IFPRI Discussion Paper 01030 October 2010 Examining Relationships between Customary and State Institutions in Ghana’s Decentralized System Cory Belden Development Strategy and Governance Division INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE The International Food Policy Research Institute (IFPRI) was established in 1975. IFPRI is one of 15 agricultural research centers that receive principal funding from governments, private foundations, and international and regional organizations, most of which are members of the Consultative Group on International Agricultural Research (CGIAR). PARTNERS AND CONTRIBUTORS IFPRI gratefully acknowledges the generous unrestricted funding from Australia, Canada, China, Denmark, Finland, France, Germany, India, Ireland, Italy, Japan, the Netherlands, Norway, the Philippines, South Africa, Sweden, Switzerland, the United Kingdom, the United States, and the World Bank. AUTHOR Cory Belden, The World Bank Short-Term Consultant Formerly an Intern in IFPRI’s DSG division Notices 1 Effective January 2007, the Discussion Paper series within each division and the Director General’s Office of IFPRI were merged into one IFPRI–wide Discussion Paper series. The new series begins with number 00689, reflecting the prior publication of 688 discussion papers within the dispersed series. The earlier series are available on IFPRI’s website at www.ifpri.org/pubs/otherpubs.htm#dp. 2 IFPRI Discussion Papers contain preliminary material and research results. They have been peer reviewed, but have not been subject to a formal external review via IFPRI’s Publications Review Committee. They are circulated in order to stimulate discussion and critical comment; any opinions expressed are those of the author(s) and do not necessarily reflect the policies or opinions of IFPRI. Copyright 2010 International Food Policy Research Institute. All rights reserved. Sections of this material may be reproduced for personal and not-for-profit use without the express written permission of but with acknowledgment to IFPRI. To reproduce the material contained herein for profit or commercial use requires express written permission. To obtain permission, contact the Communications Division at ifpri-copyright@cgiar.org. iii Contents Abstract v Acknowledgments vi 1. Introduction and Objectives 1 2. Motivation 3 3. Conceptual Framework 5 4. Methodology 10 5. Findings 11 6. Conclusions 23 References 27 iv List of Tables 1. Perceptions of traditional authorities 3 2. Regression results: Explaining the perceptions index of traditional authorities 4 3. Incentives that motivate both the state and customary institutions in Ghana 15 4. Hypotheses on the four patterns of interaction between the state and customary institutions 23 List of Figures 1. Institutional Analysis And Development framework, modified 5 2. General illustration of government and chieftaincy structures in Ghana 11 v ABSTRACT Traditional authorities are powerful leaders alongside the state in Ghana. The chieftaincy has been resilient to “modernization”—maintaining land rights, allegiance from citizens, and influence in rural communities. Nonetheless, there are few rules defining their official role in the local government structure. It is empirically acknowledged that chiefs seriously impact the development of their communities. Hence, this study looks for factors that might explain the state’s deficiency in policy regarding chiefs. This analysis combines the Institutional Analysis and Development (IAD) framework with theory on formal and informal rules. Not only does this adaptation provide additional insight, but it helps to reduce the complexity in political research on dual governance bodies. Findings suggest that the formal and informal rule sets coupled with the resources available to state and customary actors result in strong exchange organizations between the two institutions. Incentives encourage noninterference and avoidance from the state; thus, rules concerning the chieftaincy are rarely enforced or modified. If attempting to harness collaboration and mitigate conflict and collusion, the state—in partnership with the chiefs—might reconsider the lawful role and authorities of the chiefs at local levels. Based on the analysis, policy revisions are needed to improve the outcomes of the institutional arrangement; however, major changes may be difficult to achieve in the current political context. Keywords: decentralization, customary law, incentives, rules, state system, traditional authority vi ACKNOWLEDGMENTS First and foremost, I would like to thank and acknowledge Regina Birner for her endless support, insight, and assistance. Not only did she provide guidance in the design and analysis of this study, but she gave me unique opportunities to pursue my interests and ideas. I also want to thank Emelia Guo, who facilitated interviews, made contacts, and actively participated in the fieldwork. I am equally grateful to the entire Ghana Strategy Support Program team for their support, interest, and genuine thoughtfulness during my stay in Ghana. I owe a special thanks to Shashi Kolavalli, who provided extensive comments on the drafts and proposals, as well as Eva Schiffer, who introduced me to the Net-Map tool. Finally, my deepest gratitude goes to all interviewees and experts who took the time to narrate their stories. Their openness and honesty was certainly critical for this study. 1 1. INTRODUCTION AND OBJECTIVES And yet the beard has come to meet the eyebrow. – Ga Traditional Proverb Unlike the beard, whose bristled hairs come poking through the pores in adolescence, the eyebrow— forever placed on the human face—appears even before birth and often never feels the jagged edge of a razor. The eyebrow, experiencing every year of a person’s life, acquires wisdom and understanding far quicker than the unstable and late-blooming beard, whose unpredictable color and growth rate can put a damper on the morning of any hurried businessman. The eyebrow is all-seeing and never-changing. The beard is not. This proverb was originally intended to reprimand rebel youth attempting to outwit their elders. Now it serves another purpose, even years after Ghana’s independence: to remind the government that the traditional authorities have been around far longer than postcolonial rulers. Chieftaincy in Ghana, as a result of the country’s history, law, and political environment, has been—to a great extent—ignored in genuine policy debates in the last decade. Indeed, it is considered one of the most empowered traditional institutions today, particularly because the chiefs own 80 percent of all land rapidly expanding in value. Yet even despite its apparent influence and increasing power, most policies regarding the authorities of the institution have not changed since the implementation of Ghana’s Constitution in 1992. More importantly, empirical research examining the institution’s actual effects on economic development is limited. The ill-defined role of the customary leaders in modern Ghanaian government is not a new discovery. Nor is the conclusion that these leaders do indeed continue to influence both political and economic outcomes (Daanaa 2010; Logan 2009; Mahama 2009; Ubink and Quan 2008; MLGRD Ghana, 2007; World Bank 2007). In fact, a number of analysts have found that while seemingly irrelevant at the national level in Ghana, chiefs are acutely germane at the local level (Logan 2009; Daanaa 2010; 10 2010).1 Exploratory in nature, this study attempts to uncover exactly how the behaviors of contemporary chiefs affect communities in Ghana today and for what purposes the state and customary institutions interact with one another. Informant interviews and Net-Map, which is a tool to help clarify “social networks … and collect data about the perceived power of different actors within a policy field” (Schiffer and Waale 2008), were used to collect information on the relationships between the district assembly and the traditional leaders. It was hoped that by gathering detailed descriptions of projects or conflicts from multiple actors, researchers might be able to link interactions between the state and chiefs to development outcomes. This qualitative design was also used to search for possible explanations for the continued use of vague rules regarding chiefs’ authority and roles. Finally, researchers wanted to more broadly fill some This finding likely extends beyond Ghanaian boundaries, as traditional leaders are present and powerful in many communities from South Africa to Indonesia and Canada. Although there is evident need for policymakers and researchers to formally clarify the role of the traditional authorities, specifically in local settings, this has not been done in many developing countries. Decentralization— meaning here the process of moving decisionmaking powers of the central government closer to the people—implies local politics, participation, and development. Yet decentralization literature says little about the management of and interaction with traditional leaders and systems present in local governance arenas. As will be illustrated, this is a noteworthy gap in decentralization theory and practice. 1 To protect the identity of interviewees, respondents will be cited with reference numbers, instead of names, as follows: By region and district: A=First Northern district; B=First Western district; C=Second Western District; and D=Second Northern District . By institution and position: district chief executive=1, district assembly member=2, unit committee member=3, divisional chief=4, subchief=5, caretaker=6, queen mother=7, clan head=9, expert=10, mine representative=11, executive officer=12. Finally, by community: first interviewed=1, second interviewed=2. (See Section 4, “Methodology,” for additional information.) For example, a divisional chief from the Western Region, second district in the second community interviewed would be cited as C.4.2 and alphabetized accordingly in the reference list. When the overwhelming majority of interviewees are cited, no direct quote will be given. 2 of the gaps in empirical literature relating to chiefs in decentralized systems. Thus the following questions were asked: • What is the contemporary role of the traditional authorities, and what are the implications for development outcomes and local government effectiveness? • What powers do the chiefs retain, and what is the source of it? • How do state and customary actors interact, and why? • To what extent would a different formal structure or policy make better use of the opportunities and reduce possible conflict? Given the complex institutional arrangements in Ghana, an adapted version of the Institutional Analysis and Development (IAD) framework is used to structure and analyze the findings. Because this study addresses dual governing institutions, additional theory is incorporated into the IAD framework. Recent scholarly work has highlighted the need to separate formal rules or institutions from those informal in political research (Helmke and Levitsky 2004; Knight 1992; Lauth 2000). However, informal rules are often framed as the customary norms or traditional body in a given society. This poses problems for research in countries like Ghana, where both strong state and customary bodies coexist. In this context, both institutions maintain their own sets of formal and informal rules, which are particularly acute at the local level. Thus in order to analyze the effectiveness of decentralization policy and local government structure, we separate rules into four categories. This provides a much more comprehensive approach to the analysis. Given such complexity in rules and interactions between the state and customary actors, we also integrate incentives and exchanges into the framework. Douglass North’s transaction cost theory of exchange as well as social exchange theory help to clarify the incentives that shape the interactions between the two institutions. Broadly, our findings suggest that scholars and policymakers should stimulate a more open dialogue on the official role of chiefs in local government systems. For one, chiefs clearly influence development outcomes and local government structure in Ghana. More importantly, the dual incentive structures that motivate both the state and customary bodies illustrate why state formal rules have not been modified in recent years. In essence, because incentives lock the coexisting governing institutions in a competitive arrangement, state and customary actors will either retain or exchange their assets in attempts to maintain power over the other. Though some exchanges result in negative development outcomes, revising law could reduce asset maximization for some actors. As a result, these actors steer clear of invoking policy change. Collaboration between the two institutions appears the most beneficial for communities in need of development. However, collaborative efforts occur only under certain conditions. After providing a brief history and some background on decentralization theory and its applicability to Ghana, the conceptual framework and methodology are reviewed in depth. Subsequently, the findings on resources, formal and informal rules in both the state and the chieftaincy, and incentive structures are discussed. We then explore the patterns of interaction—collaboration, noninteraction, collusion, and conflict within the chieftaincy—and the conditions under which they occur. Finally, we present hypotheses that might help to predict the interactions and exchanges between the state and the customary institution. Policy implications are briefly discussed in the conclusion. 