Democratic governance and its effects on Rwandan service delivery

(1)

Master Thesis Political Science: Political Economy

Research Project: The African Renaissance and the Politics of Development  

Democratic Governance and its Effects on Rwandan Service

Delivery 

 

Victor Sebastian Fickinger (11274239)

Supervisor: Dr. Michael Onyebuchi Eze 

Second Reader: Dr. Silvia Aru 

 

21 June 2019

 

Word Count: 17,522

 

(2)

Abstract:

Across the board, the world of policy makers and development institutions have championed the so-called good governance approach. Within this approach, the concept of democratic governance has gained in traction. It is believed that approaches pertaining to democratic governance are most conducive to developmental policies that tackle poverty, and ineffective service delivery. Since the 1994 genocide, Rwanda has made it a developmental priority to adopt many of these principles. Yet, how do they work in the context of Rwandan political economy? How do the dynamics of Rwandas political, economic and social life affect the application of democratic governance? How will this affect Rwandan development in the long-run? These are just a few of the implicit questions addressed in the course of the thesis. This research examines three key tenants of democratic governance, namely democratisation, decentralisation and social accountability and attempts to delineate how and why they have affected Rwandan service delivery. An in-depth analysis of Rwandan development policy, post genocide, will be undertaken to illuminate the relationship between Rwandan governance and effective service delivery.  

(4)

Abbreviations: 

Bundesministerium für Wirtschaftliche Zusammenarbeit und Entwicklung (BMZ) 

Citizen Report Card (CRC) 

Development Economics Vice-Presidency (DEC) 

Democratic Governance (DG) 

Effective Governance (EG) 

European Union (EU) 

Foreign Direct Investment (FDI) 

Gross Domestic Product (GDP) 

Governance Indicators (GI) 

Government of Rwanda (GoR) 

High-Income Country (HIC) 

Home-Grown Solution (HGI)

International Monetary Fund (IMF) 

Institute for Policy Analysis and Research (IPAR-Rwanda) 

Limited Access Order (LAO) 

Land Locked Developing Country (LLDC) 

Ministry of Finance and Economic Planning (MINECOFIN) 

Ministry of Local Government (MINALOC) 

National Institute for Statistics (NISR) 

National Decentralisation Plan (NDP) 

Open Access Order (OAO) 

Tri-Star Investments (CVL) 

United Nations (UN) 

United Nations Development Programme (UNDP) 

United Nations Economic and Social Commissions for Asia and the Pacific (UN ESCAP) United Nations Human Rights Office of the High Commissioner (OHCHR) 

World Bank Developmental Report (WDR 2004) 

(5)

1. Introduction

In the 30 years, the debate on development aid practices has shifted considerably. In response to the realisation that ‘bad’ governance was a key symptom of underdevelopment or stagnant growth, key developmental institutions shifted their focus in the late 1980’s (Carothers and Brechenmacher, 2014). The ensuing agenda, namely the ‘good governance’ approach was popularised throughout the 1990’s as a result of prominent advocacy by institutions such as the World Bank, the UN and recently, the EU (Kaufmann et al. 1999; Price, 2018; Hackenesh, 2016). Across the board, the approach has gained traction and been applied with a varying degree in a multitude of contexts. Broadly speaking, ‘good governance’ refers to a developmental approach that relies on normative underpinnings such as ‘inclusivity, accountability, and the rule of law’ (Price, 2018; UN ESCAP, 2009; OHCHR, 2000). However, the success of this approach is contested and heavily debated, especially within the context of a subcategory of ‘good governance’, namely the ‘democratic governance’ (DG from hereon) approach (Burnell, 2000; Santiso, 2003, as cited in Hackenesch, 2016, p. 12). As the literature review and theoretical framework will show, there are three main approaches that have become mantras of ‘democratic governance’. These are democratisation, decentralisation and social accountability. There is a growing uncertainty concerning the general applicability and effectiveness of such democratic governance practices when taking into account the relative success of regimes such as Rwanda. It is no secret that Rwanda has experienced some of the most remarkable development patterns in the last two decades; its rise to one of the most prosperous African nations is well noted. This is especially the case in regards to the countries ability to reduce poverty, as well as pursue effective public goods and services delivery (UNDP, 2019). Therefore, the Rwandan case provides valuable insights into the larger discussion on the relationship between political regimes, social and economic development. To specify, this also extends to the discussion on the relationship between governance regimes and the provision on public goods and services (Hackenesch, 2016). 

This research attempts to delineate the broader debate on development and in doing so, illuminates the relationships between democratic forms of governance and efficient public goods and services provision. As a hybrid form of 'developmental state’, the unique Rwandan case provides us with a critical insight to the discussion, as it seems to challenge the orthodox approach to a considerable degree (Hackenesch, 2016; Booth and Cammack, 2012). An in-depth analysis of the Rwandan success story in regards to service delivery and development will provide us with a valued critique of the ‘democratic governance’ approach. Due to the fact that approaches pertaining

(6)

to democratic governance have been applied in Rwandan development policy and service delivery, this thesis aims to answer the following broader research question: ‘how and why do democratic governance approaches affect Rwandan service delivery?’ In order to answer the main question, the thesis attempts to answer the following supporting research questions: 

- How and why have democratisation approaches affected Rwandan service delivery? - How and why have decentralisation approaches affected Rwandan service delivery?  - How and why have social accountability approaches affected Rwandan service delivery? 

Although there has been a steadily growing body of work around the developmental aid debate, as introduced above, the literature on ‘developmental states’ and governance dimensions to service delivery is relatively new. As such, an in-depth look at the Rwandan case will be undertaken, in order to outline the dynamics between the governance regime and service delivery, as well as development more broadly.  

This thesis will be structured in the following manner; to begin with, the literature review is organised in a two-fold manner. In the first step, the research will outline the main contesting arguments surrounding the aid debate. This debate comprises of three contending schools of thought. They are the ‘optimist’, ‘pessimist’ and ‘multi-path philosophy’ (Carothers and Brechenmacher, 2014). The aid debate requires attention because it summarises the broader understanding of how various forms of governance affect development. Key concepts such as ‘good governance’, ‘democratic governance’ as well as ‘developmental state' are introduced here. Moreover, this discussion is necessary because it lays the foundations to later delve into the more specific case of governance dimensions of effective service provision. The second step will discuss the literature on governance dimensions to service delivery as well as outlining essential concepts and theories such as, ‘social accountability’, ‘rent-seeking’ and ‘decentralisation’, amongst others. This allows us to fully contextualise the ensuing case study on Rwanda where these concepts will be assessed. These two chapters give us a full overview of the contending theories, analytical concepts and most importantly, types of approaches towards effective public service delivery and development.  

