie was out of the bottle. Mugabe and ZANU-PF were vulnerable and unable either to make a successful transition or to find a path back to their old domination. Zimbabwe slipped into a netherworld characterized by the collapse of the rule of law and the systematic erosion of most formal political, economic, admini-

The Pathologies of Drought Relief in Zimbabwe

117

strative, and social institutions. In the ensuing decade, the inconclusive face-off between ZANU-PF and the MDC entrenched a political and social environment in which insecurity reigned. As a result, disorder and short-term political considerations prevailed right up to 2004, when Mugabe sent shockwaves through drought-relief operations, ordering all aid agencies to stop their work in the country and asserting that a healthy harvest season was on the horizon. Relief During Political Collapse

In early 2000, just as Zimbabweans were getting ready for the referendum vote, the country had a relatively healthy food balance with an anticipated total grain surplus of some 146,308 Mt expected at the end of the 2000-2001 season. However, three developments altered the country’s food security status, providing insecure incumbents with yet another opportunity to use relief to strengthen their position. First, the vulnerability of Zimbabwe’s agricultural system to drought had increased at the turn of the century because of what Rukuni et al. describe as “the real and perceived failures of experimentation with liberal policies in the form of the economic structural adjustment programme.”162 Significantly, the austerity measures of the reform period undermined many of the support programs that had produced the agricultural miracle of the first decade of independence.163 Second, the politically-motivated land occupations in the aftermath of ZANU-PF’s referendum defeat shocked an already weakened agricultural sector. The chaotic early phase of the radical land reform program saw many commercial farms remain unploughed as white farmers, fearing that their land would be occupied and resettled, held back from investments in new planting.164 Finally, drought returned in 2001 and persisted until 2004. Thus, after 2001, Zimbabwe could no longer grow enough food to meet its domestic requirements. According to the Famine Early Warning System Network, Zimbabwe had a grain deficit in excess of 1 million Mt in 2002 and an estimated 7.18 million people, or 52 percent of the total population, were in need of drought relief if widespread starvation was to be prevented. Under political conditions reminiscent of the early independence period (narrow bases of support and the presence of strong challengers), insecure incumbents again opted to respond with food aid to a drought that would leave the country facing its severest food emergency in close to two decades. Amid their political spiraling, Mugabe and ZANU-PF incumbents returned to food aid to breathe life into their collapsing rural electoral coalition. They also reintroduced the political selection of

118

Domestic Politics and Drought Relief in Africa

beneficiaries and a reliance on ZANU-PF personnel in the administration of relief at the local level. Most notably, the post-2000 food aid program in Zimbabwe relied on war veterans, the key actors in the occupation of farms, to identify and distribute food rations to the rural population. As the political stalemate between ZANU-PF and the MDC intensified and the economic collapse deepened, further entrenching political insecurity for incumbents, the politicization of food assistance expanded beyond the distribution of food aid. The government leveraged its control of the grain marketing system through the GMB, amid reports of individuals being asked to produce ZANU-PF party cards to be permitted to buy grain from local depots.165 This marked an explicit return to the ZANU-PF strategy of the early 1980s, when party membership or support was a prerequisite for relief. Also, as in 1982-84, the distribution of patronage in this way was targeted to individuals, villages and locales that were viewed to be loyal to the ZANU-PF government. To increase the efficacy of his political use of access to food, Mugabe and his ZANU-PF government declared the drought broken in May 2004166, challenging an April 2004 report by the Zimbabwe Vulnerability Assessment Committee (ZimVAC), a food security evaluation team mandated by the regional body, Southern Africa Development Community, suggested that the threat of famine remained, as 2.3 million Zimbabweans needed food assistance to survive until the next harvest. In the coming months, it became clear that Mugabe’s declaration of a healthy harvest and his orders to stop the distribution of relief were intended to restore his monopoly over food and provisioning in Zimbabwe in the run-up to the 2005 elections. Anticipating another strong electoral challenge from Morgan Tsvangirai and the MDC, Mugabe wanted to clear the food-insecure rural areas of all other actors, domestic and international, so that he could effectively deploy the partisan war veterans to identify beneficiaries and distribute food rations to his party’s advantage. In the absence of alternative sources of food, war veterans strategically targeted food aid to electorally important rural constituencies.167 This strategy systematically kept marketing depots in districts deemed loyal well supplied with grain, but those areas that had shown support for the opposition in previous elections were understocked.168 As the country geared up for parliamentary elections in 2005, Mugabe and his ZANU-PF government were intent on showing the rural population the real price for supporting the opposition Politically-manipulated drought relief, fraud, intimidation, and violence won those elections for Mugabe’s party.169 ZANU-PF captured 78 of the 120 contested seats to the MDC’s 41. Combined with the 30

The Pathologies of Drought Relief in Zimbabwe

119

non-constituency members that Mugabe, as president, could appoint to parliament, the gains in the elections gave him the two-thirds majority. After the March 2005 elections, incumbents quietly conceded that the country was facing a serious threat of famine from the poor 2004 harvest and permitted the World Food Programme to resume its aid distributions across the country.170 Thus, the return of political vulnerability for Mugabe and ZANU-PF after 2000 produced the expected change in programs of relief: insecure incumbents once again sought to use food aid and other components of drought relief to breathe life into their collapsing rural electoral coalition. As the political stalemate between ZANU-PF and the MDC persisted and the economic collapse deepened, further entrenching political insecurity for incumbents, the quest for the slightest advantage also intensified. Summary and Comparisons with Kenya

This chapter revealed that over the course of his thirty-plus years in power, Robert Mugabe had to respond to dearth under various micropolitical conditions. It demonstrated that he assumed power in the early 1980s as an insecure prime minister operating under an unstable political settlement, which generated strong incentives to find immediate measures to assist him and his political party in fashioning stable incumbency. As in Kenya, insecure incumbents in Zimbabwe used the political opportunity created by the 1981-1983 drought to adopt a relief program, which, while it responded to the threat of famine, also explicitly addressed Mugabe’s desire to build support, eliminate challengers, and legitimate ZANU. Thus, the program of relief adopted by insecure incumbents in Zimbabwe provided food aid to able-bodied adults in the rural areas, it imposed both complementary and exclusionary interventions in the market for food, and it used ZANU cadres at the local level to administer relief and gave them discretion in identifying beneficiaries of food aid. In Zimbabwe, as in Kenya, this selective distribution of food aid using ZANU cadres permitted the new rulers to build, expand, and consolidate their rural support, while interventions in the food market kept the cities fed and quiet. The drought and the relief program incumbents adopted extended the power of the state and the ruling party over provisioning. Beginning in 1985, as incumbent instability and insecurity gave way to a more steady and assured hold on power, Mugabe and his ZANU government became less motivated by immediate political gains in their drought relief decision-making. Secure incumbents started to articulate

120

Domestic Politics and Drought Relief in Africa

their preference for cost-effective drought relief programs that did not engender dependency. I showed that this change in the interests guiding incumbents in their resolution of the four challenges produced by provisioning crises led to quite different relief programs when droughts in 1986-1987 and 1991-1992 brought threats of famine to Zimbabwe. During this period of political stability and security, Mugabe and his government shifted away from food aid for able-bodied adults in favor of a food-for-work program, a policy that introduced self-selection as the targeting mechanism. In addition to changing the general policy and the terms of access, secure incumbents also chose not to rely on party officials to administer relief. In 1987-1988 and again in 1992-1993, the food-for-work program was administered by line staff in the Department of Social Welfare. In the third period, incumbents in Zimbabwe achieved excessive security and insulation from political vulnerability, and I demonstrated the negative consequences of this for drought relief. The thinning of political challenges between 1995 and 1999 produced few incentives for comprehensive relief, and excessive security created a political context in which incumbents could pander to narrow sectional interests in drought relief policy-making. Thus, in 1995, the government adopted the Grain Loan Scheme to aggrandize a fairly narrow constituency, chiefs and kraal heads, at the expense of the larger rural population. In so doing, excessive insulation led to the adoption of a relief policy that was minimal in reach and regressive in outcome. The seemingly impenetrable hegemony that Mugabe and ZANU enjoyed for almost 15 years collapsed in dramatic fashion beginning in early 2000. In a short period of time, the protracted and polarizing political contest between Mugabe and his ZANU-PF and Morgan Tsvangirai and his MDC produced the systematic collapse of most formal political, economic, administrative, and social institutions. This new political environment of instability and insecurity again bred strong interests to quickly fashion popular bases of support using whatever policy instruments were available. As in the early 1980s, a drought between 2002 and 2004 provided Mugabe and ZANU-PF with an additional policy opportunity to defeat their challengers, Tsvangirai and the MDC. The beleaguered incumbents turned to a highly politicized form of food aid as a relief strategy in hopes of again using the visibility of this form of patronage to win over rural supporters. It is clear from this case study that the strength or weakness of Mugabe’s incumbency over his three-plus decades in power was the central factor determining the type of relief he extended to able-bodied adults. The analysis therefore supports the argument that drought relief

The Pathologies of Drought Relief in Zimbabwe

121

programs change in accordance with the interests generated by prevailing micro-political conditions, as I argued in my analysis of Kenya in Chapter 3. Because of the political similarities between Kenya and Zimbabwe, the insights from these two cases may not be conclusive. Therefore, in the next chapter, I examine Botswana, which, as a political and economic outlier, provides an opportunity to check whether the mechanisms associated with the strength or weakness of political incumbency affect drought relief policy-making in other contexts and within other types of governments. Notes 1

“Plea to Media on Drought Victims,” Chronicle (July 2, 1992). Sky News interview of Robert Mugabe, U.K., May 24, 2004. 3 “Zimbabwe Predicts Good Harvest,” Herald (May 12, 2004). 4 “Zim Will Not Require Food Aid: Mangwana,” ibid. (May 14, 2004). 5 Zimbabwe Vulnerability Assessment Committee, Zimbabwe Rural Food Security and Vulnerability Assessments – April 2004 Preliminary Report. This estimate was consistent with the food insecurity trend evident in the Famine Early Warning System Network’s Zimbabwe Monthly Food Security Updates for the first four months of 2004, and was subsequently confirmed in Zimbabwe Vulnerability Assessment Committee, Zimbabwe Rural Food Security and Vulnerability Assessments – June 2005 Report, Report No. 5. 6 See Hunger Project, “Statement on the Situation in Zimbabwe.” http://www.thp.org/what_we_do/key_initiatives/honoring_africa_leadership/ laureate_list/zimbabwe_statement. 7 Sen, Development as Freedom, pp. 169-84. 8 Mandaza, “The Political Economy of Transition,” pp. 1-20; Preston, “Stalemate and the Termination of Civil War,” pp. 65-83; Treverton and Levy, Rhodesia Becomes Zimbabwe; Stedman, The Lancaster House Constitutional Conference on Rhodesia; Soames, “From Rhodesia to Zimbabwe,” pp. 405-19. 9 For a fuller discussion of the elements of the Lancaster House Constitution of 1979, see, for instance, Mandaza, “The State in Post-White Settler Colonial Situation,” pp. 21-74. 10 For the growth and development of the nationalist movement in Zimbabwe, see Sibanda, The Zimbabwe African People’s Union, 1961-1987; Nkomo, Nkomo; Sithole, Zimbabwe; Sibanda, “Early Foundations of African Nationalism,” pp. 25-49; Bhebe, “The Nationalist Struggle, 1957-1962,” pp. 50115; Sithole, “The General Elections, 1979-1985,” pp. 75-97; Slater, “The Politics of Frustration”; Ranger, The African Voice in Southern Rhodesia, 18981930; Sithole, African Nationalism. 11 Rich, “Legacies of the Past?,” pp. 42-55. 12 Quoted in Sithole, “The General Elections, 1979-1985,” p. 85. 13 Rich, “Legacies of the Past?,” pp. 42-55. 14 “Mugabe Home Attacked,” Herald (February 7, 1980). 2

122

Domestic Politics and Drought Relief in Africa

15 “Second Bid to Kill Mugabe: Near Miss by Bomb in Road,” ibid. (February 11, 1980). 16 Rich, “Legacies of the Past?,” pp. 42-55. 17 Soames, “From Rhodesia to Zimbabwe,” pp. 405-19. 18 “Mugabe Romps in with 62pc of Total Poll,” Herald (March 6, 1980). 19 For an examination of the ideological leanings of the nationalist parties, see Moore, “The Ideological Formation of the Zimbabwean Ruling Class,” pp. 472-95. In this analysis Moore suggests that the supposed Marxist orientation of Mugabe and ZANU was exaggerated. By 1980, many of the party ideologues had been progressively purged from positions of authority and power and replaced by more pragmatic members. Moore’s account is not inconsistent with the main thesis of Astrow, Zimbabwe. 20 Meredith, The Fate of Africa, p. 627. 21 See “Coup Would Have Failed,” Herald (August 12, 1980); “Walls: Government Could Take Action,” ibid. (August 16, 1980); “Ministry Heads Refute ‘Coup Plot’ Allegations,” ibid. (August 20, 1980). Recent obituaries also mention the coup plot and the Rhodesian forces’ attempts to assassinate Mugabe. See Alan Cowell, “Peter Walls, General in Zimbabwe, Dies at 83”; Alfred Mulenga, “Lt-Gen Walls gave Zim a Torrid Time.” 22 Astrow, Zimbabwe, pp. 127-30, addresses the build-up of ZAPU’s conventional army in the 1970s and the anxiety and distrust this created within the ZANU leadership. 23 Weitzer, “In Search of Regime Security,” p. 537. 24 “Mugabe in Crucial Talks,” Herald (March 5, 1980); “Coalition: Mugabe and Nkomo Meet,” ibid. (March 6, 1980). Joshua Nkomo also talks about this period in his autobiography, Nkomo. 25 Quoted in Herald (April 18, 1980). 26 Quoted in Meredith, Our Votes, Our Guns, pp. 42-43. 27 Meredith, The Fate of Africa, p. 618. See also Gaidzanwa, “Drought and the Food Crisis in Zimbabwe,” p. 250. 28 Meredith, The Fate of Africa, p. 618. Taylor also cites these early postive incentives for white farmers in his Business and the State in Southern Africa, pp. 101-10, as does Bratton, “The Comrades and the Countryside,” pp. 174-202. 29 “Parties Attack Nkala Stand,” Herald (July 2, 1980). 30 Quoted in “I Aim to Crush Nkomo, Nkala Tells Rally,” ibid. (July 7, 1980). 31 “PF Women React to Minister,” ibid. (July 10, 1980); “Tell Us If It’s Policy – PM,” ibid. (July 9, 1980). 32 “Parties Clash at Inyati Mine,” ibid. (July 18, 1980). 33 “Ex-Guerillas Fight 4-hour Gun Battle in Bulawayo: Man Killed, 34 Injured in Rioting,” ibid. (November 10, 1980); “Ceasefire Ends Fierce Battle in Entumbane,” ibid. (November 11, 1980). 34 “Battle Toll: 43 Dead and 400 Injured,” ibid. (November 12, 1980); “Mpilo Crammed with Wounded,” ibid. (November 12, 1980); “Entumbane is Quiet: More Bodies Found,” ibid. (November 13, 1980); “Police Deny Rumors of Massacres,” ibid. (November 13, 1980); “Bulawayo Fighting Death Toll Rises to 55,” ibid. (November 14, 1980). The clash seemed to be a reaction to Enos Nkala’s rally warning that “from today the PF [ZAPU] has declared itself the enemy of ZANU.” See “Nkala Warns PF at Rally,” ibid. (November 10, 1980). A few days after the clashes, the Mayor of Bulawayo openly blamed the

The Pathologies of Drought Relief in Zimbabwe

123

fighting on Nkala and ZANU’s electioneering in the area. See “Ministers Blamed for Violence,” ibid. (November 14, 1980). 35 Alexander, “The Unsettled Land,” p. 585. 36 Chitiyo, “Land Violence and Compensation,” provides estimates of the number of ZANU, ZAPU, and former Rhodesian forces in the early 1980s. 37 “Gwanda Arms Caches Found,” Herald (February 18, 1980). It should be noted here that it is likely that all parties had engaged in stashing arms at strategic locations around the country out of a concern for the ceasefire arrangement. Having lost the election, ZAPU was then at a disadvantage because ZANU could use its control of state security machinery to expose the other parties. 38 Quoted in Nkomo, Nkomo, p. 2. Soon afterwards, Nkomo was placed under house arrest but after an attack on his residence by security agents, he escaped to Botswana and then, in March of 1983, to London. 39 The dynamics behind the emergence of this movement, the government’s response to it, and its end, including its labeling as a dissident movement, are well analyzed by Ranger, Voices from the Rocks; Alexander, “Dissident Perspectives on Zimbabwe’s Post-Independence War,” pp. 151-82. 40 I coded these insurgent attacks from newspaper coverage in The Herald and The Chronicle between 1980 and 1983. 41 Catholic Commission for Justice and Peace and the Legal Resources Foundation, Breaking the Silence, Building True Peace. 42 Ibid. The CCJP study also conservatively estimated that over 350 civilians went missing and 3,000 more were detained, of whom roughly 380 were tortured, 250 raped, and about 1,500 suffered another form of injury. More recent estimates suggest that over 20,000 civilians were killed or went missing at the hands of state agents. 43 Astrow, Zimbabwe, p. 175, from a report in the Financial Times (London, March 20, 1980); “Labour Stoppages Hit Six Companies,” Herald (March 19, 1980); “Labour Unrest Hits 9 More Firms,” ibid. (March 20, 1980). 44 See Government of Zimbabwe, Ministry of Labour, Labour and Economy, Vol. 1, p. 23. 45 “Strikes: Kangai Sympathetic, but…,” Herald (March 22, 1980); “Stoppages are Inexcusable, says Mugabe,” ibid. (March 26, 1980); “Call to Curb Strikes,” ibid. (May 6, 1980). 46 For a fuller discussion of early labor strife and the new government’s response to it, see Astrow, Zimbabwe, pp. 175-84. 47 “Let Army Back up the Police,” Herald (May 30, 1980). 48 See Government of Zimbabwe, Ministry of Labour, Labour and Economy, Vol. 1; Raftopoulos and Phimister, eds., Keep on Knocking. In fact, at its formation, the ZCTU was led by one of Mugabe’s nephews. 49 Sanders and Davies, “The Economy, the Health Sector and Child Health in Zimbabwe Since Independence,” pp. 723-31. 50 For a discussion of reforms in the health and education sectors, see Stoneman and Cliffe, Zimbabwe, esp. ch. 14. 51 Government of Zimbabwe, Working Group, The Children’s Supplementary Feeding Programme; Government of Zimbabwe, Ministry of Health, Planning for Equity in Health.

124

Domestic Politics and Drought Relief in Africa

52 See Zvobgo, “Education and the Challenge of Independence,” pp. 31954; and Sanders and Davies, “The Economy, the Health Sector and Child Health in Zimbabwe Since Independence,” pp. 723-31. 53 Central Statistical Office, Government of Zimbabwe, Quarterly Digest of Statistics. 54 For analyses of the government’s pricing and marketing policies, see Muir-Leresche and Muchopa, “Agricultural Marketing,” pp. 299-320; Herbst, State Politics in Zimbabwe; Mumbengegwi, “Continuity and Change in Agricultural Policy,” pp. 203-22; Gaidzanwa, “Drought and the Food Crisis in Zimbabwe,” pp. 250-57; and Bratton, “The Comrades and the Countryside,” pp. 174-202, 55 Government of Zimbabwe, Zimbabwe; Economic Intelligence Unit, Zimbabwe Country Profile. 56 Bratton, “The Comrades and the Countryside,” pp. 174-202. 57 Herbst, State Politics in Zimbabwe, pp. 82-109. 58 “Rainfall Stays Below Normal,” Herald (November 12, 1981); “Rainfall Deficits Continue to Grow,” ibid. (November 19, 1981); and a month later, “And Still the Country Waits for the Rains,” ibid. (December 24, 1981); “Children are Going Hungry Says Report: Drought Relief Team to be Set Up,” ibid. (April 30, 1982). 59 Leys, “Drought and Drought Relief in Southern Zimbabwe,” pp. 258-74. 60 Interview with former Department of Social Welfare official, February 18, 2004. 61 Ibid. 62 Government of Zimbabwe, Zimbabwe’s Experience in Dealing with Drought, 1982 to 1984; Government of Zimbabwe, Working Group, The Children’s Supplementary Feeding Programme in Zimbabwe; Drèze, “Famine Prevention in Africa,” pp. 123-72; Leys, “Drought and Drought Relief in Southern Zimbabwe,” pp. 258-74. 63 “Maize Harvest Exceeds Hopes by 100,000 Tonnes,” Herald (October 7, 1980); “Maize Record Signals End of Lean Years,” ibid. (November 20, 1980); “Six Months Needed to Bring in Record Maize Harvest, say Grain Sellers,” ibid. (March 6, 1981). 64 “What is the Drought Fund Used For?,” The Insider. The fund was to be used to cover the costs of the 1987-88 intervention, before it became the subject of a major corruption scandal. 65 Leys, “Drought and Drought Relief in Southern Zimbabwe,” p. 262. 66 Weiner, “Land and Agricultural Development,” p. 71. 67 Drèze, “Famine Prevention in Africa,” pp. 123-72; Leys, “Drought and Drought Relief in Southern Zimbabwe,” pp. 258-74. 68 This account of food aid distribution draws on interviews with rural residents in Buhera, Sanyati, and Mount Darwin, April-May 2004. 69 Leys, “Drought and Drought Relief in Southern Zimbabwe,” p. 269. 70 See report by Berkeley et al., Zimbabwe: Wages of War, pp. 136-38; and Alexander, “The Unsettled Land,” pp. 581-610. 71 Sithole, “The General Elections, 1979-1985,” pp. 75-98; and Kriger, “ZANU(PF) Strategies in General Elections, 1980-2000,” pp. 1-34. 72 Government of Zimbabwe, Registrar-General, 1985 General Election Report.

The Pathologies of Drought Relief in Zimbabwe

125

73 Sithole, “The General Elections, 1979-1985,” pp. 75-98. See also Sylvester, “Zimbabwe’s 1985 Elections,” pp. 229-55. 74 Kriger, “ZANU(PF) Strategies in General Elections, 1980-2000,” pp. 134. See also Caute, “Mugabe Brooks No Opposition,” pp. 138-44. 75 See “ZAPU Leadership Backs Nkomo on Unity,” Herald (November 18, 1987); “No Unity Pact Yet, says ZAPU Official,” ibid. (December 12, 1986); “Dabengwa and Four Others Released from Detention,” ibid. (December 5, 1986); “Unity Talks are Only in Advanced Stages,” ibid. (October 20, 1986). 76 See Banana, ed., Turmoil and Tenacity. 77 “Amnesty Gazetted,” Herald (May 5, 1988). For a critical assessment of the blanket amnesty, see Amnesty International, Zimbabwe. 78 Moyo, “The Dialectics of National Unity and Democracy in Zimbabwe,” pp. 83-102. 79 Chikuhwa, A Crisis of Governance. 80 Ibid., p. 39. 81 See Jackson and Rosberg, Personal Rule in Black Africa. 82 Chikuhwa, A Crisis of Governance. 83 Address at the University of Zimbabwe, July 22, 1989, quoted in Sylvester, “Unities and Disunities in Zimbabwe’s 1990 Election,” p. 386. 84 Moyo, Voting for Democracy. 85 For a comprehensive discussion of the factors behind ZUM’s failed challenge, see Kriger, “ZANU(PF) Strategies in General Elections, 1980-2000,” pp. 1-34; Moyo, Voting for Democracy; Sylvester, “Unities and Disunities in Zimbabwe’s 1990 Election,” pp. 375-400. 86 “40,000 Will Need Drought Assistance,” Chronicle (March 4, 1986); “130,000 ‘Need Food Aid’” ibid. (March 7, 1986); “Drought Hits Mat. South Schools,” ibid. (May 30, 1987); “Food Crisis in Nkayi,” ibid. (June 17, 1987); “Drought: Thousands of People in Rural Areas May be Translocated,” ibid. (June 19, 1987); “Drought Hits Sugar Field,” ibid. (June 23, 1987); “6,000 in Need of Drought Aid in Beitbridge,” Herald (February 17, 1986). 87 Herald (April 1986). 88 Ibid. (August 26, 1986). 89 Government of Zimbabwe, Zimbabwe’s Experience in Dealing with Drought, 1982 to 1984; Government of Zimbabwe, Ministry of Labour, Memorandum on Drought Relief, 1986. 90 Munro, Poverty and Social Safety Nets in Zimbabwe, 1990-99, p. 168. 91 Interview with former Department of Social Welfare official, February 18, 2004. 92 Munro, Poverty and Social Safety Nets in Zimbabwe, 1990-99, p. 179. 93 Ibid; Kinsey et al., “Coping with Drought in Zimbabwe,” pp. 89-110. 94 “What is the Drought Fund Used For?,” The Insider. 95 Benson and Clay, The Impact of Drought on Sub-Saharan African Economies; Clay, “Aid and Drought,” pp. 199-220; de Waal, Famine Crimes. 96 de Waal, Famine Crimes, p. 58. 97 Government of Zimbabwe, Zimbabwe; Economic Intelligence Unit, Zimbabwe Country Profile. 98 Herbst, State Politics in Zimbabwe. 99 Mumbengegwi, ed., Macroeconomic and Structural Adjustment Policies in Zimbabwe.

126

Domestic Politics and Drought Relief in Africa

100 Government of Zimbabwe, Report of the Public Service Review Commission, Vols. 1-2; Maphosa et al., Public Service Training Needs and Resources in Zimbabwe. 101 Thompson, Drought Management Strategies in Southern Africa. 102 See Government of Zimbabwe, Second Five-Year National Development Plan, 1991-1995. 103 Rukuni, “Revisiting Zimbabwe’s Agricultural Revolution,” p. 11. 104 Lennock, Paying for Health; Renfrew, ESAP and Health; Gibbon, Structural Adjustment and the Working Poor in Zimbabwe; Balleis, A Critical Guide to ESAP; and Zimbabwe Textile Workers Union (ZTWU) and Zimbabwe Congress of Trade Unions (ZCTU), Report on Structural Adjustment and Its Impacts on Workers. 105 “Teachers’ Strike Spreads to Rural Areas,” Herald (May 5, 1990); “Strike Grounds AirZim Fleet,” ibid. (July 2, 1990); “Civil Servants Urged to Speak With One Voice,” ibid. (July 13, 1990); “ZCTU Chief Calls for Speedy Solution to Pay Disputes,” ibid. (July 30, 1990); “Employees Late for Work in Drivers’ Bus Go-Slow,” ibid. (September 17, 1991); “6,500 Workers at Triangle on Strike,” ibid. (May 5, 1992); “Mahachi Bans ZCTU Demo,” ibid. (June 11, 1992); “ZCTU Meets to Discuss Ban on Demo,” ibid. (June 16, 1992); “Chitauro to Meet Labour Unions,” ibid. (June 12, 1992); “Police Urge Workers not to Demonstrate Against Government,” ibid. (June 13, 1992); “ZCTU Chief Calls for Dialogue with State,” ibid. (June 15, 1992); “6 ZCTU Demonstrators in Court,” ibid. (June 16, 1992). 106 “Farmers Expected to Deliver 8 800 tons of maize to GMB,” ibid. (May 1, 1992); “No End in Sight for Masvingo Maize Crisis,” ibid. (May 1, 1992); “100 000 Face Hunger,” Chronicle (October 14, 1992); “Mammoth Exercise to Secure Food,” ibid. (August 3, 1992); “140 000 Cattle Die in Masvingo,” Sunday News (August 20, 1992); “Villagers Face Death as Hunger Looms,” Chronicle (August 21, 1992); “Drought Has Destroyed Family Ties,” Sunday News (November 8, 1992); “Peasants Resort to Wild Fruits,” Chronicle (January 13, 1992); “Children Collapse Because of Hunger,” Sunday News (December 13, 1992); “120 000 Need Food,” Chronicle (July 4, 1992); “Agritex Reports Total Maize Failure,” ibid. (April 16, 1992). 107 Government of Zimbabwe, The Drought Relief and Recovery Programme, 1992/93; Marquette, “Current Poverty, Structural Adjustment, and Drought in Zimbabwe,” pp. 1141-49; Sachikonye, “Zimbabwe,” pp. 88-94; Stoneman, “The World Bank Demands Its Pound of Zimbabwe’s Flesh,” pp. 9496; Stack, Drought Forecasts and Warnings in Zimbabwe; Munro, “Zimbabwe’s Drought Relief Programme in the 1990s,” pp. 125-41; Zimbabwe Women’s Resource Centre and Network, Women and Drought; Meldrum, “Drought, Death and Dissidents,” pp. 32-35; Organization of Rural Associations for Progress, “The 1992 Drought”; Government of Zimbabwe, Medical Officer, Matabeleland North Province, Drought Monitoring – Matabeleland North Province, May-August 1992; Government of Zimbabwe, Ministry of Lands, Agriculture and Rural Resettlement, National Early Warning Unit, Cereal Grains Availability in Communal Areas; Kinsey et al., “Coping with Drought in Zimbabwe,” pp. 89-101. 108 For a discussion of these skeptics, see Thompson, Drought Management Strategies in Southern Africa.