3 2. MOTIVATION A brief political history of chieftaincy in Ghana, a review of the current perceptions concerning chiefs today, and their relevance in decentralization practices are all important aspects of this discussion. Previous debates on each of these matters have bounced between ideologies of the “modernists” and the “traditionalists” (Keulder 1998), and between the “trivialists” and the “romanticizers” (Oomen 2000). Chiefs have also been called the source of gender inequalities (Ribot 2001; Vijayalakshmi 2002). The rise of local government in developing countries provides new reasoning for their exclusion, including that the empowerment of alternative authorities undermines the function and legitimacy of the new democratic ones (Ribot 2003; Ntsebeza 2007). Nonetheless, the chiefs continue to influence society and the government. This is a serious matter as the lack of correspondence between formal institutions and society’s norms or values will likely result in weak and ineffective institutions (North 1990a). Traditional leaders were responsible for the daily administration of their people (Mahama 2009) long before colonialists hit the rich soil of the Gold Coast. Yet as early as the mid-1800s, their duties were congealed as the British bestowed upon them an “official” role within the colonial government (Buah 1980). Chiefs collected taxes, managed community affairs, and liaised with the colonialists on matters of trade (Owusu 2006), using their local knowledge for colonial benefit. Some suggest that even the growth of rural capitalism further intensified this role, as it accentuated indigenous social and political hierarchies (Boone 2003). However, in the mid-1900s, the nationalist movement began. Kwame Nkrumah, the first president of independent Ghana, established a decentralized nationalist administration (Mahama 2009; Boone 2003) to curtail the power of the chiefs, who were now his political opponents. Coupled with this strategy were growing theoretical suppositions that modern (political) office would become more important than traditional leadership (Steegstra 2005). As power tensions rose, the ideological separations of modern government and traditional leadership solidified (Ribot 2001). After a series of coups from 1966 to 1981, Lieutenant J. J. Rawlings of the Provisional National Defense Council ascended to power, suspending both the constitution and all politics. In the early 1990s, Rawlings began to revolutionize legal order by revising the constitution and creating the Local Government Act (LGA). This was done without chiefly participation, and thus their political party prohibition was maintained (Koranteng and Larbi 2008). In 1996, when Rawlings suggested that chiefs finally be permitted to participate in partisan politics again, public pushback was immense. This example demonstrates the normative separation of politics and traditions so evident in the brief historical account given above; some findings suggest that the widely shared image of the “good chief” is always nonpartisan (Lentz 1998). As demonstrated, this nonpartisan image originates from the power struggle during the independence era. Reinforcing this notion is the role of chiefs as vote brokers (Ubink and Quan 2008), which certainly incentivizes the government to enforce nonpartisanship. Current public perceptions are also relevant. As noted previously, many have confirmed that contemporary traditional leaders influence their communities and development. This is also well represented in surveys and media. Table 1 presents important findings from a chieftaincy study in Ghana that surveyed almost 3,000 people (Abotchie 2006) in 2006. As displayed, 90 percent of respondents still believe that chiefs can contribute to economic development. Table 1. Perceptions of traditional authorities Perceptions of chiefs, in Ghana % Chiefs unify the community. 90 Chiefs should continue their traditional roles. 77 Chiefs contribute to ethnic conflict. 60 Chiefs contribute to national unity. 80 Chiefs can contribute to economic development. 90 Chiefs do not have respect of the community. 43 Source: Abotchie 2006. 4 Another recent study conducted in six African countries (not including Ghana) found that the general public views traditional authorities and elected leaders as “two sides of the same coin” (Logan 2009).2 Table 2 Using factor analysis on data from an Afrobarameter survey, Logan creates indexes for perceptions of various governance bodies like Parliament and the executive branch. She also generates an index for traditional leaders. This index includes topics about how trustworthy, attentive, and corrupt the chief is in the respondents’ perspective. Through regression analysis (see ), Logan finds that the perceptions of traditional authorities are positively linked to perceptions of other government leaders. In fact, it is strongest when compared to the local government officials (Logan 2009). Most surprisingly, positive perceptions of traditional authorities are related to democratic principles, like voting. Table 2. Regression results: Explaining the perceptions index of traditional authorities (Constant) B (unstandardized) Index perceptions of executive government 0.065** Index perceptions of Parliament 0.069** Index of perceptions of local government 0.200*** Urban or rural (1/0, rural excluded) 0.090 Vote matters 0.024* Work for a candidate or party 0.047* Source: Logan 2009. Note: * P < 0.05. ** P < 0.01. *** P < 0.001. Thus if chiefs can contribute to economic development and are positively associated with democracy and local government, what does this mean for decentralization efforts? Ghana was formally decentralized in 1992, and its success is frequently debated. Although the decentralized system allocates some administrative and fiscal responsibilities to the district assemblies, they are actually quite limited in their decisionmaking powers (Ayee 2004; Crook and Manor 1998). District assemblies are not permitted to borrow money from the federal government without the approval of the Ministry of Finance (Ayee 2004), to transfer staff without consulting central authorities (Boone 2003), or to demand that bureaucratic agencies such as the Ministry of Education or the Ministry of Health take action on projects (Koranteng and Larbi 2008; Ayee 2004). The assemblies have minimal funds as well. The insufficient district assembly common fund and poorly yielding taxes leave little revenue for projects (Koranteng and Larbi 2008; Ayee 2004). As a result of these inadequate authorities and resources, elected assembly members are often left to one strategy to achieve community development: liaising and bonding with the district chief executive (DCE). The appointed DCE retains considerable decisionmaking power in the assembly; thus one of the only ways of ensuring that a project comes to one’s community is by supporting his political party. As will be illustrated, the constraints of the assembly members help to explain why and how chiefs with power and resources can influence development efforts and alter local government structures. Finally, those studying the shortcomings and successes of Ghana’s decentralization do not often examine the real impact of the customary authorities in the local government structure. This is surprising, as many central governments and development agencies integrate chiefs into the state administration by way of decentralization processes (Alexandre 1970; Lund 1998; Mawhood 1983). Moreover, there seems to be gaps in empirical literature on how chiefs in general distort social welfare and elected official accountability. For example, although decentralization outcomes are difficult to quantify, scholars agree that there are some factors that either make or break them. Social welfare distortions—where certain groups receive special treatment (Bardhan and Mookherjee 2006)—are considered some of the breaking factors. Yet none of the widely acknowledged social welfare distortions include the effects of traditional authorities or preexisting governance bodies. Two additional obstacles to the success of decentralization are poor incentive structures and ineffective accountability mechanisms; these are both reflected in the state and customary bodies in Ghana. 2 It should be noted that only respondents who said that they have a traditional leader participated in the survey. 5 3. CONCEPTUAL FRAMEWORK Upon indentifying the acuteness in which traditional authorities influence the local government and development outcomes, we chose a framework that captured the various institutional arrangements and the factors that influence them. The IAD framework provides an adaptable backdrop for the study. The framework has undergone extensive changes and additions since its creation in 1982 (Ostrom 1999) and still proves useful for institutional analysis today. We made significant changes to the framework, as seen in Figure 1, most of which are compatible with the authors’ original intentions. Figure 1. Institutional Analysis and Development framework, modified Source: Adapted from Ostrom, Gardner, and Walker 1994; Ostrom 1999. Helmke and Levitsky (2004) recently suggested that most comparative research on institutions focuses primarily on formal rules and institutions, ignoring those informal. This is a problem in governance research as “scholars who fail to consider [these] rules of the game risk missing many of the important incentives and constraints that underlie political behavior” (Helmke and Levitsky 2004, abstract). Yet there are additional scholarly assumptions that hinder thorough analysis of political institutions. Customary institutions, from the Western perspective, are typically considered informal. Yet this is not the case in Ghana. On the contrary, customary institutions clearly maintain both formal and informal rule sets just like the modern state. Hence, in this context and in other colonial settings where customary institutions are empowered, comprehensive analysis requires differentiation between both state and customary institutions, as well as formal and informal rules. Given this fact, we replace the general rules-in-use in the IAD framework with these more specific rule sets: state formal, state informal, customary formal, and customary informal. The IAD framework can be combined with many theories (Ostrom 1999). Given this flexibility, we also add incentives to the framework. In this study, these incentives are grounded in transaction cost theory of exchange and social exchange theory. As will be demonstrated, the findings suggest that 6 depending on the resources available and the rules-in-use in a given