Following this, the theoretical framework will be presented, in order to situate the critical discussion on the democratic governance approaches for Rwanda. Through the insights from North, Wallis and Weingast (2009), the logics of ‘natural states’, ‘mature natural States’ and ‘open access orders’ will provide the basis of the theoretical framework. The theoretical framework also includes

(7)

a further conceptualisation of decentralisation, social accountability and democracy as forms of governance pertaining to service delivery. Subsequently, the methodological considerations for the research will be given. This will include an elaborate operationalisation of the concepts at hand. Furthermore, within the methodological chapter, the case selection is presented. The case of Rwanda as a hybrid form of ‘developmental state’ is elaborated on in this section, in order to showcase the motivation for research, as well as highlight its importance. For the case study itself, the first section will illuminate the historical and contemporary context of Rwandan development and its service delivery in the post-genocide era. Due to the fact that Rwandan development policy incorporates aspects of the democratic governance approaches in an overlapping manner, the following considerations have to be made; the individual policies pertaining to Rwandan development, namely the ‘Vision 2020’, the ‘National Decentralisation Plan’ (NDP), ‘Imihigo’, ‘Umuganda’, and ‘Girinka’ will be assessed.  

Following the analysis, an evaluation and presentation of findings will ensue. Here, I will argue that Rwanda as a ‘developmental state’ has appropriated forms of the democratic governance approach to a relatively successful degree. Nevertheless, it will be argued that premature adopted approaches pertaining to democratic governance do not necessarily affect Rwandan service delivery positively in the long run. Ultimately, the addressing of supporting research questions will pave the way for a nuanced critique of the democratic governance (DG, from hereon) approach. The conclusion and summary of arguments will finalise the thesis, as well as comment on the limitations of the research.  

This research is of imperative importance for several reasons. Understanding the true dynamics of governance within Rwanda can illuminate stark gaps within the widely applied democratic governance approach (a sub-type of ‘good governance’). Not only will this research allow for the critique of the broader ‘good governance’ approach, but also outline takeaways for Rwandan service delivery and the challenges that might lay ahead. Secondly, this research will provide further empirical evidence for the growing amount of literature which aims to highlight the fact that the international aid community must be very careful in promoting its ideologically-run approaches; not only can premature implementation of ‘western-style’ approaches have adverse effects, they can trap developing countries within a cycle of perpetual underdevelopment (North, Wallis and Weingast, 2009; Khan, 2012a; Booth and Cammack, 2013). This can result in the unintended ‘removal’ of essential institutional and societal prerequisites for sustainable development. This is aptly described by Chang (2002) as ‘kicking away the ladder’. It can also have an affect in disturbing the balance between sustainable governance and development, resulting in

(8)

conflict spillover (Collier, 2007; 2010). All in all, the Rwandan case will give new impetus to the questions surrounding how development practices and policies could be shaped for the future. Lastly, this case study allows us to understand how Rwanda could pose as a model for ‘doing development differently' (Booth and Chambers, 2012).

2. Literature Review: The Three Schools of Thought

This section outlines three schools of thought (hereon referred to as ‘camps’) in order to situate the ensuing argumentation on effective public goods provision within the broader developmental aid debate. The following three camps showcase what has been said about the relationship between governance and development. The camps comprise of the ‘optimist’, ‘pessimist’ and ‘multiple path philosophy’. (Carothers and Brechenmacher, 2014) Part of the ‘optimist’ camp introduces the ‘good governance’ approach, as the concept falls within this category. Following the classification of these camps, the relationship between governance and public service provision will ensue, in order to contextualise the case of Rwandan service provision.  

The debate on the complex relationship between forms of governance, institutional set-up, economic and social development is controversial to say the least. One school of thought, namely the ‘optimist’ camp, is by far the most conspicuous. Undoubtedly, this camp is the most popularised, and therefore most conventionally applied throughout the world of developmental policy-making. The ‘optimist’ camp can broadly be defined by the understanding that there is a positive correlation between governance and economic progress. To elaborate, this understanding argues that inclusive institutions and democratic governance are both imperative and conducive towards achieving high levels of socioeconomic development. This form of thinking originates from the post-Cold War era of policy making in the 1990’s, and has continuously gained in popularity. (Carothers and Brechenmacher, 2014) While it is stated that these form of policies were introduced on ‘axiomatic’ terms rather than based on empirical findings, considerable empirical research has emerged to substantiate the broader claim. Most notably, economists from the World Bank such as Aart Kraay, Daniel Kaufmann and Pablo Lobaton (Kaufmann, Kraay and Lóbaton, 1999) have offered very plausible claims which support the notion that inclusive forms of governance influences development positively. In Kaufmann and Kraay’s landmark study, where they measured governance quality changes in relation to the development, they made use of certain Governance Indicators (GI’s). These were related to concepts such as: ‘voice and accountability,

(9)

graft’ (Kaufmann et al. 1999, as cited in Carothers and Brechenmacher, 2014, p. 18). As the ensuing

conceptualisation will show, these concepts clearly link to various underpinnings of the ‘good governance’ approach. In addition, scholars such as Conzelmann (2003); Faust (2004); and Andrews (2013) support and identify principles such as ‘efficiency’, ‘predictability’, ‘accountability’, ‘levels of corruption’, in their understanding of ‘good governance’ (Conzelmann, 2003; Faust, 2004; Andrews, 2013; as cited in Hackenesch, 2016, p. 12). 

‘Good governance’ itself is an essentially contested concept (Kohler-Koch and Rittberger, 2006, as cited in Hackenesch, 2016, p. 12). It has been rigorously debated and been applied differently across different contexts. Due to the fact that the concept does not enjoy universal acceptance, its application is relatively flexible. This can provide problems in terms of its operational application (OHCHR, 2000). The World Bank for example, works with the “Good Governance Practice” which entails the approach of building ‘capable, efficient, open, inclusive,

and accountable institutions’ (Price, 2018). According to the UN ESCAP (United Nations 1

Economic and Social Commissions for Asia and the Pacific), ‘good governance’ is an ‘ideal’ that encompasses for 8 key principles listed below. (UN ESCAP, 2009)

Figure 1: Good Governance Characteristics (Adopted from: UN ESCAP, 2009, p. 3) 

In order to steer clear of overlapping argumentation and provide clarity for what type of ‘good governance’ is most conducive to development, distinctions between the notions of ‘effective governance’ and ‘democratic governance’ (EG and DG, respectively) must be made. For the European Union’s Directorate General for External Relations (2016), these distinctions are made

(10)

use of (Hackenesch, 2016). Although they are applied differently across the spectrum of academia and the developmental aid community, the EU conceptualises them in a two-fold manner; the ‘broad’ and ‘narrow’ definition of ‘good governance’ (Börzel et al. 2008; Leftwich 1994; Robinson and Hout 2009, as cited in Hackenesch, 2016, p.12). The broad understanding that is ‘DG’, stems from scholars and ‘practitioners of democracy promotion’, who emphasise the normative dimensions dealing with the political aspects and components of democratic governance and human rights (Burnell, 2000; Santiso, 2003, as cited in Hackenesch, 2016, p. 12). To elaborate, concepts such as ‘horizontal and vertical’ accountability, the separation of powers, political rights, civil rights, and electoral regimes fall under this category. The 'narrow’ understanding, ‘EG’, stems from the developmental aid community and its policy-related world, largely influenced by the aforementioned scholars and most notably from the World Bank itself (Kaufmann, Kraay and Lobatón, 1999). Focal points to this understanding are the ‘efficient and effective apparatus and functioning of the state’ (Hackenesch, 2016, p. 12). Understanding the broader developmental aid debate and its relation to the provision of public goods, necessitates a clear understanding of the popularised ‘good governance’ approach. For the course of this paper, when speaking of ‘good governance’, both DG and EG are meant. 