The Pathologies of Drought Relief in Zimbabwe

127

109 “Drought Declared a National Disaster,” Herald (March 7, 1992); “Drought Relief Teams Set Up,” Chronicle (April 11, 1992); “No-one Will Starve – President,” ibid. (August 9, 1992). 110 “President to Assess Drought in Provinces,” Herald (June 27, 1992). 111 Government of Zimbabwe, The Drought Relief and Recovery Programme, 1992/93; Munro, “Zimbabwe’s Drought Relief Programme in the 1990s,” pp. 125-41. 112 Government of Zimbabwe, The Drought Relief and Recovery Programme, 1992/93. 113 Ibid. 114 “Mealie-Meal Price Rise Off,” Herald (June 22, 1992). 115 Government of Zimbabwe, Social Dimensions of Adjustment. 116 See Government of Zimbabwe, Ministry of Health, National Nutrition Unit, Child Supplementary Feeding Programme, Status Report; Government of Zimbabwe, The Drought Relief and Recovery Programme, 1992/93; “State to Launch Feeding Scheme Countrywide,” Herald (May 5, 1992); “Ministry Probes Scheme to Feed Children,” ibid. (May 22, 1992); Munro, “Zimbabwe’s Child Supplementary Feeding Programme,” pp. 242-61. 117 Government of Zimbabwe, The Drought Relief and Recovery Programme, 1992/93. 118 Thompson, Drought Management Strategies in Southern Africa. 119 Tendler, Good Government in the Tropics. 120 Mugabe, “The Unity Accord,” p. 351. 121 See Sachikonye, “The Context of the Democracy Debate,” pp. 43-60; Sithole, “Should Zimbabwe Go Where Others Are Coming From?,” pp. 71-82; Moyo, “The Dialectics of National Unity and Democracy in Zimbabwe,” pp. 83-102. 122 Sithole, “Zimbabwe’s Eroding Authoritarianism,” p. 127. 123 Ibid. 124 Sylvester, “Whither Opposition in Zimbabwe?,” p. 403. 125 In their comprehensive study of the 1995 elections, Behind the Smokescreen, Makumbe and Compagnon argue that the poor showing by opposition parties was a product of intimidation and repression by ZANU-PF. Opposition candidates had property destroyed, followers were routinely harassed by the Central Intelligence Organization (CIO), and members of ZANU’s Youth League threatened the destruction of property and livelihoods of anyone who ventured to attend opposition political rallies. 126 Herald (March 22, 1995). 127 Interview with official from the Ministry of Public Service, Labour and Social Welfare, March 10, 2004. 128 Ibid. 129 Ibid. 130 Ibid. Successive droughts would have led to the cancellation of all loans. 131 Munro, Poverty and Social Safety Nets in Zimbabwe, 1990-99. 132 Interview with official from the Ministry of Public Service, Labour and Social Welfare, March 10, 2004; interview with officials from the Ministry of Local Government and National Housing, April 5, April 8, and May 6, 2004. 133 Interviews with rural residents in Buhera, Sanyati, and Mount Darwin, April-May 2004.

128

Domestic Politics and Drought Relief in Africa

134

Munro, Poverty and Social Safety Nets in Zimbabwe, 1990-99, p. 171. Interviews with officials from the Ministry of Public Service, Labour and Social Welfare, March 10, 2004, and from the Grain Marketing Board, June 15, 2004. 136 Ministers Florence Chitauro and John Nkomo, cited in Munro, Poverty and Social Safety Nets in Zimbabwe, 1990-99, p. 172. 137 I thank LeBas, “Polarization as Craft,” pp. 419-38, for the insight on polarization and stalled transition. See also LeBas, From Protest to Parties. 138 Bond and Manyanya, Zimbabwe’s Plunge. 139 See Tsanga, “Progress Towards Democracy, Or Revenge of Civil Society”; and Masiwa, “Zimbabwe: Count Down to Year 2000.” 140 Tsanga, “Progress Towards Democracy, Or Revenge of Civil Society”; Sithole, “Fighting Authoritarianism in Zimbabwe,” pp. 160-69. 141 I was an undergraduate student at the University of Zimbabwe at the time of the founding of the NCA in 1997. I experienced the excitement of the mobilization, but was also aware of the limitations of the NCA—it did not yet incorporate workers or the rural population in any formal way. 142 Quoted in Sithole, “Fighting Authoritarianism in Zimbabwe,” p. 163. 143 Chitiyo, “Land Violence and Compensation,” examines the contest within ZANU-PF over the definition of war veteran and the splits this struggle produced within the party. See also Financial Gazette (November 3, 1997). 144 Herald (September 17, 1997). 145 Bond and Manyanya, Zimbabwe’s Plunge. 146 Kriger, “Zimbabwe Today,” pp. 443-50. 147 Bond and Manyanya, Zimbabwe’s Plunge. 148 See, for instance, Mandaza, “Reflections on the Referendum”; Ihonvbere, “Constitution Making in Africa”; Tsanga, “Progress Towards Democracy, Or Revenge of Civil Society”; Sithole, “Fighting Authoritarianism in Zimbabwe,” pp. 160-69. 149 The national conference path to democratic transitions in Francophone Africa is examined by Bratton and van de Walle, Democratic Experiments in Africa. See also Robinson, “The National Conference Phenomenon in Francophone Africa,” pp. 575-610. 150 Sithole, “Fighting Authoritarianism in Zimbabwe,” p. 163. 151 See Sachikonye, “The People Have Decided”; Goncalves, “A Rare Defeat for Mugabe.” 152 Sithole, “Fighting Authoritarianism in Zimbabwe,” pp. 160-69. 153 Quoted in Goncalves, “A Rare Defeat for Mugabe,” p. 5. 154 Quoted in ibid., p. 5. 155 Ibid. 156 The immediate debate this generated is illustrated by Rutherford, “Zimbabwean Farm Workers”; Baregu, “The Urgent Need for Land and Agrarian Reform in Southern Africa”; Mushonga, “Zimbabwe.” 157 See Tendi, Making History in Mugabe’s Zimbabwe; Ranger, “Nationalist Historiography, Patriotic History and the History of the Nation,” pp. 215-34. See also Mugabe, Inside the Third Chimurenga, for how the embattled government reframed its exclusive claims to power. 158 Tendi, Making History in Mugabe’s Zimbabwe, p. 12. 159 See Kriger, “Zimbabwe Today,” pp. 443-50. 160 Sithole, “Fighting Authoritarianism in Zimbabwe,” p. 166. 135

The Pathologies of Drought Relief in Zimbabwe

129

161 See, among others, Kriger, “Zimbabwe Today,” pp. 443-50; and Alexander, “Zimbabwean Workers, the MDC & the 2000 Election,” pp. 385406. 162 Rukuni, “Revisiting Zimbabwe’s Agricultural Revolution,” p. 13. 163 Rukuni et al., Zimbabwe’s Agricultural Revolution Revisited. 164 Richardson, The Collapse of Zimbabwe in the Wake of the 2000-2003 Land Reforms. 165 Human Rights Watch, Not Eligible. 166 “Zimbabwe Predicts Good Harvest,” Herald, (May 12, 2004); “Zim Will Not Require Food Aid: Mangwana,” ibid. (May 14, 2004). 167 See accounts by Zimbabwe Peace Project, Partisan Distribution of Food and Other Forms of Aid; Zimbabwe Peace Project, Partisan Distribution of Food and Other Forms of Aid at District Level. 168 Human Rights Watch, The Politics of Food Assistance in Zimbabwe. 169 Human Rights Watch, Not a Level Playing Field; International Crisis Group, Post-Election Zimbabwe: What Next?; Andrews and Morgan, Zimbabwe After the 2005 Parliamentary Election; Chan, “The Old Fox Eludes the Hunt.” 170 “Mugabe Allows UN to Increase Food Aid,” Guardian (London, June 2, 2005).

5 The Evolution of Drought Relief Policy in Botswana

I am forced by circumstances and problems which affect us all to address you today on the drought situation in Botswana. – President Seretse Khama declaring Botswana drought-stricken, May 29, 1979 Now, in April 1983, it is clear that Botswana remains drought-stricken. … The drought relief programme will be continued throughout 1983 until 1984. – President Quett Masire extending drought relief for a second year, April 21, 1983

In the early 1990s, incumbents in Botswana took the unprecedented step of incorporating drought relief and rehabilitation programs into their standing development plans.1 First, the government introduced a 3-4 percent increase in annual development expenditures to its 1991 to 1997 development plan.2 In addition to setting aside money to pay for future relief efforts, incumbents in Botswana also committed to a program of relief: labor-based projects for adults. Thus, all subsequent drought relief interventions in Botswana (between 1992 and 1993, 1995 and 1997, and 2001 and 2005) have offered adults protection from famine through expansive, publicly-funded labor-based relief projects.3 Establishing a standing Famine Code in Botswana through the creation of an endogenous drought relief fund and the writing of a relief program into the country’s development plan removed the uncertainty over whether government would respond to future droughts and tied incumbents to a set general relief policy. Thus, this standing commitment curtailed incumbents’ discretion in drought relief policy-making,

131

132

Domestic Politics and Drought Relief in Africa

which previous chapters have shown allow them to fashion relief in accordance with their prevailing micro-political conditions. How did Botswana come to develop this standing Famine Code? To answer this question, I show that it is informative to examine how political incumbency came to be institutionalized in Botswana. The analysis provides evidence that, even under democracy, the strength or weakness of incumbency determined how politicians responded to drought, including their decision in the early 1990s to entrench a program of drought relief. In Botswana, as in Kenya and Zimbabwe, insecure incumbents responded to drought by adopting food-aid programs for able-bodied adults and relying on incumbent-controlled personnel and administrative structures to identify beneficiaries and implement relief at the local level. In addition, the Botswana case demonstrates that the manner in which incumbents consolidate their security affects subsequent programs of relief. I show that, because incumbents invested in the consolidation of democracy and built a strong developmental state apparatus to manage the economy, they created positive externalities for other policy areas, including drought relief. Specifically, stable institutionalized rule generated long time horizons and a strong planning ethos within the bureaucracy in Botswana, factors that, in turn, generated strong interests within the state to develop standing administrative blueprints for responding to droughts. This chapter underscores the effects of institutionalized security, especially when compared with the personalized security we saw in Zimbabwe, which left drought relief policy-making open to the whim of the executive. The Colonial Origin of Institutional Insecurity in Botswana, 1885-1966

Spurred by geopolitical interests and the quest for a secure route into the interior of central Africa, colonial expansion into what became the Bechuanaland Protectorate commenced in 1885.4 Early economic assessments of the territory were bleak, with no known or anticipated mineral deposits and much of it falling within the Kalahari Desert and its semi-arid periphery, which was ill-suited for commercial agricultural investments. Thus, “Bechuanaland’s only apparent value was as a place to pass through [from the Cape Colony] on the way to the more lucrative trade of the African interior.”5 Such assessments had a direct effect on colonial state building, leading Britain to adopt a stripped-down version of its indirect rule strategy.6 Colonial authorities established a nominal state apparatus that superimposed a cheap, weak, and skeletal admini-

The Evolution of Drought Relief Policy in Botswana

133

strative structure over the preexisting system in which eight separate Tswana chiefs exercised juridical, executive, and legislative power at the local level.7 Scholars contend that this form of colonial rule, which some have described as benign neglect, produced minimal disruptions to traditional authority structures in Botswana.8 But it also meant that the administration of the protectorate was fragmented, which, I argue, constrained political development and laid the foundations for the subsequent institutional vulnerability of incumbents immediately after independence. Initial moves toward a unified administrative structure came in 1920, when colonial officials established the Native Advisory Council (later renamed the African Advisory Council) which consisted of the eight chiefs as permanent ex officio members, seven British officials, and representatives of Africans who did not belong to the Tswana tribes.9 Europeans in the protectorate were not subject to rule by traditional authorities but were governed under a civil code formulated by the protectorate administration in consultation with the European Advisory Council, also established in 1920. In 1949, officials made another overture at unifying the colony’s administration by creating the Joint Advisory Council, a body that brought members of the African and European Advisory Councils together. However, these early reforms did not deepen or extend the reach of the colonial state; they only provided colonial officials with a bit more information and gave chiefs a nominal sense of being part of a national government, while retaining a great deal of local authority. Officials could not sustain support for broader changes that might have conflicted with Britain’s long-established goal to administer Botswana on the cheap and potentially transfer the territory to the Union of South Africa. Thus, British rule in Botswana remained fragmented, with a central administration that had no real capacity to exercise effective control over its territory.10 However, political agitation in South Africa in the 1950s provided the impetus for colonial officials to engage in more thorough state reform, altering the pace and depth of centralized state-building in Botswana. These reforms were the first attempt to establish the basic framework of a modern political and bureaucratic administration quite distinct from the tribally-fragmented system the British had relied on during the period of benign neglect. While economic reforms were narrow, aiming solely at budgetary self-sufficiency, the political reforms were ambitious.11 In early 1961, colonial officials adopted a constitution providing for the creation of a partially-elected consultative Legislative Council (Legco), a quasi-cabinet in the form of an Executive Council, and an ex-

134

Domestic Politics and Drought Relief in Africa

panded African Advisory Council that was intended to generate broad legitimacy for the first two bodies. The Legco consisted of 34 members: ten colonial officials, ten Europeans elected by the white population, ten Africans selected by the African Advisory Council, one elected Asian representative, and three unofficial members nominated by the Resident Commissioner.12 The African Advisory Council, itself constituted by a series of elections at the ward and tribal levels with chiefs as ex officio members, served as an electoral college for the African members of the Legco. Election to the Legco was restricted to Africans already on the council. Significantly, chiefs were not given ex officio membership to the Legco; they had to be elected like any other African on the Advisory Council. Only one sitting chief, Bathoen II of the Ngwaketse, was chosen to serve on the Legco.13 Colonial officials then appointed two Africans (Seretse Khama and Bathoen II) and two Europeans (Russell England and David Morgan) to the quasi-ministerial Executive Council. Late colonial state reconstitution then paved the way for the formation of political parties, independence negotiations, independence elections in 1965, and finally independence itself the next year. The first party formed in the protectorate was the Bechuanaland (later Botswana) People’s Party (BPP) in 1960. Its three founding officers—K.T. Motsetse (president), P.G. Matante (vice-president) and Motsamai Mpho (secretary general)—had all lived and worked in South Africa in the 1950s and been drawn into the nationalist mobilization sparked by the revival of the African National Congress (ANC) and the formation in 1959 of the more radical and militant Pan Africanist Congress (PAC). Matante was a member of the PAC, while Mpho had returned to Botswana after detention and trial for his activities as a member of the ANC.14 The experience and ideas earned from political activity in South Africa gave the BPP’s officers an early advantage over other Africans still largely deferential to traditional authority. However, the left-leaning ideology and Africanist position that had served the ANC and the PAC respectively in economically advanced and racially polarized South Africa were ill-suited for political mobilization in underdeveloped and racially less-polarized Bechuanaland. Importantly, ideological differences stemming from its roots in South African nationalist politics, along with an internecine competition for leadership, led to divisions within the BPP. Within a couple of years of its formation, the BPP’s leadership split, and by 1963, there were three factions. In the competition for control of the the party, Motsetse and Matante expelled Mpho. Mpho refused to accept his expulsion and responded by creating his own party and giving it the name BPP (Mpho). In 1964 Mpho transformed his faction into the Bechuanaland

The Evolution of Drought Relief Policy in Botswana

135

Independence Party. Motsetse and Matante then split in 1963, and Matante formed his own BPP faction. The formation of the BPP and its rapid factionalization unsettled many of the Africans on the Legco and the Advisory Council. These elected officials, bureaucrats, capitalists, and traditional elites were materially and politically threatened by the BPP’s socialist pronouncements and demands for independence now. Thus, in November of 1961, Seretse Khama and other Africans on the Legco formed the Bechuanaland Democratic Party (BDP) as a moderate and pro-capitalist alternative to the BPP. Khama was elected president, Archelus Tsoebebe vice-president, and Quett Masire, a farmer, its secretary general. Khama, the Oxford-educated lawyer whom the British had forced to abdicate his Bamangwato chieftaincy because he had married a white woman in 1948, combined both a traditional following from the Bamangwato (the largest of the Tswana tribes) with his experience of the modern political institutions that colonial officials were creating.15 In April 1963, the British government authorized the Resident Commissioner to establish a constitutional committee to begin independence negotiations. The committee was composed of: the Resident Commissioner, as chair; three chiefs, to represent traditional interests; two protectorate officials, three Europeans, and one Asian; and three representatives each from the BDP, the BPP (Mpho), and BPP (Motsetse). The conference carved out the basic outline for a modern government system for Botswana by redefining the role and position of traditional authorities. In this regard, as noted by Colclough and McCarthy: The 1963 constitutional conference marked the beginning of the power struggle between the traditional ascriptive authority of the chiefs and the younger, more educated leaders of the BDP. In the coming years, this struggle was to overshadow that between the political parties, especially after the BDP had asserted its dominant position [among the political parties].16

The shared interest among modernizers from the new parties to subordinate traditional authority promoted the subsequent development of institutionalized multi-party politics. It “worked to reduce the possibility of all out conflict between the parties. For the parties politics [was] not and [was] not seen as a zero-sum game.”17 Modern state-builders, led by Seretse Khama, resisted attempts by chiefs to create “an upper chamber of a bicameral legislature … possessing powers similar to those of the House of Lords.”18 United in their “belief that political leadership should be based on achieved rather

136

Domestic Politics and Drought Relief in Africa

than ascribed status,”19 modernizers offered chiefs a choice between a few reserved seats in a unicameral legislature or positions on a purely advisory Council of Chiefs.20 Far outnumbered in the negotiations, the chiefs opted for what they saw as the better of the two options, the advisory council. The council was later renamed the House of Chiefs in an attempt to make it sound more important than it was.21 In truth, it was subservient and subordinate to the institutions of the modern state, which were also under construction; the House of Chiefs had no legislative, veto, or delaying powers. The constitutional conference then established a mixed governmental system that combined parliamentary features, such as mutual dependence between parliament and the head of government, with presidential elements, such as veto powers for the head of government. Botswana’s independence constitution established a legislature with 31 representatives to be filled by single-member constituency elections under plurality. Four political parties contested the first general elections in 1965: the BDP led by Seretse Khama; two factions of the People’s Party led by K.T. Motsetse and Philip Matante; and Motsamai Mpho’s Botswana Independence Party. The two BPP factions leaned left: they concentrated their campaign in the towns where they stressed redistributive policies and the rapid localization of a bureaucracy that was about 85 percent expatriate in composition, and they had a less permissive attitude of the future role of chiefs in politics.22 This strategy was disconnected from the social and political reality of Botswana in 1965; only 10 percent of the country’s population lived in towns in 1965— hardly a viable electoral block. Also, the People’s Party calls for rapid redistribution and localization unsettled the traditional and bureaucratic elites, whose support was critical. Commenting on Matante and BPP’s approach as a whole, BDP co-founder Quett Masire notes that “he mainly preached South African politics, addressing the question that was not raised, which was what to do in South Africa. He did not answer the question that was raised, which was: How could we make Botswana successful after independence?”23 The BDP adopted a moderate stance: not threatening growth prospects by adopting redistributive policies, not sacrificing effective administration by rapid localization, and being temperate regarding reforms to chiefly power and authority.24 The BDP focused its campaign in the rural constituencies where over 90 percent of the population lived. Describing its campaign strategy, Masire stresses that the BDP “knew the positions being taken by the People’s Party on a variety of issues. Having observed these developments, we concluded that running to ex-

The Evolution of Drought Relief Policy in Botswana

137

tremes really didn’t pay, and we thought moderation would be a virtue for us. … We meant to cling to the golden mean.”25 These positions enabled the BDP to get support from a broad crosssection of the electorate in Botswana, including large cattle owners, Europeans, and many of the chiefs. Thus, the party swept 28 of the 31 seats, taking 80.4 percent of the vote. Matante’s faction of the BPP won the remaining 3 seats with 14.2 percent of the vote.26 Khama and his new government wasted no time, accelerating the modern institution building on which their incumbency would depend even through the transitional period from March 1965 to September 1966 that preceded independence. Post-Independence Insecurity

As in Kenya and Zimbabwe, early insecurity in Botswana was a legacy of the preceding regime. The weak institutional position of Khama and the BDP was apparent at independence. As Herbst highlighted, the shallowness of the state’s infrastructural presence—for instance, there were no more than 12 kilometers (roughly 7.5 miles) of tarred roads in the country27—meant incumbents had limited capacity to broadcast their power.28 The winners of the independence elections of 1965 had to construct—and then legitimate—modern administrative institutions, if they were to secure their position from the threat posed by traditional authorities. On the economic side, the country attained independence with a GDP per capita of only about US$310 (in 2005 terms). Also, as Parson shows, Botswana’s value as a source of labor for mines and industry in South Africa meant that there had been very little investment internally in basic economic infrastructure.29 In 1965, only 12 percent of the total labor force of 251,000 earned their wages within the country. While benign neglect had meant limited colonial penetration of African political and economic institutions, it had also brought chronic underdevelopment. Political developments in the post-independence period also contributed to the vulnerability of the new incumbents. Victory for the BDP in the 1965 elections led to the formation of Botswana National Front (BNF), a decidedly more radical political party than any of the factions of the BPP. Led by Kenneth Koma, a young Motswana who had earned a doctorate in political science from Russia, the immediate objectives of the BNF were to save Botswana from a neo-colonial BDP regime.30 The threat posed to incumbents by the formation of the BNF increased in 1969 when Chief Bathoen II of the Bangwaketse, the most experienced

138

Domestic Politics and Drought Relief in Africa

of the chiefs, resigned his position as hereditary chief and joined the BNF. Bathoen’s decision to enter electoral politics on the side of the opposition was significant for incumbents who believed “that the greatest potential threat to their continued rule would be discontented traditional leaders who, in alliance with disaffected elements of the nonSetswana speaking minority, could create an effective, rural-based opposition movement.”31 Like Seretse Khama and the BDP, Bathoen provided traditional support and legitimacy to the BNF. In the 1969 general elections, he crushed the sitting vice-president, Quett Masire of the BDP, in the Kanye South constituency, winning over 71 percent of the votes to his rival’s 28 percent.32 Bathoen’s influence and traditional stature also helped the BNF win all the parliamentary seats in the Ngwaketse-dominated Southern District of Botswana and secure 10 out of 24 seats on the Southern District Council, where in the 1966 elections the BDP had won all 24 seats. After the 1969 polls, Bathoen was elected president of the BNF. However, in subsequent elections, the party struggled to reconcile its provincial Ngwaketse support with the socialist rhetoric espoused by its founder, Kenneth Koma. Although Bathoen’s political career eventually stalled, his open challenge to Khama and the BDP was followed by a period of widespread expressions of popular dissatisfaction with the government. By the mid-1970s, incumbents who had not yet consolidated their institutional power were for the first time forced to contend with frequent political and economic protests. They came from rural communities dislocated by the rapid expansion of the mining sector, urban workers largely ignored by BDP’s development policies, and younger segments of the population (high school and university students) impatient with the slow pace at which the bureaucracy was being localized.33 Expectedly, the BNF and other opposition parties took advantage of the crisis to drum-up their criticism of Khama’s government.34 Fearful of growing agitation, in January 1977, the government announced that it was increasing recruitment to the Paramilitary Police Unit (PMU)—effectively Botswana’s army since independence. In the coming months, this modest initiative escalated into legislative proposals to establish a proper standing national army. Incumbents moved quickly to introduce the Botswana Defense Forces Bill, which would transform the PMU into a freestanding, professionally trained and equipped modern national army. After two weeks of debate, the “army bill received full support from BDP MP’s” and passed into law.35 The Botswana Defense Force was pased on April 18, 1977. The commissioner of police, Mompati Marafhe, was named its

The Evolution of Drought Relief Policy in Botswana

139

first commander and the president’s son, Ian Khama, deputy commander.36 During final debate on the bill, a BDP member of parliament, G. Mosinyi, indicated that he had credible information that “youths in Botswana were being taught tactics to overthrow the government”37 by the BNF. Other BDP MPs concurred that something needed to be done to counter the threat posed by subversive elements within the country. Despite the BDP’s efforts to intimidate the BNF, the political confrontation between the two continued and escalated. The following year, President Khama withdrew the passports of 17 BNF members headed for a conference in Havana, Cuba, on the grounds that the group would get military training there.38 At the 17th annual conference of the BDP in May 1978, Khama warned the party against threats posed by the BNF: Allow me at this juncture to say something about our democracy and how I see it developing in the future. When I addressed you at Mochudi last year I made it clear that we could not continue to take our democracy for granted and expect to continue to be a democratic nation. While I do not doubt our commitment to the survival of our democratic system of government as long as the B.D.P. is in power I nevertheless feel that you should be made aware of the growing threats to our democratic way of life which are being posed by some of our opponents, the B.N.F. in particular.39

A few months later, the confrontation between the BNF and the government intensified further when President Khama declared John K. Modise, a leading member of the BNF whose mother was South African, an undesirable immigrant and deported him to apartheid South Africa.40 Early Responses to Insecurity in Botswana

Political and administrative reforms after the 1965 elections accelerated the modern institution-building project on which the security of incumbents was dependent. One such reform was the 1965 Chieftainship Act,41 which gave incumbents powers to appoint and remove chiefs, regents, deputy-chiefs, sub-chiefs, and headmen. The act also defined the powers and functions of traditional authorities in Botswana’s new political context and provided for their remuneration by the state. The act “strip[ped] the Chief of all his legislative powers and most of his executive authority,”42 thus legally subordinating traditional authorities and entrenching the progressive monopolization of power by the new modern rulers of Botswana.

140

Domestic Politics and Drought Relief in Africa

The government gradually assumed chiefs’ customary powers to allocate tribal land, transferring them to Tribal Land Boards and other new institutions incumbents created with the Tribal Land Act of 1968.43 An amendment in 1975 gave the president the power to change any land board decision. These reforms were in line with the incumbents’ goal to develop a modern state, as the BDP’s 1965 elections manifesto had expressed: The Bechuanaland Democratic Party stands for a gradual but sure evolution of national state in Bechuanaland, to which the tribal groups will, while they remain in existence, take a secondary place. This is an unavoidable development, an evolutionary law to which we must yield to survive, or resist and disappear as a people.44

But even as they undercut the authority of chiefs, incumbents damped opposition to reform by including them in new, modern institutions at the local level. For instance, the Local Government (District Councils) Act of 196545 created directly-elected local councils, which replaced tribal authorities completely. Chiefs were made ex-officio chairs of the new councils, but their old legislative and executive powers were stripped. Responsibility for local service provision—functions formerly carried out by chiefs through traditional authorities—was shifted to these elected councils. When elections to them were first held in 1966, the BDP secured majority control of 10 of the 12 councils, averaging about 85 percent of the seats.46 As articulated by Seretse Khama, this democratization of local government was a critical part of the modern institution building: This is an age of self-determination, and the elected representatives of the people must assume more and more responsibility for the welfare of the people in the areas they represent. The tribal chiefs still have much traditional authority under the Chieftainship Law; as members of the House of Chiefs they are in a position to share in the legislative authority of the Central Government, and as Chairmen of the Local Councils they have ample opportunity to use whatever ability and experience they possess to guide the councils and serve their tribes in co-operation with the people’s elected representatives. I am sure you and I will agree that this should be the role of a chief in this age, our age, and I am satisfied that there cannot, and there will not be any going back in this process of making local government democratic.47

The Evolution of Drought Relief Policy in Botswana

141

After democratizing local government, incumbents then centralized control of the district councils by making their appointed representatives at the local level, called District Commissioners (DC), ex-officio members of the elected councils and giving them the authority to audit council finances.48 The government then created District Development Committees (DDC), chaired by the DC and composed of central government representatives from other service ministries, to spearhead and coordinate government development projects at the local level.49 DDC were supported by elected Village Development Committees (VDC) in carrying out their centrally-mandated functions. Terms for VDCs were set at a relatively short two years, which seemed designed to embed and inculcate democracy very rapidly among Batswana, while also weakening them in any confrontation with the center. To complete the transformation of District Councils into modern democraticallyconstituted and bureaucratically-run entities, in 1973 incumbents established a Unified Local Government Service.50 The service centralized the recruitment and appointment of council administrative staff, including council secretaries. This was a responsibility that had formerly resided in the council itself. In strengthening their institutional position, incumbents in Botswana weakened the financial position of traditional leaders. Through the Local Government Tax Act of 1965, the government transferred the power to levy local taxes from chiefs to district councils. In September 1967, Khama’s government used its parliamentary majority to pass the Mineral Rights in Tribal Territories Act, which vested rights to all subsoil minerals found on tribal land in the national government.51 Prior to the act, these rights had belonged to the tribe and been under the administration of chiefs. Incumbents leveraged revenues from mineral resources to invest in legitimacy-generating transfers (infrastructural development, service delivery, and drought relief). To complete the financial subordination of traditional authorities to the center, in 1968 incumbents passed the Matimela (impounded cattle) Act, which stripped chiefs of their traditional claim to ownership of stray cattle found in their territory.52 This prerogative had enabled chiefs and members of the traditional elite to accumulate large herds, and through the custom of mafisa or kgamelo (cattle-loaning for draught power, milk, and sometimes meat) to cement bonds of allegiance and reciprocity with the people.53 After 1968, rights to stray cattle were transferred to the popularly-elected district councils. As chiefs and other traditional leaders beneath them (dikgosana) had done before, councils used their new resources to cement and legitimate themselves among rural Batswana.