situation, state and customary actors will exchange resources with one another according to the strength of the incentives present in their interaction. These exchanges are not limited to tangible objects like consumer products (North 1990b). In this context, actors can exchange income for values, votes, promises, and expectations. The causes for these exchanges are split into two main variables: the actor’s motivation and his or her environment (North 1990b). We find that these motivations (or incentives) are generated by the rules-in-use and resources available (or the environment). The combination of the three variables—the resources available, the rules-in-use, and incentives—thereby result in specific patterns of interaction between the two institutions. These patterns of interaction, which highlight how the state and customary institutions interact, are the focus of this study due to their influence on development outcomes and local government performance. The following section describes in more detail the uses and intentions of the variables in this adapted framework. Resources Available3 The resources available are defined as the assets that each actor possesses in a given interaction. In the original IAD framework, these resources—called physical conditions and community attributes— represent the overarching environment that sets the scene for the actors’ behaviors and motivations (Ostrom 1999). In this study, resources are limited to only those that each actor holds in his possession; they are either retained or exchanged for another resource in the action arena. Resources are defined this way because of their relevance in explaining why and how customary and state institutions interact with one another. (Although the overarching environment indeed affects the interactions as well, our findings were not sufficient to include in the analysis.) The specifics of these political, monetary, positional, and pardoning resources will be addressed when discussing the findings. Rules-in-Use Ostrom, Gardner, and Walker (2004) provide a useful paradigm regarding the rules-in-use in the IAD framework. Acknowledging that some rules are hard to conceptualize, the authors provide “rule configurations” that help sort the many rules used in a given society. These rule configurations include entry and exit rules, authority rules, information rules, and a number of others (Ostrom 1999). However, because this analysis examines two prominent institutions, rules will only be categorized into the four sets displayed in Figure 1. Most of the current literature on rules and institutions assumes that “actors’ incentives and expectations are shaped primarily…by formal rules” (Helmke and Levitsky 2004, 725). Yet informal rules certainly influence the incentives. Using a matrix, Helmke and Levitsky illustrate that formal– informal institutional interactions produce complementary, accommodating, competing, and substitutive outcomes. The effectiveness of the formal rules and the extent to which the informal ones diverge from them can help to predict these outcomes (Helmke and Levitsky 2004). While these four outcomes are useful when studying one institution, they are not as effective when studying two, both of which maintain formal and informal rules. Moreover, because most of the formal state rules discussed in this study appear ineffective or unenforced, the matrix is not as helpful. North (1990b) suggests that both types of rules and their enforcement define the opportunities of the actors in a given interaction. This enforcement becomes important to our findings. 3 These resources are generated from observations and interviews, as well as supporting literature discussed within the paper. 7 Formal Rules State formal rules are the constitutional, legislative, and administrative rules (Sabatier 2007) in a society. They are the “procedures [or regulations] that are created, communicated, and enforced through channels widely accepted as official” (Helmke and Levitsky 2004), and are also known as the “rule of law.” State formal rules may be founded on the customary preexisting rules, or created without them (Helmke and Levitsky 2004, 727). This is why customary formal rules are important in colonial contexts. In Ghana’s case and in others, the state institution was created seemingly alongside of these rules. As a result, the chieftaincy stood and still stands as a dual governing institution with its own set of rules. Thus the customary formal rules today are the rules still used within the chieftaincy. These rules are important because they structure the intrainstitutional behavior of the traditional leaders. Informal Rules Informal rules are equally as important as the formal rules. States often lack the ability to enforce their formal rules (Tsai 2007). This leaves an open window for powerful informal rules, which can be defined as traditions and other norms of behavior (Dia 1996; Pejovich 1999) that are enforced through unofficial channels (Helmke and Levitsky 2004). These rules are typically more difficult to change through policy intervention because they are deeply embedded in the codes of conduct in all society (North 1990a; Ribot 2001). Helmke and Levistsky (2004) define these rules quite specifically, distinguishing them from organizations, political parties, and weak institutions. But for our purposes, informal rules will not be so clearly delineated. This is done for two reasons: (1) defining them so specifically does not add value to the analysis, and (2) most of the informal rules—even though some are simply violations of the formal rules—are ingrained in norms and values of the society. This means that even if the formal rules are enforced, the informal rules that allow leaders to participate in unpopular or illegal activities might continue to influence interactions and encourage unconstructive exchanges (Helmke and Levistsky 2004). The state informal rules and customary informal rules, as they interact with the formal, influence development outcomes and consequences. It should be stated here that these informal rules, while pertinent to both exchanges and interactions, will not be part of our final hypotheses in the conclusion. Incentives The resources available to state and customary actors coupled with the rules-in-use in their environment generate incentives that prompt exchanges between the customary and state actors. These incentives are constant in every interface, yet they vary in strength depending on the context. The four incentives present and influencing all interinstitutional exchanges and interactions in Ghana are mutual threat, moral ethics, political reward, and resource gain. Before discussing each of these in the findings, some theoretical review is necessary. Social exchange theory and North’s transaction cost theory of exchange exemplify the importance of these incentives in generating exchange organizations and reinforcing dual governance bodies. These theories provide the foundation for why the state and customary institutions coexist and interact with (or avoid) one another, and they are discussed here and in Section 5.4, “Patterns of Interaction.” A Darwinian lens assumes that inefficient institutions are weeded out over time, hence leaving humanity with a gradual evolution of more efficient forms of economic, political, and social organization (Alchian 1950). However, this does not appear to be the case, as inefficient institutions that stagnate or reduce economic productivity continue to function in many contexts. North (1990a) and Ostrom, Gardner, and Walker (1994) argue that imperfect information, ideology, path dependence, and high enforcement or monitoring costs prolong the survival of these institutions. Thus these topics become essential to our analysis. Path dependence partially explains the existence of dual institutions. In colonial contexts, traditional and state agents are almost always equally present in the daily lives of citizens; this coexistence occurs because “the initial set of institutions … created interest groups with a stake in the existing constraints” (North 1990a, 99). Yet interest groups are not the only ones embracing both the state and customary bodies. Even despite possible inefficiencies, the public will ideologically rationalize their 8 function, thereby reinforcing the incentives (North 1990) for leaders to maintain the status quo. These incentives that reinforce the institutional structure also prompt exchange between actors. Action Arena The action arena is simply where the incentives, rules-in-use, and resources produce an interaction between the state and the customary institution. Examples are used to support our findings on the patterns of interaction. These examples are the action arenas, in which interactions and exchanges occur between actors. Patterns of Interaction Patterns of interaction are the focus of this study. In each community visited, four interactions between the state and customary institutions were visible. The rules-in-use, resources available, and incentives produced collaboration, noninteraction, or collusion between state and customary actors. Though it is not an interaction between the two institutions, conflict within the chieftaincy is also included due to its prevalence and development consequences. Each of these interactions results in state and customary actors either exchanging or retaining their resources. The following theoretical underpinnings help to explain why these exchanges occur. Yet before introducing the patterns of interaction further, it must be clarified that although chiefs gave land or finances to the community or state in all of the communities visited, contribution is not considered a pattern of interaction. This is because contribution results in highly divergent and unpredictable development outcomes. More importantly, chiefs can contribute and collude, contribute and collaborate, or contribute and conflict with one another. This indicates that contribution is more a behavior of a chief that can be integrated into any interaction with the state body. North, in a sort of “utilitarianism” approach, suggests that when parties own their assets, “they have incentives to maximize the potential gains from exchange” (North 1990b, 187) while minimizing cost (Molm, Takahashi, and Peterson 2000). In Ghana, each resource in the adapted IAD framework can be considered an asset owned by the state or customary actor involved in the interaction. Thus, state and customary actors form exchange organizations in an attempt to maximize the utility of their assets. These exchanges usually involve complementary resources (Koranteng and Larbi 2008). For example, an elected state official might turn a blind eye to a chief engaging in illegal behavior in order to gain the chief’s community voter support. Neither the chief nor the state official could easily achieve this by using only personal assets. However, unlike in pure economic exchange, where prices are set and exchange processes are clear, the actors in social exchanges cannot flawlessly measure their benefits and risks. Blau (1986) explains that exchange in this context involves the principle that while one person does another a favor and a general expectation of some future return exists, its exact nature is not always stipulated in advance (Blau 1986). Social exchange—defined as the exchange of activity, tangible or intangible, and more or less rewarding or costly, between at least two persons (Homans 1961)—certainly pertains to this discussion. Homans’s social exchange and North’s institutional exchange seem to parallel each other in premise, as both suggest that interactions are produced by one actor’s behavior, reinforcing the other’s and vice versa. Finally, interactions and exchanges result in both positive and negative economic and social outcomes. These exchanges have transaction costs, some of which are significant. Monitoring and enforcing institutional exchanges can be quite expensive (North 1990b; Ostrom, Gardner and Walker 1994); thus in many political settings they are not monitored. This is likely the case in countries with traditional and state systems. For one, trying to effectively monitor all of the interactions between the chiefs and state would likely be difficult and costly, not to mention that the incentives to avoid close supervision of these exchanges are high. This has serious consequences for relationships between the institutions. Without supervision, the state and the customary institutions face even more uncertainty in their exchanges, which leads to self-interested and competitive behavior. Thus it can be assumed that customary and state actors will cheat, steal, or conceal information from each other if it leads to a higher 9 payoff in a given exchange (North 1990b). Moreover, this uncertainty encourages the actors to not interact with one another if the risk is too great. Outcomes Outcomes are a main focus in the original Ostrom, Gardner, and Walker (1994) framework. While they will not be discussed in-depth in this paper, some outcomes pertaining to the effects of the customary and state interactions will be mentioned when reviewing the patterns of interaction. The outcomes most relevant to this study are community development, equity in the district, and community cohesion.4 4 Community cohesion is considered an increase or decrease in community conflict. 