To extend the ‘optimist’ camp’s argumentation, let us take a short look at the relationship between a certain form of governance (democracy) and its role in development. Without confusing development with trade, or development with globalisation, there is a clear correlation between democratic governance and development. It has been stated that democracies trade more between one another, and are more attractive to Foreign Direct Investment (FDI) (Mansfield et al. 2000; Moon, 2015, as cited in Hackenesch, 2016, p. 13). Another strong line of arguments on the relationship between democracy and globalisation are put forth by Barry Eichengreen and David Leblang (2006). Their findings suggest that a positive relationship between democracy and globalisation exists (Eichengreen and Leblang, 2006, p. 21). In fact, they argue that the longer or more ‘mature’ a democracy is, the more likely it will be open to trade. (Eichengreen and Leblang, 2006, p. 19) Across the board, these arguments seem to be substantiated. In fact, the number of democratic countries that are not performing well economically is relatively small. Nonetheless, it is crucial to respect the arguments’ temporal considerations; these arguments usually refer to later

stages of development. A similar argument is made by the conceptualisation of the 'open access

orders’ (OAO’s) and ‘natural states’ (North, Wallis and Weingast, 2009). As parts of the ensuing discussion for the ‘pessimist’ camp will show, various issues can arise if certain countries prematurely adopt policies that subject them to financial and global trade openness. This also paves

(11)

the framework to understanding the path of Rwanda’s development in the last 25 years.  

The pessimist camp is broader in nature and can be defined in two separate ‘groups’. The first group can be identified by its critique of the orthodox framework mentioned above, as well as its focus on structural factors. Essentially denouncing the importance of governance, this part of the camp stresses structural conditions such as geographical location, technological and cultural aspects. The three most commonly understood hypotheses (‘geography’, ‘lack of expertise’, and ‘culture’), as postulated by the scholars Daron Acemoglu and James A. Robinson (2012), will be outlined shortly in order to provide an overview for the first side of the pessimist camp.  

Most notably supported by Jeffrey Sachs (2003) and Jared Diamond (1997), the geographical hypothesis establishes a link between a countries or a regions geographical location and economic prosperity. For example, Jeffrey Sachs (2003) supports the notion that ‘hot countries’ have developmental disadvantages due to their proneness to tropical diseases such as Malaria, as well as possessing unproductive soil, due to their tropical climate (Sachs, 2003; Diamond, 1997, as cited in Acemoglu and Robinson, 2012, p. 102). This thinking finds its roots in the controversial ‘environmental determinism’ literature, famously argued for by enlightenment-era philosophers such as Montesquieu and Machiavelli (Johnston, 2017). In relation to geography and global trade, a considerable amount of research points towards the structural disadvantage that (some) landlocked countries have. This is due to the fact that land locked developing countries (LLDC’s) face the 2

complex challenge of being disadvantaged in a twofold manner; firstly, they are disadvantaged by having to overcome the costly and inefficient access to international trade as they have no direct access to coastlines or deep-sea ports. Secondly, they are already considered ‘poor’ and therefore experience a doubled plight in overcoming their developmental challenges (Arvis et al., 2011, p. 1). While the geographical hypothesis certainly holds value in explaining various challenges to prosperity and poverty reduction, the argument does not hold across all contexts and certainly does not provide for a ‘general’ hypothesis. A glance at the contemporary example of Burundi and Rwanda’s divergent developmental paths show just how important institutional legacies are in

addition to the geographical context that both countries share (Biedermann, 2015, as cited in

One must take note of the fact that global trade itself is not flawless and does not necessitate sustainable development,

2

across the board. For example, André Gunder Frank’s ‘Dependency Theory’ (1967) which introduced the ‘core’ and ‘periphery’ dichotomy, argues that when opening to the international global markets, ‘periphery’ countries are largely constrained to producing primary commodities for ‘core’ countries, not only because they can be relatively ‘factor-endowed’ to primary commodities. In this analysis, core countries were associated to the ‘global north’. Frank argues that this results in a regressive cycle of exploitation, out of which 'periphery’ countries cannot break loose from. (Frank, 1967, as cited in Willis, 2005, pp. 67-69) For an elaboration on the ‘Dependency Theory’ see Immanuel Wallerstein’s work on ‘World Systems Theory’(Wallerstein, 1974).

(12)

Achtenhagen and Brundin, 2016, p. 143). The second hypothesis (‘lack of expertise’), essentially 3

argues that the incompetence or shortcomings of politicians in creating better policies is the reason why some countries do not develop. The hypothesis is summarised well by Acemoglu and Robinson (2012): ‘world inequality exists because we or our rulers do not know how to make poor countries

rich.’ (Acemoglu and Robinson, 2012, p. 127). However, as the authors note later, this form of thinking also suggests that all one has to do is implement ready made solutions that ‘work’. This would in turn ignore the ‘incentives and constrains they (politicians) face from the political and

economic institutions in their societies’ (Acemoglu and Robinson, 2012, p. 133). The last

hypothesis, namely, the ‘culture hypothesis’ relies on argumentation based around culture, ethics, beliefs and values (Acemoglu and Robinson, 2012, p.97). While levels of trust or opportunities for cooperation between various actors are influenced by these broader indicators of ‘culture’, one must also take into account that inefficient institutions can produce certain ethics, beliefs or values. The application of these three hypotheses in regards to case study of Rwanda will be discussed in the case selection. Nevertheless, while the three larger hypotheses do provide valuable insights to understanding underdevelopment, or inequality, single hypotheses cannot explain the majority of cases because certain important questions would remain unanswered. The ‘hypotheses’ definitely hold explanatory power for various developmental issues. However, failing to see the complexity of developmental challenges by misinterpreting the real root causes or relying on single-headed and generic arguments premised on ‘determinisms’, may well lead to ‘slippery-slope’ fallacies, or even the so-called ‘uncertainty principle’.  4