142

Domestic Politics and Drought Relief in Africa

Incumbents also transferred service delivery from chiefs to the new institutions they were building. For instance, in 1967 the government of Botswana adopted a Supplementary Feeding Programme, under an agreement with the World Food Program (WFP). The WFP agreed to provide food to feed pre-school and primary school children, pregnant and lactating mothers, and later, TB patients.54 To administer and distribute food under this program, the government established the Institutional Food Programme (IFP) in Ministry of Local Government and Lands, instead of relying on traditional authorities. With headquarters in the capital, five line-of-rail depots and seven depots located in districts across the country, IFP was responsible for receiving WFP food, transporting the rations to district depots, and subsequently, to the distribution points around the country (clinics and primary schools). Each year, pre-school children would be entitled to 330 days of rations, expectant mothers 120 days, nursing mothers 180 days, TB patients 365, and primary school children would be entitled to 200 days of rations a year.55 On the economic front, Botswana’s new rulers aimed at selfsufficiency, affirming their “determination to make the country a financially viable entity in the shortest possible period.”56 This guiding state ethos set Botswana on a different trajectory of economic management from Zimbabwe where, as we saw in Chapter 4, incumbents faced immediate consumption demands from a restive working-class. In the short term, to achieve their objective of self-sufficiency, Botswana’s new rulers had to operate under strict budget controls. These hard fiscal constraints reinforced institution building and the need for careful planning, leading to the ascendancy of a bureaucratic ethos within the state. In pursuit of budgetary self-sufficiency, the new government invested in cattle ranching, the only viable economic sector at independence. The government formed the Botswana Meat Commission (BMC) by nationalizing the abattoir at Lobatse and building two more, one in Maun and the other in Francistown.57 This marketing board, the equivalent of the NCPB in Kenya and the GMB and CMB in Zimbabwe, had monopsony purchasing power of cattle in the country. By statute, the BMC was required to buy all cattle that ranchers brought to market. However, unlike its Kenyan and Zimbabwean counterparts, the BMC was not permitted to make profits. This meant that it had little incentive to offer ranchers prices far below those on the world market. In the event of a difference between the prices paid to farmers and returns earned through beef exports, the BMC was required to return all surpluses to farmers after deducting its operating costs. Its annual reports indicate that producers received bonuses in 23 of 28 years between 1966

The Evolution of Drought Relief Policy in Botswana

143

and 1994.58 The positive effects of good institutions and political support for cattle ranching were immediately apparent. Scholars estimate that prior to the development of a vibrant mineral sector in Botswana, “livestock products provided one-third of Botswana’s export earnings and agriculture, including livestock and crop production and meat processing, contributed about one-third of Gross Domestic Product.”59 In addition to developing the livestock sector, the government also encouraged mining companies to explore the country.60 Mine prospecting led to the discovery of copper and nickel deposits at SelebiPhikwe, coal at Marupule, and kimberlite industrial and gem quality diamonds at Orapa, Letlhakane, and later, at Jwaneng. But Khama warned government officials: We have all been elated by the recent announcement that Orapa constitutes a major new diamond discovery…This announcement is not an open sesame… Whilst hopes are high that mineral development will come to Botswana, let me sound a note of caution. Exploitation of the minerals depends basically on our ability to plan and administer developments properly.61

Incumbents’ commitment to planning and careful budgeting persisted as a guiding state ethos even as revenues from the mineral sector increased: From now on, as I warned this House a year ago, careful budgeting, rigorous control of expenditure and sound administration will be needed, not because a donor insists on it, nor because next year’s grant depends on it; but because, without determined financial discipline, resources will be dissipated. If we do not practise rigorous selfdiscipline, the momentum of development will suffer.62

Careful management of the new mineral resources led to rapid and sustained growth in Botswana.63 GDP per capita in 2005 US dollars grew from $340 in 1968 to $755 in 1973, the year Botswana achieved budgetary self-sufficiency. In the next period, between 1973 and 1980, incumbents used state revenues to fund rural development projects as they moved to legitimate the new institutions they had built by transferring tangible benefits to the Batswana.64 For instance, in November 1973, incumbents in Botswana adopted the Accelerated Rural Development Programme (ARDP)—the first major post-independence program of development for rural areas.65 Under it, the government built over 60 new clinics and health posts across the country, at a cost of US$75.31 million.66

144

Domestic Politics and Drought Relief in Africa

Thus, in Botswana, state institutions were created under scarcity in the 1960s and 1970s, but then legitimated and consolidated with revenues from mining in the 1980s and 1990s. In so doing, it seems that Botswana successfully followed a capital-intensive model of modern state-formation.67 While common applications of Tilly’s framework to Africa, for instance Herbst’s States and Power in Africa, contend that African states are on the whole weak because the process of state formation did not involve the extensive bureaucratic development necessary for successful warmaking, few have considered the capitalintensive alternative. The growth of Botswana’s mineral sector strengthened the ability of incumbents to use material transfers to the rural population to secure the position of many of the new institutions they had created at the local level, so much so that some scholars contend that the BDP government spent “its way out of political trouble.”68 Botswana’s example suggests that, in the absence of warmaking as the primary imperative for bureaucratic growth, there are conditions under which economic expansion (capital) might produce a viable state. But prior to this consolidation of a capital-intensive mode of state formation, vulnerability dictated how Khama and his government responded to drought in 1978. I show in the next section that, like insecure incumbents in Kenya and Zimbabwe, Khama’s government settled on direct food aid for adults and relied on institutions created by the executive to implement relief on the ground, as both measures could assist them in consolidating their position. Vulnerability and the Choice of Food Aid in Botswana

After about a decade of normal rainfall, Botswana received erratic and less-than-adequate rainfall over the 1978-1979 growing season. A joint WFP/FAO assessment of the drought estimated that Botswana faced a grain deficit of between 110,000 and 140,000 Mt over the 1979-1980 consumption year.69 The seriousness of the country’s food situation was underscored when the Botswana Agricultural Marketing Board (BAMB) notified parliament that it had enough stocks to feed the country for only three months.70 Unease over this situation brought close attention to BAMB’s day-to-day operations. Botswana’s daily newspaper ran a story claiming that BAMB had been defrauded of about 1,000 bags of corn.71 Authorities denied the story, but it became clear that, as in Kenya and Zimbabwe, the legitimacy of the new order and its government rested on how well incumbents addressed the crisis. In early February 1979, agitated members of parliament quizzed Khama’s government on its drought contingency plans. Vice President

The Evolution of Drought Relief Policy in Botswana

145

Masire tried to calm MPs by reporting that the government had taken appropriate steps, including carrying out a study, to form such plans.72 However, incumbents had largely ignored recommendations from the drought study they had commissioned in 1976 when many agricultural experts were warning of an impending drought in the country. In response to these predictions, the government, through the British Ministry of Overseas Development, engaged a drought consultant, Stephen Sandford of the Overseas Development Institute in London. Sandford was to provide an “analysis of drought relief measures currently proposed by the Ministry of Agriculture … [with emphasis upon] the hatching of fresh drought relief measures and identification of major relief measures that would benefit from more detailed study.”73 In his final report, submitted in May 1977, Sandford concluded that Botswana should establish permanent institutions to deal with drought. The government needed, he wrote, “an overall strategy and sets of institutions and procedures in which all aspects of drought can be dealt with in a coherent way and not piecemeal.”74 The drought relief expert made four recommendations. First, he suggested that the government come up with an overall national strategy in which it defined its objectives in dealing with drought, including human relief, crop production, the livestock sector, and rural water supplies. As to human relief specifically, Sandford contended that the government should recognize that “drought tips [people] over the edge into near-famine conditions because of its effect on their incomes rather than on their food supply.”75 For Sandford, this meant that smoothing incomes through cash-paying labor-based relief should be a key component of the government’s strategy. Second, Sandford recommended the creation of a drought early-warning system with semi-automatic relief triggers, thus removing control of when drought was declared and when relief would begin from incumbents. Third, he stressed that the government must define clear lines of responsibility for officials to monitor drought and implement responses. Finally, Sandford proposed that incumbents create standing drought relief coordinating committees, in which district-level responsibility for relief resided in District Development Committees with the district commissioners as chairs. At the national level, he recommended the creation of an Inter-Ministerial National Drought Committee, which would consist of the Ministries of Agriculture, Health, Works and Communication, Natural Resources and Water Affairs, and Finance and Development Planning, as well as the Botswana Meat Commission, the Botswana Agricultural Marketing Board, and the Bank of Botswana.

146

Domestic Politics and Drought Relief in Africa

The consultant incumbents hired proposed programs of relief and a structure of administration similar to those advocated by technocrats in Kenya and Zimbabwe; like them, he stressed the establishment of standing relief institutions and a focus on smoothing incomes through labor-based relief over smoothing consumption with food aid.76 However, consistent with what we have come to expect of insecure incumbents, leaders in Botswana rejected Sandford’s recommendations.77 As a result, when drought struck in 1979 and relief became necessary to avert famine, the government was unprepared. Gooch and MacDonald note that although leaders recognized the need for a well-planned relief program, by 1979 no policy decisions had been made and no responsibilities had been defined for a wide range of urgent matters: what exactly constituted a drought; when relief would start; the type of relief to be provided to able-bodied adults; what the composition of rations would be or their quantity and source; or how domestic water would be provided.78 Every aspect of drought relief was open for incumbents to shape in ways consistent with their immediate political interests. Because insecure incumbents had refused to tie their hands by committing to a standing administrative structure and program of relief, their early steps were inevitably ad hoc. For instance, incumbents announced that they were allocating US$322,653 (in 2005 terms) for the construction of six new BAMB grain depots at Ghanzi, Kasane, Serowe, Kanye, Tahane, and Lenyeletse.79 This infrastructural expansion of BAMB would increase grain storage capacity, but it could have had no effect on the immediate threat of famine. Incumbents also somewhat hurriedly put together an Inter-Ministerial Drought Working Group in April 1979 to discuss the drought situation and recommend a program of relief.80 As Khama’s government moved to address drought in Botswana, the interests generated by weak institutional incumbency and the BDP’s continued preoccupation with the political challenges posed by the BNF shaped the government’s relief program more than its democratic politics or the professionalism of its bureaucracy. A month after the first meeting of the Inter-Ministerial Working Group, President Khama declared the country drought-stricken and initiated government relief in the form of food aid in the rural areas81 The provision of food aid was to occur in three stages. The first would provide food to vulnerable Batswana—children under five, expectant and lactating mothers, TB patients, and all primary school children in registered schools—while Stages II and III targeted able-bodied adults. Like insecure incumbents in Kenya and Zimbabwe, incumbents in Botswana also took advantage of the mobilizational, credit-claiming,

The Evolution of Drought Relief Policy in Botswana

147

and legitimation opportunities created by distributing food to adults. Over the course of the relief program, Khama and his ministers crisscrossed the country addressing kgotlas and other public gatherings on the government’s relief program. With parliamentary and council elections scheduled for October 20, 1979, food aid distribution became a key campaign issue.82 The 1979-1980 relief effort also included a scheme that provided funds for the purchase of old and weak cull cattle at roughly US$154 a head (in 2005 terms). The first objective of the program was to inject cash into drought-affected areas. To avoid affecting the overall quality of Botswana’s successful beef export sector, incumbents decided to use the cull cattle as additional food in the relief program. Thus, district officials were directed to slaughter one beast per village or school per week.83 The imprint of weak institutional power was evident in the administrative institutions and selection mechanisms the government adopted for Stages II and III of the food aid program. Similar to their insecure counterparts in Kenya and Zimbabwe, vulnerable incumbents in Botswana delegated decisions about access to free food and the implementation of the adult food aid program to central government agents at the local level, to appointed district commissioners, and to the new elected councils. By doing so, incumbents advanced toward their goal of transferring all service delivery at the local level from traditional authorities to modern institutions. Prior to independence in 1966, chiefs and tribal authorities had been responsible for the limited human relief measures adopted by the colonial state.84 Thus, successfully responding to drought through these new institutions held out the prospect of aiding in their legitimation. At the national level, government officials established three executive committees to implement and monitor drought relief measures across the country: an Inter-Ministerial Drought Committee (IMDC), chaired by the Ministry of Finance and Development Planning; an Agriculture and Water Committee (AWC), chaired by the Ministry of Agriculture; and the Food and Nutrition Committee (FNC), chaired by the Ministry of Local Government and Lands. The IMDC was the primary service and coordinating organ for the relief program, while the agriculture and water committee was responsible for “handling all livestock, arable land and water relief measures.”85 Food aid fell under the FNC. In this regard incumbents placed responsibility for delivering food to Batswana through the same government ministry that was responsible for the new system of local administration introduced in 1965.

148

Domestic Politics and Drought Relief in Africa

At the local level, the responsibility for distributing food aid to adults was delegated to district councils. Incumbents then made the two central government appointees to councils, district commissioners and council secretaries, “joint managers with responsibility for all aspects of the drought relief programme in their respective districts.”86 These officials were responsible for monitoring the effects of drought in their districts, declaring and implementing relief Stages II and III, and making special provisions for the relief of the destitute.87 To assist them in their tasks, District Drought Committees and Village Drought Committees were established across the country. District commissioners and council secretaries chaired the District Drought Committees. Chiefs did not play a significant role in any element of the government’s drought relief program. Despite the suggestion of administrative criteria for adult qualification for free food aid, with a threshold of a 40-percent malnutrition rate among children under 5 for Stage II rations, and a 60-percent malnutrition rate for Stage III relief, in practice district commissioners declared Stages II and III without relying on nutritional data. There was also a suggestion that districts attempt to exclude from government aid households with alternative sources of support, such as those with large numbers of cattle, or with food reserves of their own. However, selection and targeting proved too politically costly for district officials. Thus, district commissioners expanded the distribution of free food aid to able-bodied adults so widely that it became universal in some districts. Commissioners in 6 of the 9 rural districts declared Stage II conditions88 and in 3 of those 6 food rations were distributed to all adults. Another district used development funds collected from local taxes and the sale of stray cattle to purchase and distribute aid to everyone, even though only a small area qualified for Stage II relief. Through district commissioners, councils also exercised a great deal of discretion in identifying beneficiaries for the cattle purchase scheme. Incumbents’ directive was that councils should implement the cattle scheme using their knowledge of local ownership patterns to distribute the purchases equitably.89 Councils were to decide by themselves from whom among the local cattle owners they would buy drought-affected cattle, purchases that became another important form of protection from drought for the rural Batswana. In the first year of the scheme, the government approved close to US$400,000 (in 2005 terms) for the purchase of cattle and then increased the amount available to US$1.56 million (in 2005 terms) the following year.90 Incumbents also allocated each district council additional discretionary funds for use during the relief period. In all, incumbents allocated about US$1.1 million for discretionary relief

The Evolution of Drought Relief Policy in Botswana

149

with the understanding that the money be used by councils to purchase additional food, clothing, and medical supplies or cover the cost of logistical support. Discretionary funds added to the patronage at the disposal of incumbents and their new institutions at the local level during an election year. Table 5.1 below summarizes the per capita district-bydistrict discretionary allocations. Table 5.1: District Discretionary Allocations for Relief, 1979-1980 District

District Population

Total Amount Allocated in Pula

Total Amount in 2005 USD

Per capita allocation in 2005 USD

Kweneng

117,127

P44,000

$135,266

$1.15

Kgatleng

44,461

P24,000

$73,781

$1.60

Kgalagadi

24,059

P15,000

$46,113

$1.92

North East

36,636

P7,000

$21,520

$0.60

South East

30,648

P15,000

$46,113

$1.50

Southern

104,182

P50,000

$153,711

$1.50

Central

323,329

P120,000

$368,907

$1.14

North West

75,997

P60,000

$187,453

$2.46

Ghanzi

19,096

P15,000

$46,113

$2.41

775,535

P350,000

$1,078,980

Total

Source: MacDonald and Austin, A Human Drought Relief Programme for Botswana.

The food aid for adults in the 1979-1980 drought relief campaign and the administrative institutions used to implement it fell well short of the expert recommendations incumbents received from Sandford in 1977, but they were consistent with the political objectives of the new leaders. Thus, even in democratic Botswana, insecurity determined how incumbents responded to drought. It led Khama’s government to reject establishing a standing drought administrative structure or committing to

150

Domestic Politics and Drought Relief in Africa

a program of relief predicated on smoothing incomes for adults. Without these prior commitments, the response in 1979-1980 was subject to the immediate political interests generated by weak institutional incumbency and anxieties about the BNF in the run-up to elections in October 1979. In the next section, I show that the form of relief offered to adults changed in predictable ways following the resolution of the BNF threat and the institutionalization of modern state structures in Botswana. The Consolidation of Institutional Incumbency in Botswana, 1980 and Beyond

Results of the 1979 elections, held at the height of the 1979-1980 drought relief program, strengthened incumbents’ position against challenges mounted by the BNF. The BDP won 29 of the 32 parliamentary seats contested, with an average winning margin in each constituency of over 52 percent.91 The BDP’s performance was even more impressive in council elections, where the central role played by district councils in administering key components of the relief program paid off. For the first—and only—time since the creation of elected councils, the BDP secured majority control of all of them. Many opposition candidates attributed the BDP’s success to the drought relief program. For instance, Motsamai Mhpo of the Botswana Independence Party accused the BDP of using the 1979-1980 relief intervention to win rural votes.92 In 1979, Mhpo lost his Okavango seat to a BDP candidate, managing only 36 percent of the vote to the BDP’s 60.28 percent in a constituency that he had won with 56 percent of the vote in 1969 and 52 percent in 1974.   Election gains in 1979 addressed the proximate vulnerability generated by the militancy of the BNF in the late 1970s. However, secure incumbency in Botswana still depended on the public’s acceptance of the modern state. Since independence, as scholars have suggested, Botswana had benefited from Seretse Khama’s ability to elicit deference to the new political order under construction.93 Yet questions remained regarding whether the modern state system had become institutionalized94; that is, whether political behavior in Botswana was now shaped and constrained by the formal and informal rules of the new state institutions incumbents created in the early period of independence. Khama’s death in office, from natural causes, in July 1980, and the regulated and peaceful ascension of his vice president proved that Botswana’s modern structure of government had become institutionalized.95 While executive successions in other African countries have generated regime instability and discord,96 including Moi’s succession of Kenyatta, Quett Masire’s ascension to the presidency in Botswana was

The Evolution of Drought Relief Policy in Botswana

151

quick, constitutionally governed, and collectively supported by all political figures in the country. As Huntington described institutionalization, the succession in Botswana confirmed that the new modern institutions created since independence to replace traditional authorities were coherent and adaptable, could respond to complexity, and were now fully autonomous from traditional ascriptive power. Within the stipulated seven days, Botswana’s parliament met and unanimously elected Masire, a co-founder of the BDP, vice president, and Minister of Development Planning, as the new president.97 Parson rightly adds that “this smooth changing of the guard may indicate the presence of a successful process of parliamentary succession and its underpinning of legal-rational legitimacy.”98 While Seretse Khama had benefited from his traditional status in the early period of modern state building, Masire’s peaceful ascension to the presidency affirmed that modern rational-legal legitimacy was entrenched in the country. Thus, the Office of the President responded to South African questions about the future of the country without Khama by declaring that: “Botswana was not a one-man band.”99 After succeeding Khama in 1980, Masire led Botswana through a period of unparalleled economic growth and political stability. Despite drought from 1981 to 1987, Botswana’s GDP per capita more than doubled from $1,380 in 1980 to $2,820 by 1990. During Masire’s tenure, incumbents persisted with the strategy of maintaining and expanding services to enhance the legitimacy of the central government. Thus, through rapid economic growth, “state allocated economic resources underpinned the folding of traditional authority systems into the legal-rational one and the hegemony of the latter over the former by 1987.”100 Economic success and the continued expansion of service delivery translated to the political realm. Masire led the BDP to success in the 1984 elections, retaining and consolidating his position as president. Then in the 1989 elections, the BDP won 91.18 percent of the seats contested—the largest share of parliamentary seats won by the party.101 The consolidation of incumbency under an institutionalized modern political structure changed the interests that guided incumbents during drought relief policy-making, as was evident early in Masire’s presidency. In his first address to parliament as president in November 1980, Masire stressed that in the future “we must husband our resources more carefully and devote them to the most productive uses we can find.”102 The weight placed on careful planning by the new president affected how incumbents responded to two experts hired by the government, Toby Gooch and John MacDonald, to review the 1979-1980 drought relief program.103 The two concluded that, although the program had

152

Domestic Politics and Drought Relief in Africa

“unquestionably saved the lives of countless people,”104 the government had been slow to address logistical problems with the delivery of food, incumbents adopted a poor organizational structure, and the distribution of food aid to adults was wasteful. To address the problems associated with the food delivery system, Gooch and MacDonald recommended making those involved in the distribution of food responsible to district authorities, not incumbents at the center. Beyond the delivery of food itself, they proposed that the government establish an organization responsible for monitoring food supplies and demand. More importantly, the two experts revived Sandford’s earlier proposals to smooth incomes through labor-based relief for adults in Botswana. Changed micro-political conditions led incumbents to respond differently now to expert recommendations on drought relief than they had when Sandford reported. First, incumbents organized a government seminar, chaired by the administrative secretary in the Office of the President, to discuss the 1979-1980 experience and to develop a program for the future. Opening the seminar, the president’s administrative secretary noted that unlike the 1978 drought symposium, which merely expressed intent on the part of government, their session would be able to draw upon experience in making specific recommendations for an improved drought relief program in Botswana.105 Then, between June 1981 and January 1982, incumbents sent the proposals drawn up by Gooch and MacDonald to the districts for wide public discussion at kgotlas before the cabinet announced its decision endorsing the recommendations.106 The Introduction and Consolidation of Labor-Based Relief in Botswana

The ascendancy of instrumental technocratic interests in drought relief policy-making led to the adoption of labor-based relief in 1982 and its subsequent institutionalization in 1991. After a partial recovery from the 1978-1979 crisis, drought returned to Botswana in 1981-1982, and persisted for six years.107 On April 1, 1982, President Masire declared Botswana drought-stricken and initiated a national relief program.108 However, this time the programs of relief offered and the administrative structures used to implement it were more closely in line with many of the recommendations drought experts had offered. For instance, incumbents retained food aid for vulnerable groups but removed it for ablebodied adults,109 and they included measures for water provision and livestock relief.110 Because drought was not broken for the entire country

The Evolution of Drought Relief Policy in Botswana

153

until 1990, each component of the relief program was renewed with each successive failure of rains in Botswana.111 Consistent with the new interest in limiting waste—husbanding state resources more carefully—secure incumbents replaced food aid for able-bodied adults with labor-based relief. As in Zimbabwe during periods of secure incumbency, the labor-based relief element provided “temporary employment on local development projects (e.g. road maintenance, small dams, and general construction),” as an alternative to food aid.112 However, unlike Zimbabwe’s labor-based relief program, which paid participants in food rations, Botswana’s offered a drought subsistence wage of US$14.42 a week to workers and US$19.22 to supervisors.113 The primary objective of the wage-based relief program was to compensate rural Batswana for income lost due to crop failure. As summarized in Table 5.2, labor-based wages compensated Batswana for as much as a third of the income lost due to poor agricultural performance during the prolonged drought. Table 5.2: Income from Crop Failure Replaced by Labor-based Relief Transfers, 1982-1985 Year

1982

1983

1984

1985

Crop Production (Mt)

17,220

14,425

6,925

20,000

Shortfall from Normal Production

42,780

45,575

53,075

40,000

Imputed Value of Lost Income (in 2005 USD)

$18.0m

$20.35m

$21.26m

$13.76m

Income Replaced Through Labor-based Relief (in 2005 USD)

$3.6m

$6.1m

$7.0m

$5.12m

20

30

33

37

Labor-based Wages as Percent of Lost Income

Source: Rural Development Unit, Ministry of Finance and Development Statistics.

154

Domestic Politics and Drought Relief in Africa

As rain gradually returned beginning in 1986, incumbents added the Accelerated Rainfed Arable Programme (ARAP) to the ongoing relief program. ARAP was intended to help small-to-medium-sized farmers recover from the drought. It was a 5-year “project with an initial estimated operational budget of about P30 million . . . to be [used for] a once off assistance scheme which would assist all groups of farmers engaged in dryland farming.”114 The objectives of the program were threefold: to “increase arable production through dryland farming with the ultimate goal of reducing grain deficits and achieving food grain self-sufficiency in the short and long run”; to “enhance rural development and welfare by increasing rural incomes received directly or indirectly from arable production”; and to “optimize rural income distribution by concentrating on all farming households and creating productive and remunerative employment in the lands areas, and therefore reducing rural urban migration.”115 As summarized in Table 5.3, ARAP contained eight packages under which farmers could benefit from government assistance. According to Amis, of all the elements of the 1982-1990 relief and recovery program, ARAP was the only one that produced negative distributional outcomes. Many of its components, in particular ploughing, fencing and water development, became easy routes by which wealthy, well-placed farmers benefited from government aid in ways that they could not under the other programs of the 1982-1990 relief effort.116 Changes to the form of aid offered to adults in 1982 produced concomitant changes to the administrative structures incumbents relied on to implement relief. At the level of the central government, a permanent Inter-Ministerial Drought Committee (IMDC) was established in 1982 to monitor drought and coordinate relief programs. Its chairmanship went to the Ministry of Finance and Development Planning. The other members of the IMDC were drawn from five key line ministries: Local Government and Lands, Education, Health, Mineral Resources and Water Affairs, and Agriculture. Food distribution to vulnerable groups was handled by a new organ, the Food Resources Department (FRD), which replaced IFP and the Food and Nutrition Committee from the 1979-1980 relief effort.117 To obviate the confusion that had characterized the distribution of food in the earlier period, the FRD developed a single “Manual on Human Relief” which was sent to all district- and village-level officials involved in the drought relief program. 118 The manual detailed the groups to be covered by the feeding program and indicated where the vulnerable were to receive their rations.

The Evolution of Drought Relief Policy in Botswana

155

Table 5.3: Summary of ARAP Components and Types of Assistance, 1985-1989 ARAP Component Ploughing

1

Farmer could plough for him/herself or hire someone with draught power and would be paid P50.00 per hectare

Row planting

Weeding

1

Farmer could hire a contractor or do the operation himself for a grant of P20.00 per hectare

1

Government paid the farmer, or the person hired by the farmer, P10.00 per hectare for weeding

Fertilizer provision

Seed Provision Destumping

1

Type of Government Assistance

1

1

Farmers were assisted with fertilizer (2:3:2 compound) sufficient for 3 ha, i.e., 4 bags/ha for a 2 total of 12 bags (1 bag = 50kgs) Farmers were to be provided with enough seed to plant up to a maximum of 10 hectares Farmers were paid P50.00/ha irrespective of the 3 number of stumps removed per hectare

Fencing

The maximum ceiling for the ARAP fencing 4 package was P1,200.00

Water Provision

The maximum ceiling for the package was P20,000.00 per group/community

Support given for a maximum of 10 hectares.

2

During the 1987/88 season farmers were assisted with sufficient fertilizer for a maximum of 4 ha, i.e., 200 kg of a 2:3:2 compound. 3

In 1987/88, the government changed the payment for destumping to take into account the amount of work done. Thus, for the same 10 ha maximum farmers are paid P30.00/ha for 20 stumps or less, P40.00/ha for 21-30 stumps and P50.00/ha 7 for more than 30 stumps. 4

Under this package, the P1,200.00 went to the purchase of fencing materials and paying the fence erector. Source: Montshiwa and Mhlanga, Accelerated Rainfed Arable Programme [ARAP], Annual Report 1987-88; Amis, Financial Efficiencies in Drought Relief.

Through the FRD, the Ministry of Local Government and Lands handled food-aid imports, local purchases, and the distribution of food to schools and health centers. The Ministry of Education oversaw the feeding of primary school children, while the Ministry of Health was responsible for monitoring the nutrition situation and, when necessary, organizing

156

Domestic Politics and Drought Relief in Africa

on-site feeding of children at health centers. Finally, the Ministry of Mineral Resources and Water Affairs tackled problems associated with water supply, while the Ministry of Agriculture administered ARAP through its agricultural extension staff. At the local government level, responsibility for the administration of the drought relief program was transferred to District Drought Committees chaired by an elected district administrator—a key change from 1979-1980, when incumbents relied on appointed district commissioners to administer relief at the local level. The other members of DDC were district-level bureaucrats from the Ministries of Agriculture, Health, Mineral Resources and Water Affairs, and the Treasury. Each district drought committee set up a Drought Office whose function was to vet labor-based relief projects submitted by Village Development Committees (VDC), implement projects approved by the DDC, and process payments for labor-based relief for the community. Individual works projects were identified by VDC and administered and implemented by a Drought Relief Technical Officer assigned to a VDC. Thus in principle, projects were to be selected after consultation with the villages and directed towards the construction of local infrastructure. The duration of these projects ranged from 3 weeks to a maximum of 8 months. Districts were encouraged to halt all projects at the end of the year to encourage and enable participants to plough their own fields. Jobs were also frequently rotated amongst the participants, resulting in over 30,000 Batswana gaining employment from the program each year. In 1982-1983, incumbents spent $4.8 million on works projects that included 2,500 kilometers of road improvements, 520 traditional huts, 68 dams, 19 soil reclamation projects, 17 wells, 230 pit latrines, 26 village clean-up projects, 27 clearing projects, 20 gardens, and 23 kgotla improvement projects.119 The funds for projects were funneled to VDC through the FRD but administered by the district council at the district level.120 Between 1982 and 1989, when laborbased relief projects were terminated for most parts of the country, the program had involved a staggering 14,810 projects, which had provided jobs for 295,856 people for an average employment period of 76 days. Immediately after the 1982-1990 response, institutionalized incumbents took the extraordinary step of incorporating drought relief and rehabilitation programs into their overall development planning. The impetus for mainstreaming future relief interventions came from senior officials, especially those in the Ministry of Finance and Development Planning, who noted that government spending on relief between 1982 and 1990 was largely unplanned.121 As a percentage of total government development expenditures between 1982 and 1990, spending on the

The Evolution of Drought Relief Policy in Botswana

157

drought relief and recovery program grew from 3.4 percent in 19821983, reached a peak of almost 17 percent of development expenditures in 1985-1986, and remained above 10 percent of total development expenditures until 1989 (see Table 5.4 below). Table 5.4: Rate of Spending on Drought Relief, 1982-1989 Financial Year

Drought Relief Spending (2005 USD in millions)

Total Development Expenditure (2005 USD in millions)

Drought Relief Spending as Percent of Botswana’s Development Expenditure

Drought Relief Spending as Percent of Ministry of Agriculture’s Development Expenditures

1982/83

10

308

3.37%

7.70%

1983/84

21

249

8.50%

31.10%

1984/85

44

295

14.93%

54.40%

1985/86

40

236

16.89%

81.80%

1986/87

54

388

14.03%

91.00%

1987/88

92

579

15.84%

69.90%

1988/89

82

703

11.63%

52.40%

Source: Amis, Financial Efficiencies in Drought Relief.