10 4. METHODOLOGY In 2008, the International Food Policy Research Institute and World Bank (2010) conducted a large quantitative household survey in six districts and three regions in Ghana.5 After careful consideration, we selected two districts that were surveyed in this household questionnaire. This was done with the intention of making available additional important data (that is, poverty levels, age dispersion, citizen participation in meetings) that could not be captured in this qualitative case study. The two districts are located in the Northern and Western regions. These regions were selected based on their resources (the Northern region being resource poor and the Western, resource rich) in order to observe how revenue affects chief and state interactions. The interview guide was made to focus quite specifically on only interactions between the government and traditional authorities. After initial scoping interviews were conducted, we realized that focusing on areas with interesting cases (like conflict between actors) could provide more insight on this topic. Therefore we selected two additional neighboring districts, one in each region, to learn more about state and chief interactions. A total of eight communities—two in each district—served as the centers of each case, where the corresponding district assembly member, subchief, unit committee members, and other key informants were interviewed. Complementary interviews were conducted with various district executives and experts as well. More than 40 interviews were conducted, and 13 Net-Maps were collected. 5 This survey collected information from approximately 1,000 households, community members, and service providers. The data included questions on water supply, agriculture extension, political activities, assets, and other relevant information. Both male and female household members were interviewed. 11 5. FINDINGS Formal and Informal Rules-in-Use In Ghana, state and customary institutions are more or less parallel structures. To help visualize their structure, Figure 2 provides a generalized diagram illustrating their similarities.6 Please note that the physical boundaries of the state and chieftaincy are different from one another.7 Figure 2. General illustration of government and chieftaincy structures in Ghana The four rule sets described in the conceptual framework are discussed following. Source: Author’s creation. State Institution Formal Rules: The Constitution, Chieftaincy Act, and Local Government Act The two most important legal documents summarizing the state’s formal rules regarding traditional leaders are the 1992 Ghana Constitution and 1971 Chieftaincy Act. The 1993 Local Government Act, while imperative to the functions of the local government, does little to address the role or authority of the chiefs in the decentralized system. The Ghana Constitution first states, “Parliament shall have no power to enact any law which … detracts or derogates from the honor and dignity of the institution of chieftaincy” (Ghana Parliament, 1992, 270.1). It also prohibits chiefs from participating in active partisan politics (Ghana Parliament, 1992, 276.1). The divisional and paramount chiefs that sit in the National House of Chiefs (NHC) and Regional Houses of Chiefs (RHC) oversee the activities and land transfers in their traditional areas; they are also to “undertake the progressive study, interpretation and codification of customary law … and compile the customary laws and lines of succession applicable to each stool or skin” (Ghana Parliament, 1992, 271.3). Currently, there are 193 Traditional Councils and 10 Regional Houses, which are state institutions run by civil servants (Ghana Parliament, 1971, 6.1, 1.1). However, they are not always treated as such. In 2006, all civil servants except those working in the Traditional Councils were given a salary raise (Anamzoya 2008). Briefly, although Ghana also has a state judicial system, the National, Regional, and Traditional Councils are responsible for performing judiciary functions through the state as well. These functions include but are not limited to resolving chieftaincy disputes and sanctioning people for committing crimes like theft and adultery (Ghana Parliament, 1971, 12). 6 It should be noted that not all traditions mirror this exact structure. The example does not include other types of traditional authorities such as caretakers. 7 Accurate maps of the traditional areas are not available because borders have not been demarcated. 12 The Ghana Constitution also frames property rights and the processes for collecting lease rents. The Office of Administration of Stool Lands (OASL), a state-run agency, is to “collect rents, royalties, dues, [etc.]” (Ghana Parliament, 1992, 267) for all land rents and distribute them to the appropriate beneficiaries. It should be noted here that a “stool land” is the territory of a divisional chief. There are typically a number of stools in a “traditional area,” which the paramount chief oversees. The revenues are dispersed based on a constitutional formula: 10 percent remains with the OASL for administrative purposes, 25 percent goes to the stool (or the Traditional Council) for its maintenance, 20 percent goes to the traditional authority (or divisional chief), and the remaining 45 percent is allocated to the district assembly in which the stool lands are situated (Ghana Parliament, 1992, 267.7). Notably, the Ghana Constitution and other laws do not require the traditional authorities to publicly share how they use these land rents and royalties. The OASL and the Lands Commission are also to consult the chiefs “in all matters relating to [the stool’s] administration and development” (Ghana Parliament, 1992, 267). It should be mentioned here that these rents and royalties are not the only revenues that chiefs receive for leasing land to others. They also receive “drink money,” which will be discussed shortly. According to the 1993 Local Government Act, the elected district assembly members, just like the chiefs, are not permitted to align with a political party. However, the appointed DCE, who oversees and provides the final say for the decisions of the district assembly, is allowed to participate and openly support political parties. This is important to note when discussing the resources available to state actors. Beyond the DCE and civil servants, the district assembly includes “one elected representative from each electoral area within the district … and other members not being more than thirty percent, who are appointed by the President in consultation with the traditional authorities and other interest groups in the district” (Ghana Ministry of Local Government, 1993, 242.A, italics added). The last clause in this article is the only reference to the actual role of traditional authorities at the local level. Article 10 defines the duties of the district assemblies (which often mimic the role patterns of the traditional authorities). They are responsible for the following8 • Supporting productive activity and social development in the district : • Initiating programs for the development of basic infrastructure • Initiating joint participation with any other people to execute development plans • Monitoring the execution of projects and evaluating their impact on the people’s development and the local and district economy • Cooperating with the appropriate national and local security agencies for the maintenance of security and public safety in the district (Ghana Ministry of Local Government, 1993, 10.1– 10.7) Informal Rules That Encourage Violation of the Formal: Drink Money and Partisanship Chiefs also maintain informal property rights alongside the formal. These informal property rights are considered rules of the state because the state is the body that has authority to change these rights and rules. “Drink money” is a payment that a land leaser makes to the chiefs in addition to paying rent to the OASL (Ubink and Quan 2008). This payment was originally intended to show appreciation for the good stewardship of the landowner (or chief). Before land increased in value, this appreciation actually came in the form of drinks. One or two bottles of Schnapps were sufficient to show gratitude (Ubink and Quan 2008). However, as demand for land has risen, so has the price of drinks. Chiefs now collect an up-front cash payment from companies, foreign investors, and even nonindigenous citizens, which now represents the market price for land leases (Wily and Hammond. 2001). The OASL does not account for this money, and the chiefs are not lawfully required to disclose the amount they receive from leasers. In effect, it has become income of the traditional authorities. 8 These are only the functions of the district assembly that are applicable to the analysis. 13 Since independence, Ghana has developed partisanship that has led to the politicization of numerous nonpolitical activities. Today the country is a two-party system with serious voting blocs. Informal rules generated by socioeconomic and geographical factors reinforce these voting blocs (Akramov and Asante 2009; Fridy 2007). Ethnicity influences partisanship as well. Populous groups like the Ewe consistently support the National Patriotic Party (NPP); most Ashantis espouse the National Democratic Congress (NDC) (Akramov and Asante 2009; Fridy 2007). Party politics also influence how the district assembly performs. Partisanship encourages the DCE to assist assembly sympathizers more often than those in the opposition. Even when district assembly nominees lose in elections, if they are party activists, there is that evidence they are frequently appointed to the assembly (Ghana and Development Partner 2007). These partisan rules, particularly in district leadership, result in the politicization of many community development activities. Finally, although formal state rules do not exempt chiefs from criminal punishment, there appears to be a strong hesitancy to sanction them when they commit a crime. This is partially due to the “noninterference” constitution clauses and the respect that chiefs retain in communities (Crook 2003). Moreover, the customary consequences of sanctioning an accused chief are grave: Their breaking the law results in their automatic destoolment (or impeachment). Thus politicians and police are cautious to arrest or report chiefs for illegal activities. Due to their role as vote brokers, chiefs can also damage the careers of politicians who allow their arrest. Customary Institution Formal Rules: Succession, Traditions, and Rank Group cultures and traditions are complex in Ghana. With more than 100 ethnic divisions and stool lands (Salm and Falola 2002), traditional systems vary greatly. While many customary formal structures today follow those in Figure 2, there are still vast differences between groups. Some traditions have kings and others do