In contrast to the first group of the ‘pessimist’ camp, the second ‘group’ acknowledges the importance of ‘governance’. However, it clearly distances itself from the ‘western-style’ principles expressed by the ‘optimist’ camp. Due to the fact that ‘western-style’ principles expressed by the optimist camp relate most closely to notions such as ‘democracy’ and ‘accountability’, as well as having more normative connotations, these principles link most closely to ‘democratic governance’ distinction. Therefore, this second ‘group’ can mostly aptly be described in relation to the notion of ‘effective governance' discussed in the previous section. A prominent line of thinking that encompasses this is the theory of the 'developmental state’ (Chang, 2011). A key point in this line of

This is also the case for the different outcomes of development for Haiti’s and the Dominican Republic. While both

3

countries have a similar geographical location, they are incomparable in regards to the prospects for development. (CFR, Labrador, 2018)

Inspired by the Heisenberg Uncertainty Principle, Ravi Kashyap first coined the ‘Uncertainty Principle of the Social

4

Sciences’ as: ‘Any generalization in the social sciences cannot be both popular and continue to yield accurate

predictions, or in other words, the more popular a particular generalization in the social sciences, the less accurate will be the predictions it yields.’ (Kashyap, 2014, pp. 1-2).

(13)

research is the explanation for the sudden and successful path to development, without democratic or inclusive institutions/governance. This argumentation originated from the developmental successes of the so-called ‘Asian Tigers’. The ensuing emergence of the ‘developmental state’ approach therefore substantiates that successful developmental paths do not necessarily have to follow the principles of democratic governance. To elaborate, leading scholars in the field such as Adrian Leftwich (1993) famously brought forth arguments undermining the democratic governance approach (Leftwich, 1993). In regards to the ‘Asian Tigers’ (Taiwan, Singapore, and South Kore), all three nations experienced incredible periods of GDP growth, as well as producing effective bureaucracies and institutions. All the while, the ‘Tigers’ forewent the ‘western-style’ democratic governance approaches as well as very clearly distancing themselves from the Washington Consensus (Rodrik and Wacziarg, 2005, p. 975). Notably, the case of South Korea reveals a further 5

counter-argument against the ‘optimist’ camp. Taking the previous reasoning a step further, it has been argued that it was precisely the abstinence of democratic participation which allowed South Korea to develop so quickly and strongly; ‘in fact, scholars have argued that the success of these

efforts specifically depended on limiting citizen participation in the political process and isolating state institutions from popular pressure and accountability mechanisms’ (Carothers and

Brechenmacher, 2014, p. 19). As the case-study will show, this line of thinking will be a valuable contribution in understanding how and why Rwanda has developed to its current form. In fact, Rwanda falls partly within this categorisation of ‘developmental states’. Another important caveat to this ‘camp’s' argumentation is presented by scholars such as David Booth (2011) and Mushtaq Khan (2006). Similar to the ‘open access order’ notion (North, Wallis and Weingast, 2009), Booth and Khan position themselves critically to the ‘democratic governance’ approach. According to them, the premature realisation of ‘democratic-like’ institutions or forms of governance might exacerbate impending collective action problems. In fact, they state these to be at the heart of governance challenges across Sub-Saharan Africa (Booth, 2012, as cited in Carothers and Brechenmacher, 2014, p. 19).

In short, the Washington Consensus refers to a set of 10 desirable policies advocated for by the policy-makers in

5

Washington in the 1980-1990’s. Intended to be understood as instruments rather than objectives or outcomes, the policies comprised of: (1) fiscal discipline, (2) increased public expenditure on social services and infrastructure, (3) tax reform to broaden tax bases and reduce marginal tax rates, (4) market-determined interest rates, (5) unified and competitive exchange rates, (6) import liberalisation, (7) openness to foreign direct investment, (8) privatisation, (9) deregulation, and (10) secure property rights (Williamson, 1990; 1999). Although not all policies are strictly neoliberal, the term is considered a mantra for neoliberalism and market fundamentalism, often with adverse effects on the receiving countries.

(14)

To round off the discussion, the third camp combines aspects of the two previously mentioned schools of thought and is argued for by the likes of Kunal Sen (2012); Sue Unsworth (2010); Brian Levy (2014); and Merilee Grindle (2004). Coined as the ‘multiple path philosophy’, this approach showcases the manner in which the differing schools overlap, pointing at the need to move beyond the dichotomous nature of the debate. In essence, this camp purports that it is context which matters most, because different governance models in different contexts will produce varied outcomes. With the underlying thought that set-templates (Washington Consensus, Good Governance) cannot be applied uniformly across all contexts, Merilee Grindle (2004) famously coined the term ‘good enough governance’ (Grindle, 2004, p. 525). Generally speaking, the ’good enough governance’ refers to the need of re-evaluating policy approaches pertaining to governance and institution building, in order to encompass for local contexts (Grindle, 2004, p. 526). Scholars such as Brian Levy (2014; 2016) argue in similar fashion, stating that we should ignore rigidly applied templates, which he describes as ‘governance maximalisms’ (Levy, 2016, p.12). Sue Unsworth (2010) falls into the same camp as well, and stresses the fact that development thinking should: ‘move beyond the formal, rule-base institutions’ (Unsworth, 2010, as cited in Carothers and Brechenmacher, 2014, p. 20).  

While there is considerable overlap between the schools of thought, the previous section has attempted to delineate the broader debate on development, positioning the main scholars in three camps, respectively. Clearly, there seems to be no clear consensus or uniform acceptance of one approach over the other. The relationship between governance, institutions and developmental approaches remains an incredibly complex and contentious debate. As the case selection will show, when viewed separately, the previously mentioned camps are insufficient to deal with the unique case of Rwanda. Nevertheless, one pertinent issue stands out. Namely, the issue of premature adoption of ‘democratic governance’, and a reliance on universally applied templates. Across the board, these approaches seem to provide the most fissures in the debate. Therefore these issues will be addressed throughout the course of this thesis. The Rwandan case can most aptly be understood through the argumentations used within the second ‘pessimist’ camp and the ‘multi-path philosophy'. Having outlined the larger debate on the three schools of thought, I can now take the second step, and look into the relationship between governance dimensions and effective service delivery.  

(15)

2.1 Literature Review: Governance Dimensions of Service Delivery

Having delineated the broader development aid debate in regards to the three camps, the first stepping-stones for the ensuing discussion on the governance dimensions of service provision have been laid. Hereby, we can develop an understanding for how effective service delivery is portrayed in the contemporary discourse. Although it is a relatively new, the growing literature at hand has been widely popularised and therefore been highly influential as a driving force for policies pertaining to growth enhancing measures, poverty reduction and development at large (Kelsall, 2016). In relation to the previous section, the literature on the governance dimensions to service provision ties well into the previously mentioned second group of the ‘pessimist’ camp. That is because both deal with the question of what types or forms of governance are most conducive for effective public service delivery and development. Additionally, this literature poses as a critique to the democratic governance approach and the optimist argumentation.  