Unplanned spending on drought relief was particularly problematic for the Ministry of Agriculture, which was forced to devote as much as 91 percent of its annual allotted development budget to relief. In all, average spending on drought relief and recovery in these years was over 12 percent of Botswana’s total development expenditure. Officials in the Ministry of Finance and Development Planning also noted that because of the emergency spending on drought relief, the government exceeded its target for total development expenditure in the 1985-to-1991 period by 16 percent.122 Relief had saved lives, but planners in government were concerned that if it remained ad hoc, it would become unaffordable and unsustainable. In response, secure incumbents supported plans to institutionalize drought relief policies in Botswana. As a first step in the rationalization process, incumbents built into the 1991-1997 Development Plan (NDP7) a 3-to-4 percent increase in annual development expenditures to fund drought relief programs.123 According to Thompson, the flexible ceiling

158

Domestic Politics and Drought Relief in Africa

on the resources incumbents were willing to set aside for drought relief was the strongest indication of their intention to “target the needy, rather than offering blanket coverage as most programmes did in the 1980s.”124 In addition to this ceiling, cash-paying labor-based projects for ablebodied adults were written into the standing drought relief contingency plans developed by incumbents in 1991. Thus, in NDP7, incumbents established workfare programs during “nondrought years to create infrastructure and employment for the rural population … [However] the scale of the programs [would be expanded] … during drought periods.”125 By writing drought relief contingency planning into NDP7, incumbents in Botswana also enabled the consolidation of many of the drought relief institutions created during the 1982-1990 relief program. In effect, the contingency planning in NDP7 turned the key drought relief organs—the Inter-Ministerial Drought Committee, the Early Warning Technical Committee, the Food Resources Department, ten District Drought Committees, and over 580 Village Development Committees—into permanent statutory institutions. As a result, they continued to meet and maintain their capacity to respond to drought even in non-drought years. Drought relief planning and implementation become a permanent and long-term responsibility for officials assigned to these bodies, including those who, during the ad hoc period, had complained that drought relief responsibilities were taking them away from their “real” work. The institutionalization of drought relief contingency planning in Botswana meant that incumbents met all subsequent droughts in Botswana (between 1992 and 1993, 1995 and 1997, and 2001 and 2005) with greater policy and institutional preparedness.126 Responses have followed NDP7’s clear guidelines: free food to vulnerable groups, laborbased relief for everyone else, and administration through standing relief institutions. As in 1982-1990, adults self-selected to labor on the public works projects chosen by Village Development Committees and administered by District Drought Committees. Beginning in 1992, there were two types of works projects for drought-affected adults seeking to benefit from government aid. One type paid Batswana to help fulfill the national development plan, NDP7, including work on water projects, building agricultural stations, constructing viewing sites in game parks across the country, clearing bush for fire breaks, and erecting cattle kraals and fences at markets.127 In addition to these centrally determined projects, adults could receive a drought subsistence wage by working on projects selected by Village Development Committees or District Drought Committees.

The Evolution of Drought Relief Policy in Botswana

159

These typically involved rural road maintenance. with the daily wage set at $2.90—intentionally below the requirements of a normal breadbasket for a family. This met the requirement that relief work not “compete with the rural labor market, such as cattle herding, shop assistants or hawkers.”128 Thus, in contrast to Mugabe and ZANU who denied that Zimbabwe was drought stricken so that they could use their monopoly over access to food as a political tool in the 2005 election, institutionally secure incumbents in Botswana responded to the region-wide drought by expanding the country’s standing labor-intensive public works projects, establishing a Livestock Feed Subsidy and Cattle Purchasing Project, and preparing rural Batswana for recovery by offering them free seed packages.129 To implement their relief program, incumbents relied on the five standing institutions: the Inter-Ministerial Drought Committee; the Early Warning Technical Committee (EWTC); the Food Resources Department; District Drought Committees; and Village Development Committees. The IMDC coordinated all elements of the program, while the EWTC provided it with regular updates on food production and consumption, as well as issuing food security reports to all levels of government. The FRD, within the Ministry of Local Government, was responsible for administering the food aid component of relief during the drought and the supplementary feeding program to vulnerable groups in non-drought years.130 The FRD had the logistical responsibility for transporting food rations from over 20 depots, with a capacity of close to 25,000 Mt in 1992, to feeding points (schools and clinics) scattered around the country. The administration of labor-based relief at the local level was the responsibility of the ten District Drought Committees, with support from the hundreds of Village Development Committees. Summary and Comparisons with Kenya and Zimbabwe

The objective of this chapter was to test whether drought-relief policy decisions are tied to the strength or weakness of political incumbency even in African countries with different political, economic, and agricultural conditions from Kenya and Zimbabwe. In the past, scholars who have addressed Botswana’s impressive record of responding to drought have stressed the country’s distinctive features: the national recognition of drought as a problem; the country’s proximity and close links with the highly developed foreign trade and transport infrastructure of South Africa; its relatively robust communications and transport systems; its functioning market economy and remarkable economic growth; and,

160

Domestic Politics and Drought Relief in Africa

finally, the presence of a competent and professional bureaucracy.131 These scholars also contend that, above all, Botswana’s responsiveness and the programs of relief it has adopted are a product of its democratic political system.132 As one scholar insists, “the politics of famine prevention in Botswana are intimately linked with the accountability of the ruling party to the electorate, the activism of the opposition, the vigilance of the press, and […] the rising demands for public support on the part of the affected populations.”133 Thus, the features that set Botswana apart from Kenya and Zimbabwe (namely, multi-party democracy, sustained growth, and an effective bureaucracy) have generally produced narratives predicated on its exceptionalism in explaining its response to drought in 1979-1980, the shift to labor-based relief in 1982, and the subsequent consolidation of work-based relief in 1991. It is in this vein that Holm and Morgan contend that political openness explains not only Botswana’s success, but also, and sadly, why “other African countries will not follow Botswana’s lead.”134 By focusing on the policy-making interests generated by strong or weak incumbency, this chapter demonstrated that Botswana is not an exception. In democratic Botswana, as in de jure one-party state Kenya and de facto one-party state Zimbabwe, the often different and sometimes similar programs of relief adopted in each country were determined by the strength or weakness of incumbency. In all three countries, vulnerable incumbents saw drought as a political opportunity to use patronage, in the form of food aid programs, to bolster their control of the state, while improvements in the strength of political incumbency resulted in shifts to labor-based relief in both authoritarian Zimbabwe and democratic Botswana. This chapter demonstrated that incumbent insecurity (or security) was as central to how incumbents responded to drought in Botswana as it was in the other two countries, and that programs of relief often contributed to the stability and vulnerability of incumbents in ways that shaped subsequent responses to drought. However, in addition to confirming the general relationship between weak or strong incumbency and programs of relief offered to adults across Africa, the analysis of Botswana revealed that how incumbents build stable tenure for themselves is not inconsequential. This chapter illustrated that, in Botswana, early post-independence conditions encouraged incumbents to secure their position by building strong modern political and administrative institutions. Once institutionalized incumbency was established, it created the stable long-term policy-making conditions under which incumbents could support the adoption of a standing administrative commitment to respond to droughts and to do so through cash-paying labor-based relief programs for adults.

The Evolution of Drought Relief Policy in Botswana

161

But this different trajectory in Botswana is consistent with my general argument. Incumbents’ responses were tied to assessments of their vulnerability. Stable and institutionalized incumbency had the effect of channeling and institutionalizing challenges to incumbents through the electoral system. BDP dominance in these freely-contested polls created the political space for secure incumbents to support a standing and predictable response to droughts. Thus, despite establishing and consolidating democracy, drought relief policy selection in Botswana was determined by the same micro-political considerations as in Kenya and Zimbabwe. Political incumbency proves to be informative. Notes 1 Republic of Botswana, National Development Plan 6, 1985/91; and Republic of Botswana, National Development Plan 7, 1991/97. 2 Republic of Botswana, National Development Plan 7, 1991/97, p. 59. 3 Letter from the Permanent Secretary, Ministry of Agriculture, A34/6XVII(32)PS (April 21, 2005); Thompson, Drought Management Strategies in Southern Africa. 4 In Southern Africa, Britain faced competition from the Germans in Southwest Africa (what became Namibia), whose interest in eastward expansion into the interior of the continent threatened to cut off Crown access to the heart of Africa. The British at the Cape were also concerned about the westward expansion of the Transvaal Republic under Paul Kruger. 5 Vengroff, Botswana, p. 25. See also Acemoglu et al., “An African Success Story”; and Sillery, Founding a Protectorate. 6 Lange, Lineages of Despotism and Development, pp. 143-54, offers a recent description of British colonialism in Botswana. See also Rey, Monarch of All I Survey; Samatar, An African Miracle; du Toit, State Building and Democracy in Southern Africa, pp. 17-73; Picard, The Politics of Development in Botswana; Parson, Botswana. 7 In practice, a gathering of the tribe (kgotla) provided nominal checks on the exercise of authority by traditional chiefs. For a comprehensive account of the traditional system of Tswana governance, see Schapera, Tribal Innovators; and Schapera, “The Tswana.” 8 Lange, Lineages of Despotism and Development, pp. 139-68; Acemoglu et al., “An African Success Story”; Englebert, State Legitimacy and Development in Africa, pp. 105-11; du Toit, State Building and Democracy in Southern Africa, pp. 17-73. 9 Picard, The Politics of Development in Botswana; Parson, Botswana; Proctor, “The House of Chiefs and the Political Development of Botswana,” pp. 59-79. The name of this body was later changed to the African Advisory Council. 10 This is what Jackson and Rosberg describe as empirical statehood in “Why Africa’s Weak States Persist,” pp. 1-24.

162

Domestic Politics and Drought Relief in Africa

11 Parson, Botswana; Colclough and McCarthy, The Political Economy of Botswana. 12 Parson, “Succession, Legitimacy, and Political Change in Botswana, 1956-1987,” pp. 97-142; Picard, The Politics of Development in Botswana; Proctor, “The House of Chiefs and the Political Development of Botswana,” pp. 59-79. 13 The other African members on the Legco were: Seretse Khama, Quett Masire, Goareng Mosinyi, Archelus Tsoebebe, Tsheko Tsheko, Leetile Raditladi, Letlole Mosielele, Dr. Silas Modiri Molema, and Joe Gugushe. 14 See Picard, “From Bechuanaland to Botswana,” pp. 3-25; Parson, Botswana; Polhemus, “Botswana Votes,” pp. 397-430; Nengwekhulu, “Some Findings on the Origins of Political Parties in Botswana,” pp. 47-75; and Wiseman, “Multi-Partyism in Africa,” pp. 70-79. 15 Acemoglu et al., “An African Success Story”; Samatar, An African Miracle; Henderson, “Seretse Khama,” pp. 27-56. For a full account of his marriage to Ruth Williams and the challenges it presented for colonial officials and the Bamangwato chieftancy, see, among others, Lange, Lineages of Despotism and Development, pp. 151-54; Parsons, “Seretse Khama and the Bangwato Succession Crisis, 1948-1953,” pp. 73-95. 16 Colclough and McCarthy, The Political Economy of Botswana, p. 36. Proctor, “The House of Chiefs and the Political Development of Botswana,” pp. 59-79, also stresses the importance of this cleavage in shaping inter-party relations in Botswana. 17 Wiseman, “Multi-Partyism in Africa,” p. 75. 18 Proctor, “The House of Chiefs and the Political Development of Botswana,” p. 61. See also Picard, The Politics of Development in Botswana. 19 Wiseman, “Multi-Partyism in Africa,” p. 75. 20 Proctor, “The House of Chiefs and the Political Development of Botswana,” p. 63. 21 The House of Chiefs was to have 15 members: 8 chiefs from the legallyrecognized Tswana Tribes; 4 elected sub-chiefs; and 3 specially-elected members to represent non-Tswana groups such as the Bakalanga, Bayei, and Basarwa. 22 Parson, “Succession, Legitimacy, and Political Change in Botswana, 1956-1987,” pp. 97-142. 23 Masire, Very Brave or Very Foolish?, p. 46. 24 See Bechuanaland Democratic Party, Election Manifesto: This Is What We Stand For. 25 Masire, Very Brave or Very Foolish?, p. 41. 26 Republic of Botswana, Supervisor of Elections, Report on the General Elections, 1969. For explanations of the BDP victory in the 1965 election, see Masire, Very Brave or Very Foolish?; Parson, “Succession, Legitimacy, and Political Change in Botswana, 1956-1987,” pp. 97-142; Polhemus, “Botswana Votes,” pp. 397-430; and Vengroff, Botswana. 27 See Herbst, States and Power in Africa. 28 Acemoglu et al., “An African Success Story.” 29 Parson, Botswana. Parson contends that Botswana developed as a labor reserve for South African mines and industry. See Parson, “The ‘Labour Reserve’ in Historical Perspective,” pp. 40-57.

The Evolution of Drought Relief Policy in Botswana

163

30 Polhemus, “Botswana Votes,” pp. 397-430. Vengroff, Botswana, p. 88, also points to the BDP’s victory as the reason for the formation of the BNF. 31 See Picard, “Bureaucrats, Elections and Political Control,” p. 181. 32 Republic of Botswana, Supervisor of Elections, Report on the General Elections, 1969. 33 See “Angry Students Disrupting Admin,” Botswana Daily News (April 1, 1975); “Angry Strikers Refuse to Listen to Vice-President,” ibid. (July 30, 1975); “S-Phikwe Calm After Angry Mass Strike,” ibid. (July 31, 1975); “Illegal Strikes in Botswana: President’s Worry,” ibid. (July 31, 1975); “Miners’ Strike Results in Bitter War of Words,” ibid. (August 26, 1975); “H.E. Warns Political Agitators; Will Take Action if Necessary,” ibid. (August 26, 1975); “H.E. Warns Misguided Elements,” ibid. (December 22, 1975); “Kgari says 625 Were Sacked After Mine Riot,” ibid. (March 12, 1976); “Opposition Calls for Probe Into Mine Strike,” ibid. (March 16, 1976); “Kwelagobe Says Statements are Irresponsible, Malicious,” ibid. (March 23, 1976); “Matante’s Motion on S-P Mine Strike Thrown Out,” ibid. (March 30, 1976); “Students Demo,” ibid. (November 23, 1976); “Campus Situation Reaches an Apogee,” ibid. (November 24, 1976). 34 “Opposition Calls for Probe Into Mine Strike,” ibid. (March 16, 1976); “Kwelagobe Says Statements are Irresponsible, Malicious,” ibid. (March 23, 1976); “Matante’s Motion on S-P Mine Strike Thrown Out,” ibid. (March 30, 1976); “OLD CLOWNS..! Minister Slams into BPP Leaders on Demonstration,” ibid. (February 1, 1977). 35 “MP’s Give Blessing to Army Bill,” ibid. (March 31, 1977). 36 “Youth Taught Coup Tactics?,” ibid. (April 19, 1977). 37 Quoted in ibid. 38 “Government Withdraws Passports from 17,” ibid. (June 21, 1978); “BNF Denies Charges,” ibid. (June 29, 1978); “BNF Protests Passports Withdrawal,” ibid. (July 26, 1978). 39 Khama, “Speech Opening the 17th Annual Conference of the Botswana Democratic Party,” May 5, 1978. 40 Good, Diamonds, Dispossession & Democracy in Botswana, p. 40. Section 24 of Botswana’s Immigration Act confers on the president the power to impose special restrictions on aliens, including prohibited immigrant status, as he may consider necessary in the public interest. 41 Government of Bechuanaland, Chieftainship Act, No. 26 of 1965. 42 Proctor, “The House of Chiefs and the Political Development of Botswana,” p. 70. See also Gillet, “The Survival of Chieftaincy in Botswana,” pp. 179-85; Tordoff, “Local Administration in Botswana, Part I,” pp. 172-83. 43 Republic of Botswana, Tribal Land Act, 1968. Tribal Land Boards consisted of chiefs as ex officio members, and centrally-appointed officials. Colclough and McCarthy, The Political Economy of Botswana; Frimpong, “A Review of the Tribal Grazing Land Policy in Botswana,” pp. 1-16. 44 See Bechuanaland Democratic Party, Election Manifesto: This Is What We Stand For. 45 Government of Bechuanaland, The Local Government (District Councils) Act, No. 27 of 1965. The immediate effects of the act are analyzed by Tordoff, “Local Administration in Botswana, Part I,” pp. 172-83. 46 Vengroff, Botswana, pp. 87-88.

164

Domestic Politics and Drought Relief in Africa

47 Khama, “Presidential Address to the Sixth Annual Conference of the Botswana Democratic Party, Gaborone,” March 25, 1967. 48 Parson, Botswana. Picard, “Bureaucrats, Elections and Political Control,” pp. 176-205, discusses how district commissioners were used as extensions and representatives of the Office of the President at the local level. District commissioners also exercised judicial powers at the local level as they headed district internal security teams that reported directly to the president. 49 Picard, The Politics of Development in Botswana; Picard, “District Councils in Botswana,” pp. 285-308; Tordoff, “Local Administration in Botswana, Part II,” pp. 293-304. 50 Parson, Botswana. 51 Republic of Botswana, Mineral Rights in Tribal Territories Act, Act 31 of 1967. See Acemoglu et al., “An African Success Story,” for a discussion of the importance of this act in Botswana’s subsequent stable political and economic development. 52 Tordoff, “Local Administration in Botswana, Part I,” pp. 172-83. 53 Parson, “Cattle, Class and the State in Rural Botswana,” pp. 236-55. For an analysis of cattle ownership patterns in Botswana, see also Samatar, An African Miracle; and Samatar and Oldfield, “Class and Effective State Institutions,” pp. 651-68. 54 Republic of Botswana, Botswana Project 324/II Quarterly Progress Report, 1st January, 1982 to 31st March, 1982. 55 Ibid. 56 Parson, Botswana; Republic of Botswana, Transitional Plan for Social and Economic Development, 1966-70. 57 Samatar and Oldfield, “Class and Effective State Institutions,” pp. 65168; Parson, “Cattle, Class and the State in Rural Botswana,” pp. 236-55. 58 Botswana Meat Commission, Annual Reports, 1966 to 1994. 59 See McGowan and Associates, A Study of Drought Relief and Contingency Measures Relating to the Livestock Sector of Botswana, p. 1. 60 Colclough and McCarthy, The Political Economy of Botswana, p. 231. 61 Khama, “Presidential Address to the First Meeting of the Third Session of the First National Assembly,” December 9, 1968. 62 Khama, “Speech on the People’s Progress at the Opening of the Third Session of the Second Session of Parliament,” December 3, 1971. 63 Acemoglu et al., “An African Success Story”; Love, “Drought, Dutch Disease and Controlled Transition in Botswana Agriculture,” pp. 71-83; Lewis, “The Potential Problems of Diamond-Dependent Development,” pp. 14-28. 64 Dunning, Crude Democracy, pp. 258-67, contends that natural resource wealth in Botswana promoted democracy. 65 ARDP is discussed by Tsie, “The Political Context of Botswana’s Development Performance,” pp. 599-616; Good, “Interpreting the Exceptionality of Botswana,” pp. 69-95; Colclough and McCarthy, The Political Economy of Botswana, pp. 231-35; and Chambers, Botswana’s Accelerated Rural Development Programme. Cf. Holm, “Liberal Democracy and Rural Development in Botswana,” pp. 83-102, who questions the efficacy of ARDP in addressing rural poverty. 66 According to Colclough and McCarthy, the government of Botswana financed 55 percent of these expenditures by drawing on the growing surpluses

The Evolution of Drought Relief Policy in Botswana

165

from the mineral sector. The remaining 45 percent of the cost of ARDP was covered by rural development aid from Norway and Sweden. 67 See Tilly, Coercion, Capital, and European States, for a discussion of the paths to state formation followed by European countries. 68 Parson, Botswana, p. 51. 69 United Nations Development Programme, Report of the WFP/FAO Mission on Drought Relief in Botswana. This study also placed the value of crops lost to drought at P6.5 million (almost $20 million in 2005 USD). 70 “BAMB Stocks,” Botswana Daily News (January 12, 1979); “BAMB Maintains ‘House Keeping’ Stock,” ibid. (February 5, 1979). 71 See “BAMB Robbed of Bags of Corn,” ibid. (January 25, 1979); and “BAMB Denies Swindled Story,” ibid. (January 30, 1979). 72 “Measures Taken to Arrange for Drought,” ibid. (February 5, 1979). 73 Sandford, Dealing With Drought and Livestock in Botswana, p. A1. 74 Ibid., p. 110. 75 Ibid., p. 56. See also Vierich and Sheppard, Drought in Rural Botswana. 76 Other experts made similar proposals when it became clear that the country was in the midst of a major drought over the 1978-79 growing season. See United Nations Development Programme, Report of the WFP/FAO Mission on Drought Relief to Botswana; Austin, A Drought Contingency Plan for Botswana; Sheppard and Clement-Jones, Coping With Drought in Botswana. 77 Kreysler, “Nutritional Surveillance in Botswana,” pp. 221-31. 78 Gooch and MacDonald, Evaluation of 1979/80 Drought Relief Programme. 79 “BAMB Development Costs P104,954 for 1979/80,” Botswana Daily News (February 9, 1979). 80 “Body Meets to Talk on Drought,” ibid. (April 11, 1979). The working group was chaired by the Ministry of Finance and Development Planning. 81 “President Declares Botswana Drought-Stricken From June,” ibid. (May 29, 1979). 82 “We’re Trying Hard to Fight Drought,” ibid. (June 15, 1979); “Government Committed to Help in Drought,” ibid. (June 26, 1979); “VP Talks on Drought Relief Measures, Education,” ibid. (August 7, 1979); “Government Combats Drought, Foot and Mouth,” ibid. (August 16, 1979); “President Addresses 10,000 at Letlhakane,” ibid. (August 21, 1979); and “Foot and Mouth, Drought Will be Conquered – H.E.,” ibid. (October 8, 1979). 83 Gooch and MacDonald, Evaluation of 1979/80 Drought Relief Programme. 84 Government of Bechuanaland, Drought Relief Legislation, 1962, 196364 [S.598/8, S.594/1]. In response to drought in 1964, the colonial administration used chiefs and tribal authorities to distribute 50,000 bags of grain to affected Batswana. See Minutes of the First Meeting of the Drought Relief Committee (Lobatsi; September 18, 1964). 85 MacDonald and Austin, A Human Drought Relief Programme for Botswana, p. 10. 86 Ibid., p. 11. 87 Ibid.; Republic of Botswana, Summary of Drought Relief Proposals. 88 “Kgatleng Declared Stage Two,” Botswana Daily News (December 5, 1979); “Bobirwa Worse Hit by Drought,” ibid. (January 3, 1980); “Over 32,600 Bakwena get Food Supplies,” ibid. (February 1, 1980); “Okavango Drought is

166

Domestic Politics and Drought Relief in Africa

Now Alarming,” ibid. (February 15, 1980); “Many Get Ration Under Drought Relief Program,” ibid. (February 27, 1980). 89 Gooch and MacDonald, Evaluation of 1979/80 Drought Relief Programme; MacDonald and Austin, A Human Drought Relief Programme for Botswana; Republic of Botswana, Summary of Drought Relief Proposals. 90 Gooch and MacDonald, Evaluation of 1979/80 Drought Relief Programme, p. 71. 91 Republic of Botswana, Supervisor of Elections, Report to the Minister of Public Service and Information on the General Election, 1979. 92 “BDP: 17 Years of Famine Relief Victory – Mpho,” Botswana Daily News (April 18, 1983). 93 See, for instance, Acemoglu et al., “An African Success Story”; Henderson, “Seretse Khama,” pp. 27-56. 94 See Huntington, Political Order in Changing Societies, pp. 12-24, for the classic statement on institutionalization as involving the adaptability, complexity, autonomy, and coherence of state institutions. In recent years, this understanding of institutionalization has been used by scholars to explain the resilience of authoritarianism in China; see Nathan, “China’s Changing of the Guard: Authoritarian Resilience,” pp. 6-17. 95 See Parson, “Succession, Legitimacy, and Political Change in Botswana, 1956-1987,” pp. 97-142; Picard, The Politics of Development in Botswana. 96 See, among others, Londregan et al., “Ethnicity and Leadership Succession in Africa,” pp. 1-25; Hughes and May, “The Politics of Succession in Black Africa,” pp. 1-22; Sylla, “Succession of the Charismatic Leader,” pp. 11-28; Anise, “Trends in Leadership Succession and Regime Change in African Politics Since Independence,” pp. 507-24. 97 “Dr. Masire is Now President,” Botswana Daily News (July 19, 1980). 98 Parson, “Succession, Legitimacy, and Political Change in Botswana, 1956-1987,” p. 97. 99 Henderson, “Seretse Khama,” pp. 27-56. 100 Parson, “Succession, Legitimacy, and Political Change in Botswana, 1956-1987,” pp. 98-99. 101 Republic of Botswana, Ministry of Presidential Affairs and Public Administration, Report to the Minister of Presidential Affairs and Public Administration on the General Election, 1989. 102 Quoted in Botswana Daily News (November 18, 1980). 103 Gooch and MacDonald, Evaluation of 1979/80 Drought Relief Programme. 104 Ibid., p. 12. 105 “Admin. Secretary Opens Drought Relief Seminar,” Botswana Daily News (May 18, 1981). 106 Tabor, Drought Relief and Information Management. 107 “Good Rains Bring Happy Things: Botswana is now Drought Free,” Botswana Daily News (June 15, 1981); “Ngamiland Shows Signs of Recovery From Drought,” ibid. (July 7, 1981). At the beginning of 1982, Batswana expressed uneasiness over insufficient rain: “Lack of Rain Causes Concern to Farmers,” ibid. (January 20, 1982); “Farmer’s Hopes Fade as Dry Spell Continues,” ibid. (January 28, 1982); “Drought: A Grave Concern,” ibid. (February 25, 1982). See Bhalotra, The Drought of 1981-87 in Botswana;

The Evolution of Drought Relief Policy in Botswana

167

Republic of Botswana, Ministry of Agriculture, Botswana Agricultural Statistics, Yearly Reports, 1980 to 1990. 108 Botswana Daily News (April 2, 1982). 109 Republic of Botswana, Food Resources Department, Manual on Human Relief Programme (1987); Republic of Botswana, The Drought Situation in Botswana; Republic of Botswana, Rural Development Unit, Report on the National Food Strategy; Tabor, Drought Relief and Information Management; Republic of Botswana, Food Resources Department, Manual on Human Relief Programme (1982). 110 Drèze, “Famine Prevention in Africa,” pp. 123-72; Republic of Botswana, The Drought Situation in Botswana; Republic of Botswana, Rural Development Unit, Report on the National Food Strategy; Hay et al., A SocioEconomic Assessment of Drought Relief in Botswana. 111 Rockliffe-King, Drought, Agriculture and Rural Development, provides the most complete year-by-year breakdown of relief elements in Botswana. 112 Amis, Financial Efficiencies in Drought Relief, p. 6; Hay, “Famine Incomes and Employment,” pp. 1113-25; Holm and Cohen, “Enhancing Equity in the Midst of Drought,” pp. 31-38. 113 Rockliffe-King, Drought, Agriculture and Rural Development; Amis, Financial Efficiencies in Drought Relief; Tabor, Drought Relief and Information Management. 114 Montshiwa and Mhlanga, Accelerated Rainfed Arable Programme [ARAP], Annual Report 1987-88, p. 1. 115 Ibid. 116 Amis, Financial Efficiencies in Drought Relief. 117 Obstacles faced by IFP in handling food distribution during the 1979-80 drought led to calls for its abolition, or at least its removal from the Ministry of Local Government and Lands to the Ministry of Finance and Development Planning. According to Tabor, Drought Relief and Information Management, a series of negotiations within government between August and October 1981 led to a decision to keep the department responsible for food relief in the Ministry of Local Government and Lands, though a decision on whether it would remain known as IFP was pending. The final decision on the name of the department responsible for food aid came on May 1, 1982 when the Food Resources Department was formed. 118 Republic of Botswana, Food Resources Department, Manual on Human Relief Programme (1982 and 1987). 119 Tabor, Drought Relief and Information Management. 120 Amis, Financial Efficiencies in Drought Relief. 121 See evaluations of the 1982-90 relief program by Rockliffe-King, Drought, Agriculture and Rural Development; Buchanan-Smith, Drought, Income Transfers and the Rural Household Economy; Amis, Financial Efficiencies in Drought Relief; Boers et al., Botswana, Drought Relief; Hay et al., A Socio-Economic Assessment of Drought Relief in Botswana. 122 See Thompson, Drought Management Strategies in Southern Africa; Republic of Botswana, National Development Plan 7, 1991/97; Republic of Botswana, National Development Plan 6, 1985/91. 123 Republic of Botswana, National Development Plan 7, 1991/97, p. 59. 124 Thompson, Drought Management Strategies in Southern Africa, p. 26.

168

Domestic Politics and Drought Relief in Africa

125 Teklu and Asefa, “Factors Affecting Employment Choice in a LaborIntensive Public Works Scheme in Rural Botswana,” pp. 175-86. 126 Letter from the Permanent Secretary, Ministry of Agriculture, A34/6XVII(32)PS (April 21, 2005). 127 Republic of Botswana, National Development Plan 7, 1991/97. 128 Thompson, Drought Management Strategies in Southern Africa, p. 31, citing interviews with N. J. Manamela, deputy coordinator of rural development, Ministry of Finance and Development Planning. 129 Letter from the Permanent Secretary, Ministry of Agriculture, A34/6XVII(32)PS (April 21, 2005). 130 Thompson, Drought Management Strategies in Southern Africa. 131 See de Waal, Famine Crimes; Love, “Drought, Dutch Disease and Controlled Transition in Botswana Agriculture,” pp. 71-83; Thompson, Drought Management Strategies in Southern Africa; Drèze, “Famine Prevention in Africa,” pp. 123-72; Hay, “Famine Incomes and Employment,” pp. 1113-25; Hay et al., A Socio-Economic Assessment of Drought Relief in Botswana; Holm and Morgan, “Coping With Drought in Botswana,” pp. 463-82. 132 For the role of democracy in shaping responsiveness and responses in Botswana, see Sen, Development as Freedom; de Waal, Famine Crimes; Thompson, Drought Management Strategies in Southern Africa; Drèze, “Famine Prevention in Africa,” pp. 123-72; Hay, “Famine Incomes and Employment,” pp. 1113-25; Holm and Morgan, “Coping With Drought in Botswana,” pp. 463-82. 133 Drèze, “Famine Prevention in Africa,” p. 153. 134 Holm and Morgan, “Coping With Drought in Botswana,” p. 481.

