TY  - THES
T1  - Two worlds in agricultural policy making in Africa? Case studies from Ghana, Kenya, Senegal and Uganda
A1  - Mockshell,Jonathan
Y1  - 2016/06/07
N2  - There has been a renewed interest among donors and domestic policy makers in promoting agricultural development in Africa. Such renewed interest is evident in initiatives, such as the Comprehensive Africa Agricultural Development Program (CAADP), the Alliance for a Green Revolution in Africa (AGRA), the New Alliance for Food Security and Nutrition, and TerrAfrica. Yet, the choice of policy instruments that are most suitable to promote agricultural development in Africa remains subject to a contested debate. The main objective of this cumulative thesis is to analyze the role that the policy beliefs of different actors play in determining policy choices and policy implementation, a topic that has been largely neglected in the agricultural economics and political science literature on the political economy of agricultural policies in Africa.
While some policy actors argue that agricultural development requires strong state support, others criticize state-focused instruments and favor market-oriented strategies. Examples of such unresolved debates regarding the role of the state versus the private sector include controversies about input subsidies, import taxes, price stabilization, and buffer stock programs. On the ground, one can observe that African governments implement input subsidy programs and reverse the abolishment of parastatal organizations, in spite of the prevailing critique of such policy instruments by some donor organizations. As the literature reviews included in this thesis show, the dominant explanations in the agricultural economics literature for the choice of such agricultural policies have mainly been based on the rational choice paradigm, such as interest group theories and voter-politician models. Explanations in the political science literature have focused on the role of politician’s self- interest in the form of patronage. Neither of these strands of literature has paid attention to the role of policy beliefs that different actors have and the influence of such beliefs on policy choices. The main rationale for this thesis is the proposition that a narrow focus on self-interest in the prevailing explanations of agricultural policy choices in Africa limits their value. This is especially the case, if the goal of policy research is to identify strategies on how long-standing controversies can be resolved with a view to moving towards more effective policy implementation.
This thesis is based on case studies conducted in four countries: Ghana, Kenya, Senegal, and Uganda. The countries were selected because they are similar in the following aspects: They are largely agrarian-based, they implemented Structural Adjustment Programs in the 1980s, and they moved towards increasing support to agriculture in recent years. However, they differ in important characteristics, especially regarding the nature of their political system. These characteristics make them good case study examples for analyzing contested agricultural policy debates. By combining different types of case studies, the thesis covers the entire spectrum from policy formulation to policy implementation.
The thesis begins with an introduction, which presents the rationale for the study, provides background information on the agricultural policy context in the four case study countries, and reviews the relevant literature in the fields of agricultural economics and political science. After the introduction, the thesis presents two case studies that deal with policy formulation. They focus on the policy discourse in three of the four case study countries, namely, Senegal, Ghana, and Uganda. In these case studies, the roles of policy discourses and policy beliefs in the policy process are examined. The case studies combine the Advocacy Coalition Framework with discourse analysis and narrative policy analysis to identify the predominant policy beliefs of different coalitions of actors. Based on interviews with a wide range of policy-makers and stakeholders, two main coalitions were identified in each country. In the case study of Ghana and Uganda, these coalitions were labeled “donor discourse coalition” and “domestic discourse coalition.” In Senegal, the labels “agricultural support coalition” and “agricultural support critique coalition” were used. As the cluster analysis showed, the donor coalition and the agricultural support critique coalition mostly comprise international financial institutions, the finance ministry, and academia. Members from the agricultural ministry, Parliament, political parties, and civil society groups mostly form the domestic coalition and the agricultural support coalition. The study finds strikingly similar patterns of policy beliefs across the three case study countries. Donors and domestic policy makers held fundamentally different policy beliefs regarding the question: What does it take to develop smallholder agriculture? While domestic policy actors tend to believe that transforming smallholder agriculture requires public intervention to provide modern inputs at subsidized rates, members of the donor coalition tend to believe that these interventions are market distorting and only motivated by political patronage. The narrative policy analysis of the discourses suggests that the agricultural support narrative has a convincing story-line with a clear beginning (low productivity caused by lack of inputs), middle (providing subsidized inputs), and end (increased productivity). In contrast, the agricultural support critique is mostly presented in the form of a non-story (focusing on what should not be done without providing a convincing alternative story-line of what should be done).