In order to build up the argumentation, I will outline one of the most influential policy reports related to effective public service delivery. The World Bank’s ‘World Developmental Report 2004’ (WDR 2004, from hereon) is understood to have sparked the debate on effective service provision, and therefore will be outlined in a critical light. With its seminal influence, the WDR 2004, titled: ‘Making Services Work for Poor People’ provided a breakthrough research report on the topics of governance and service provision (WDR, 2003). It argued that ‘chains of accountability’ between service providers and service receivers were an essential marker to improve substandard, ineffective or at times, detrimental service provision (WDR, 2003). In order to contextualise the findings, services that were considered in the report were public goods such as, ‘health, education, water, electricity, and sanitation’. These public goods were described as lacking ‘direct’ accountability opportunities between providers to consumers (WDR, 2003, p. 6). As a means to rectify the lack of accountability opportunities, the highly influential report provided a two-fold analogy. This analogy was aimed at commanding accountability from service providers, through either a ‘long’ or ‘short route’. The WDR report classifies the ‘long route’ as the chain of accountability in which the citizens (as clients) influence policy makers, who in return influence the service or good providers (WDR, 2004, p. 6). There are however considerable shortcomings for this principle, such as the disadvantages ‘poor people’ have in relation to formulating ‘collective objectives’, or the difficulty of policy makers to ensure the effective service delivery by service providers (WDR, 2004, p. 7). To visualise this chain of accountability, the two-fold analogy can be understood as the ‘accountability triangle’ shown below.

(16)

Figure 2: Accountability Triangle (Adopted from: WDR, 2003, p. 6)

As the graphed ‘accountability triangle’ shows, the ‘short route’ is characterised by the absence of policymakers in the accountability chain, essentially providing citizens/clients with a more ‘direct’ link to the service providers.When speaking in the context of developmental policies , the ‘long route’ analogy gave rise to solutions such as public sector reform, anti-corruption measures and election support for governments who were possibly more efficient or trustworthy. For the ‘short route’, solutions such as participatory budgeting and community scorecards became increasingly relevant. A common thread for these solutions was that they aimed at improving the flow of information to citizens. This tactic is guided by the premise that citizens require accurate information to hold both policymakers and service providers accountable (Kelsall, 2016). The extent to which these ‘routes’ were applied in Rwanda will be elaborated on in the case study.  

Along with the other annual reports, the WDR 2004 remains a highly influential piece of research, and has permeated through to a large range of developmental approaches. This is furthermore substantiated by the manner through which the academic branch of the World Bank, the Development Economics Vice-Presidency (DEC) exerts its influence. This phenomenon has been coined as ‘paradigm maintenance’ (Wade, 1996). The reason this is important, is that ideologically 6

driven research, and perpetuation of adverse ideas by influential actors can be highly problematic. So, simple terms, what did the WDR report get ‘wrong’? Surprisingly, one prominent critique stems from the senior director of the DEC, Shanta Devarajan (2014). Devarajan argues that there are three main flaws with the ‘accountability triangle’; firstly that the two links for the ‘long route’ are not symmetric. Secondly, that the WDR’s ‘short route’ is essentially ‘the market’, and failing to classify

For further insights on the DEC’s influence, and how adverse policies permeate developmental thinking and

6

(17)

it as such could lead to the possibility of ‘overlooking’ common market failures such as ‘asymmetric information and oligopolies’ (Devarajan, 2014). Lastly, Devarajan argues against the WDR framework’s suggestion that absenteeism among individuals who are ‘supposed’ to provide goods and services, such as doctors and teachers, is inevitably due to flawed incentives. To elaborate, the WDR suggested that a teacher faces discouraging incentives such as, (i) being paid for work regardless of showing up, (ii) having better opportunities outside of their profession, (iii) and possessing the political clout to pressure the government in order to ‘maintain the status quo’ (Devarajan, 2014). Devarajan rightly points that while these skewed incentives are prevalent in other parts of the world such as Sweden and France, absenteeism is not as common-place there. Without falling into the argumentative trap of discussing ‘cultural’ attitudes that might incentivise or de-incentivise certain behaviour, this does point towards the necessary re-evaluation of public policy in certain contexts. The effects of these measures will therefore be assessed in the case-study.   As mentioned earlier, Tim Kelsall (2016) summarises the manner through which the WDR has been applied and outlined the larger debate on its shortcomings. What the following literature shares in common is the fact that they all advocate to moving beyond the use of the ‘short’ or ‘long route' analogy. As previously mentioned, Brian Levy (2014) advocated for an approach to development which ‘goes by the grain’, positioning him in the ‘multi-path philosophy’ camp. In his book titled: ‘Working with the grain: Integrating governance and growth in development strategies’ (2014), Levy states that while the WDR 2004 was a breakthrough text, it does have considerable shortcomings. Specifically within the distinction between supply-side and demand-side approaches (Levy, 2014, p. 10). Tim Kelsall (2016) summarises this well: ‘Essentially, Levy

shares the perspective on using collective action theory to find a ‘middle way’ between the long and short route of accountability. He postulates that collective action can address challenges in the service delivery performance by: ‘(i) it can create space within difficult environments; (ii) it can include both governmental and non-governmental actors; and (iii) it has a wide application range’.

(Levy, 2014, as cited in Kelsall, 2016) Another critique to the WDR report 2004 comes from the Development Research Center (2011). It notes that governance programs that focus on either the ‘short’ or ‘long’ routes are not exactly successful, and that the ability to use both, or find a multi-path solution is most advantageous. In essence, it argues for both ‘top-down’ and ‘bottom-up’ approaches that encompass for state responsibility and accountability from both horizontal and vertical angles (DRC, 2011, as cited in Kelsall, 2016). David Booth and Diana Cammack (2012; 2014) argue in similar fashion, and lay out fundamental challenges to the WDR 2004 that they have discovered. In their research on governance challenges across a variety of African countries, they

(18)

highlight that governance ‘needs’ in Sub-Saharan Africa have to break free from the ‘good governance’ template. Within their research motivation, they refer to the need for an alternative 7

agenda, distancing themselves from the previously described ‘optimist’ camp. In doing so, they also criticise the contemporary approach, which they describe as ‘dangerously content-free’. (Booth and Cammack, 2012, p. 7) They take heed to move beyond the understanding that the relationships are complex and that context-specific solutions are necessary: ‘there is an urgent need to go beyond the

mantra that the institutional challenges of development are complex and that every country is special. In particular, there is a need to spell out what country reformers, and the development agencies that support them, should be doing differently’. (Booth and Cammack, 2012, p. 25) This

makes it difficult to attribute the authors solely to the pessimist or multi-path philosophy camp previously described. More importantly however, are the main takeaways from their research; while they do acknowledge the importance of the WDR 2004, they also point out that, in attempts to tackle the challenges for better service provision (and governance at large), the developmental industry has been fixated on the wrong framework. Namely, the ‘principle-agent’ framework. 8

According to them, this framework does not encompass for collective action problems, which they argue to be at the heart of developmental challenges.  9

The above mentioned literature, its critical analysis of the WDR 2004 (World Bank, 2003) and the discussion on key concepts pertaining to social accountability touch upon a variety of key issues within development. Having discussed the specific theories pertaining to governance provisions of service delivery and development at large, the template to situate the theoretical framework within the larger context has been outlined. The following section will further delineate the analytic concepts pertaining to the of ‘democratic governance’, as well as laying out the theoretical framework for the specific research question: ‘how and why do democratic governance approaches affect Rwandan service delivery?’