not. Some traditions are matrilineal and others are patrilineal. In most cases, the divisional or paramount chief has authority over the land; however, in others, clans or family heads retain this power (Immigration and Refugee Board of Canada 2002). It should be noted here that despite slight differences in who owns or oversees the land, OASL revenues always go to the divisional chief, who is the authority over the stool land. It is then up to the chief’s discretion if and how these revenues are dispersed to subchiefs. On the other hand, drink money is more flexible. This payment typically goes to whatever chief manages the land.9 As a result of traditional differences, customary laws also vary. Even before colonialism, stools had their own unwritten constitutions that outlined the qualifications of the chief, who is permitted to sit as king, and how the gates should rotate (Immigration and Refugee Board of Canada 2002; Brukum 2004). Succession is a key feature of this formal customary law. In most traditions, succession rotates through “gates,” or a cycle of successors. For example, if a divisional chief died and his title was open, a royal family member from the appropriate gate would be selected to fill his position. These gates rotate each time a traditional position becomes available. The number of gates is not absolute; rather it is defined by the number of royal families in a given area. As will be illustrated, some succession is automatic, and in other cases it is not. As a result, forms of succession greatly impact chiefly conflict. These divergent customs and intertraditional formal rules make state regulation difficult. However, there have been attempts by the NHC to legalize some of these norms, such as succession, for the purpose of preventing conflict (Yaro 2010). These laws have not been passed through Parliament (Daanaa 2010). Nonetheless, it can be assumed that drink money goes to higher-ranking chiefs when transactions involve expensive land or large investments, even if a subchief oversees that property. Policing the community is also considered a customary formal rule because the state gives some judicial authorities to the chieftaincy. This can be quite useful where state police are not stationed in communities (A.12 2010). In one community, a subchief fines individuals who commit crimes like theft and adultery. Interviewees agreed that the community respects these fines “because it keeps the youth 9 This was observed during the fieldwork. 14 under control” (C.3.1 2010). Policing the community is important in this study because it gives the chiefs authority and an elevated social status. Informal Rules That Encourage Violation of the Formal: Politicization and Partisanship Partisanship is also present in the customary institution’s informal rules. Chiefs’ participation in party affairs is well known, but both state and customary actors will avoid admitting this.10 Finally, royalty dispersion is also considered an informal rule as distributions change based on traditional leaders’ personalities and ideologies. As aforementioned, it appears that to whom and how divisional chiefs give royalties and drink money varies throughout the stools. Thus it can be assumed that informal rules within each stool dictate this dispersion. Mahama actually states that “a disturbing percentage of traditional authorities are actively involved in politics” (2009, 12). Though quick to deny this finding, almost every interviewee stated that the public knows which party the chief supports. Although referendums have repeatedly considered allowing chiefs to participate in politics, there are strong arguments against this change. Shouting, “Every government is my government!” a subchief affirmed that if he were to support one party, then when the opposition came into power, the district assembly would stop assisting him and his community (C.5.1 2010). Also contributing to chiefly partisanship is the personal history of a chief. We encountered a number of chiefs who, preceding their enstoolment, were political activists or district assembly members. Thus they carried a strong political history into their traditional position. Resources Available Again, the resources available to both customary and state actors serve as assets that are used to maximize potential in exchanges and interactions.11 Incentives that Prompt Resource Exchange In Ghana, it appears that political resources are often in the hands of the chiefs. Because chiefs continue to sway voters, the traditional authorities will “sell” votes in exchanges with the state. Both state and customary actors own monetary resources. Royalties and revenues, as well as government allowances, go to both institutions; depending on context and scope, these finances can be used in an exchange as well. Possessing positional resources is equivalent to having a higher-than-normal social status. Being popular, seen, and respected in communities appears to be important to both state and customary leaders. At the local level, elected district assembly members seem to typically retain a higher social status compared with the subchiefs because citizens expect much more from them. Thus the local state actors often control this resource. If a subchief contributes to a district assembly project, the chief usually receives praise from the district assembly member, which in turn increases the chiefs’ visibility in the community. In abstract form, this can be viewed as a financial-for- positional status exchange. Finally, the ability to turn a blind eye or to pardon an individual can also be viewed as a resource. It is an asset typically owned by the state, as it is up to state actors to enforce the rule of law. If a chief is involved in illegal activity, a state actor can pardon the chief in exchange for votes or monetary resources. The rules-in-use and the resources available to actors generate the incentives for interaction (or noninteraction) between the state and customary institutions. The following paragraphs describe these incentives that encourage patterns of interaction between the two bodies. These incentives are derived from economic, sociological, and anthropological concepts; yet they are also tailored to fit Ghana’s political and institutional context. Essentially, these incentives serve as an implicit chessboard for every state and customary interaction: In every opportunity for exchange, the traditional and state authorities weigh their potential gains against their risks in consideration of the four incentives below. 10 This was observed during the fieldwork. 11 These resources are defined based on what appeared important to chiefs and state actors during the fieldwork. 15 Resource Gain Resources themselves are only one part of the environment that creates the motivation for interaction; another incentive is the actual act of obtaining resources. This incentive is well known in economic theory. The state actor is motivated to interact with the customary institution based on the resources the chief has available. If these resources are complementary, the state actor will exchange his or her resources for those owned by the customary actor. The same goes for the customary actor looking to increase or expand resources. Political Reward Political reward is fairly straightforward and is also founded in economic principles. State actors are incentivized by reelection and votes; traditional authorities seek higher ranks in the NHC, RHC, or Traditional Councils. Subchiefs might also receive supplementary royalties or support from divisional chiefs if they make an active attempt to empower the customary institution. Both the state and customary institutions consider these political rewards when they interact with one another. Moral Ethics Moral ethics also serve as an incentive for both the state and customary institutions. This incentive provides a more sociological approach to understanding why actors make certain decisions. All interviewees, particularly when discussing traditional authorities, spoke of maintaining cultural respect and stewardship in their communities or districts. This desire to make ethical decisions and uphold morals is apparent in both bodies. It is likely that these ethics, or the desire to appear morally committed, stems from traditional and perhaps religious belief systems. Thus historically and still today, many communities expect their subchiefs to be good stewards (Mahama 2009). This same expectation translates to state actors—bringing projects and development to the community is a primary duty of the district assembly member. Both state and traditional leaders who do not uphold (or at least do not appear to uphold) these cultural moral obligations are not typically viewed as popular leaders. Mutual Threat Both institutions face an internal threat in the current institutional structure: each other. This mutual threat is not always acknowledged, nor is leading to an impending institutional death for either the state or the customary institution. However, the threat does serve as an incentive to keep one another in check and is quite visible in the attitudes of interviewees. In Ghana, it can be assumed that the state democratic government has the upper hand over the chieftaincy. However, because the traditional leaders retain most of the property rights, they are still a source of competition for the state. The public seems to view both institutions as a source of authority; thus in a sense, there is a constant possibility that one institution might become far more powerful than the other. Table 3 summarizes the incentives that serve as a foundation for exchanges and interactions between the state and customary institutions. Table 3. Incentives that motivate both the state and customary institutions in Ghana Source: Author. Incentive State Customary Resources Looking for political or monetary resources Looking for positional, pardon, or monetary resources Political Reward Motivated by reelection and votes Motivated by higher rank in chieftaincy Moral Ethics Morally maintain elected leadership position Morally maintain traditional stewardship Mutual Threat Customary institution still has authority State institution has more authority 16 Patterns of Interaction After analyzing the findings, it became evident that particular patterns of interaction between the state and customary institutions were observed in every community. Resources, rules, and incentives that generate the four interactions are described below, using examples from the field as supporting evidence. Before discussing these interactions, it is important to briefly note that trends in the behaviors of the chiefs were also apparent. These positive roles include contributing resources, mobilizing the community, resolving disputes, and giving land. On the other hand, negative behaviors like gate disputes and bad land transfers led to stalled projects, reduced investment, and corruption. These behaviors and their consequences will be further visible in the interactions between the state and customary actors. Collaboration Defining Collaboration The first pattern of interaction observed between state and customary institutions is collaboration. Collaboration occurs under specific circumstances; however, it seems like the most difficult interaction to explain. Collaboration is viewed as an important part of district development; however, interviewees could not provide practical examples of it. The DCEs and civil servants in both regions verbalized the necessity of the chiefs; one stated: “Traditional authorities are major stakeholders. We want to consult them—no matter what we do, we must consult them” (B.12 2010). Another explained that “politicians need to keep a good rapport with chiefs and build a relationship with them” (C.12 2010); when DCEs enter communities, they meet with the chiefs to inform them of their purpose (A.1 2010). Most chiefs feel that they have a good relationship with the district assembly. They often describe their relationship with state actors as “cordial.” In one case, a clan head felt that the he and his assembly member are like father and son (A.9.1 2010). All