Moving from policy formulation to decision-making on policy adoption, the third case study examines how divergent policy beliefs are translated into policy programs. The analysis focuses on two major agricultural programs in Ghana: the Block Farms Program and the agricultural investment program developed under CAADP, called Medium Term Agriculture Sector Investment Plan (METASIP). A participatory mapping of the process leading to the policy programs (called Process Net-Map) was combined with in-depth interviews conducted with policy experts. The empirical results reveal two divergent policy processes: (1) The METASIP policy process that involved broad stakeholder consultation, but where donors were considered the most influential policy actors in the process. (2) The policy process leading to the Block Farms Program. In contrast to the METASIP process, donors did not feature as policy actors in this policy process, which was also less consultative. Domestic policy makers, including the ruling political party, Parliament, and the president, played key roles in this policy process. A fundamental difference between the two programs relates to the policy orientation: The Block Farms Program takes a public sector approach, while METASIP stresses private sector participation in agricultural service provision. Thus, this case study indicates that basic differences in policy beliefs between donors and domestic policy makers are translated into different policy programs through parallel policy processes, which are not connected to each other. Therefore, the case study suggests that “two worlds” (donors and domestic policy makers) exist not only with regard to policy beliefs, but also with regard to policy processes and the resulting policy choices.
After having covered policy formulation and policy adoption, the last two case studies deal with policy implementation. They focus on animal health services, because this is a policy field where the case study countries had already implemented policy reforms during the structural adjustment period. These reforms largely followed donor prescriptions to reduce government involvement and promote privatization. The two case studies aimed to analyze the effects of these reforms in relation to prevailing policy beliefs regarding the potential positive and negative effects of privatization. Data for the two case studies were collected through surveys of livestock keepers in a marginal livestock production area in the northern region of Ghana and in a high-potential dairy production area in Kenya. Key informant interviews with livestock policy experts were also conducted in Ghana. The analysis showed that the policy reform of liberalization introduced new players into the animal health service delivery systems. A multinomial logit model was applied to determine the factors that influence households’ choice of service providers. In Ghana, a low potential region (semi-arid, remote) was examined, and the analysis revealed that in this region government para-vets were preferred to community animal health workers and private para- vets. This choice was attributed to the relatively low performance of the community animal health workers, resulting from their limited training. In areas with few or no government para-vets, farmers resorted to self-treatment or to selling sick animals for human consumption, which has undesirable health implications. In Kenya, a high-potential area with intensive dairy production was examined. The results expectedly suggest that service delivery was generally better than in Ghana, as a private service provider market had indeed emerged. In this respect, the experience corresponded to the policy beliefs of those policy actors who had pursued this reform. However, the study found that poorer framers in these areas also face challenges in accessing qualified service providers. Thus, the findings indicate that privatization of livestock services had problematic results for farmers in marginal areas, as well as for poor farmers in high-potential areas. These experiences in the liberalization of the livestock sector may be one contributing factor to why domestic policy makers did not change their policy beliefs regarding the need for government intervention in support of smallholder farming.
The final two chapters of the thesis discuss the findings of the five case studies in a cross-cutting perspective and derive implications from for policy processes that aim at developing smallholder agriculture in Africa. As the case studies show, the policy beliefs of donors and domestic policy makers are not easily reconciled, especially since there is a tendency among policy actors on both sides to have a positive self-representation and a negative other representation, which can easily lead to “blame games”. What seems rather problematic for bridging the gap between the “two worlds” is the view identified in the donor discourse that domestic policy makers only pursue government-focused programs as a strategy to stay in political power or to create opportunities for corruption. There would be no contradiction in accepting that domestic policy makers genuinely believe in the need for better physical access of smallholders to inputs, while acknowledging that such instruments have political advantages as well. The study suggests that if donors could accept genuine concerns of domestic policy makers, it would be easier to engage in a more productive policy dialogue, such as on how to make input-subsidy programs more effective or how to address service needs of poor livestock keepers. Based on the proposition that a more consensus-oriented approach will ultimately lead to more effective agricultural policies, the study concludes that it is critical to find strategies to promote such a dialogue between donor coalitions and domestic coalitions. Such strategies may aim at engaging policy actors through deliberation, building new coalitions, promoting policy-oriented learning, and involving policy brokers to find new alternative solutions to reach a consensus. Considering the similarities found across the four case study countries, the study also concludes that the findings are relevant for other African countries.

KW  - Diskursanalyse
KW  - Agrarpolitik
KW  - Afrika 
CY  - Hohenheim
PB  - Kommunikations-, Informations- und Medienzentrum der Universität Hohenheim
AD  - Garbenstr. 15, 70593 Stuttgart
UR  - http://opus.uni-hohenheim.de/volltexte/2016/1211
ER  -