In their study, spanning across the course of five years, David Booth and Diana Cammack (2012) researched the

7

economic transformations in the post independence periods for seven African countries. These were Rwanda, Tanzania, Côte d’ivoire, Ghana, Kenya, Malawi and Uganda (Booth and Cammack, 2012, p. 25).

Principle agent problems can be described in the following manner: ‘A principal-agent problem exists where one party

8

to a relationship (the principal) requires a service of another party (the agent) but the principal lacks the necessary information to monitor the agent’s performance in an effective way. It is assumed that the principal wants the service, so that the difficulty to be overcome is distinctly about the agent’s compliance and the ‘information asymmetry’ that affects it.’ (Booth and Cammack, 2012, p. 9).

In the context of David Booth and Diana Cammack’s research (2012), Kelsall (2014) describes the problems

9

accordingly: ‘The problems are caused by ‘asymmetries of information’ that prevent the principal (in this case the

service-receiving public) from getting the agent (policymakers and service providers) to do a good job. Consequently, there has been a profusion of programmes to provide citizens with better information about services, with a view to helping them hold service providers to account’ (Kelsall, 2014).

(19)

3. Theoretical Framework: 

Much has been said about the contentious relationships that make up the debate on governance, development and effective service delivery. In order to realise a specific research question, the concept of democratic governance (as a subcategory of good governance) must be clearly defined and operationalised. Within this conceptualisation, there will be a critical analysis of three key tenants of democratic governance. These are democratisation, decentralisation and social accountability. Additionally, the theoretical underpinnings of the ‘natural state’ and ‘open access order’ will be explained. The central research question and respective subquestions are outlined at the end of this section.  

3.1 Good Governance and Democratic Governance Revisited 

As touched upon in chapter one, the good governance and democratic governance agenda is difficult to define (Kohler-Koch and Rittberger, as cited in Hackenesch, 2016, p. 12). Since its application varies across different contexts, it has been defined as an ‘ideal’, indicative of normative underpinnings such as the eight key principles listed by the United Nations Economic and Social Commissions for Asia and the Pacific (UN ESCAP): ‘Consensus oriented, accountable,

participatory, transparent, follows the rule of law, responsive, effective and efficient, equitable and inclusive’ (UN ESCAP, 2009, p. 3). As mentioned earlier, the World Bank’s ‘Good Governance

Practice’ works with an approach that aims to build institutions that are ‘‘capable, efficient, open,

inclusive, and accountable institutions’ (Price, 2018). Due to the fact that the multitude of

definitions can pose issues for the agenda’s operational application, the ensuing discussion will be making use of David Booth and Diana Cammack’s (2012) understanding of democratic governance, because it matches the scope of the research question most closely.  

Firmly rooted within the practice of ‘democratic governance’ lie three ‘magic bullets’ that Booth and Cammack identify (Booth, 2012; Booth and Cammack, 2013). While these are no new solutions to tackling inefficient public goods provision, the authors argue that they have been blindly applied, with no gearing towards context specificity. The three ‘magic bullets’ have arguably become mantras in the contemporary developmental approaches and comprise of (1) democratic decentralisation, (2) rediscovering demand and (3) social accountability (Booth and Cammack, 2014, pp. 63-72). Due to the scope of the research at hand, this following segment will only be

(20)

discussing democratic decentralisation and social accountability.  10

3.2 Democratic Decentralisation

The first key concept to assess is democratic decentralisation. Also referred to as devolution, the approaches pertaining to this concept are widely promoted to be conducive in tackling obstacles in the effective provision of public goods and services. The distribution of power to local forms of governance such as elected districts or municipal governments comes to mind here. As an example, certain measures of democratic decentralisation are pushed by actors such as the World Bank. (World Bank, 1997; ch7, as cited in Booth, 2012, p. 66). However, scholarly debate on the effectiveness of democratic decentralisation is varied. For example, Booth and Cammack (2012) argue that while democratic decentralisation ‘may be intrinsically desirable’, the concept sits on shaky theoretical foundations. (Booth and Cammack, 2012, p. 67) Daniel Treisman (2007) provides rigour to this argument, finding no theoretical foundations to the supposition that democratic decentralisation is more effective than centralised systems in relation to public goods and services provision. Surprisingly, Treisman concludes that there is no substance to democratic decentralisation being less effective either (Treisman, 2007, pp. 270-289). In addition to the 11

questionable theoretical foundation of the concept, scholars such as Crook, Manor and Sverrisson (1998; 2003) provide arguments in regards to its empirical foundation (Crook and Manor, 1998; Crook and Sverrisson, 2003, as cited in Booth, 2012, p. 67). Their arguments are summarised well by Booth (2012): ‘the impacts of democratic decentralisation on outcomes such as the effectiveness of poverty reduction efforts are indeterminate.’ What matters more are factors such as the ‘political complexion of the central government’ and its effectiveness in ‘capturing local power’ (David Booth, 2012, p. 67)

. 

 

The WDR 2004, and the ensuing ‘long’ and ‘short’ route have already been discussed at length in the previous

10

section.

‘In short, it is hard to reach any general conclusions about whether political – or administrative, or fiscal –

11

decentralization will improve or impair the quality of government and economic performance. They will have many effects, driving in different directions on different dimensions. These effects depend on numerous conditions, many of which are difficult to disentangle in theory and to identify in practice. As one would expect, empirical studies have found almost no solid, general results about the consequences of decentralization. Decentralizing government in a particular place and time is very much a leap into the dark.’ (Treisman, 2007, p. 274).  