in all, the relationships between the government officials and the traditional authorities are quite amiable and informal. Yet from the interviewees’ perspectives, the chiefs only collaborate in community development projects if the state requests their assistance. In this regard, it appears that collaboration is largely viewed as “contribution” in Ghana. This means that when a chief gives land or revenues, he is perceived as collaborating (A.9.1 2010; B.2.1 2010). As aforementioned, we distinguish between these two actions because chiefs’ contributions can have both positive and negative outcomes simultaneously. However, in collaboration, outcomes are always positive. Development activities seem to increase as subchiefs become more actively involved in each of their communities,12 Examples of Collaboration while information sharing results in more efficient projects seen in these collaborative cases. Thus, collaboration in our view is a steady process of interaction—where both the state actor and traditional authority use time and resources toward a joint development purpose. In the first collaboration example provided below, all interviewees reported that the subchief, district assembly, and unit committee members formally meet every three weeks, demonstrating a high-level partnership unseen in the other cases. This would be an example of collaboration. The opposite case is more typical, where meetings are not frequent or standardized (C.1 2010). Two field cases illustrate collaboration. The first one describes the processes of achieving electrification in the community. Beginning in 2002, the district assembly member and community requested their Member of Parliament (MP) and a mining company providing development assistance to provide the necessary equipment for electricity. After the equipment was received, the Ministry of Energy hired a contractor to erect the poles, and five years later the infrastructure was complete (C.2.1 2010; C.3.1 2010). However, after only three months of electricity, the Electricity Company of Ghana (ECG) pulled the plug. Evidently, the community was missing one small but very important piece of the infrastructure—the meters. Without the ability to measure the electricity used, ECG refused to provide the 12 Subchief involvement was common throughout the district. 17 energy. Thus the district assembly member and the unit committee pressured the ministry and the district assembly to finance the missing meters, but after three years, the situation had not improved. Resources and energy exhausted, and the district assembly member went to share her woes with the community’s subchief. Following their discussion, the subchief provided fuel money and other needed equipment to the district assembly member to assist in achieving electrification. He and his elders also sent letters and made personal visits to the regional ministry (C.5.1 2010; C.2.1 2010). As a result, all of the interviewees expect the meters to come to the community this year. This was not the only donor and advocacy work the subchief had taken part in since the district assembly member contacted him. He also purchased two computers for the school after overhearing that the government encouraged all communities to have them available to students. More importantly, he levied the community for donations to purchase two more: After publicly contributing five cedi (US$3.50), he asked the parents whose children attend the school to contribute the same. Collaboration was similarly apparent in other communities within the same stool and district. In another case, the subchief also mobilized his community to donate five cedi toward roofing materials (D.2.2 2010). This was after the district assembly provided funding for only half of the needed equipment. Before beginning the interviews in this particular community, we met with both the chiefs and the district assembly member in the same room. When we asked to interview the leaders separately, both stated that separation was unnecessary because they always work together and share all information. Explaining Collaboration through Rules and Resources The only existing state requirement on collaboration between the state and the customary institutions in the district is found in the LGA. As aforementioned, district assembly appointees are supposed to be selected in the consultation of the chiefs, which implies collaborative efforts between the two bodies. However, this process rarely happens. In a Western district, one of the assembly appointees was the queen mother. The chiefs were not consulted in her selection (as the law requires); rather she received a letter from the government requesting her participation (B.6.1 2010). When asked whether she would give information to the district assembly about the Traditional Council’s activities, she said, “No—information will only be passed the other way around” (B.7.1 2010). This seems like a common mind-set for both the assembly and traditional authorities. Collaboration implies information sharing; thus it is actually avoided in most circumstances. This makes the functional role of the traditional appointee on the district assembly questionable. It also exemplifies the power struggle, or mutual threat, between the two institutions. Based on the above and other findings, consistent and genuine collaboration between the state and customary institutions seems difficult to achieve. Constitutional state formal rules also further deter collaboration. While not one interviewee felt the role of the chief in development should be ignored, district executives and assembly members were quick to say that they are not to involve themselves in the affairs of the chiefs. Moreover, some laws that might induce collaboration are irrelevant in today’s government structure. Although Ghana’s Constitution requires that the Lands Commission consult the chiefs on the development of their stool, this is no longer the central government’s responsibility. Since decentralization, the district assemblies do the majority of stool development. There are additional explanations for why the state and chiefs choose to collaborate. When chiefs have few resources, they seem more likely to work together with their district assembly member if she has few resources as well. In the electrification example described above, the subchief said that while he collects a “small token” from the Traditional Council, he does not receive any distribution from the divisional’s “top-secret” OASL royalties (C.5.1 2010). It was not entirely evident, then, how he receives the money to contribute to projects, although he did add that his major source of his income is farming. While chiefs with higher resources do contribute to development, no cases were found in which they collaborate with the district assembly. This will be illustrated in other patterns of interaction. A final explanation for collaboration comes from the chieftaincy. Where collaboration occurs, it appears that strong customary formal rules that encourage good stewardship and oversight are also present. In the electrification example, the divisional chief had high expectations of his subchiefs. The 18 subchief explained that the divisional chief expects him to donate to projects like these if financially capable (C.5.1 2010). In other cases where the divisional chief did not expect certain behavior or interact with his subchiefs, contribution and collaboration were limited. Noninteraction Defining Noninteraction The opposite of collaboration, noninteraction occurs when the state and customary actors, although living and working in the same community, do not involve themselves with the other. It is the most common interaction between two institutions at the local level. Noninteraction is not a problem by nature; however, it is a problem when the development of a district is directly and negatively affected. Illustrations of noninteraction and its consequences are presented below. Examples of Noninteraction The first example of noninteraction is also a conflict within the chieftaincy; however, it is discussed here because of its implications for the district and the assembly member. In the particular community in this example, there are 12 clans with 12 clan heads, who rotate as the subchief. The purpose of rotating the subchief title among clan heads is to maintain some overarching order in the community. However, the subchief does not have more authority over land compared with the other clan heads. This case involves a telephone company, which constructed two network towers over the course of three years within the community. The towers fell within the land boundaries of one of the clan heads (A.9.1 2010). The prospectors who came to lease land for the first pole went directly to the subchief, as companies typically do. When the subchief realized that these prospectors were not surveying his land, he went to appropriate clan head to alert him of their request (A.2.1 2010). After confirming the site for construction, the clan head never heard from the telephone company again. Additional network prospectors approached the clan head’s predecessor some years later. In this case, the clan head formalized and signed a compensation contract (or drink money agreement) with the company, a step that he knew did not transpire with the original pole. Thus the clan head realized that he should have been compensated for the land the first time (A.9.1 2010; A.2.1 2010). Suspecting that the subchief unfairly appropriated the compensation, the clan head went to the police and tried to bring him to the state court (A.3.1 2010; A.5.1 2010; A.9.1 2010). The subchief was not arrested for the accusation, and eventually the case was given to the King of Gonja because it was considered a customary dispute. At the time of interviewing, sanctions had not ensued. The most important aspect of this case is the lack of state involvement. When the police were first called to the situation, the district assembly member was also, but only as an observer (A.2.1 2010). When asked about land acquisition and the conflict concerning this case, an executive officer explained that foreign companies must come to the district offices for a building permit, but neither he nor the other district officers are involved in land transfers or drink money agreements (A.1 2010). Not only does this leave the district without any control of the land, but it also restricts community development. This conflict was a serious one, dividing the community and damaging clan relationships. Moreover, because of the conflict, the district assembly member feels that his attempts to bring development to his community are somewhat ineffective. With clans in dispute, consensus on project sites and building processes is always difficult to achieve. This is further challenging because clans organize around political parties. Due to these difficulties, the district assembly often chooses not to interact with the clan heads unless he needs to request land. This kind of criminal behavior within the chieftaincy frequently spreads into private and state affairs as well. A district assembly member in another case elaborated on the difficulties of ensuring transparency between the chiefs and the companies (D.2.2 2010). Recently, he and others had discovered a document proving that a mining company had been giving large sums of money to an unknown person (assumed to be a chief) for years. This money, according to the agreement, was supposed to be used for “development purposes” (D.2.2 2010). Yet no one in the district government had seen these funds or the 19 projects from them. Because state actors do not interact with the customary institution when land or drink money is transferred, there is little oversight over these processes. Another example of noninteraction is quite different. A divisional chief who receives gold mining royalties in the Western region contributes significant funds to development projects in the district. However, he did not channel these funds through the district assembly; rather, he constructed his own buildings and led his own projects. Given the extensive