6 Conclusion

The argument presented in this book challenges the conventional wisdom that African governments lack the technical capacity and political will to respond to the threat of famine drought produces, or that they are necessarily incompetent and irrational if they do intervene. More often than not, the attention paid to Africa in famine studies revolves around such failures of governments to protect their citizens. The common image sketched by scholars, analysts, and journalists writing on drought and famine in Africa is one defined by pictures of emaciated children in Somalia, the Sudan, and Ethiopia. Or analyses highlight the external intervention and aid necessary to make up for the inadequacies of domestic governments. This book shows that both these views of famine relief across Africa are inadequate. Though the jury is still out on whether the excessive focus on the worst cases succeeds in shocking or shaming us into action or simply numbs us into believing that there is little that can be done to aid Africa, what is clear is that this bias has concealed the other side of the story— the story that tells of governments across Africa, including in Botswana, Kenya, and Zimbabwe, that have, since gaining independence, responded to protect their populations from famine. They have, over the last forty years, demonstrated the capacity, will, and competence to respond to drought and threats of famine, but they have done so by adopting programs of relief shaped by the vulnerability of their incumbent executives. Through a sustained focus on such cases, this book has revealed that domestic government-initiated and -funded relief programs vary across countries, as well as within countries over time. In my search for answers to why relief programs sometimes offered free food aid to ablebodied adults or provided protection through labor-based relief at other times, I concentrated on the significance of political vulnerability. Despite the presence of conditions and factors that scholars previously

169

170

Domestic Politics and Drought Relief in Africa

thought would induce governments to respond to drought in specific ways, I showed that it is in fact variation in the strength of political incumbency that accounts for the types of relief programs that African governments offer to citizens during dry times. Where and when drought occurred in states controlled by insecure incumbents, governments aggressively intervened in the food market to control the urban food supply, offered free food aid to able-bodied rural adults, and delegated the selection of food-aid beneficiaries and the distribution of aid to leaders’ political appointees. In pursuit of political security, vulnerable incumbents in Kenya, Zimbabwe, and Botswana cast relief broadly to generate wide political support and expand their governing coalitions, or else selectively restricted it to key constituents to increase the value of allegiance. Thus political vulnerability prompted incumbents to favor patronage-generating drought relief interventions, enabling weak rulers to direct aid politically and thereby secure incumbency. In drought-afflicted states whose incumbents had built stable personal or institutional order, the approach to drought relief was quite different. When incumbents were insulated from the threat of coups or subversion, secure within robust majority-generating governing coalitions, or free from the immediate pressures of political and economic unrest, government relief programs involved limited interventions in the urban food market, aid to rural adults often came in the form of participation in public works projects, and the administration of relief was delegated to bureaucracies, not political appointees. The emergence of stable tenure brought substantive revisions to the relief programs governments offered to able-bodied adults, not accidentally but as a consequence of a change in the interests guiding drought relief decisionmaking. During periods of secure tenure, support for patronagegenerating relief programs eroded and was replaced by concerns for efficiency, cost-effectiveness, and avoiding dependency. It became politically palatable for incumbents to limit their interventions in the urban market for food and to offer rural residents relief through public works projects implemented by bureaucrats. These dynamics of incumbent security and insecurity were central to the drought relief programs adopted in Kenya, Zimbabwe, and Botswana. Not only did political vulnerability determine programs of relief, but programs of relief contributed to the political standing of incumbents in all three countries. Thus, the political determination of programs of relief was linked to subsequent levels of security and its durability. In Kenya, his response to famine in 1984 enabled a weak President Moi to establish a rural governing coalition beyond the Rift Valley, calmed

Conclusion

171

restive urban areas, quieted social unrest in the rural areas, and weakened potential challengers. He accomplished this by insisting on food aid for the rural population, importing food for the urban market, using the Provincial Administration to administer relief at the local level, and pushing through central control of all stages of provisioning. These interventions contributed to Moi’s consolidation of power and position in the aftermath of the attempted coup in 1982. In Zimbabwe, between 1982 and 1984 President Mugabe leveraged his control of policy-making to impose a relief program that enabled him to construct personalized incumbency out of the instability of the early post-war period. The funneling of food aid through Mugabe’s political party (ZANU) at the local level enabled him to outbid other political parties for the support of the rural population, while market interventions foreclosed the possibility of an urban-based challenge. Finally, even in democratic Botswana, food aid was used in 1979-1980 to legitimate the new modern institutional order that incumbents were creating at the local level. So too was transferring the provision of relief from chiefs and traditional institutions to district commissioners and district councils. Both measures permitted leaders at the center to complete the erosion of traditional authority and to replace it with the modern democratic power structure that secured their own incumbency. Food aid made the construction of stable tenure possible for insecure incumbents in each of the three countries. Threats to political order were addressed by expansive programs of relief that bound key constituents to incumbents through patronage. But there was a cost to these interventions, particularly when incumbents distributed aid too broadly to offer adequate levels of support for the most needy. Ironically, programs of relief that were the most efficacious in resolving political vulnerability for incumbents provided the least per-capita protection for the most severely afflicted populations. At other times, though, vulnerable incumbents went to the other extreme, as was the case with Mugabe after 2000, when food aid was so narrowly targeted to core support groups that it failed to address the prevailing food insecurity. In either case, the political interests generated by insecurity led incumbents who were in pursuit of immediate political gain to favor sub-optimal forms of relief. What emerges is a consistent pattern in which changes to the political status of incumbents had a direct effect on the programs of drought relief governments offered to able-bodied adults. The extent of government intervention in the urban market for food, the provision of food aid or labor-based relief to the rural population, and the institutions and personnel incumbents entrusted to administer relief on the ground

172

Domestic Politics and Drought Relief in Africa

were connected to the weakness or strength of political incumbency. Relief programs were fashioned strategically to assist in the immediate construction of secure rule for weak leaders or designed with costeffectiveness and sustainability in mind when secure tenure permitted incumbents to weigh future benefits more seriously. It is not surprising that incumbency and drought relief programs were so connected, as creating stable political order and dealing with crises in provisioning are two of the major and most persistent challenges leaders in postindependence Africa have had to resolve.1 The more unsettled the political environment was for incumbents, the more aggressively they used responses to food crises to tip the balance in their favor and legitimate their control of state institutions. Thus, the status of political order informed how governments responded to drought and other food crises, and programs of relief in turn contributed to the construction and consolidation of diverse forms of political order, be they personal (as in Kenya and Zimbabwe) or institutional (as in Botswana). But the analysis also suggests that there are limits to the efficacy of food aid programs in addressing insecurity. In all three countries, food aid worked well in the initial construction of governing coalitions. In Kenya, Moi successfully used food aid to extend support outside the Rift Valley and stabilize his tenure. In 1982, Robert Mugabe used food aid to consolidate ZANU’s grip on the rural areas as he outbid other parties for the support of the rural population. And democratic incumbents in Botswana benefited from the distribution of food aid to the rural population as they sought to legitimate the modern administrative and political institutions they had created at the local level. However, what worked well in constructing new governing coalitions was less effective as an instrument for resuscitating support in the face of new challenges. For instance, the analysis reveals that Mugabe’s post-2000 attempt to use food aid to forestall the collapse and disintegration of his incumbency was less successful, leading him to manipulate food aid with brutal determination in advance of the 2005 elections, feeding his supporters and starving his opponents. My argument about the relationship between the forms of relief adopted and the status of incumbency builds on the idea that the interests that animate policy-making are determined by the non-trivial trade-offs between the political and instrumental goals leaders hold. This argument is consistent with established approaches to the political economy of policy choice in developing countries, which hold that incumbents’ political interests determine the extent to which policy is guided by economic rationality or impelled by expedient conside-rations.2 In line with this literature, I show that political expediency played a deter-

Conclusion

173

minative role in drought relief policy choices, with insecure leaders using the distribution of food aid, intervention in the market, and differing administrative organs as devices to attract political support, nullify opposition, and strengthen their hold on power. Weak incumbents provided free food to able-bodied adults not because it was the easiest form of relief to administer or because they were ignorant of the wider effects of drought, but because it was strategically beneficial for them to leverage free food to reward supporters and punish challengers. The interests generated by the strength or weakness of incum-bents’ tenure determined whether famine relief to able-bodied adults involved free food aid or labor-based relief. Importantly, and perhaps surprisingly, incumbents in both democracies and authoritarian regimes were similarly motivated when insecurity prevailed. Figure 6.1 summarizes the relationship between incumbent vulnerability, the interests that motivate disaster relief, and the subsequent decisions that incumbents make regarding relief, access, market interventions, and administrative structures. But for all the similarities in the effects of political vulnerability on drought relief programs, long-term trajectories diverged according to how secure incumbency was built. In Zimbabwe, Robert Mugabe very quickly constructed secure tenure in the early post-conflict period around three non-institutional pillars: the centralization and personalization of executive authority and power; the elimination or cooption of potential challengers; and the creation of a de facto one-party state. In other words, Mugabe weakened all other political institutions in order to fashion stable tenure for himself and ZANU. The consequence of an incumbency built on personal power was that drought relief policymaking in Zimbabwe never became institutionalized, even during periods of secure incumbency. It remained ad hoc in form and, as such, subject to Mugabe’s immediate political whims. Sometimes, his political interests coincided with the adoption and implementation of costeffective and efficacious programs of relief, as was the case in 1986 and again in 1992 when he favored food-for-work as the primary form of relief for rural adults. But his excessive insulation from any political challenge between 1995 and 1999 engendered interests and an approach to policy-making that undermined the provision of adequate relief. Critically, an overlypersonalized incumbency combined with the absence of any institutional checks on Mugabe’s influence in policy-making enabled him to use drought relief to cast himself as the protector of traditional values by supporting chiefs’ calls for relief in the form of community-based loans.

Settling on levels of market intervention to augment the general policy of relief

Incumbents motivated by monopoly control of the market for food as this maximizes political gains from relief !

!

175

The administration of relief is delegated to political functionaries or ad hoc committees that are under the direct control of insecure incumbents

Incumbents motivated by the desire to ensure cost-effective and efficient implementation of relief ! The administration of the relief is delegated to standing bureaucratic structures with little interference or involvement by politicians

Page 1

Establishing a system of delivering relief to designated beneficiaries

Incumbents motivated by desire to leverage administration for political gain

Likely to see government supporting the market by acquiring food on its behalf, but leaving pricing and distribution to the regular system

2:29 PM

4. Implementation:

!

6/11/12

Likely to see incumbents suspending free market and replacing it with price controls and also taking over food distribution

Incumbents motivated by the desire for smooth operation of the market for food

fig6-1b.qxd:Layout 1

3. Market Intervention

176

Domestic Politics and Drought Relief in Africa

On Mugabe’s order, Zimbabwe’s relief program was again changed, but this time with significant regressive outcomes for the rural population. The form of political order created in Zimbabwe (personal) and the trajectory of drought relief (persistently ad hoc and contingent on Mugabe’s immediate political goals) stand in contrast to Botswana. There, early post-colonial conditions encouraged incumbents to seek secure rule by investing in strong state institutions, sound economic planning, and multi-party politics. Although this post-independence quest for secure incumbency took a decade and a half to complete, its creation of secure tenure undergirded by strong institutions affected government responses to drought in positive and enduring ways. Unlike Zimbabwe, where security under personalized power finally led to a wavering in the government’s commitment to adequate relief, secure incumbency in Botswana created the political space for leaders to establish standing relief institutions and programs that would insulate drought relief policy-making from immediate political manipulation. General Implications

My argument reveals that the aid community has not fully recognized the distinctive political dynamics that affect domestic programs of relief. Because scholars, policy analysts, and journalists assume domestic governments necessarily lack capacity and good program design, common prescriptions for dealing with famine in Africa are exclusively technical: building the administrative capacity of government and improving relief program design.3 But, as this book has shown, conventional wisdom and the recommendations that flow from it are sometimes inadequate. Purely technocratic approaches to the threat of famine have largely failed because they have neglected the most important factors determining famine responses: the broader political context within which incumbents operate and, above all, the interests that motivate them. The consequences of that neglect are evident in Ethiopia’s experience a quarter-century after Band Aid and Live Aid. As Gill notes, while international famine aid saved lives in 1984, efforts to build capacity, improve program design, or tackle the underlying poverty in the country have been less successful.4 Ethiopia is still vulnerable to devastating famines, as we saw as recently as 2003. Insights from this work suggest that policy-makers and the international community need to pay attention to the micro-political conditions in which incumbents operate if we are to understand how they respond to threats of famine caused by drought.

Conclusion

177

In developing my argument about the factors that shape the form of drought relief offered to citizens across Africa, I bracketed questions about which governments were more likely to respond to the threat of famine. My selection bias was useful for uncovering the factors that determine how leaders respond to drought, but it concealed variation in responsiveness. I return to this question to suggest the broader implication of my argument. I noted earlier in Chapter 3 that just as President Moi was marshalling his government to respond aggressively to drought in Kenya in the early 1980s, Mengistu Haile Mariam’s military regime, the Derg, was ignoring the threat of famine in Ethiopia, and when it did respond following the international humanitarian intervention, Mengistu used starvation as an instrument of war against his enemies.5 On the ground, some governments ignore or respond inadequately to droughtrelated threats of famine while others, as this book has shown, take notice of drought and adopt sufficient measures to prevent famine. Why did Moi respond preemptively to drought but Mengistu did not? More generally, what induces governments to respond to threats of famine? Two conventional arguments address government responsiveness to threats of famine. The first focuses on the configuration of political regimes, suggesting that the institutions in a democracy are sufficient for famine prevention, especially a free press with its whistle-blowing function and regular elections with their credible threat of punishment for incumbents. Most notably, Sen observes that: Famines have occurred in ancient kingdoms and contemporary authoritarian societies, in primitive tribal communities and in modern technocratic dictatorships, in colonial economies run by imperialists from the north and in newly independent countries of the south run by despotic national leaders or by intolerant single parties. But they have never materialized in any country that is independent, that goes to elections regularly, that has opposition parties to voice criticisms and that permits newspapers to report freely and question the wisdom of government policies without extensive censorship.6

And he contends that “the absence of famines holds even for those democratic countries that happen to be very poor, such as India, Botswana or Zimbabwe.”7 For this reason, scholars, policy analysts, and journalists writing on famine in Africa have placed their trust in democratic politics based on a plural society, government accountability to the electorate through regular free and fair elections, and an independent and unmolested press, as the surest way to protect citizens from starvation.

178

Domestic Politics and Drought Relief in Africa

However, other scholars contend that the liberal optimism inherent in Sen’s argument is not justified. Specifically, de Waal contends that the presumed link between the civil and political rights that come with liberalization on the one hand, and protection from famine on the other hand, is tenuous and rests on other preconditions, such as regime perception of the most vulnerable as worthy of consideration and the wide acceptance of publicity and debate in the political domain.8 Thus, the second argument jettisons the generalizing tendencies of a regime theory of government responsiveness to threats of famine, insisting instead that, even in democracies or liberalizing countries, protection from famine is contingent upon the emergence of a right to relief “guaranteed by the political process.”9 Although suggestive, neither argument can explain the difference between Kenya under Moi and Ethiopia during the Derg. When drought struck these two nations in the early 1980s, neither exhibited the key features associated with democratic authority: free or fair elections to select leaders who then decide on policy; institutionalized constraints on the powers of the executive; and guarantees of civil liberties, including a free and open press. As I discussed earlier, Kenya was a de facto oneparty state under Kenyatta before Moi removed all pretenses of competitive politics by converting the country into a de jure one-party system in 1982. Ethiopia was equally autocratic. Following the overthrow of Emperor Haile Selassie and the abolition of the monarchy in 1974, Mengistu’s military committee governed Ethiopia to the exclusion of meaningful political rights and civil liberties.10 With the new rulers aspiring to build a Marxist-Leninist regime based on mass participation through a single state-sanctioned party, political com-petition was proscribed. Thus, in the year that incumbents were faced with the threat of famine, Kenya and Ethiopia each had a POLITY score of -7, signaling similar levels of autocratic authority. Neither country had rights to relief guranteed by the political system in the form of a famine code. Despite these similarities on the key factors that Sen and de Waal claim determine responsiveness, outcomes differed. Kenya responded to drought in 1984 but Ethiopia did not. Famine was averted in Kenya but it was not in Ethiopia. My argument about the determinants of programs of relief may shed light on this variation in responsiveness between two authoritarian regimes. Although neither Ethiopia nor Kenya were democratic when drought struck and neither country possessed a famine code, political conditions on the ground in the two countries differed a great deal. Significantly, Daniel arap Moi was an insecure incumbent with strong political incentives to find ways to create stable order, and the food

Conclusion

179

crisis produced by drought in 1983 provided him with a highly visible opportunity to secure his position. Thus, I conjecture that the logics of prevailing political insecurity not only induced Moi to respond to drought, but, as I have illustrated, determined the programs of relief he offered. It is likely that political vulnerability, even when not embedded in democratic institutions, induces incumbents to respond to threats of famine as Kenya did. When drought hit Ethiopia, Mengistu’s grip on power was more secure. He had insulated himself from immediate political challenges by tightening the political space in Ethiopia and overseeing successive purges of both former members of the Imperial cabinet and Derg comrades judged to be disloyal or insufficiently committed to the revolution. I conjecture that for this reason, Mengistu and his military council faced few political incentives to respond to drought until it was too late to prevent starvation. Images of the suffering in Ethiopia caught the world’s attention through Bob Geldof’s Band Aid campaign of the mid-1980s. The famine killed thousands and retarded development for over a decade. Thus, it is possible that secure autocratic and non-institutional tenure generated a political environment in which drought and the famine it produced could be ignored by incumbents in Ethiopia. The most-similar case comparison of Kenya and Ethiopia above is suggestive, but not conclusive. Expanding further to more difficult cases, does thinking about responsiveness in terms of the vulnerability or security of tenure produced by micro-political factors account for responsiveness in liberalized and liberalizing African countries? Political changes across Africa since 1989 provide an opportunity to explore this question. As Bratton and van de Walle show, the decade that began with the collapse of the Soviet Union in 1989 ushered in an unprecedented wave of democratic transitions across Africa.11 From Benin to Zambia, many countries that had seemed to be immune to political pluralism inaugurated open politics through processes that ranged from national constitutional conventions to popular protests against authoritarian leaders. Mobilization led to political liberalization and an increase in the frequency of elections: 26 elections before 1989; 59 between 1990-1993; 74 between 1994-1998; and 73 elections between 1999-2003.12 Lindberg finds that 56 percent of these 232 elections were free and fair, suggesting real improvements in political freedom. Figure 6.2 summarizes the change in democratic and autocratic authority structures across Africa between 1960 and 2009.

ff-art-blankp.qxd:Layout 1

6/4/12

11:55 AM

180

Page 1

Conclusion

181

Among scholars, policy analysts, and journalists who argued that democratic government provides public guarantees against absolute deprivation, political liberalization not only extended the prospect of accountable government, but also brought optimism that Africa’s ageold problems of catastrophic droughts and famines would finally be addressed. The end of history was thus also meant to usher in greater government responsiveness to the frequent natural-disaster-induced threats of famine in Africa. For scholars and policymakers convinced by this liberal optimism, persistent food crises and famines in liberalizing Ethiopia, Malawi, and Mozambique13 produced disquiet about the real effect of democratization on famine relief in Africa. Improvements in political freedom across Africa have not brought with them greater protection from famine. It seems that the extension and expansion of political rights and civil liberties have fallen short of offering citizens automatic protection from starvation. Why has the presumed positive relationship between political rights and protection from famine not held in post-reform Africa? The argument of this book relating political vulnerability to drought response is suggestive in regard to this question and raises avenues for further research. For instance, I surmise that in liberalizing regimes, governments will respond to the threat of famine if democracy institutionalizes a minimal level of electoral vulnerability. Yet there remain few examples in Africa of competitive multiparty elections in which executives have been robustly challenged, parliamentary majorities nominal, and margins of victory small.14 As Bratton and van de Walle’s work reveals, a disconcerting feature of political liberalization in Africa is that it has not led to the broad institutionalization of election-based vulnerability for incumbents. Significantly, seat shares in parliament for winning parties have remained disturbingly high with parliamentary seats obtained by winning parties as high as 70 percent.15 Presidential candidates have also been decidedly under-challenged, winning about 63 percent of the vote between 1990 and 1994.16 Liberalization across Africa seems to have transferred super-majorities from one party to another or from former authoritarian leaders managing the transition process to their challengers.17 It is possible that the failure of the post1989 liberalizations to institutionalize the prospect of losing elections for incumbents is what explains why famine persists in some countries. Thus, future research on responsiveness to threats of famine in Africa needs to go beyond a narrow focus on the question of democratic versus authoritarian regimes and consider a second, the vulnerability and stability of incumbents under either regime type.

182

Domestic Politics and Drought Relief in Africa

Notes 1

I borrowed the political order formulation from Zolberg, Creating Political Order. For discussions of the centrality of dealing with food crises, see Schatzberg, Political Legitimacy in Middle Africa; and Bates, Beyond the Miracle of the Market. 2 Geddes, Politician’s Dilemma; Bates and Krueger, eds., Political and Economic Interactions in Economic Policy Reform; Meier, ed., Politics and Policy Making in Developing Countries; Bratton, “The Comrades and the Countryside,” pp. 174-202; Grindle and Thomas, “Policy Makers, Policy Choices, and Policy Outcomes,” pp. 213-48; Ames, Political Survival. 3 von Braun et al., Famine in Africa. 4 Gill, Famine and Foreigners. 5 Ibid.; de Waal, Evil Days; Henze, Ethiopia in Mengistu’s Final Years, Vols. 1-2; de Waal, Famine Crimes. 6 Sen, Development as Freedom, pp. 152-53. 7 Ibid., p. 178. 8 de Waal, Famine Crimes, pp. 8-12. 9 Ibid., p. 11. 10 Henze, Ethiopia in Mengistu’s Final Years, Vols. 1-2. 11 Bratton and van de Walle, Democratic Experiments in Africa. Lindberg, Democracy and Elections in Africa, extends Bratton and van de Walle’s analysis of events between 1990 and 1994 to a 13-year time period, from 1990 to 2003. 12 Lindberg, Democracy and Elections in Africa, pp. 62-64. 13 See Gill, Famine and Foreigners; Devereux and Tiba, “Malawi’s First Famine, 2001-2002,” pp. 143-77; Lautze and Maxwell, “Why Do Famines Persist in the Horn of Africa?: Ethiopia, 1999-2003,” pp. 222-44. 14 Ghana’s 2008 elections are a notable exception. In these elections, the average winner took about 58 percent of the vote; close to 36 percent of the constituencies had competitive races, with winning margins at or below 10 percent; and there was alternation in the party in power at the center. 15 See variable LEGSEAT in Bratton and van de Walle, Political Regimes and Regime Transitions in Africa, 1910-1994. 16 See variable PRESVOTS in ibid. 17 Bratton and van de Walle, Democratic Experiments in Africa.

Acronyms

ANC ARAP ARDP AWC BAMB BDP BMC BNF BPP CMB DC DDC DSW ESAP EWTC FAO FNC FRD GEMA GLS GMB GSU IFP IMDC IMF KADU KANU KFA KGGCU KMC

African Nationalist Congress Accelerated Rainfed Arable Programme Accelerated Rural Development Programme Agriculture and Water Committee Botswana Agricultural Marketing Board Botswana Democratic Party Botswana Meat Commission Botswana National Front Botswana People’s Party Cotton Marketing Board District Commissioner District Development Committee Department of Social Welfare Economic Structural Adjustment Programme Early Warning Technical Committee Food and Agricultural Organization Food and Nutrition Committee Food Resources Department Gikuyu, Embu and Meru Association Grain Loan Scheme Grain Marketing Board General Service Unit Institutional Food Programme Inter-Ministerial Drought Committee International Monetary Fund Kenya African Democratic Union Kenya African National Union Kenyan Farmers’ Association Kenyan Grain Growers Co-operative Union Kenya Meat Commission

183

184

Domestic Politics and Drought Relief in Africa

KPU Legco MDC MP Mt NCA NCPB NDP NFRF NGO PA PAC PF PMU PRC UNICEF USAID VDC WFP ZANU ZAPU ZCTU ZUM

Kenya People’s Union Legislative Council Movement for Democratic Change Member of Parliament Metric Tone National Constitutional Assembly National Cereals and Produce Board National Development Plan National Famine Relief Fund Non-Governmental Organization Provincial Administration Pan-Africanist Congress Patriotic Front Paramilitary Police Unit People’s Redemption Council United Nations Children’s Fund United States Agency for International Development Village Development Committee World Food Programme Zimbabwe African National Union Zimbabwe African People’s Union Zimbabwe Congress of Trade Unions Zimbabwe Unity Movement

Bibliography

Abbott, John. “Agricultural Marketing Boards in the Developing Countries,” Journal of Farm Economics, Vol. 49, no. 3 (August 1967): 705-22. Acemoglu, Daron, et al. “An African Success Story: Botswana,” in Dani Rodrik, ed., In Search of Prosperity: Analytical Narratives on Economic Growth. Princeton: Princeton University Press, 2003, 53-79. Africa Confidential, Vol. 36 (April 28, 1995). Alam, S.M. Shamsul. Rethinking the Mau Mau in Colonial Kenya. Basingstoke: Palgrave Macmillan, 2007. Alexander, Jocelyn. “Dissident Perspectives on Zimbabwe’s Post-Independence War,” Africa: Journal of the International African Institute, Vol. 68, no. 2 (1998): 151-82. ----------. “The Unsettled Land: The Politics of Land Redistribution in Matabeleland, 1980-1990,” Journal of Southern African Studies, Vol. 17, no. 4 (December 1991): 581-610. Alexander, Peter. “Zimbabwean Workers, the MDC & the 2000 Election,” Review of African Political Economy, Vol. 27, no. 85 (September 2000): 385-406. Almond, Gabriel A. “The Return to the State,” American Political Science Review, Vol. 82, no. 3 (September 1988): 853-74. Ames, Barry. Political Survival: Politicians and Public Policy in Latin America. Berkeley: University of California Press, 1987. Amis, Philip. Financial Efficiencies in Drought Relief. Report on the Evaluation of the Drought Relief and Recovery Programme, 1982-1990, Vol. 4. Gaborone: Ministry of Finance and Development Planning, 1990. Amnesty International. Zimbabwe: The Toll of Impunity. London: Amnesty International, 2002. Anderson, David M. “‘Yours in Struggle for Majimbo’: Nationalism and the Party Politics of Decolonization in Kenya, 1955-64,” Journal of Contemporary History, Vol. 40, no. 3 (July 2005): 547-64. Andrews, Carole, and Bryn Morgan. Zimbabwe After the 2005 Parliamentary Election. House of Commons Library Research Paper 05/58 (August 18, 2005). Anise, Ladun. “Trends in Leadership Succession and Regime Change in African Politics Since Independence,” African Studies Review, Vol. 17, no. 3 (December 1974): 507-24.

185

186

Domestic Politics and Drought Relief in Africa

Anyango, George, et al. “Drought Vulnerability in Central and Eastern Kenya,” in Thomas Downing et al., eds., Coping With Drought in Kenya: National and Local Strategies. Boulder: Lynne Rienner, 1989, 169-210. Astrow, André. Zimbabwe: A Revolution That Lost Its Way? London: Zed Press, 1983. Austin, James E. A Drought Contingency Plan for Botswana. Gaborone: Ministry of Finance and Development Planning, 1979. Aykroyd, Wallace. The Conquest of Famine. London: Chatto & Windus, 1974. Ball, Nicole. “Understanding the Causes of African Famine,” Journal of Modern African Studies, Vol. 14, no. 3 (September 1976): 517-22. Balleis, Peter. A Critical Guide to ESAP, Silveira House Social Series. Gweru: Mambo Press, 1993. Banana, Canaan S., ed. Turmoil and Tenacity: Zimbabwe, 1890-1990. Harare: The College Press, 1989. Baregu, Mwesiga. “The Urgent Need for Land and Agrarian Reform in Southern Africa,” Southern African Political and Economic Monthly, Vol. 13, no. 8 (May 2000). Barkan, Joel D., and Frank Holmquist. “Peasant-State Relations and the Social Base of Self-Help in Kenya,” World Politics, Vol. 41, no. 3 (April 1989): 359-80. Barrientos, Armando, and David Hulme, eds. Social Protection for the Poor and Poorest: Concepts, Policies and Politics. Basingstoke: Palgrave Macmillan, 2008. Bates, Robert H. Beyond the Miracle of the Market: The Political Economy of Agrarian Development in Kenya. Cambridge: Cambridge University Press, 1989. ----------. Markets and States in Tropical Africa: The Political Economy of Agricultural Policies. Berkeley: University of California Press, 1981. ----------. “Pressure Groups, Public Policy, and Agricultural Development: A Study of Divergent Outcomes,” in Robert H. Bates and Michael F. Lofchie, eds., Agricultural Development in Africa: Issues of Public Policy. New York: Praeger, 1980, 170-217. Bates, Robert H., and Anne O. Krueger, eds. Political and Economic Interactions in Economic Policy Reform: Evidence from Eight Countries. Oxford: Blackwell, 1993. Bates, Robert H., and Michael F. Lofchie, eds. Agricultural Development in Africa: Issues of Public Policy. New York: Praeger, 1980. Bechuanaland, Government of. Chieftainship Act, No. 26 of 1965, Bechuanaland Government Gazette Supplement (November 5, 1965). ----------. Drought Relief Legislation, 1962, 1963-64 [S.598/8 S.594/1]. Botswana National Archives and Records Service. ----------. The Local Government (District Councils) Act, No. 27 of 1965, Bechuanaland Government Gazette Supplement, (November 5, 1965). Bechuanaland Democratic Party. Election Manifesto: This Is What We Stand For. Gaborone: Bechuanaland Democratic Party, 1965. Bennett, George, and Carl Rosberg. The Kenyatta Election: Kenya 1960-61. Oxford: Oxford University Press, 1961. Benson, Charlotte. “Drought and the Zimbabwe Economy, 1980-1993,” in Helen O’Neill and John Toye, eds., A World Without Famine? New Approaches to Aid and Development. London: Macmillan, 1998, 241-73.