(21)

3.3 Social Accountability 

On the other hand, social accountability, as understood by World Bank scholars such as Carmen Malena, Reiner Forster and Janmejay Singh (2004; 2014), is another ‘magic bullet’ that Booth and Cammack discuss (2013). Being a cornerstone to the good governance agenda, and championed by institutions such as the World Bank, the concept of social accountability has become a panacea in the development community. According to Malena et al., (2004) social accountability is: ‘an

approach towards building accountability that relies on civic engagement, i.e., in which it is ordinary citizens and/or civil society organizations who participate directly or indirectly in exacting accountability’ (Malena, Forster and Singh, 2004, as cited in Booth and Cammack, 2013, p. ). This

approach is understood to have three broad effects, namely, ‘improved governance’, ‘increased

development effectiveness’, and ‘empowerment’ (Malena, Forster and Singh, 2004, pp. 5-7). In

regards to the scope of this research, there will be no delineating the intricacies and applications of this concept any further. It will be assessed through the critical lens that Booth and Cammack (2012; 2013) have provided. What is most important to mention is the link to democracy and therewith, the democratic governance agenda. According to Malena, Forster and Singh (2004) there are several (inter)linkages between social accountability and democracy. Integral to the notion of democracy lies the phenomena of social accountability which is based ‘upon the active involvement

of citizens in exacting accountability from their elected representatives and leaders’. (Manela,

Forster and Singh, 2004, p. 6) Improving governance and ‘deepening democracy’ is seen as a result of social accountability mechanisms (Reuben, 2002; 2003, as cited in Manela, Forster and Singh, 2004, p.7). Normatively speaking, this argument links well with one of the arguable core points to democracy. In the broadest sense of the term, democracy allows for the ‘bottom-up’ nature of preference expression to elites through consensual decision making. Essentially, it allows elites to be held accountable and fosters a certain degree of political legitimacy (Collier, 2010, pp. 1-18). Matching this supposition with the previous ‘magic bullet’ of democratic decentralisation shows a clear linkage to the concept of democracy. In fact, according to Booth (2012), democracy is ‘supposed to be the solution’, for good governance (Booth, 2012, p. 50). As previously discussed, this form of good governance can be understood as ‘democratic governance’ (Hackenesch, 2016, p. 14) This begs the question: What is the case for democratic governance? The theoretical underpinnings of Douglass North, John Wallis, and Barry Weingast (2009) allow for a deeper discussion of the matter at hand. Matched with Paul Colliers (2007) work on the ‘bottom billion’,

(22)

the differing logics of the ‘natural’ and ‘open access state’ show that context matters, and that democratic governance might not be the most conducive approach to targeting effective service delivery.

3.4 The logics of the ‘Natural State’ and ‘Open Access Order’ 

To begin with, North, Wallis & Weingast (2009) make relatively vigorous claims in supposing that both the economics and political science disciplines have failed to illustrate the relationship between democratic conditions and open economies (North, Wallis & Weingast, 2009, p. 269). In their work titled: ‘Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History’ (2009) they spell out various important points in regards to this relationship. The arguments deal with economic and political history, and have made an impact on the study of development (Bates, 2010, p. 752). Imperative to their work are the two forms of 'social orders’ they identify, namely: the ‘natural state’ and ‘open access order’. 

The ‘open access order’ (OAO) illustrates a society, which controls the problem of political violence through ‘open access’ and competition. To elaborate, ‘open access’ societies are marked by impersonal relationships between elites, who are subject to regulation and the rule of law (North, Wallis & Weingast, 2009, p. 152). An essential point is that economics and politics can be separated to a certain degree. In short terms, this means that an organisation can be politically strong without economic clout and vice versa. This is not the case for the ‘natural state’ (also referred to as the ‘limited access order (LAO). An important point to infer is that the authors explicitly acknowledge that there should be no generalisation for both categories of societies (North, Wallis & Weingast, 2009, p. 41-49). In fact, differences between three forms of natural states are drawn out. In ‘fragile natural states’, such as Iraq, Somalia and Afghanistan, political coalitions are prone to sudden disbandment. Therefore the main goal for elites is to survive (North, Wallis & Weingast, 2009, p. 42). ‘Basic natural states’ enjoy moderately more durability and stableness because of the emergence of public law institutions. These institutions are ones that: ‘structure aspects of the state,

its internal relations, and its relations with members of the dominant coalition’ (North, Wallis &

Weingast, 2009, p. 43). True to the historical approach for their work, the authors use the example of the Roman Republic. To specify, the period after which Brutus and Collanitus ‘validated’ the public institution, by not becoming king and staying true to the newly-established rule of only being in office for a set time (Livy, 1998, pp. 57-60, as cited in North, Wallis & Weingast, 2009, p. 44).

(23)

The ‘mature natural state’ analogy take this a step further, possessing private institutions which are able to durably support coalitions (North, Wallis & Weingast, 2009, p. 47) These three examples show that differentiating between these natural states is marked by their ‘degree’ rather than their ‘kind’ (North, Wallis & Weingast, 2009, p. 47). Nevertheless, to continue we need to understand the dynamics of dominant coalitions in ‘natural states’. The ‘natural state’s’ society is characterised by having a coalition of elites who create rents for themselves (rent-seeking), which in simple terms means that they share the economic and political power with themselves. While the creation of 12

distributed rents amongst elites is a crucial aspect to solving the problem of violence in natural states, it does not necessarily always produce a stable environment (North, Wallis & Weingast, 2009, pp. 49-51). Elite coalitions are free from checks and balances from third parties, and therefore are prone to break down more easily when the balance of political and economic interests change dramatically. When these balances change, this results in a need to renegotiate privileges and rents 13

amongst elites. If these negotiations fail, violence can become more likely (North, Wallis & Weingast, 2009, p. 21). The development or ‘transition’ to open access orders can spur these dramatic changes in the balance of interests. This falls in line with the second ‘pessimist’ camp 14

described earlier. In fact, in terms of rapid development, this can lead to what Chang (2002) describes as ‘kicking away the ladder’ (Chang, 2002, as cited in Booth, 2012, p. 23). In relation to the developmental path between ‘natural states’ and 'open access orders’, the authors outline a so-called ‘transition proper’, namely the conditions through which the transformation happens. While they are referred to as ideal types, certain categorisations between the forms of states is warranted. Especially in regards to Rwanda, as the case selection will show.  

Within the framework of development policy, ‘Rent-Seeking’ (also known as ‘privilege seeking’) refers to the

12

acquiring of economic or political ‘rents' (benefits), which are not shared with society. The idea was originally introduced by Gordon Tullock (1967), and labeled by Anne Krueger (1974) in her highly influential work titled ‘The Political Economy of the Rent-Seeking Society’ (Krueger, 1974). A key takeaway from her work is that rent-seeking behaviour is a natural part of competitive behaviour (Krueger, 1974, pp. 291-293).

All societies are subject to sudden, and unexpected shocks. Internal factors such as unintended power shifts within a

13

coalition, or external factors such as unpredictable changes in prices or climate disasters are examples given by the authors (North, Wallis & Weingast, 2009, p. 21).

In regards to the development from ‘matured natural states’ to open OAO’s, North, Wallis and Weingast outline three

14

mechanisms: (1) ‘Rule of law for elites’, (2)’ Support for perpetually lived organisations (including the state), both public and private’, (3) ‘Consolidated political control of the organisations with violence capacity including military and police forces’ (North et al., 2009, p. 26).