funds that the divisional chief poured into development projects in this stool territory, it became essential to ask where exactly this money was being used. All interviewees stated that these funds remain within the divisional chief’s community, implying that the 20–30 communities within the stool land receive little if any assistance from his benevolent deeds (B.2.1 2010). When asked whether subchiefs or stool residents could request and receive assistance from the divisional chief, most respondents said “sometimes”; however, there were no concrete examples of this assistance. Within the divisional chief’s community, these development funds were put towards development projects like community center construction, furniture for classrooms, teachers’ salaries (B.3.1 2010). Scholarships, like the development projects, are also exclusive. They are only given to those students who are doing well in school and whose mother is from the Wassa area (B.7.1 2010). Like in the previous example, the customary actor does not interact with the district assembly member (B.2.1 2010). The district assembly member, in clear agitation, explained that he is “completely uninformed” and never consulted about how the chief spends his royalties for development projects. According to him, the divisional chief purposely circumvents him and gives the money to the Unit Committee members. Because of this exclusion, he no longer lives within his own electoral area. When asked why he thought the divisional chief chose to sidestep him from this process, he replied, “Because the Unit Committee members are the royals” (B.2.1 2010). Thus, while the divisional chief is collaborating with these state actors, they are likely more allegiant to the customary institution because of their royalty. Although the divisional chief’s community benefits from his royalties, inequality in the district increases. This is not an immediate problem, but because the district assembly is not yet capable to perform a proper needs assessment, some communities like this one receive far more assistance than others. As stated by both state and customary actors, this will eventually result in conflict between communities. Moreover, the district assembly is often unaware of the divisional chief’s activities. Lack of interaction allows for overlap in projects, resulting in wasted resources. However, in more positive light, this noninteraction does produce the “leadership effect”—in which the traditional and local state leaders positively reinforce each others’ development efforts (Logan 2009). Despite his frustrations, the district assembly member in same community as the wealthy divisional chief personally visits nearby gold mining companies and requests assistance. In the past two years, this has included 7.5 miles of road construction, 36 computers, teachers’ quarters, and a clinic (B.2.1 2010). Explaining Noninteraction through Rules and Resources First of all, the findings demonstrate that most noninteraction occurs when resources are asymmetric. In the first example, the resources for the district assembly member are assumed to be higher than those of the clan heads. Because they are not part of the main tradition in the area, the clan heads do not receive any OASL royalties from the traditional council or divisional chiefs. Moreover, whereas most communities have only one subchief who oversees the land and collects drink money, this community has 12 clan heads, which limits monetary resources for each head. The district assembly member also seems to carry a higher status in the community as well; this is likely due to his attempts to unify the community for development purposes. In the second example, the divisional chief has many resources. As a result of Ghana’s damp and tropical climate, his resources stretch far beyond gold, which also results in foreign investment and increased drink money for the chief. Although the chiefs rarely discuss their royalty figures and are not legally required to publicize how they are used, this particular divisional chief did. Four interviewees confirmed that he receives approximately 120,000 cedi (or US$83,000) per year (B.3.1 20 2010; B.6.1 2010).13 State rules seem fairly ineffective in reducing the negative effects and prevalence of noninteraction. Ghana’s Constitution encourages noninterference. Thus, even if a conflict between the chiefs leads to community conflict and stalled projects, the state actor will not intervene. These consequences are even more serious when the noninteraction breeds corruption between companies and chiefs. Another challenge stemming from state formal rules is the restricted ability of the district assembly to become involved in the development activities of the chiefs. While one could argue that the chiefs are the owners of the land and their royalties—thus they should not be required to share information—if their activities meddle with those of the local government, it is logical that sharing information would enhance efficiency. Chiefs can hire their own contractors and lead their own projects. Clan heads can authorize a telephone pole. With the consent of the traditional authority, factories are opened and subsequently abandoned, all without the district assembly’s involvement. To put this figure into perspective, the Ghanaian gross national income per capita is a little under US$700 (United Nations 2008). Clearly, his financial resources are much higher than those of the unpaid district assembly member. Customary formal rules are also ineffective in noninteraction cases. Divisional chiefs are not required to distribute their royalties, which limits the subchiefs’ ability to contribute and keeps resources in a single community (B.7.1 2010; B.6.1 2010). Moreover, customary rules do not always prevent and deal with conflict effectively, as seen in the first example. Informal rules also affect noninteraction. In the first case, the clans were politicized and aligned with certain political parties. This made interaction with the district assembly difficult (A.1 2010; A.5.1 2010). Clearly, informal property rights that leave decisionmaking to the chiefs reinforce noninteraction if chiefs have incentives to keep drink money figures secret. Conflict within the Chieftaincy Defining Conflict within the Chieftaincy Conflict within the chieftaincy comes in many forms. Gate disputes, drink money, and boundary disagreements lead to conflict within the customary institution. These conflicts are often resolved through the traditional judicial system. In both regions, interviewees narrated these conflicts and their consequences. However, we do not need to present specific examples of this interaction because conflicts between state and customary actors are narrated in other parts of the paper. Yet the consequences and causes of conflict within the chieftaincy do necessitate discussion. There are a number of serious consequences to chiefly conflict. As already illustrated, in some instances, chieftaincy disputes cause public servants to request transfers and investors to leave (D.2.2 2010). In a Northern district, a chief’s participation in mobilizing community labor seemed positive until a conflict arose between him and another chief. Out of fear of being identified as supporting one chief, community members refused to provide communal labor for the market project (D.2.2 2010). Conflicts also distract the youth. In the north, more than 40 youth rallied to defend the chief that they supported. Not only did this put the communities on the verge of riot, but it also resulted in reduced farming activity (A.2.1 2010). Perhaps more seriously, some of these conflicts lead to opportunities for state and customary collusion, as it raises the traditional authorities’ incentives to partner with politicians under the table. Explaining Conflict within the Chieftaincy through Rules and Resources Customary formal rules clearly play a role in creating and sustaining these conflicts. For one, some traditions make conflict easier than others. For example, in the Northern region, the Gonja tradition rotates gates for patrilineal succession, but that succession is debatable (A.1 2010). This means that if the king considers the candidate at the gate “incapable of doing his job,” he can nominate another royal (D.4.1 2010). In addition, all people in the royal family who want to become a chief must go through 13 Though the divisional chief was not present, his caretaker and the queen mother were interviewed. 21 years of service to achieve eligibility. Traditions observed in the Western region are very different. Not only is the tradition matrilineal, but chiefs are enstooled based on seniority. Thus, when a chief position opens, the eldest son in the appropriate gate takes the position—no questions asked (B.7.1 2010). This results in longer chieftaincy reigns in the west, which might reduce gate openings. Reduced gate openings mean less opportunity for conflict. More importantly, although debated succession uses democratic principles, findings suggest that candidate contestation in the north is not always public. Thus the paramount chief may not make the most equitable decisions when choosing a candidate. Purely economic motives also increase the severity of the conflicts. The increase in land valuation and foreign investment serve as strong incentives for intertraditional disputes. This becomes a serious problem when state oversight is negligible: District assembly members will only interfere in chief conflicts when there is a threat of violence (A.3.1 2010). Collusion Defining Collusion Collusion between the state and customary institutions occurs between many different actors. For the purposes of this study, collusion is considered any relationship or exchange between the traditional and state bodies that violates the formal state law and results in negative outcomes. It also frequently involves corruption. Two examples are discussed below. Examples of Collusion The first example involves police and the chief in collusion, yet the federal government was mentioned as a colluder as well. In the Western region, illegal mining is a common and problematic practice. Though interviewees called these activities “illegal mining,” it is actually the practices used to mine that are illegal. The Ghanaian mine operators who come to the region actually have licenses to mine; however, they frequently use cheap, illegal practices that are hazardous to both the environment and the community. Though these practices are criminal, the local police do little about the problem. According to the district assembly member, this is because the subchief, who receives mining revenues from the operators, pays the police to allow the activities to continue (C.2.1 2010). Thus the police pardon the chief and the mine operators in return for cash. The subchief in this case also explained that the miners pay hefty fees to the government for their licenses to operators. In his opinion, this makes the mining a win– win situation for both the central government and the chiefs (C.5.2 2010). In this case, the district assembly member had no control over the situation or its negative consequences. The second example illustrates colluding behavior between the federal government and a divisional chief. In a Northern district, the local police and district assembly member interfered in a quickly escalating chieftaincy dispute. For unknown reasons, the military arrived and removed only one of the chiefs from the area. When the district assembly member was asked why only this chief was removed, he replied, “Probably to protect him” (D.2.1 2010). Subsequently, this chief was enstooled as the divisional chief, although many interviewees felt that he did not maintain rights to the stool. It was later found that not only did this chief sympathize with the party in power, but he was also the uncle of the DCE (D.4.1 2010). Explaining Collusion through Rules and Resources Informal rules are essential to encouraging and allowing collusion between the two institutions. Partisanship motivates politicians to gain chiefs’ support, which means that state officials—both federal and local—might pardon traditional authorities in exchange for their political assistance. Other informal rules permit collusion and corruption as well. Because the consequences for criminal action within the chieftaincy are so severe (like automatic destoolment) (C.12 2010), both the public and district officials are wary of enforcing legal behavior. 