Bibliography

187

Benson, Charlotte, and Edward Clay. The Impact of Drought on Sub-Saharan African Economies: A Preliminary Examination. World Bank Technical Paper No. 401. Washington, D.C.: World Bank, March 1998. ----------. Understanding the Economic and Financial Impacts of Natural Disasters. World Bank Disaster Risk Management Series No. 4. Washington, D.C.: World Bank, 2004. Berkeley, Bill, et al. Zimbabwe, Wages of War: A Report on Human Rights. New York: Lawyers Committee for Human Rights, 1986. Berry, Leonard, et al. Eastern Africa Country Profiles: Kenya. Worcester: International Development Program, Clark University, 1980. Berry, Sara S. “The Food Crisis and Agrarian Change in Africa: A Review Essay,” African Studies Review, Vol. 27, no. 2 (January 1984): 59-112. ----------. No Condition is Permanent: The Social Dynamics of Agrarian Change in Sub-Saharan Africa. Madison: University of Wisconsin Press, 1993. Bhalotra, Y.P.R. The Drought of 1981-87 in Botswana. Gaborone: Botswana Meteorological Services, 1987. Bhebe, Ngwabi. “The Nationalist Struggle, 1957-1962,” in Canaan S. Banana, ed., Turmoil and Tenacity: Zimbabwe, 1890-1990. Harare: The College Press, 1989, 50-115. Bienen, Henry, and Nicolas van de Walle. Of Time and Power: Leadership Duration in the Modern World. Stanford: Stanford University Press, 1991. ----------. “Time and Power in Africa,” American Political Science Review, Vol. 83, no. 1 (March 1989): 19-34. Blix, Gunnar, et al., eds. Famine: A Symposium Dealing with Nutrition and Relief Operations in Times of Disaster. Uppsala: Almqvist & Wiksells, 1971. Bloom, David, and Jeffrey D. Sachs. “Geography, Demography, and Economic Growth in Africa,” Brookings Papers on Economic Activity, Vol. 2 (1998): 207-72. Boers, W. Botswana, Drought Relief: Labour Based Relief Programme: Technical Report. Gaborone: SNV Botswana, 1990. Bond, Patrick, and Masimba Manyanya. Zimbabwe’s Plunge: Exhausted Nationalism, Neoliberalism and the Search for Social Justice. Pietermaritzburg: University of Natal Press, 2003. Borton, John. “Overview of the 1984/85 National Drought Relief Programme,” in Thomas Downing et al., eds., Coping with Drought in Kenya: National and Local Strategies. Boulder: Lynne Rienner, 1989, 24-64. Borton, John, and Robin Stephenson. Disaster Preparedness in Kenya. London: Relief and Development Institute, 1984. Botswana, Republic of. Botswana Project 324/II Quarterly Progress Report, 1st January, 1982 to 31st March, 1982: World Food Programme Projects. Gaborone: Government Printers, 1982. ----------. The Drought Situation in Botswana. Gaborone: Ministry of Finance and Development Planning, 1985. ----------. Mineral Rights in Tribal Territories Act, Act 31 of 1967. Gaborone: Government Printers, 1967. ----------. National Development Plan 7, 1991/97. Gaborone: Ministry of Finance and Development Planning, 1991. ----------. National Development Plan 6, 1985/91. Gaborone: Ministry of Finance and Development Planning, 1985.

188

Domestic Politics and Drought Relief in Africa

----------. Summary of Drought Relief Proposals. Gaborone: Government Printers, June 1979. ----------. Transitional Plan for Social and Economic Development, 1966-70. Gaborone: Government Printers, September 1966. ----------. Tribal Land Act, 1968. Gaborone: Government Printers, 1968. Botswana, Republic of, Food Resources Department. Manual on Human Relief Programme for Use by District Relief Committees and all Extension Officers as a Reference Document. Gaborone: Ministry of Local Government and Lands, 1987. Botswana, Republic of, Food Resources Department. Manual on Human Relief Programme for Use by District Relief Committees and all Extension Officers as a Reference Document. Gaborone: Ministry of Local Government and Lands, 1982. Botswana, Republic of, Ministry of Agriculture. Botswana Agricultural Statistics. Yearly Reports, 1978 to 1990. Gaborone: Agricultural Statistics Unit, Division of Planning and Statistics, Ministry of Agriculture, 1979 to 1991. Botswana, Republic of, Ministry of Presidential Affairs and Public Administration. Report to the Minister of Presidential Affairs and Public Administration on the General Election, 1989. Gaborone: Government Printers, 1989. Botswana, Republic of, Rural Development Unit. Report on the National Food Strategy. Gaborone: Ministry of Finance and Development Planning, 1985. Botswana, Republic of, Supervisor of Elections. Report to the Minister of Public Service and Information on the General Election, 1979. Gaborone: Government Printers, 1979. Botswana, Republic of, Supervisor of Elections. Report on the General Elections, 1969. Gaborone: Government Printers, 1969. Botswana Daily News. Gaborone, Botswana, 1966 to 2005. Botswana Guardian. Gaborone, Botswana, 1983 to 2005. Botswana Meat Commission. Annual Reports, 1966 to 1994. Gaborone: Government Printers, 1967-1995. Branch, Daniel, and Nic Cheeseman. “The Politics of Control in Kenya: Understanding the Bureaucratic-Executive State, 1952-78,” Review of African Political Economy, Vol. 33, no. 107 (March 2006): 11-31. Bratton, Michael. “The Comrades and the Countryside: The Politics of Agricultural Policy in Zimbabwe,” World Politics, Vol. 39, no. 2 (January 1987): 174-202. ----------. “Drought, Food and the Social Organization of Small Farmers in Zimbabwe,” in Michael H. Glantz, ed., Drought and Hunger in Africa: Denying Famine a Future. Cambridge: Cambridge University Press, 1987, 213-44. Bratton, Michael, and Nicolas van de Walle. Democratic Experiments in Africa: Regime Transitions in Comparative Perspective. New York: Cambridge University Press, 1997. ----------. “Neopatrimonial Regimes and Political Transitions in Africa,” World Politics, Vol. 46, no. 4 (July 1994): 453-89. ----------. Political Regimes and Regime Transitions in Africa, 1910-1994. Ann Arbor: Inter-university Consortium for Political and Social Research, 1997.

Bibliography

189

Buchanan-Smith, Margaret. Drought, Income Transfers and the Rural Household Economy. Report on the Evaluation of the Drought Relief and Recovery Programme, 1982-90, Vol. 3. Gaborone: Ministry of Finance and Development Planning, 1990. Buchanan-Smith, Margaret, and Thomas Downing. “Drought and Famine in Africa: Time for Effective Action,” Food Policy, Vol. 17, no. 6 (December 1992): 465-67. Callaghy, Thomas. The State-Society Struggle: Zaire in Comparative Perspective. New York: Columbia University Press, 1984. Cammack, Paul. “Bringing the State Back In?,” British Journal of Political Science, Vol. 19, no. 2 (April 1989): 261-90. Campbell, David J. “Strategies for Coping with Severe Food Deficits in Rural Africa: A Review of the Literature,” Food and Foodways, Vol. 4, no. 2 (1990): 143-62. Carter, Gwendolen M., and E. Philip Morgan, eds. From the Frontline: Speeches of Sir Seretse Khama. Stanford: Hoover Institution Press, 1980. Catholic Commission for Justice and Peace and the Legal Resources Foundation. Breaking the Silence, Building True Peace: A Report on the Disturbances in Matabeleland and the Midlands, 1980-1988. Harare: CCJP/LRF, March 1997. Caute, David. “Mugabe Brooks No Opposition,” The Nation, Vol. 241, no. 5 (August 31, 1985): 139-44. Chambers, Robert. Botswana’s Accelerated Rural Development Programme, 1973-6: Experience and Lessons. Gaborone: Government Printer, 1977. Chan, Stephen. “The Old Fox Eludes the Hunt,” World Today, Vol. 61, no. 4 (April 2005): 22-23. Cheeseman, Nic. “Political Linkage and Political Space in the Era of Decolonization,” Africa Today, Special Issue: Creating the Kenya PostColony, Vol. 53, no. 2 (Winter 2006): 3-24. Chikuhwa, Jacob. A Crisis of Governance: Zimbabwe. New York: Algora Publishing, 2004. Chitiyo, Tapera Knox. “Land Violence and Compensation: Reconceptualising Zimbabwe’s Land and War Veterans’ Debate,” Track Two Occasional Paper, Vol. 9, no. 1, Cape Town: Centre for Conflict Resolution, May 2000. Chiwewe, Willard. “Unity Negotiations,” in Canaan S. Banana, ed., Turmoil and Tenacity: Zimbabwe, 1890-1990. Harare: The College Press, 1989, pp. 242-87. Chronicle. Bulawayo, Zimbabwe, 1980 to 1995. Clapham, Chistopher, ed. Private Patronage and Public Power: Political Clientelism in the Modern State. New York: St. Martin’s Press, 1982. Clay, Edward. “Aid and Drought: Responding to the Human and Economic Consequences of Natural Disasters,” in Helen O’Neill and John Toye, eds., A World Without Famine? New Approaches to Aid and Development. London: Macmillan, 1998, 199-220. Clough, Marshall. Mau Mau Memoirs: History, Memory and Politics. Boulder: Lynne Rienner, 1998. Cohen, John M., and David B. Lewis. “Role of Government in Combating Food Shortages: Lessons From Kenya, 1984-85,” in Michael H. Glantz, ed.,

190

Domestic Politics and Drought Relief in Africa

Drought and Hunger in Africa: Denying Famine a Future. Cambridge: Cambridge University Press, 1987, 269-96. Colclough, Christopher, and Stephen McCarthy. The Political Economy of Botswana: A Study of Growth and Distribution. New York: Oxford University Press, 1980. Corbett, Jane. “Famine and Household Coping Strategies,” World Development, Vol. 16, no. 9 (September 1988): 1099-1112. “The Coup Broadcast,” Race and Class, Vol. 24, no. 3 (Winter 1983): 325-26. Cowell, Alan. “Peter Walls, General in Zimbabwe, Dies at 83,” New York Times, July 22, 2010, A25. Crook, R.C. “Patrimonialism, Administrative Effectiveness and Economic Development in Cote d'Ivoire,” African Affairs, Vol. 88, no. 351 (1989): 205-28. Currie, Kate, and Larry Ray. “The Pambana of August 1 – Kenya’s Abortive Coup,” Political Quarterly, Vol. 57, no. 1 (January 1986): 47-59. ----------. “State and Class in Kenya: Notes on the Cohesion of the Ruling Class,” Journal of Modern African Studies, Vol. 22, no. 4 (December 1984): 559-93. Daily Nation. Nairobi, Kenya, 1964 to 1985. Dashwood, Hevina. “The Relevance of Class to the Evolution of Zimbabwe’s Development Strategy, 1980-1991,” Journal of Southern African Studies, Vol. 22, no. 1 (March 1996): 27-48. Davies, Susanna. “Are Coping Strategies a Cop Out?,” IDS Bulletin, Vol. 24, no. 4 (1993): 60-72. de Sardan, J.P. Olivier. “A Moral Economy of Corruption in Africa?,” Journal of Modern African Studies, Vol. 37, no. 1 (1999): 25-52. de Waal, Alex. Evil Days: Thirty Years of War and Famine in Ethiopia. New York: Human Rights Watch, 1991. ----------. Famine Crimes: Politics & the Disaster Relief Industry in Africa. Oxford: James Currey, 1997. ----------. Famine That Kills: Darfur, Sudan, 1984-1985. Oxford: Clarendon Press, 1989. de Wilde, J. “Price Incentives and African Agricultural Development,” in Robert H. Bates and Michael F. Lofchie, eds., Agricultural Development in Africa: Issues of Public Policy. New York: Praeger, 1980, 46-66. deGrassi, Aaron. “‘Neopatrimonialism’ and Agricultural Development in Africa: Contributions and Limitations of a Contested Concept,” African Studies Review, Vol. 51, no. 3 (December 2008): 107-33. Derrick, Jonathan. “West Africa’s Worst Year of Famine,” African Affairs, Vol. 83, no. 332 (July 1984): 281-99. Devereux, Stephen. “Goats Before Ploughs: Dilemmas of Household Response Sequencing During Food Shortages,” IDS Bulletin, Vol. 24, no. 4 (1993): 52-59. ----------, ed. The New Famines: Why Famines Persist in an Era of Globalization. London: Routledge, 2007. ----------. “Social Pensions in Namibia and South Africa,” IDS Discussion Paper 379. Brighton: IDS, 2001. Devereux, Stephen, and Zoltan Tiba. “Malawi’s First Famine, 2001-2002,” in Stephen Devereux, ed., The New Famines: Why Famines Persist in an Era of Globalization. London: Routledge, 2007, 143-77.

Bibliography

191

Downing, Thomas, et al., eds. Coping with Drought in Kenya: National and Local Strategies. Boulder: Lynne Rienner, 1989. Downing, Thomas, et al. “Drought in Kenya,” in Thomas Downing et al., eds., Coping With Drought in Kenya: National and Local Strategies. Boulder: Lynne Rienner, 1989, 3-23. Drèze, Jean. “Famine Prevention in Africa: Some Experiences and Lessons,” in Jean Drèze and Amartya Sen, eds., The Political Economy of Hunger, Vol. 2, Famine Prevention. Oxford: Clarendon Press, 1990, 123-72. ----------. “Famine Prevention in India,” in Jean Drèze and Amartya Sen, eds., The Political Economy of Hunger, Vol. 2, Famine Prevention. Oxford: Clarendon Press, 1990, 13-122. Drèze, Jean, and Amartya Sen. Hunger and Public Action. Oxford: Clarendon Press, 1989. ----------. India: Economic Development and Social Opportunity. Oxford: Clarendon Press, 1995. ----------., eds. The Political Economy of Hunger, Vol. 2, Famine Prevention. Oxford: Clarendon Press, 1990. du Toit, Pierre. State Building and Democracy in Southern Africa: Botswana, Zimbabwe, and South Africa. Washington, D.C.: United States Institute of Peace, 1995. Dunning, Thad. Crude Democracy: Natural Resource Wealth and Political Regimes. New York: Cambridge University Press, 2008. East African Standard. Nairobi, Kenya, 1980 to 1984. Economist Intelligence Unit. Zimbabwe Country Profile: Annual Survey of Political and Economic Background (1987–2000). London: EIU, 2000. ----------. Quarterly Economic Review of Angola, Guinea Bissau, Cape Verde, Sao Tome, Principe: Annual Supplement 1984. Economist Intelligence Unit, 1984. EM-DAT: The OFDA/CRED International Disaster Database, Centre for Research on the Epidemiology of Disasters, Université Catholique de Louvain, Brussels, Belgium. www.emdat.be. Englebert, Pierre. State Legitimacy and Development in Africa. Boulder: Lynne Rienner, 2000. Evans, Peter B., et al., eds. Bringing the State Back In. New York: Cambridge University Press, 1985. Farmer, Graham. “Rainfall Variability in Tropical Africa and Kenya,” in Thomas Downing et al., eds., Coping With Drought in Kenya: National and Local Strategies. Boulder: Lynne Rienner, 1989, 82-93. Fellows, Lawrence. “Soviet Guns Due for Kenya Army – Defense Chief Says Britain Will Retain Training Role,” New York Times, April 30, 1965, 2. Financial Gazette. Harare, Zimbabwe, 1990-2000. Financial Times. London. January-December, 1980. Frimpong, Kwame. “A Review of the Tribal Grazing Land Policy in Botswana,” Pula: Botswana Journal of African Studies, Vol. 9, no. 1 (1995): 1-16. Gaidzanwa, Rudo B. “Drought and the Food Crisis in Zimbabwe,” in Peter Lawrence, ed., World Recession and the Food Crisis in Africa. London: James Currey, 1986, 249-57. Gallup, John, et al. “Geography and Economic Development,” in Boris Pleskovic and Joseph E. Stiglitz, eds., Annual World Bank Conference on Development Economics. Washington, D.C.: World Bank, 1998.

192

Domestic Politics and Drought Relief in Africa

Gautam, Madhur. “Managing Drought in Sub-Saharan Africa: Policy Perspectives.” Invited panel session on “Drought: Economic Consequences and Policies for Mitigation,” at the IAAE Conference, Gold Coast, Queensland, Australia, August 12-18, 2006. Geddes, Barbara. Politician’s Dilemma: Building State Capacity in Latin America. Berkeley: University of California Press, 1994. Gelbach, J., and L. Pritchett. “More for the Poor Is Less for the Poor: The Politics of Targeting,” World Bank Development Research Group. Washington, D.C., World Bank, 1997. George, Alexander L., and Andrew Bennett. Case Studies and Theory Development in the Social Sciences. Cambridge, MA: MIT Press, 2005. Gertzel, Cherry. The Politics of Independent Kenya, 1963-8. Evanston: Northwestern University Press, 1970. ----------. “The Provincial Administration in Kenya,” Journal of Commonwealth Political Studies, Vol. 4, no. 3 (1966): 201-15. Gertzel, Cherry, et al., eds. Government and Politics in Kenya. Nairobi: East African Publishing House, 1969. Gibbon, Peter, ed. Structural Adjustment and the Working Poor in Zimbabwe. Uppsala: Nordiska Afrikainstitutet, 1995. Gill, Peter. Famine and Foreigners: Ethiopia Since Live Aid. Oxford: Oxford University Press, 2010. Gillet, Simon. “The Survival of Chieftaincy in Botswana,” African Affairs, Vol. 72, no. 287 (April 1973): 179-85. Glantz, Michael H. “Drought and Economic Development in Sub-Saharan Africa,” in Michael H. Glantz, ed., Drought and Hunger in Africa: Denying Famine a Future. Cambridge: Cambridge University Press, 1987, 37-58. ----------., ed. Drought and Hunger in Africa: Denying Famine a Future. Cambridge: Cambridge University Press, 1987. Goldsworthy, David. Tom Mboya: The Man Kenya Wanted to Forget. New York: Africana Publishing Company, 1982. Goncalves, Fernando. “A Rare Defeat for Mugabe,” Southern African Political and Economic Monthly, Vol. 13, no. 6 (March 2000). Gooch, Toby, and John MacDonald. Evaluation of 1979/80 Drought Relief Programme. Gaborone: Ministry of Finance and Development Planning, 1981. Good, Kenneth. Diamonds, Dispossession & Democracy in Botswana. Woodbridge: James Currey, 2008. ----------. “Interpreting the Exceptionality of Botswana,” Journal of Modern African Studies, Vol. 30, no. 1 (March 1992): 69-95. Gordon, David F. Decolonization and the State in Kenya. Boulder: Westview Press, 1986. Gourevitch, Peter. Politics in Hard Times: Comparative Responses to International Economic Crises. Ithaca: Cornell University Press, 1986. Grindle, Merilee S. “The New Political Economy: Positive Economics and Negative Politics,” in Gerald M. Meier, ed., Politics and Policy Making in Developing Countries: Perspectives on the New Political Economy. San Francisco: ICS Press, 1991, 41-67. Grindle, Merilee S., and John W. Thomas. “Policy Makers, Policy Choices, and Policy Outcomes: The Political Economy of Reform in Developing Countries,” Policy Sciences, Vol. 22, nos. 3/4 (1989): 213-48.

Bibliography

193

Hall-Matthews, David. “Historical Roots of Relief Paradigms: Ideas on Dependency and Free Trade in India in the 1870s,” Disasters, Vol. 20, no. 3 (September 1996): 216-30. Harsch, Ernest. “Corruption and State Reform in Africa: Perspectives from Above and Below,” African Development Perspectives Yearbook, Vol. 6 (1997/98): 65-87. Hay, Roger W. “The Concept of Food Supply System with Special Reference to the Management of Famine,” Ecology of Food Nutrition, Vol. 7, no. 2 (1978): 65-72. ----------. “Famine Incomes and Employment: Has Botswana Anything to Teach Africa?,” World Development, Vol. 16, no. 9 (1988): 1113-25. Hay, Roger W., et al. A Socio-Economic Assessment of Drought Relief in Botswana. Report prepared by UNICEF/UNDP/WHO for the InterMinisterial Drought Committee. Gaborone: Government Printers, 1985. Henderson, Willie. “Seretse Khama: A Personal Appreciation,” African Affairs, Vol. 89, no. 354 (January 1990): 27-56. Henze, Paul. Ethiopia in Mengistu’s Final Years, 2 vols. Addis Ababa: Shama Books, 2007. Herald. Harare, Zimbabwe, 1980 to 2006. Herbst, Jeffrey. State Politics in Zimbabwe. Berkeley: University of California Press, 1990. ----------. States and Power in Africa: Comparative Lessons in Authority and Control. Princeton: Princeton University Press, 2000. Hinchey, Madalon T., ed. Proceedings of the Symposium on Drought in Botswana. Gaberone: Botswana Society and Clark University Press, 1979. Holm, John D. “Liberal Democracy and Rural Development in Botswana,” African Studies Review, Vol. 25, no. 1 (March 1982): 83-102. Holm, John D., and Mark Cohen. “Enhancing Equity in the Midst of Drought: The Botswana Approach,” Journal of Social Development in Africa, Vol. 3, no. 1 (1988): 31-38. Holm, John D., and Richard G. Morgan. “Coping With Drought in Botswana: An African Success,” Journal of Modern African Studies, Vol. 23, no. 3 (September 1985): 463-82. Hornsby, Charles, and David Throup. “Elections and Political Change in Kenya,” Journal of Commonwealth & Comparative Politics, Vol. 30, no. 2 (July 1992): 179-99. Hughes, Arnold, and Roy May, “The Politics of Succession in Black Africa,” Third World Quarterly, Vol. 10, no. 1 (1988): 1-22. Human Rights Watch. Not a Level Playing Field: Zimbabwe’s Parliamentary Elections in 2005. HRW Briefing Paper, March 21, 2005. ----------. Not Eligible: The Politicization of Food in Zimbabwe, Human Rights Watch, Vol. 15, no. 17(A), October 2003. ----------. The Politics of Food Assistance in Zimbabwe. HRW Briefing Paper, August 12, 2004. Hunger Project. “Statement on the Situation in Zimbabwe,” August 2001. http://www.thp.org/what_we_do/key_initiatives/honoring_africa_leadershi p/laureate_list/zimbabwe_statement. Huntington, Samuel. Political Order in Changing Societies. New Haven: Yale University Press, 1968.

194

Domestic Politics and Drought Relief in Africa

Hyden, Goran, and Colin Leys. “Elections and Politics in Single-Party Systems: The Case of Kenya and Tanzania,” British Journal of Political Science, Vol. 2, no. 4 (October 1972): 389-420. Ihonvbere, Julius. “Constitution Making in Africa: Dismantling the Imperial State,” South African Political and Economic Monthly, Vol. 13, no. 5 (February 2000). India, Government of. The Imperial Gazetteer of India, Vol. 3, The Indian Empire: Economic, 2nd ed. Oxford: Clarendon Press, 1907. ----------. Report of the Indian Famine Commission, 1901. Calcutta: Government Printing Office, 1901. ----------. Report of the Indian Famine Commission, 1880. London: HMSO, 1880. International Crisis Group. Post-Election Zimbabwe: What Next? Africa Report No. 93 (June 7, 2005). Jackson, Robert H., and Carl G. Rosberg. Personal Rule in Black Africa: Prince, Autocrat, Prophet, Tyrant. Berkeley: University of California Press, 1982. ----------. “Personal Rule: Theory and Practice in Africa,” Comparative Politics, Vol. 16, no. 4 (July 1984): 421-42. ----------. “Why Africa’s Weak States Persist: The Empirical and the Juridical in Statehood,” World Politics, Vol. 35, no. 1 (October 1982): 1-24. Jodha, N.S. “Famine and Famine Policies: Some Empirical Evidence,” Economic and Political Weekly, vol. 10, no. 41 (October 11, 1975): 160923. Joseph, Richard. Democracy and Prebendal Politics in Nigeria: The Rise and Fall of the Second Republic. Cambridge: Cambridge University Press, 1987. Karega-Mûnene. “Polarisation of Politics in Kenya Along Ethnic Lines,” WAJIBU: A Journal of Social & Religious Concern, Vol. 18, nos. 1-2 (May-July 2003). http://web.peacelink.it/wajibu/18_issue/p3.html . Karimi, Joseph, and Philip Ochieng. The Kenyatta Succession. Nairobi: Transafrica, 1980. Keen, David. The Benefits of Famine: A Political Economy of Famine and Relief in Southwestern Sudan, 1983-1989. Princeton: Princeton University Press, 1994. Kenya, Government of. BB/1/144 Famine and Famine Relief, 1965-1973. Room 4, Shelf 3491, Box 1, Kenya National Archives, Nairobi. ----------. GO/3/2/31 Report on Famine Relief in Kenya, 1962. Room 3, Shelf 2718, Box 4, Kenya National Archives, Nairobi. ----------. Report of Enquiry into the Kenyan Meat Industry. Nairobi: Government Printers, 1956. “Kenya: Arms & Odinga,” Time, Vol. 85, no. 17 (April 23, 1965). “Kenya: The End of an Illusion,” Race and Class, Vol. 24, no. 3 (Winter 1983): 221-44. Khama, Seretse. “Presidential Address to the First Meeting of the Third Session of the First National Assembly,” December 9, 1968, in Gwendolen M. Carter and E. Philip Morgan, eds., From the Frontline: Speeches of Sir Seretse Khama. Stanford: Hoover Institution Press, 1980, 30. ----------. “Presidential Address to the Sixth Annual Conference of the Botswana Democratic Party, Gaborone,” March 25, 1967, in Gwendolen M. Carter

Bibliography

195

and E. Philip Morgan, eds., From the Frontline: Speeches of Sir Seretse Khama. Stanford: Hoover Institution Press, 1980, 10. ----------. “Speech on the People’s Progress at the Opening of the Third Session of the Second Session of Parliament,” December 3, 1971, in Gwendolen M. Carter and E. Philip Morgan, eds., From the Frontline: Speeches of Sir Seretse Khama. Stanford: Hoover Institution Press, 1980, 135. ----------. “Speech Opening the 17th Annual Conference of the Botswana Democratic Party,” May 5, 1978, in Gwendolen M. Carter and E. Philip Morgan, eds., From the Frontline: Speeches of Sir Seretse Khama. Stanford: Hoover Institution Press, 1980, 279. Khapoya, Vincent B. “Kenya under Moi: Continuity or Change?,” Africa Today, Vol. 27, no. 1 (1980): 17-32. Kinsey, Bill, et al. “Coping with Drought in Zimbabwe: Survey Evidence on Responses of Rural Households to Risk,” World Development, Vol. 26, no. 1 (1998): 89-110. Kreysler, J. “Nutritional Surveillance in Botswana as a Method of Planning the Prevention of Hunger Situations,” in Madalon T. Hinchey, ed., Proceedings of the Symposium on Drought in Botswana. Gaborone: Botswana Society and Clark University Press, 1979, 221-31. Kriger, Norma. “ZANU(PF) Strategies in General Elections, 1980-2000: Discourse and Coercion,” African Affairs, Vol. 104, no. 414 (2005): 1-34. ----------. “Zimbabwe Today: Hope against Grim Realities,” Review of African Political Economy, Vol. 27, no. 85 (September 2000): 443-50. Kyle, Keith. The Politics of the Independence of Kenya. New York: St. Martin’s Press, 1999. Lange, Matthew. Lineages of Despotism and Development: British Colonialism and State Power. Chicago: University of Chicago Press, 2009. Lautze, Sue, and Daniel Maxwell. “Why Do Famines Persist in the Horn of Africa?: Ethiopia, 1999-2003,” in Stephen Devereux, ed., The New Famines: Why Famines Persist in an Era of Globalization. London: Routledge, 2007, 222-44. Lawrence, Peter, ed. World Recession and the Food Crisis in Africa. London: James Currey, 1986. LeBas, Adrienne. From Protest to Parties: Party-Building and Democratization in Africa. Oxford: Oxford University Press, 2011. ----------. “Polarization as Craft: Party Formation and State Violence in Zimbabwe,” Comparative Politics, Vol. 38, no. 4 (July 2006): 419-38. Lennock, Jean. Paying for Health: Poverty and Structural Adjustment in Zimbabwe. Oxford: Oxfam Publications, 1994. Lewis, Stephen R., Jr. “The Potential Problems of Diamond-Dependent Development,” in Charles Harvey, ed., Papers on the Economy of Botswana. London: Heinemann, 1981, 14-28. Leys, Roger. “Drought and Drought Relief in Southern Zimbabwe,” in Peter Lawrence, ed., World Recession and the Food Crisis in Africa. London: James Currey, 1986, 258-74. Lindberg, Staffan I. Democracy and Elections in Africa. Baltimore: Johns Hopkins University Press, 2006. Lofchie, Michael F. “Political and Economic Origins of African Hunger,” Journal of Modern African Studies, Vol. 13, no. 4 (December 1975): 55167.