(24)

3.5 Contextualising Democratisation

In regards to democracy, North, Wallis and Weingast (2009) argue that in the context of an open access order, democracy allows for a political competition that subdues the potential for violence (North, Wallis & Weingast, 2009, p. 125). However, one must be aware of the fact that democracy itself is no universal template. For example, certain aspects of democracy such as competitive elections, checks & balances, social accountability and the decentralisation of power must be taken into account. This is especially the case when looking at how elections, for example work in the open access order or natural state. According to the authors, elections in open access orders allow for ‘a rich civil society, free press and open competition of an

opposition... natural states limit these aspects’ (North, Wallis & Weingast, 2009, p.145).

Additionally, the authors argue that this enables the government to provide essential public goods and services, whilst also responding to the demands of the electorate. Judging from the underlying logic of natural states on the other hand, these states are unable to facilitate this form of governance in a sustainable manner. Paul Collier (2007; 2010) argues a step further, stating that imposing elections in the context of natural states is shortsighted. Aptly described as 15

‘Democrazy’, a natural states imposition of certain democratic principles, in this case free and fair elections, could have adverse effects (Collier, 2010). This issue can be substantiated by the widely believed yet shortsighted assumption that elections equates to democracy. In fact, true or ‘deep’ democracy can actually be constrained by the phenomena of ethnic rather than instrumental voting, for example. Additionally, Collier (2010) makes comments on the issue of scale in relation to public goods provision. He says that these sorts of states have a two-fold challenge, in that these states are ‘too large and too small’ (Collier, 2010, p. 9); ‘too large’ in regards to their ethnic diversity, therefore reducing the possibility of overcoming cooperative barriers to produce public goods. Additionally, they are ‘too small’ in the sense that they 'cannot

reap the scale economies of the key public good, security’ (Collier, 2010, p. 9).  

As previously alluded to, when extending this discussion to development, the following points can be made; all previously mentioned authors argue that the translation of open access institutions on natural states have unintended impacts. While I have been discussing the concept of democracy broadly, the good governance and democratic governance agenda itself can be There is a growing body of literature with similar argumentation. As summarised by Booth and Cammack (2014),

15

‘careless promotion’ of elections and liberalising the economy can be negative. Especially in contexts with

‘inter-communal relations’ as well as unstable political settlements (Autessere, 2010; Kaplan, 2008; Lindemann, 2008; Putzel,

(25)

understood as an open access institution, and so can principles which are categorised to make up the agenda. Therefore, to complete the discussion, Collier’s (2007; 2011) argument on shortsighted implementation of elections, Chang’s (2002) analogy of ‘kicking away the ladder’ and Booth and Cammack’s (2012; 2013) argumentation on ‘premature implementation’ all provide very strong arguments critical of ‘democratic governance’. What then are the unintended impacts of translating open access institutions in natural state contexts?  

The previous debate on developmental aid, the contextualisation of the three magic bullets within the democratic governance framework as well as the preliminary discussion on effective service delivery open a pandora's box of questions. This is especially the case for ‘developmental states’ such as Rwanda. The range of insights from the previous theoretical chapter aimed to outline the set of literature and theories that are relevant to answering the following research questions: ‘how and why do democratic governance approaches affect Rwandan service delivery?’ To reiterate, the supporting questions are:  

- How and why have democratisation approaches affected Rwandan service delivery?  - How and why have decentralisation approaches affected Rwandan service delivery?  - How and why have social accountability approaches affected Rwandan service delivery? 

4. Methodology

Having outlined the relevant debates, analytical concepts, theoretical framework and identified the research question(s), this chapter will present the methodological considerations for the thesis. To allow close examination of democratic governance approaches in the Rwandan service delivery this thesis makes use of an exploratory mixed methods single case study. According to Yin (2009), a case study refers to an empirical inquiry that ‘investigates a contemporary

phenomenon in depth and within its real-life context, especially when the boundaries between phenomenon and context are not clearly evident…and in which multiple sources of evidence are used’ (Yin, 2009, p. 18). Although the proposed form of case study allows for an in-depth

exploration of ‘democratic governance’ in the Rwandan service delivery, the methodological approach itself is contested. In general terms, case studies are regarded in critical light for several reasons. Firstly, as admitted by its most prominent advocate, there is a ‘relative absence’ of methodological procedures and guidelines in comparison to other methods (Yin, 2009, pp.

(26)

14-15). Additionally, case studies are thought to lack in precision (Yin, 2003). Furthermore, although it is understood as a broader critique of qualitative research in general, there are concerns about validity, author bias and replicability for single case study analysis (Willis, 2014). Nevertheless, any methodological consideration will require a form of trade-off, impacting the research both positively and negatively. On the basis thereof, there are considerable advantages that case studies have too. As previously mentioned, the ability to delve into an in-depth study of a ‘new’ and ‘contemporary’ case is highly relevant for this particular research. The literature on governance provisions to service delivery described above is relatively new and has great contemporary relevance both within academia and the policy-related realm. Additionally, Yin (2018) outlines several reasons where case studies are appropriate: ‘(a) the focus of the study is

to answer “how” and “why” questions; (b) you cannot manipulate the behaviour of those involved in the study; (c) you want to cover contextual conditions because you believe they are relevant to the phenomenon under study; or (d) the boundaries are not clear between the phenomenon and context.’ (Yin, 2003, as cited in Baxter and Jack, 2008, p. 545).  

The research question at hand focuses on the ‘how’ and ‘why’ questions. In order to answer the broader question: ‘how and why do democratic governance approaches affect Rwandan service delivery’ there is a division into various subchapters. Firstly, the governance challenges in Rwandan service delivery will be outlined, through an elaborate discussion on Rwandan development since the genocide in 1994 until now. In the second chapter the effects of democratisation, decentralisation, and social accountability mechanisms are analysed. Due to the fact that these three approaches are not clearly distinguishable and overlap to a considerable degree within Rwandan developmental policy, the following considerations have to be made; instead of analysing the concepts singularly, the policies and development plans pertaining to these approaches will be analysed in depth. Emphasis will be put on separating the concepts during the analysis, however overlap may occur. This remains an important limitation to the research and will be addressed at the end of the thesis. A further justification for the methodological consideration is due to the fact that this research is unable to ‘manipulate the

behaviour of those involved in the study’ and also aims to ‘cover contextual conditions’ that are

believe to be ‘relevant to the phenomenon under study’ (Yin 2003, as cited in Baxter and Jack, 2008, p. 4). This thesis makes use of the so-called exploratory case study. Yin (2003) describes this methodological approach as one ‘used to explore those situations in which the intervention

being evaluated has no clear, single set of outcomes’ (Yin, 2003, as cited in Baxter and Jack,

Democratic governance and its effects on Rwandan service delivery

Master Thesis Political Science: Political Economy