22 Formal state rules regarding royalty transparency also cause collusion. The lack of bank statements and financial documentation for drink money promotes confusion and reduced clarity on land transfers (A.5.1 2010). Financial transfers within the chieftaincy or from the chieftaincy to the state are not public. Even the OASL, who will provide royalty figures to the public, is slow to do so. In addition, findings suggest that the subchiefs do not typically know the royalty figures of their division, and the district assembly members do not know them for their district. Where this is the case, opportunities for collusion and illegal exchanges between actors of the two institutions are high. 23 6. CONCLUSIONS Forming Hypotheses The patterns of interaction described in Section 5.4 raise a number of important issues, the first of which serves as the building block for policy implications and conclusions. After toying with and reanalyzing the patterns of interaction and supporting examples, we found trends in the resources, rules, and incentives. It appears that certain combinations of these three variables can predict interactions between a state and a customary actor. Thus we hypothesize about the conditions under which the four interactions—collaboration, noninteraction, collusion, and conflict within the chieftaincy—occur. Informal rules are not included in these hypotheses because they did not differentiate significantly between patterns of interaction. This is not to suggest that they are irrelevant; rather, our case study did not gather sufficient information to discuss them in the hypotheses. Table 4 highlights the combinations of resources, incentives, and formal rules-in-use that produce certain interactions. Conflict within the chieftaincy is also included, as the findings suggest that it sometimes leads to collusion with the state. Table 4. Hypotheses on the four patterns of interaction between the state and customary institutions Noninteraction Collaboration Conflict w/in Chieftaincy Collusion Resources Many, asymmetric Few, symmetric Not applicable Many, symmetric Incentives Mutual threat Retain resources to keep power Exchange in hopes of more power Personal attempt to avoid threat Exchange in hopes of more power Moral ethics Maintain status quo Seen as a contributor Seen as a leader Can blame other Political reward Maintain upper hand Gain reelection, higher rank, or respect Gain chief position or land Gain more votes, higher rank, or pardon Resource gain Keep them Complemented Keep them Complemented Rules-in-use State formal Constitution upheld LGA upheld Not applicable Constitution/LGA violated Customary formal Not applicable Rules are stronger than normal Rules are violated Rules are violated Source: Author. The following paragraphs elaborate on these hypotheses. Noninteraction occurs when resources are asymmetric and where state formal rules are effective at discouraging interference. In these cases, resources are concentrated in the hand of one institutional actor, thus providing the resource-rich actor with the incentives and power to avoid engaging the opposing institution. While Ghana’s Constitution actively discourages interaction, the customary formal rules say little about it. Informal rules that reinforce nontransparency also discourage interaction between the two bodies. Collaboration occurs when resources are few and symmetric and when customary formal rules are strong. Resources must also be complementary. If resources are asymmetric, this interaction would likely become noninteraction. Yet it must also be noted that in most cases where collaboration was observed, the customary formal rules related to community stewardship were high. Subchiefs were also more accountable to divisionals. State formal rules are largely inapplicable to this pattern because they do not make any specific demands on either institution to collaborate with the other. Because the LGA encourages the district assembly to collaborate with other actors (generally) in the district, it could be said that the LGA is upheld. In the electrification case, the chief gave financial resources to the district assembly member in exchange for an increase in positional status in the community. 24 Conflict within the chieftaincy occurs when customary formal rules are weak. Resources only serve as an incentive for conflict. While customary rules are weak in this interaction, state rules do not play a role. It could be proposed that in dangerous or violent conflict, both the customary and state institutions lack the ability to enforce formal rules. This interaction is important not only because it results in negative development outcomes but also because it leads to collusion in some cases. If chiefs are in conflict, a state actor can interfere if he has complementary resources and incentives to do so. Collusion occurs when resources are asymmetric and high and when state and customary formal rules are easily violated. Resources, just like in collaboration, must also be complementary. Though incentives like moral ethics still influence the actors in this interaction, it is unlikely that rule violations will be exposed. The strength of informal rules also reduces chances of serious exposure. Partisanship and chiefly respect reinforce public acceptance of corrupt behavior. In the illegal mining example, the subchief gave financial resources to the local police in exchange for pardon. What the Patterns of Interaction Mean for Governance with Dual Institutions The hypotheses above are included in this paper to promote future research on these topics. We are not claiming that they hold true for all situations, as the interactions and exchanges between state and customary actors are highly circumstantial. Other variables not captured in this study likely influence them as well. Nonetheless, we can use these hypotheses to discuss policy implications. Bringing the discussion back to local government and development outcomes, the question remains: What interaction between the two institutions leads to the best local development outcome? Based on the findings, the answer is collaboration. In some cases, noninteraction is also effective in bringing development because it increases positive competitive behavior in the two institutions. Thus the follow-up question is, Are there ways in which policymakers can increase collaboration, or at the least, mitigate negative consequences in situations with noninteraction, conflict, or collusion? By reexamining the resources available, rules-in-use, and incentives, one can discuss how changes in these variables might shift results in patterns of interactions and outcomes. Formal and informal rules certainly play a role in generating these incentives as well as accountability within the institutions. However, informal rules have deep roots in society and are not easily changed; neither are customary formal rules, due to their cultural implications. Some rules on resources could be revised, but the resources themselves are more difficult to change through policy. Indeed, of the three variables, state formal rules-in-use are undoubtedly the quickest to modify, though this is not to say changing them is easy. The following four topics are the most prominent concerns raised by the state–customary interactions and exchanges found in this study. Considering the hypotheses, revising some of the state formal rules might lead to improved local outcomes. Transparency Lack of transparency in land transfers, drink money, projects, party politics, and even in exchanges themselves certainly leads to negative patterns of interaction. Transparency, even in cases with noninteraction, led to better development outcomes. Public disclosure helps to hold both the government and customary institutions accountable for their action and interactions or exchanges with one another. More transparent rules would have a number of positive consequences. They would reduce transaction costs and uncertainty in institutional interactions. They would also improve incentive structures, as actors could not maintain moral ethics or achieve political reward if publicly participating in activities like collusion or corruption. Various formal rules could be modified to encourage transparency. Making royalties more publicly available or fixing them at a certain rate might reduce the resource effects in the interactions. At the least, creating a code of ethics for using royalties in both institutions could lower the incentives for exchanging resources for culpable purposes. Posting who contributed to what project in communities might increase the leadership effect as well. 25 Consultative Processes Though some formal rules require the state actors to consult the chiefs on certain matters, these rules are often either unenforced or irrelevant. Moreover, it seems that there are few rules requiring that the chiefs consult the state on their activities. This leads to the inefficient use of resources. It also results in noninteraction. If one institutional actor has more resources than the other, they are incentivized to keep their resources. However, if required to consult one another—even despite differences in resources—the poor effects of noninteraction like project overlap might be reduced. More frequent meetings between the two institutions at a local level could increase information sharing. This might also increase the voice of subchiefs, which would enhance their ability to contest decisions of divisional and paramount chiefs. Formal rules could also require that chiefs’ contributions or development projects be included in the district assembly budget. This might help reduce inefficiency. Management and Roles The ability of the state to manage both the customary and its own institution is limited. While the customary institution manages some of its own affairs, because the NHC, RHC, and Traditional Councils are state bodies, some of these management responsibilities lie with the state. Difficulties managing the customary institution might be due to the ambiguous formal rules that are purposed to guide state actors. Though the LGA encourages collaboration with various actors, it says little about how the district assembly member should interact with the chief. This causes uncertainty in local state officials when problems arise with the traditional authorities in their jurisdiction. Uncertainty then makes incentives like mutual threat more apparent as well. Defining the role of the chief in local government policy might help to improve management processes. A clear cutting of the traditional authorities’ role could also reduce internal threat between the two institutions by mitigating competition. The local government could also reward chiefs who collaborate or contribute their funds or energy, which may enhance their incentives to participate. The state might also reconsider haphazard allowances; standardizing (while publicizing) allowances and payments to chiefs or councils could perhaps diminish opportunities for collusion and chiefly conflict. Enforcement and Sanctions Enforcement of formal rules is one of the main challenges in achieving positive patterns of interaction. As demonstrated, informal rules play a major role in allowing nonenforcement. Partisanship and rules that demand devoted chiefly respect deter the state to enforce its formal rules. Thus, there are many opportunities for criminal behavior, collusion, and corruption in the dual governance structure. As stated previously, monitoring and policing every interaction between the two institutions would be impossible. Nonetheless, increasing some enforcement of the formal rules is the only way to curtail shady exchanges. Sanctioning crime and corruption would disincentivize collusion. Clearly, enforcing sanctions fo