196

Domestic Politics and Drought Relief in Africa

Lofchie, Michael F., and Stephen K. Commins. “Food Deficits and Agricultural Policies in Tropical Africa,” Journal of Modern African Studies, Vol. 20, no. 1 (March 1982): 1-25. Londregan, John, et al. “Ethnicity and Leadership Succession in Africa,” International Studies Quarterly, Vol. 39, no. 1 (March 1995): 1-25. Longhurst, Richard. “Household Food Strategies in Response to Seasonality and Famine,” IDS Bulletin, Vol. 17, no. 3 (1986): 27-35. Love, Roy. “Drought, Dutch Disease and Controlled Transition in Botswana Agriculture,” Journal of Southern African Studies, Vol. 20, no. 1 (March 1994): 71-83. Loveday, Alexander. The History & Economics of Indian Famines. London: G. Bell and Sons, 1914. MacDonald, John, and James E. Austin. A Human Drought Relief Programme for Botswana. Gabarone: Ministry of Local Government and Lands, 1980. Macodras, Michael W., et al. “Synoptic Features Associated with the Failure of the 1984 Long Rains in Kenya,” in Thomas Downing et al., eds., Coping With Drought in Kenya: National and Local Strategies. Boulder: Lynne Rienner, 1989, 69-81. Madras Presidency. Famine Code. Madras: Government Press, 1905. Makumbe, John Mw, and Daniel Compagnon. Behind the Smokescreen: The Politics of Zimbabwe’s 1995 General Elections. Harare: University of Zimbabwe Publications, 2000. Mallory, Walter H. China: Land of Famine. New York: American Geographical Society, 1926. Mamdani, Mahmood. Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism. Princeton: Princeton University Press, 1996. Mandaza, Ibbo. “The Political Economy of Transition,” in Ibbo Mandaza, ed., Zimbabwe: The Political Economy of Transition, 1980-1986. Dakar: CODESRIA, 1987, 1-20. ----------. “Reflections on the Referendum,” South African Political and Economic Monthly, Vol. 13, no. 6 (March 2000). ----------. “The State in Post-White Settler Colonial Situation,” in Ibbo Mandaza, ed., Zimbabwe: The Political Economy of Transition, 1980-1986. Dakar: CODESRIA, 1987, 21-74. ----------., ed. Zimbabwe: The Political Economy of Transition, 1980-1986. Dakar: CODESRIA, 1987. Mandaza, Ibbo, and Lloyd Sachikonye, eds. The One-Party State and Democracy: The Zimbabwe Debate. Harare: SAPES Trust, 1991. Maphosa, N., et al. Public Service Training Needs and Resources in Zimbabwe. Education Division Documents No. 23, Swedish International Development Authority (SIDA), 1985. Marcussen, Henrik S., ed. Improved Natural Resource Management: The Role of Formal Organisations and Informal Networks and Institutions. Roskilde: International Development Studies, Roskilde University, 1996. Marquette, Catherine. “Current Poverty, Structural Adjustment, and Drought in Zimbabwe,” World Development, Vol. 25, no. 7 (July 1997): 1141-49. Marshall, Monty, and Keith Jaggers. Polity IV Project: Political Regime Characteristics and Transitions, 1800-2007. Dataset Users’ Manual. Center for Systemic Peace, February 2009.

Bibliography

197

Masire, Quett Ketumile. Very Brave or Very Foolish? Memoirs of an African Democrat, Stephen R. Lewis, Jr., ed. Gaborone: Macmillan Botswana, 2006. Masiwa, Nyasha. “Zimbabwe: Count Down to Year 2000,” South African Political and Economic Monthly, Vol. 12, no. 4 (April 1999). Maxon, Robert M., and Peter Ndege. “The Economics of Structural Adjustment,” in B.A. Ogot and W.R. Ochieng’, eds., Decolonization and Independence in Kenya, 1940-93. Athens: Ohio University Press, 1995, 151-86. McGowan and Associates. A Study of Drought Relief and Contingency Measures Relating to the Livestock Sector of Botswana. Albury, Australia: McGowan & Associates, 1979. Médard, Jean-François. “Patrimonialism, Neopatrimonialism and the Study of the Post-colonial State in Sub-Saharan Africa,” in Henrik S. Marcussen, ed., Improved Natural Resource Management: The Role of Formal Organisations and Informal Networks and Institutions. Roskilde: International Development Studies, Roskilde University, 1996, 76-97. ----------. “The Underdeveloped State in Tropical Africa: Political Clientelism or Neo-patrimonialism?,” in Christopher Clapham, ed., Private Patronage and Public Power: Political Clientelism in the Modern State. New York: St. Martin’s Press, 1982, 162-92. Meier, Gerald M., ed. Politics and Policy Making in Developing Countries: Perspectives on the New Political Economy. San Francisco: ICS Press, 1991. Meilink, Henk. Food Consumption and Food Prices in Kenya: A Review, Report No. 21/1987. Food and Nutrition Planning Unit, Ministry of Planning and National Development, Nairobi, Kenya, 1987. Meldrum, Andrew. “Drought, Death and Dissidents,” Africa Report, Vol. 38, no. 1 (January 1993): 32-35. Meredith, Martin. The Fate of Africa: A History of Fifty Years of Independence. New York: Public Affairs, 2005. ----------. Our Votes, Our Guns: Robert Mugabe and the Tragedy of Zimbabwe. New York: Public Affairs, 2002. Moene, K., and M. Wallerstein. “Targeting and Political Support for Welfare Spending,” Economics and Governance, Vol. 2, no. 1 (2001): 3-24. Moi, Daniel. “Speech Announcing the ‘First’ Cabinet,” November 28, 1979. http://www.statehousekenya.go.ke/hist/1979.htm. Montshiwa, P., and J. Mhlanga. Accelerated Rainfed Arable Programme [ARAP], Annual Report 1987-88. Gaborone: Crop Production Division, Department of Agricultural Field Services, Ministry of Agriculture, 1988. Moore, David B. “The Ideological Formation of the Zimbabwean Ruling Class,” Journal of Southern African Studies, Vol. 17, no. 3 (September 1991): 472-95. Moris, Jon R. “Indigenous versus Introduced Solutions to Food Stress in Africa,” in David E. Sahn, ed., Seasonal Variability in Third World Agriculture: The Consequences for Food Security. Baltimore: Johns Hopkins University Press, 1989, 209-34. Morton, Andrew. Moi: The Making of an African Statesman. London: Michael O’Mara Books, 1998.

198

Domestic Politics and Drought Relief in Africa

Moyo, Jonathan N. “The Dialectics of National Unity and Democracy in Zimbabwe,” in Ibbo Mandaza and Lloyd Sachikonye, eds., The One-Party State and Democracy: The Zimbabwe Debate. Harare: SAPES Trust, 1991, 83-102. ----------. “State Politics and Social Domination in Zimbabwe,” Journal of Modern African Studies, Vol. 30, no. 2 (January 1992): 305-30. ----------. Voting for Democracy: Electoral Politics in Zimbabwe. Harare: University of Zimbabwe Publications, 1992. Moyo, Sam. “The Land Question,” in Ibbo Mandaza, ed., Zimbabwe: The Political Economy of Transition, 1980-1986. Dakar: CODESRIA, 1987, 165-201. ----------. The Land Question in Zimbabwe. Harare: SAPES Books, 1995. Mugabe, Robert G. Inside the Third Chimurenga. Harare: Department of Information and Publicity, Office of the President and Cabinet, 2001. ----------. “The Unity Accord: Its Promise for the Future,” in Canaan S. Banana, ed., Turmoil and Tenacity: Zimbabwe, 1890-1990. Harare: The College Press, 1989. “Mugabe Allows UN to Increase Food Aid,” Guardian (London, June 2, 2005). Muir-Leresche, K., and C. Muchopa. “Agricultural Marketing,” in Mandivamba Rukuni et al., Zimbabwe’s Agricultural Revolution Revisited. Harare: University of Zimbabwe Publications, 2006, 299-320. Mulenga, Alfred. “Lt-Gen Walls gave Zim a Torrid Time,” Mmegionline, Vol. 27, no. 121, August 13, 2010. http://www.mmegi.bw/index.php? sid=10& aid=4096&dir=2010/August/Friday13. Mumbengegwi, Clever. “Continuity and Change in Agricultural Policy,” in Ibbo Mandaza, ed., Zimbabwe: The Political Economy of Transition, 1980-1986. Dakar: CODESRIA, 1987, 203-22. ----------., ed. Macroeconomic and Structural Adjustment Policies in Zimbabwe. New York: Palgrave Macmillan, 2002. Munemo, Ngonidzashe. “Political Incumbency and Drought Relief in Africa,” in Armando Barrientos and David Hulme, eds., Social Protection for the Poor and Poorest: Concepts, Policies and Politics. Basingstoke: Palgrave Macmillan, 2008, 264-81. Munro, Lauchlan T. Poverty and Social Safety Nets in Zimbabwe, 1990-99. Doctor of Philosophy Thesis, Institute of Development Policy and Management, University of Manchester, 2001. ----------. “Zimbabwe’s Child Supplementary Feeding Programme: A Reassessment Using Household Survey Data,” Disasters, Vol. 26, no. 3 (2002): 242-61. ----------. “Zimbabwe’s Drought Relief Programme in the 1990s: A Reassessment Using Nationwide Household Survey Data,” Journal of Contingencies and Crisis Management, Vol. 14, no. 3 (September 2006): 125-41. Mushonga, Alan. “Zimbabwe: Origins and Resolution of the Land Crisis,” Southern African Political and Economic Monthly, Vol. 13, no. 8 (May 2000). N’Diaye, Boubacar. “How Not to Institutionalize Civilian Control: Kenya’s Coup Prevention Strategies, 1964-1997,” Armed Forces & Society, Vol. 28, no. 4 (Summer 2002): 619-40.

Bibliography

199

Nathan, Andrew J. “China’s Changing of the Guard: Authoritarian Resilience,” Journal of Democracy, Vol. 14, no. 1 (January 2003): 6-17. Ndemo, Salim P. Epitome of State Power: The Provincial Administration in Kenya. Nairobi: Regional Institute for Cooperatives, Development and Management, 2007. Nengwekhulu, R. “Some Findings on the Origins of Political Parties in Botswana,” Pula: Botswana Journal of African Studies, Vol. 1, no. 2 (June 1979): 47-75. Nkanata, J. Workshop on Provision of Livestock Production Research Information for Use by Extension Workers in the Training and Visit Programme. Muguga: Kenya Agricultural Research Institute, 1985. Nkomo, Joshua. Nkomo: The Story of My Life. London: Methuen, 1984. ----------. “The Significance of National Unity and the Future,” in Canaan S. Banana, ed., Turmoil and Tenacity: Zimbabwe, 1890-1990. Harare: The College Press, 1989, 300-304. Nordlinger, Eric A., et al. “The Return to the State: Critiques,” American Political Science Review, Vol. 82, no. 3 (September 1988): 875-901. Nuemann, Charlotte, et al. “Impact of the 1984 Drought on Food Intake, Nutritional Status, and Household Response in Embu District,” in Thomas Downing et al., eds., Coping With Drought in Kenya: National and Local Strategies. Boulder: Lynne Rienner, 1989, 211-30. Odinga, Oginga. Not Yet Uhuru: An Autobiography. New York: Hill and Wang, 1967. Ogot, B.A. “The Politics of Populism,” in B.A. Ogot and W.R. Ochieng’, eds., Decolonization and Independence in Kenya, 1940-93. Athens: Ohio University Press, 1995, 187-213. Ogot, B.A., and W.R. Ochieng’, eds. Decolonization and Independence in Kenya, 1940-93. Athens: Ohio University Press, 1995. Okoth-Ogendo, H.W.O. “The Politics of Constitutional Change in Kenya Since Independence, 1963-69,” African Affairs, Vol. 71, no. 282 (January 1972): 9-34. Omaar, Rakiya, and Alex de Waal. “Land Tenure, the Creation of Famine and the Prospects for Peace in Somalia,” African Rights Organization, Discussion Paper No. 1. London: African Rights, 1993. O’Neill, Helen, and John Toye, eds. A World Without Famine? New Approaches to Aid and Development. London: Macmillan, 1998. Organization of Rural Associations for Progress. “The 1992 Drought and the Structural Adjustment Programme in Zimbabwe: Preliminary Report.” Bulawayo: ORAP, 1992. Orvis, Stephen W. “Bringing Institutions Back Into the Study of Kenya and Africa,” Africa Today, Vol. 53, no. 2 (Winter 2006): 95-110. “Pambana: Organ of the December Twelve Movement,” Race and Class, Vol. 24, no. 3 (Winter 1983): 322-25. Pankhurst, Alula. Resettlement and Famine in Ethiopia: The Villagers’ Experience. Manchester: University of Manchester Press, 1992. Pankhurst, Richard. The History of Famine and Epidemics in Ethiopia Prior to the Twentieth Century. Addis Ababa: Relief and Rehabilitation Commission, 1993. Parson, Jack. Botswana: Liberal Democracy and the Labor Reserve in Southern Africa. Boulder: Westview Press, 1984.

200

Domestic Politics and Drought Relief in Africa

----------. “Cattle, Class and the State in Rural Botswana,” Journal of Southern African Studies, Vol. 7, no. 2 (April 1981): 236-55. ----------. “The ‘Labour Reserve’ in Historical Perspective: Toward a Political Economy of the Bechuanaland Protectorate,” in Louis A. Picard, ed., The Evolution of Modern Botswana. Lincoln: University of Nebraska Press, 1985, 40-57. ----------. “Succession, Legitimacy, and Political Change in Botswana, 19561987,” in Jack Parson, ed., Succession to High Office in Botswana: Three Case Studies. Athens: Ohio University Center for International Studies, 1990, 97-142. Parsons, Neil. “Seretse Khama and the Bangwato Succession Crisis, 19481953,” in Jack Parson, ed., Succession to High Office in Botswana: Three Case Studies. Athens: Ohio University Center for International Studies, 1990, 73-95. Passmore, P. “Famine Relief: Introduction,” in Gunnar Blix et al., eds., Famine: A Symposium Dealing with Nutrition and Relief Operations in Times of Disaster. Uppsala: Almqvist & Wiksells, 1971, 92-94. Picard, Louis A. “Bureaucrats, Elections and Political Control: National Politics, The District Administration and the Multi-Party System in Botswana,” in Louis A. Picard, ed., The Evolution of Modern Botswana. Lincoln: University of Nebraska Press, 1985, 176-205. ----------. “District Councils in Botswana – A Remnant of Local Authority,” Journal of Modern African Studies, Vol. 17, no. 2 (June 1979): 285-308. ----------., ed. The Evolution of Modern Botswana. Lincoln: University of Nebraska Press, 1985. ----------. “From Bechuanaland to Botswana: An Overview,” in Louis A. Picard, ed., The Evolution of Modern Botswana. Lincoln: University of Nebraska Press, 1985, 3-25. ----------. The Politics of Development in Botswana: A Model for Success? Boulder: Lynne Rienner, 1987. Polhemus, James H. “Botswana Votes: Parties and Elections in an African Democracy,” Journal of Modern African Studies, Vol. 21, no. 3 (September 1983): 397-430. Preston, Matthew. “Stalemate and the Termination of Civil War: Rhodesia Reassessed,” Journal of Peace Research, Vol. 41, no. 1 (January 2004): 6583. Proctor, J.H. “The House of Chiefs and the Political Development of Botswana,” Journal of Modern African Studies, Vol. 6, no. 1 (May 1968): 59-79. Raftopoulos, Brian, and Ian R. Phimister, eds., Keep on Knocking: A History of the Labour Movement in Zimbabwe, 1900-1997. Harare: Baobab Books, 1997. Ranger, Terrence. The African Voice in Southern Rhodesia, 1898-1930. London: Heinemann Educational, 1970. ----------. “Nationalist Historiography, Patriotic History and the History of the Nation: The Struggle Over the Past in Zimbabwe,” Journal of Southern African Studies, Vol. 30, no. 2 (June 2004): 215-34. ----------. Voices from the Rocks: Nature, Culture and History in the Matopos Hills of Zimbabwe. Oxford: James Currey, 1999. Ravallion, Martin. Markets and Famines. Oxford: Clarendon, 1985.

Bibliography

201

Renfrew, Anne. ESAP and Health. Harare: Mambo Press, 1992. Rey, Charles. Monarch of All I Survey: Bechuanaland Diaries, 1929-37, Neil Parsons and Michael Crowder, eds. London: James Currey, 1988. Rich, Tony. “Legacies of the Past? The Results of the 1980 Election in Midlands Province, Zimbabwe,” Africa: Journal of the International African Institute, Vol. 52, no. 3 (1982): 42-55. Richardson, Craig. The Collapse of Zimbabwe in the Wake of the 2000-2003 Land Reforms. Lewiston: The Edwin Mellen Press, 2004. Robinson, Paul. “Reconstructing Gabbra History and Chronology: Time Reckoning, the Gabbra Calendar, and the Cyclical View of Life,” in Thomas Downing et al., eds., Coping With Drought in Kenya: National and Local Strategies. Boulder: Lynne Rienner, 1989, 151-68. Robinson, Pearl. “The National Conference Phenomenon in Francophone Africa,” Comparative Studies in Society and History, Vol. 36, no. 3 (July 1994): 575-610. Rockliffe-King, Geoffrey. Drought, Agriculture and Rural Development: Policy Options. Report on the Evaluation of the Drought Relief and Recovery Programme, 1982-90, Vol. 2. Gaborone: Ministry of Finance and Development Planning, 1990. Rodrik, Dani, ed. In Search of Prosperity: Analytical Narratives on Economic Growth. Princeton: Princeton University Press, 2003. Rukuni, Mandivamba. “Revisiting Zimbabwe’s Agricultural Revolution,” in Mandivamba Rukuni et al., Zimbabwe’s Agricultural Revolution Revisited. Harare: University of Zimbabwe Publications, 2006. Rukuni, Mandivamba, et al. Zimbabwe’s Agricultural Revolution Revisited. Harare: University of Zimbabwe Publications, 2006. Rutherford, Blair. “Zimbabwean Farm Workers: The Struggle for Utopia,” Southern Africa Report, Vol. 15, no. 2 (February 2000): 21-25. Sachikonye, Lloyd. “The Context of the Democracy Debate,” in Ibbo Mandaza and Lloyd Sachikonye, eds., The One-Party State and Democracy: The Zimbabwe Debate. Harare: SAPES Trust, 1991, 43-60. ----------. “The People Have Decided: Zimbabwe’s Referendum Result,” Southern African Political and Economic Monthly, Vol. 13, no. 6 (March 2000). ----------. “Zimbabwe: Drought, Food and Adjustment,” Review of African Political Economy, Vol. 19, no. 53 (March 1992): 88-94. Sahn, David, ed. Seasonal Variability in Third World Agriculture: The Consequences for Food Security. Baltimore: Johns Hopkins University Press, 1989. Samatar, Abdi Ismail. An African Miracle: State and Class Leadership and Colonial Legacy in Botswana Development. Portsmouth, NH: Heinemann, 1999. Samatar, Abdi Ismail, and Sophie Oldfield, “Class and Effective State Institutions: The Botswana Meat Commission,” Journal of Modern African Studies, Vol. 33, no. 4 (December 1995): 651-68. Sanders, David, and Robert Davies. “The Economy, the Health Sector and Child Health in Zimbabwe Since Independence,” Social Science & Medicine, Vol. 27, no. 7 (1988): 723-31.

202

Domestic Politics and Drought Relief in Africa

Sandford, Stephen. Dealing With Drought and Livestock in Botswana: A Report Prepared for the Government of Botswana. Gaborone: Government Printers, May 1977. Sanger, Clyde, and John Nottingham. “The Kenya General Election of 1963,” Journal of Modern African Studies, Vol. 2, no. 1 (March 1964): 1-40. Schapera, Isaac. Tribal Innovators: Tswana Chiefs and Social Change, 17951940. New York: Humanities Press, 1970. ----------. “The Tswana,” in Daryll Forde, ed., Ethnographic Survey of Africa: Southern Africa, Part 3. London: International African Institute, 1953. Schatzberg, Michael G. Political Legitimacy in Middle Africa: Father, Family, Food. Bloomington: Indiana University Press, 2001. Sen, Amartya. Development as Freedom. New York: Anchor Books, 1999. ----------. “Famines as Failures of Exchange Entitlements,” Economic and Political Weekly, Special Issue: Population and Poverty, Vol. 11, no. 31/33 (August 1976): 1273-80. ----------. Poverty and Famines: An Essay on Entitlement and Deprivation. Oxford: Clarendon Press, 1981. ----------. “Starvation and Exchange Entitlements: A General Approach and Its Application to the Great Bengal Famine,” Cambridge Journal of Economics, Vol. 1, no. 1 (1977): 33-59. Shaw, D. “Triangular Transactions in Food Aid: Concept and Practice: The Example of the Zimbabwe Operations,” IDS Bulletin, Vol. 14, no. 2 (April 1983): 29-31. Shepherd, Jack. The Politics of Starvation. New York: Carnegie Endowment for International Peace, 1975. Sheppard, Christopher, and Debbie Clement-Jones. Coping With Drought in Botswana: Household Strategy and Government Policy. Interim Report. Gaborone: Rural Sociology Unit, Ministry of Agriculture, 1979. Sibanda, Eliakim M. The Zimbabwe African People’s Union, 1961-1987: A Political History of Insurgency in Southern Rhodesia. Trenton: Africa World Press, 2005. Sibanda, Misheck. “Early Foundations of African Nationalism,” in Canaan S. Banana, ed., Turmoil and Tenacity: Zimbabwe, 1890-1990. Harare: The College Press, 1989, 25-49. Sillery, Anthony. Founding a Protectorate: History of Bechuanaland, 18851895. The Hague: Mouton and Co., 1965. Sithole, Masipula. “Fighting Authoritarianism in Zimbabwe,” Journal of Democracy, Vol. 12, no. 1 (January 2001): 160-69. ----------. “The General Elections, 1979-1985,” in Ibbo Mandaza, ed., Zimbabwe: The Political Economy of Transition, 1980-1986. Dakar: CODESRIA, 1987, 75-97. ----------. “Should Zimbabwe Go Where Others Are Coming From?,” in Ibbo Mandaza and Lloyd Sachikonye, eds., The One-Party State and Democracy: The Zimbabwe Debate. Harare: SAPES Trust, 1991, 71-82. ----------. Zimbabwe: Struggles-Within-the-Struggle (1957-1980). Harare: Rujeko Publishers, 1999. ----------. “Zimbabwe’s Eroding Authoritarianism,” Journal of Democracy, Vol. 8, no. 1 (January 1997): 127-41. Sithole, Ndabaningi. African Nationalism. Cape Town: Oxford University Press, 1959.

Bibliography

203

Sky News. Interview of Robert Mugabe, May 24, 2004. Slater, Henry. “The Politics of Frustration: The ZAPU-ZANU Split in Historical Perspective,” Kenya Historical Review, Vol. 3, no. 2 (1975): 261-86. Soames, Lord. “From Rhodesia to Zimbabwe,” International Affairs, Vol. 56, no. 3 (Summer 1980): 405-19. Solway, Jacqueline. “Drought as a ‘Revelatory Crisis’: An Exploration of Shifting Entitlements and Hierarchies in the Kalahari, Botswana,” Development and Change, Vol. 25, no. 3 (July 1994): 471-95. Southern African Development Community Secretariat. Measures to Address Food Security in the SADC Region: A Background Paper. Gaborone: SADC Secretariat, 2009. Southern Rhodesia, Government of. Drought Conditions and Food Shortages, 1942-1947 [S2998/1]. Relief from Poverty: Memorandum Prepared by H.H.D. Simmonds, Chief Native. Special Correspondent. “The Changing Face of Kenya Politics,” Transition, No. 25 (1966): 44-50. Spencer, John. The Kenya African Union. London: KPI, 1985. Stack, Jayne. Drought Forecasts and Warnings in Zimbabwe: Actors, Linkages and Information Flows. Working Paper, AEE 2/98. Harare: Department of Agricultural Economics and Extension, University of Zimbabwe, 1998. Stedman, Stephen. The Lancaster House Constitutional Conference on Rhodesia, PEW Program in Case Teaching and Writing in International Affairs, Case #341, Center for International Studies, School of International Relations, University of Southern California (1988). Stepan, Alfred C. The Military in Politics: Changing Patterns in Brazil. Princeton: Princeton University Press, 1971. Stevens, Christopher, and John Speed. “Multi-Partyism in Africa: The Case of Botswana Revisited,” African Affairs, Vol. 76, no. 304 (July 1977): 381-87. Stewart, Frances. “Adjustment with a Human Face: The Role of Food Aid,” Food Policy, Vol. 13, no. 1 (February 1988): 18-26. Stoneman, Colin. “The World Bank Demands Its Pound of Zimbabwe’s Flesh,” Review of African Political Economy, no. 53 (March 1992): 94-96. ----------., ed. Zimbabwe’s Prospects: Issues of Race, Class, State and Capital in Southern Africa. London: Macmillan, 1988. Stoneman, Colin, and Lionel Cliffe. Zimbabwe: Politics, Economics, and Society. London: Pinter Publishers, 1989. Subbarao, Kalanidhi. “Systemic Shocks and Social Protection: Role and Effectiveness of Public Works Programs,” World Bank Social Protection Discussion Paper No. 0302. Washington, D.C.: World Bank, January 2003. Sunday Mail. Harare, Zimbabwe, 1980 to 2006. Sunday News. Bulawayo, Zimbabwe, 1980 to 1995. Swainson, Nicola. “State and Economy in Post-Colonial Kenya, 1963-1978,” Canadian Journal of African Studies, Vol. 12, no. 3 (1978): 357-81. Swift, Jeremy. “Why Are Poor People Vulnerable to Famine?,” IDS Bulletin, Vol. 20, no. 2 (1989): 8-15. Sylla, Lanciné. “Succession of the Charismatic Leader: The Gordian Knot of African Politics,” Arthur Goldhammer, trans., Daedalus, Vol. 111, no. 2 (Spring 1982): 11-28.

204

Domestic Politics and Drought Relief in Africa

Sylvester, Christine. “Simultaneous Revolutions: The Zimbabwean Case,” Journal of Southern African Studies, Vol. 16, no. 3 (September 1990): 45275. ----------. “Unities and Disunities in Zimbabwe’s 1990 Election,” Journal of Modern African Studies, Vol. 28, no. 3 (September 1990): 375-400. ----------. “Whither Opposition in Zimbabwe?,” Journal of Modern African Studies, Vol. 33, no. 3 (September 1995): 403-23. ----------. “Zimbabwe’s 1985 Elections: A Search for National Mythology,” Journal of Modern African Studies, Vol. 24, no. 2 (June 1986): 229-55. Szeftel, Morris. “Clientelism, Corruption & Catastrophe,” Review of African Political Economy, Vol. 27, no. 85 (September 2000): 427-41. Tabor, Steven. Drought Relief and Information Management: Coping Intelligently With Disaster. Gaborone: Government Printers, 1983. Tamarkin, M. “From Kenyatta to Moi: The Anatomy of a Peaceful Transition of Power,” Africa Today, Vol. 26, no. 3 (1979): 21-37. ----------. “The Roots of Political Stability in Kenya,” African Affairs, Vol. 77, no. 308 (July 1978): 297-320. Tarr, S.B. “Undermining the Political Logic of African Governments’ Poor Economic Policies,” Genève-Afrique, Vol. 30, no. 1 (1992): 9-34. Taylor, Scott D. Business and the State in Southern Africa: The Politics of Economic Reform. Boulder: Lynne Rienner, 2007. Teklu, Tesfaye, and Sisay Asefa. “Factors Affecting Employment Choice in a Labor-Intensive Public Works Scheme in Rural Botswana,” Economic Development and Cultural Change, Vol. 46, no. 1 (October 1997): 175-86. Tendi, Blessing-Miles. Making History in Mugabe’s Zimbabwe: Politics, Intellectuals, and the Media. Oxford: Peter Lang, 2010. Tendler, Judith. Good Government in the Tropics. Baltimore: Johns Hopkins University Press, 1997. Terhal, Piet, and Indira Hirway. “Rural Public Works and Food Entitlement Protection: Towards a Strategy for Preventing Hunger,” in Helen O’Neill and John Toye, eds., A World Without Famine?: New Approaches to Aid and Development. London: Macmillan, 1998, 373-403. Thompson, Carol. Drought Management Strategies in Southern Africa: From Relief Through Rehabilitation to Vulnerability Reduction. Windhoek: UNICEF Policy Monitoring Unit, February 1993. Throup, David. “Elections and Political Legitimacy in Kenya,” Africa: Journal of the International African Institute, Vol. 63, no. 3 (1993): 371-96. Tilly, Charles. Coercion, Capital, and European States, AD 990-1992. Cambridge, MA: Blackwell, 1992. Tordoff, William. “Local Administration in Botswana, Part I,” Journal of Administration Overseas, Vol. 12, no. 4 (October 1973): 172-83. ----------. “Local Administration in Botswana, Part II,” Journal of Administration Overseas, Vol. 13, no. 1 (January 1974): 293-304. Treverton, Gregory, and Marc Levy. Rhodesia Becomes Zimbabwe. PEW Program in Case Teaching and Writing in International Affairs, Case # 442 (A&B), John F. Kennedy School of Government, Harvard University (1989). Tsanga, Amy. “Progress Towards Democracy, Or Revenge of Civil Society,” Southern African Political and Economic Monthly, Vol. 13, no. 5 (February 2000).

Bibliography

205

Tsie, Balefi. “The Political Context of Botswana’s Development Performance,” Journal of Southern African Studies, Vol. 22, no. 4 (December 1996): 599616. United Nations Development Programme. Report of the WFP/FAO Mission on Drought Relief in Botswana. Gaborone: UNDP, August 2, 1979. United Nations Food and Agriculture Organization, FAOSTAT. http://faostat. fao.org/site/535/DesktopDefault.aspx?PageID=535#ancor. US Agency for International Development. Cape Verde: Food for Development Program. Washington, D.C.: USAID, 1982. US Agency for International Development, Office of Foreign Disasters Assistance. OFDA Annual Report, FY 1985. Washington, D.C.: USAID, 1985. van de Walle, Nicolas. African Economies and the Politics of Permanent Crisis, 1979-1999. New York: Cambridge University Press, 2001. Vengroff, Richard. Botswana: Rural Development in the Shadow of Apartheid. London: Associated University Presses, 1977. Vierich, Helga, and Christopher Sheppard. Drought in Rural Botswana: SocioEconomic Impact and Government Policy. Gaborone: Rural Sociology Unit, Ministry of Agriculture, 1980. ViVa. Hours of Chaos and the Aftermath. Nairobi: Trend, 1982. von Braun, Joachim, et al. “Famine as the Outcome of Political Production and Market Failures,” IDS Bulletin, Vol. 24, no. 4 (1993): 73-77. von Braun, Joachim, et al. Famine in Africa: Causes, Responses, and Prevention. Baltimore: Johns Hopkins University Press, 1999. von Braun, Joachim, et al. “Labour-Intensive Public Works for Food Security in Africa: Past Experience and Future Potential,” International Labour Review, Vol. 131, no. 1 (1992): 19-33. Waruhiu, S.N. From Autocracy to Democracy in Kenya: Past Systems of Government and Reforms for the Future. Nairobi: Waruhiu and Guthuru Associates, 1994. Weber, Max. Economy and Society: An Outline of Interpretive Sociology, Guenther Roth and Claus Wittich, eds., 3 vols. New York: Bedminster Press, 1968. ----------. From Max Weber: Essays in Sociology, H.H. Gerth and C. Wright Mills, trans. and eds. New York: Oxford University Press, 1958. Weiner, Daniel. “Land and Agricultural Development,” in Colin Stoneman, ed., Zimbabwe’s Prospects: Issues of Race, Class, State and Capital in Southern Africa. London: Macmillan, 1988, 63-89. Weitzer, Ronald. “In Search of Régime Security: Zimbabwe Since Independence,” Journal of Modern African Studies, Vol. 22, no. 4 (December 1984): 529-57. “What is the Drought Fund Used For?,” The Insider (Bulawayo, Zimbabwe, February 21, 1992). Widner, Jennifer A. The Rise of a Party-State in Kenya: From “Harambee!” to “Nyayo!” Berkeley: University of California Press, 1992. Wiseman, John A. “Multi-Partyism in Africa: The Case of Botswana,” African Affairs, Vol. 76, no. 302 (January 1977): 70-79. World Bank. Managing Agricultural Production Risk. Washington, D.C.: Agriculture and Rural Development Department, World Bank, 2005.

206

Domestic Politics and Drought Relief in Africa

Young, Crawford, and Thomas Turner. The Rise and Decline of the Zairian State. Madison: University of Wisconsin Press, 1985. Zimbabwe, Government of. The Drought Relief and Recovery Programme, 1992/93. Harare: Government Publications, 1993. ----------. Report of the Public Service Review Commission, Under the Chairmanship of Professor D. Kavran, Ph.D, Vols. 1-2. Harare: Government Publications, 1989. ----------. Second Five-Year National Development Plan, 1991-1995. Harare: Government Publications, December 1991. ----------. Social Dimensions of Adjustment: A Programme of Action to Mitigate the Social Costs of Adjustment. Harare: Government Printers, 1991. ----------. Zimbabwe: The First Decade 1980-1990. Harare: Government Publications, 1991. ----------. Zimbabwe’s Experience in Dealing with Drought, 1982 to 1984. Harare: Ministry of Labour, Manpower Planning and Social Welfare, 1986. Zimbabwe, Government of, Central Statistical Office. Quarterly Digest of Statistics. Harare: Government of Zimbabwe, 1986. Zimbabwe, Government of, Medical Officer, Matabeleland North Province. Drought Monitoring – Matabeleland North Province, May-August 1992. Harare: Government Publications, December 1992. Zimbabwe, Government of, Ministry of Health, National Nutrition Unit. Child Supplementary Feeding Programme, Status Report. Harare: Government Publications, August-October 1992. Zimbabwe, Government of, Ministry of Health. Planning for Equity in Health: A Sectoral Review and Policy Statement. Harare: Government Publications, 1984. Zimbabwe, Government of, Ministry of Labour, Manpower Training and Social Welfare. Labour and Economy: Report of the National Trade Unions Survey, Vol. 1. Harare: Government Publications, 1984. Zimbabwe, Government of, Ministry of Labour, Manpower Planning and Social Welfare. Memorandum on Drought Relief, 1986. Harare: Government Publications, 1986. Zimbabwe, Government of, Ministry of Lands, Agriculture and Rural Resettlement. Policy Statement for the 1992/3 Agricultural Year. Harare: Government Publications, July 1992. Zibabwe, Government of, Ministry of Lands, Agriculture and Rural Resettlement, National Early Warning Unit. Cereal Grains Availability in Communal Areas. Harare: Government Publications, May 1992. Zimbabwe, Government of, Registrar-General. 1985 General Election Report. Harare: Government Publications, September 13, 1985. Zimbabwe, Government of, Working Group. The Children’s Supplementary Feeding Programme in Zimbabwe. Harare: Government Publications, 1982. Zimbabwe Peace Project. Partisan Distribution of Food and Other Forms of Aid: A National Report – October 2007. Harare: ZPP, 2007. ----------. Partisan Distribution of Food and Other Forms of Aid at District Level: The Case of Manicaland – September 2006. Harare: ZPP, 2006. Zimbabwe Textile Workers Union and Zimbabwe Congress of Trade Unions. Report on Structural Adjustment and Its Impacts on Workers. Harare: ZTWU and ZCTU, 1993.

Bibliography

207

Zimbabwe Vulnerability Assessment Committee. Zimbabwe Emergency Food Security Assessment Report. Gaborone: SADC FANR RVAC, September 2002. ----------. Zimbabwe Rural Food Security and Vulnerability Assessments – June 2005 Report, Report No. 5. Gaborone: SADC FANR RVAC, 2005. ----------. Zimbabwe Rural Food Security and Vulnerability Assessments – April 2004 Preliminary Report. Gaborone: SADC FANR RVAC, 2004. Zimbabwe Women’s Resource Centre and Network, Women and Drought, ZWRCN Workshop Report No. 4, Harare, December 1992. Zolberg, Aristide R. “The Structure of Political Conflict in the New States of Tropical Africa,” American Political Science Review. Vol. 62, no. 1 (March 1968): 70-87. ----------. Creating Political Order: The Party-States of West Africa. Chicago: University of Chicago Press, 1985. Zvobgo, Rugano. “Education and the Challenge of Independence,” in Ibbo Mandaza, ed., Zimbabwe: The Political Economy of Transition, 1980-1986. Dakar: CODESRIA, 1987, 319-54.

Index Accelerated Rainfed Arable Programme (ARAP; Botswana), 1–2, 154, 155(table) Access to relief, 39, 46(table); Botswana's insecure incumbents determining, 147; Botswana's universal access, 40; Kenya's food shortage response, 70; Moi leveraging, 75–77, 77–78; nonexclusion principle, 40–42; ZANU's manipulation of food aid, 97 Ad hoc institutions, 44–45, 46(table), 76 Administering relief, 44–45, 46(table), 70–71 Africa Prize for Leadership, 88 African National Congress (ANC), 134 Agricultural performance, linking relief to, 104 Agricultural sector position, 15–16 Authoritarianism, 16, 18; determining relief program, 17(fig.); differing famine relief outcomes, 178–179; projected erosion of Mugabe's, 108–109. See also Regime type Authority systems, Weberian, 5 Autocracy: deepening of Moi's, 65– 66; democratic and autocratic authority trends, 1960–2009, 180(fig.); Kenya under Moi, 63– 64. See also Regime type Band Aid, 176, 179 Bathoen II, 134, 137–138 Bechuanaland Democratic Party (BDP), 135–137, 140, 146, 150 Bechuanaland Independence Party, 134–135 Bechuanaland People's Party (BPP), 134 Bechuanaland Protectorate, 132–137 Botswana: ARAP, 1–2, 155(table), 254; complementary intervention

in food markets, 44; components of relief program, 1–2; consolidating institutional incumbency, 150–159; crop production decline, 30–31; distinctive conditions for drought response, 159–160; district commissioners, 164(n48); drought incidence frequency, 28; drought relief spending, 1982– 1989, 157(table); Famine Code, 131–132; financial efficiency of food aid, 34, 35(table); food distribution mechanism, 33; freefood distribution, 31–32; House of Chiefs, 162(n21); Immigration Act, 163(n40); income from crop failure replaced by labor-based relief, 153(table); income smoothing through labor-based relief, 37; institution building, 139–141; institutionalization of drought relief, 157–158; as labor reserve for South Africa, 162(n24); labor-based relief, 152–159; mineral wealth, 143– 144, 164(nn64,66); Nkomo's flight, 123(n38); origins of institutional insecurity, 132–137; political insecurity driving food aid choice, 146–150; POLITY score, 15; post-independence political instability, 137–139; recommended national drought relief strategy, 144–146, 151– 152; research methodology, 9–10, 11–12, 19; responsibility for food aid, 167(n117); role of incumbency strength in drought relief responsiveness, 159–161; targeting relief, 41; traditional authority, 161(n7); universal access to relief, 40; variations in

209

210

Index

drought relief programs for ablebodied adults, 11(table) Botswana Agricultural Marketing Board (BAMB), 144, 146 Botswana Defense Force, 138–139 Botswana Independence Party, 136 Botswana Meat Commission (BMC), 142–143 Botswana National Front (BNF), 137–139, 150 Bureaucracy: colonial origins of Botswana's, 133–134 Cash relief, 41 Cattle ownership (Botswana), 141– 142 Cereal Balance, 17(fig.) Change-the-constitution movement (Kenya), 56–57, 81(n25) Chieftainship Act (1965; Botswana), 139–140 Child Supplementary Feeding Programme (Zimbabwe), 94, 107 Civil society (Zimbabwe), 108–109, 112–116 Coalition building, Zimbabwe, 91– 94, 99–100, 101–102 Collapse, political (Zimbabwe), 117– 119 Colonialism: Botswana's institutional insecurity, 132–137; British and German expansion, 161(n4); determining research criteria, 10– 12, 19 Complementary intervention in food markets, 44, 46(table), 67, 70, 72–75 Conservative Alliance of Zimbabwe, 98 Constitutional reform: Kenya, 55–57, 63; Mugabe centralizing position and power, 100–101; Zimbabwean civil society's push for, 112–116 Consumption, smoothing. See Smoothing consumption Corruption, 5; drought and famine as political tool, 6; exclusionary market intervention, 43–44; Moi's free-milk policy, 60; Moi's stance on, 59; "Willowgate," 101

Costs of relief, 2, 3, 32–33, 34, 35(table), 37–38, 157(table) Coups and coup attempts: Kenya, 57, 64–66, 72; Zimbabwe, 90, 122(n21) Crop failure and crop decline, 9, 30– 31, 36, 106, 153(table) Crop surplus, 16, 19, 42, 69, 117–119 Currency devaluation (Zimbabwe), 113–114 De facto one-party state, 4, 19, 54, 109, 160, 173, 178 De jure one-party state, 4, 31, 57, 63, 83, 100, 108–109, 119, 160 December Twelve Movement, 62–63 Democracy, 16, 18; determining relief program, 17(fig.); POLITY score, 15; sufficiency of famine prevention, 177–179 Dependency syndrome, 88, 102–103, 110–120, 170 Detentions, 63–64, 123(n42) Development, economic and social (Botswana), 131–132, 143–144 Development Plan (NDP7; Botswana), 157–158 District drought commissions (Botswana), 141, 156, 158–159, 164(n48) Drought relief: administering, 44–45; defining, 6; frequency, duration, and type, 27–29. See also Food aid; Grain loans; Incumbency; Labor-based relief; Policies, drought relief Drought Relief Levy (Zimbabwe), 97 Drought study: Botswana, 144–146 Early-warning systems, 70, 145–146 Economic crisis: Kenya under Moi, 61, 62; Zimbabwe, 113–114 Economic drought, 28, 35 Economic growth (Botswana), 143, 151, 159–160 Economic performance: Botswana at independence, 137; Botswana's labor-based relief and ARAP programs, 154; Botswana's selfsufficiency goals, 142–143; income shocks, 36–37; political

Index 211

consequences of drought on incumbents, 8 Economic reforms (Zimbabwe), 105– 107, 108, 113 Economic Structural Adjustment Programme (ESAP; Zimbabwe), 105, 112 Elections: BNF's post-independence victory, 138; Botswana's food aid as campaign issue, 147; Ghana's competition and alternation, 182(n14); independent Kenya, 52, 79(n7); Moi's attempt to consolidate political security, 60; Moi's call for early elections, 67; Mugabe centralizing and consolidating power, 100–102; Mugabe's politically manipulated drought relief, 118–119; postindependence Zimbabwe, 89–90, 98–99; stability of Zimbabwe's one-party state, 109 Electoral fraud: Kenya, 79(n16) England, Russell, 134 Ethiopia, drought and famine, 4, 9– 10, 77, 176, 177, 178, 179 Ethnic cleavages and coalitions, 5; KADU formation, 78(n4); Kenyans' support for Moi, 66–67; Kenya's KADU and KANU, 52– 54; political support for Moi's food-aid plan, 72 European Advisory Council (Botswana), 133–134 Exclusion from relief, 39–42 Exclusionary market interventions, 43–44, 46(table); Kenya under Moi, 67, 73–74; Kenya's food shortage response, 70 Executive. See Incumbency External intervention, 6, 10, 31–32. See also International aid Famine and famine threat, 4; Ethiopia, 4, 9–10, 77, 176, 177, 178, 179; labor-based relief as response to, 38; market response to, 42; Moi's market intervention, 73–74; Moi's political insecurity driving crisis response, 70–71; as political tool, 6

Famine Code (Botswana), 131–132 Famine Early Warning System Network, 117 Food aid: administration of Botswana's school food program, 142, 155–156; Botswana relief program, 1–2, 11–12; Botswana's modified delivery with laborbased relief, 152–153; Botswana's political impetus for, 146–147; case study selection, 10; criticism of, 8–9; crop reserves determining feasibility of, 16; distribution mechanism, 33–34; financial efficiency of, 35(table); food sources for, 31–32; goals of relief programs, 29–30; improving Botswana's delivery, 151–152; incumbents' political security driving relief, 19–20, 146–150, 170, 172–173; Kenya, 3; labor-based programs versus, 13; Moi leveraging, 67, 75–77; Moi's NFRF, 71–72; Mugabe targeting support groups, 171; Mugabe's refusal of, 87–88; Mugabe's relief plan, 103–104; political interests affecting policy, 7; POLITY score and, 17(fig.); rational basis for, 9; smoothing consumption, 30–34, 31–34; transition to labor-based relief, 39; universal access, 40; ZANU support for, 96–97; Zimbabwe's Child Supplementary Feeding Programme, 94, 107; Zimbabwe's political collapse, 117–118 Food markets, 42–44. See also Market intervention Food prices, 32, 38, 42–44, 43(table) Food Resources Department (FRD; Botswana), 154–156, 159 Food security, 8, 32, 59–60, 68–69, 118 Food-for-work programs, 12, 17(fig.), 41, 71, 88, 96, 102–103, 103–104, 105–107 Free food distribution. See Food aid Geldof, Bob, 77, 179

212

Index

Ghana: competitive elections and electoral alternation, 182(n14) Gikuyu, Embu and Meru Association (GEMA), 58, 59, 81(n44) Githii, George, 63–64 Gooch, Toby, 151–152 Governance. See Authoritarianism; Autocracy; Democracy; Regime type Government-import food sources, 31–32 Grain Loan Scheme (GLS; Zimbabwe), 110–111 Grain loans, 10, 17(fig.), 41–42, 109– 112 Grain Marketing Board (GMB), 96– 97, 105, 111, 119 Harambee (self-help development funds), 54, 58 Hoarding food, 42–43 House of Chiefs (Botswana), 162(n21) Hunger Project, 88 Hydrological drought, 27–29 Immigration controls, 163(n40) Income, smoothing. See Smoothing income Income effects, 30 Income shocks, 35–39 Incumbency: Botswana's consolidation of institutional incumbency, 150–159; as central factor in drought relief responsiveness, 159–161; challenges of relief, 45; coherence of actions and policies, 13–14; collapse of Mugabe's, 112–117; food security, 32; free food costs, 32; incumbency status-relief form link, 3–4, 7–9, 19–20, 132, 170–173; institutional argument of relief policy, 14–15; Kenyatta securing, 57–58; micro-political conditions, 21–22; Moi's anxiety over, 63; Mugabe and ZANU's consolidation of, 98–102; Mugabe's attempt at ZANUZAPU coalition, 91–92; political and economic interests shaping

policy, 6–7; regime argument, 18; targeting relief, 41; Zimbabwe's food-for-work program, 105–107. See also Insecurity, political Insecurity, political: colonial origins of Botswana's, 132–137; driving African relief programs, 169– 170; driving Botswana's food aid choice, 146–150; drought relief during Mugabe's periods of political security, 102–105; food security affecting, 83(n101); Khama and the BDP, 137; liberalizing regimes, 181; longterm relief trajectories, 173–174; Moi's constraints increasing, 58– 67; Moi's failure to overcome, 67; Moi's manipulation of food relief, 77–78; Moi's response to drought and famine, 68–70; Moi's succession to Kenyatta, 55–57; Mugabe exploiting dearth for political security, 95–98, 119– 121; Mugabe's shifting relief stance, 88; relationships between relief programs and, 174(fig.); relief contributing to incumbents' security, 170–172; threats to Mugabe's incumbency, 90–92. See also Incumbency Institution building, Botswana, 132– 137, 145–146, 150–159, 157– 158, 160 Institutional argument, 14–15 Institutional Food Programme (IFP), 167(n117) Institutional structures, 44–45 Insulation, political (Zimbabwe), 108–112, 120 Inter-Ministerial Drought Committee (IMDC), 154 International aid, 70, 87–88, 176– 177, 179. See also External intervention International Monetary Fund (IMF), 61, 105 Joint Advisory Council (Botswana), 133–134 Kangai, Kumbirai, 93 Kariuki, P.N., 66

Index 213

Kenya: administering relief, 45; Botswana's relief program and, 2, 159–160; components of relief programs, 3; coup attempt, 64– 66; crop production decline, 30; drought, famine threat, and Moi's response, 68–71; drought incidence frequency, 29; famine relief in authoritarian regimes, 178; food distribution, 31–33; income shocks, 35–36; maize price changes, 43(table); market response to dearth, 42; Moi leveraging food relief access, 75– 78; Moi's descent into autocracy, 65–67; Moi's increasing insecurity and autocracy, 57–67; Moi's lack of policy commitment, 12; origin of autocracy and political insecurity, 51–57; POLITY score, 15; relief contributing to Moi's political security, 170–171, 172; research methodology, 9–10, 19; urban market interventions, 72–75; variations in drought relief programs for able-bodied adults, 11(table) Kenya Africa Socialist Alliance, 63 Kenya African Democratic Union (KADU), 52, 54, 78(n4), 79(nn7,9) Kenya African National Union (KANU), 52–58, 62, 66–67, 79(n7), 80(nn18,24) Kenya Farmers' Association (KFA), 74–75 Kenya Grain Growers Co-operative Union (KGGCU), 74–75 Kenya People's Union (KPU), 54 Kenyatta, Jomo, 52–58, 62, 79(n13), 80(n18) Khama, Ian, 139 Khama, Seretse, 131(quote), 134– 137, 139, 140, 144–145, 146, 150–151 Kibaki, Mwai, 64, 67 Koma, Kenneth, 137, 138 Labor-based relief: Botswana, 1–2, 11–12, 152–159; Botswana's

income from crop failure replaced by, 153(table); Botswana's institutionalization of drought relief, 152; Botswana's national relief strategy, 145–146; case study selection, 10; conditions for Botswana's responsiveness, 160; determining relief program, 17(fig.); free food aid versus, 9, 13, 34; goals of, 30; Moi's drought relief strategy, 67; Moi's insecurity quashing Kenya's, 71– 72; political interests affecting policy, 7–8; POLITY score of democracy, 15; secure incumbents choosing, 170; smoothing incomes, 35–39, 37– 38; targeting the needy, 41; transition from free food relief, 39; Zimbabwe's food-for work programs, 12; Zimbabwe's secure incumbents' support of, 102 Lancaster House Conference, 89 Land reform, 115–116, 117, 140 Legislative Council (Legco; Bechuanaland), 133–135, 162(n13) Legitimacy, regime, 63–64, 66–67, 68, 83(n101). See also Incumbency; Insecurity, political Live Aid, 176 Livestock production, 36, 142–143, 147 Local government, democratization of Botswana's, 140–141 MacDonald, John, 151–152 Made, Joseph, 87 Maize production, 29, 30, 32–33, 43(table), 72–73, 95, 106(table) Makombe, Tranos, 102–103 Mangwana, Paul, 87 Marafhe, Mompati, 138–139 Market intervention, 43–44, 46(table), 67, 69, 70, 72–75, 78, 94–95, 106–107 Marxist movements: Kenya under Moi, 62–63; Zimbabwe, 122(n19) Masire, Quett, 2, 131(quote), 135, 138, 145, 150–151, 152 Matante, P.G., 134–135, 136, 137

214

Index

Matimela (impounded cattle) Act (1968; Botswana), 141 Mboya, Tom, 53, 54–55, 79(n16), 80(n24) Mbuthia, Peter, 66 Meat production, 36, 142–143 Mengistu Haile Mariam, 4–5, 77, 177, 179 Meteorological drought, 27–29, 35 Micro-political conditions, 19–20, 21–22, 119, 152, 161 Migration, 27–28, 137 Mining (Botswana), 143–144 Modern state, authority systems and, 5 Modise, John K., 139 Moi, Daniel arap, 3, 12, 20–21, 42, 170–171, 172; background and political emergence, 51–53; building and losing political security, 51, 57–67; drought relief strategy, 68–71; ethnic coalition, 78(n4); leveraging food relief access, 75–77; military coup attempt, 64–66; political reforms, 82(n50); vice-presidency, 55–57 Morgan, David, 134 Mosinyi, G., 139 Motsetse, K.T., 134–135 Movement for Democratic Change (MDC; Zimbabwe), 105, 112, 115, 117, 119 Mpho, Motsamai, 134, 135 Mugabe, Robert, 3, 21; denial of food crisis, 87–88; exploiting dearth for political security, 119–121; grain loan scheme, 110–112; labor-based relief, 102–103; longterm relief trajectory, 173–174, 176; Mugabe exploiting dearth for political security, 95–98; political ideology, 122(n19); post-independence elections, 98– 99; ZANU-ZAPU coalition, 89– 94; ZANU-ZAPU Unity Accord, 99–100. See also Zimbabwe Multi-party democracy (Botswana), 135–136, 160 Murumbi, Joseph, 55 Muzorewa, Abel, 90, 109

National Cereals and Produce Board (NCPB; Kenya), 32, 42, 73–74 National Constitutional Assembly (NCA), 114, 128(n141) National Famine Relief Fund (NFRF; Kenya), 70, 71–72 National security: Moi's control of food relief, 73–74 Native Advisory Council (Botswana), 133–134 Natural resource wealth, Botswana's, 132–133, 141–142, 143–144, 164(nn64,66) Neocolonialism, 137–138 Neopatrimonialism, 5–6 Ngei, Paul, 56, 57, 79(n7) Njonjo, Charles, 56 Nkala, Enos, 91–92, 122(n34) Nkomo, Joshua, 89, 92, 123(n38) Non-exclusion principle of relief access, 40–42 Nyagumbo, Maurice, 101 Nyayo (footsteps), 62, 67 Odinga, Oginga, 53, 54–55, 63, 79(n13) Odinga, Raila, 63 Oil production, 60–61 Oloitipitip, Stanley, 66 One-party states. See De facto oneparty state; De jure one-party state Operation Maize, 59–60 Overseas Development Institute, 145 Pan Africanist Congress (PAC), 134 Paramilitary Police Unit (PMU; Botswana), 138 Patronage, 19–20, 61, 71, 78, 94–95 People's Party (Bechuanaland), 136 People's Redemption Council (PRC), 64–65 Personal patrimonial authority, 5 Policies, drought relief: actors and actions shaping, 6–8, 13–14, 27; challenges and alternatives, 46(table); goals of relief programs, 29–30; government intervention in food markets, 43– 44; regime argument, 14–15; structural argument, 15–16; universal access, 39–42. See also

Index 215

Food aid; Grain loans; Laborbased relief Political determination of relief programs, 170–171 Political economy, 14, 78 Political interests. See Incumbency Political liberalization, 179, 181 Political order, 5–6 Political rights, 178 Political will, 4, 169 POLITY score, 15, 17(fig.), 24(n47) Populism: Moi's attempt at political security, 59, 60, 66–67 Postcolonial modern state, 5 Price controls: Zimbabwe, 94–95, 105 Process tracing, 19 Provincial Administration (PA), 53– 54, 58, 63, 67, 71, 73–74, 75–77 Public works projects, 1–2. See also Labor-based relief Rational political interests, 8 Rational-legal authority, 5 Referendum: Mugabe's constitution, 115 Regime argument, 14–15 Regime type, 16, 17(fig.), 18, 31, 37, 177–179 Rehabilitation programs, 131–132. See also Labor-based relief Responsiveness, government, 5–6, 12, 22(n16) Rhodesian Front Party, 90, 91 Sandford, Stephen, 145–146, 152 Sandura Commission, 101 School feeding programs, 142, 155– 156 Security, political. See Incumbency; Insecurity, political Self-sufficiency, Botswana's, 142– 143 Sen, Amartya, 88, 177–178 Shava, Frederick, 102–103 Sithole, Ndabaningi, 109 Smith, Ian, 90, 91, 98 Smoothing consumption, 30–34, 45, 46(table), 70 Smoothing income, 37, 45, 46(table), 70, 145–146. See also Laborbased relief

Social conditions increasing vulnerability to drought, 4–6 Social drought, 27–28 Socialism: Botswana National Front, 137–138, 139 Sorghum prices, 106(table) South Africa, 31, 134, 137 Standing relief institutions, 44, 46(table), 146, 158, 176 State weakness, 4–5 Strikes: Kenya under Moi, 61–62; Zimbabwean insurgency, 93–94; Zimbabwe's economic crisis, 114 Structural adjustment programs, 105 Structural argument, 15–16 Student protests, 61–63, 65, 66, 137 Subsidies, 105 Succession rules: Botswana's political stability, 150–151; Moi succeeding Kenyatta, 55–57 Supplementary Feeding Programme (Botswana), 142 Taxation: Zimbabwe's Drought Relief Levy, 97 Technical capacity, 169 Technocratic approaches, 7–8, 176 Tekere, Edgar, 101, 102 Towett, Taita arap, 57 Traditional authority: Bathoen's radical politics, 137–138; Bechuanaland, 132–135, 135– 136, 161(n7); Botswana's incumbents weakening the financial position of, 141–142; Botswana's modern government stripping powers of, 139–140; Botswana's provision of food relief, 147; Weberian authority systems, 5 Transitional society, authority systems in, 5 Transportation, food, 33–34 Tsoebebe, Archelus, 135 Tsvangirai, Morgan, 114, 118 Unions: Mboya's origins, 80(n24); Moi's destruction of the KFA, 74–75; Zimbabwean insurgency, 93–94. See also Strikes United African National Congress (Zimbabwe), 90

216

Index

United Parties (Zimbabwe), 109 Unity Accord (Zimbabwe), 99–100, 101, 108 Universal access. See Access to relief Unrest, social and political: Kenya under Moi, 61–62 Urban areas, market intervention, 72– 75, 94–95 Variations in drought relief programs for able-bodied adults, 11(table) Village Development Committees (VDC; Botswana), 141, 156, 158–159 Violence: execution of Kenya's coup leaders, 66; ZANU-ZAPU clashes, 90, 122(n34); Zimbabwe's coalition government, 92; Zimbabwe's post-independence elections, 89– 90; Zimbabwe's referendum defeat, 115–116 Voter turnout: Kenya under Moi, 67 Vulnerability, political. See Insecurity, political Walls, Peter, 90–91 War veterans (Zimbabwe), 113–114, 118 Water resources, 36–37, 156 Weapons cache, 92, 123(n37) Weber, Max, 5 Wheat production, 30, 72–73, 106(table) White population, 91, 94–95, 98, 100, 115–116, 117, 134 Women: Zimbabwe's children's food program, 107 Work-for-cash programs, 96. See also Labor-based relief World Bank, 61, 105 World Food Program (WFP), 142 ZANU-PF, 99–102, 107–109, 111– 113, 115, 117, 119, 120, 127(n125) Zimbabwe: administering relief, 45; Botswana's relief program and, 2, 153, 159–160; change in electoral support for ZANU by province, 99(table); complementary intervention in food markets, 44; crop production decline, 30–31;

drought incidence frequency, 28; drought relief during Mugabe's periods of political security, 102– 105; erosion and collapse of Mugabe's incumbency, 112–117; food price controls, 94–95; foodfor-work program, 105–107; freefood distribution, 31–32, 33; freefood program costs, 33; income shocks, 36–37; labor-based relief, 12, 38, 102; long-term relief trajectory, 173–174, 176; Mugabe centralizing and consolidating power, 100–102; Mugabe exploiting dearth for political security, 119–121; political collapse, 117–119; political insulation threatening relief programs, 108–112, 109–112; political security and relief policy, 21, 95–98; POLITY score, 15, 17(fig.), 24(n47); postindependence political instability and insecurity, 89–98; relief contributing to Mugabe's political security, 171, 172; research methodology, 9–10, 11–12, 19; targeting relief, 41–42; variations in drought relief programs for able-bodied adults, 11(table); ZANU consolidating Mugabe's incumbency, 98–100. See also Mugabe, Robert Zimbabwe African National Union (ZANU), 2–3, 88, 89–94, 97, 98– 99, 120, 171; Marxist orientation, 122(n19). See also Mugabe, Robert Zimbabwe African People's Union (ZAPU), 89–94, 98–99, 102 Zimbabwe Congress of Trade Unions (ZCTU), 93–94, 105, 114 Zimbabwe Unity Movement (ZUM), 101–102 Zvogbo, Edson, 113

About the Book

Ngonidzashe Munemo challenges the conventional wisdom that African governments lack the technical capacity and political will to respond to drought and the threat of famine. Through a comparative analysis of three politically disparate countries—Botswana, Kenya, and Zimbabwe—Munemo demonstrates that differences in the ways that governments face similar droughtinduced food crisis are the result not of incompetence, but of rational political considerations. His original analysis shows why, in democracies and authoritarian regimes alike, the less effective option is so often the policy choice. Ngonidzashe Munemo is assistant professor of political science at Williams College.

217