Democratic Governance in Sub-Saharan Africa

Twenty years into sub-Saharan Africa’s democratic renewal, an understanding of democratic governance has become established among political actors in the region. Democratic governance is characterised by: legitimation of governmental power through elections, legitimation of governmental politics through transparency, participation and accountability, and legitimation of governmental work through results. This conceptual understanding has asserted itself in African reality as the following paper argues: First and foremost, a far more vigorous public discourse about democratic governance can be observed, intensifying the pressure on governments to institute reforms. Despite numerous inconsistencies and weaknesses, the increased demand for democratic governance has led to ongoing improvements, at least in a group of countries in the region (manifested in fair elections, more transparency, more participation, and more domestic accountability). There are many indications that this is also reflected in improved socioeconomic outcomes. In short, democratic governance has become a reality in a number of sub-Saharan African countries and, in terms of development policies, democratic governance matters. Key Descriptors: Governance, Democracy, Civil Society, Transparency, Accountability


Introduction
On 2 nd December 1990, 93.2 percent of the electorate in Benin voted in favour of a new democratic constitution for the country. A few weeks later, the incumbent President Mathieu Kérékou was defeated by his challenger Nicéphore Soglo in democratic elections and proceeded to vacate the presidential office voluntarily and in a peaceful manner. These events 20 years ago marked the start of the Third Wave of democratization (Huntington 1992) in sub-Saharan Africa. Since then, research in the field of African Studies has focussed extensively on the phenomenon of democratic transitions on the continent: specifically, the multifaceted transformation of sub-Saharan African governance systems that could be observed after the end of the Cold War. In the 1990s, there was general consensus that the "democratic experiment" had taken root in Africa (Bratton and van de Walle, 1997). However, questions such as how to assess progress in individual countries; under which circumstances transitional processes might genuinely lead to democracy; or whether the process is irreversible continue to arouse controversy (cf. Barkan, 2000;Carothers, 2002).
Over the past 20 years, the most influential concept in the developmental debate about political rule in Africa can be summed up with the catchphrase "good governance". The view that "getting politics right is, if not a precondition, at least a requisite of development" (Hyden, Court & Mease, 2004: 10) has become increasingly wellestablished among governments and non-state actors in Africa, but also among the international donors. Alongside the general points of agreement however, the debate has been characterised by major controversies. One of the main lines of conflict concerns the question whether good governance should also mean democratic governance (Mkandawire, 2004). This article will start by tracing the political debate and then, in a second step, will consider whether and how the concepts of good and democratic governance reflect African realities.

Democratic Governance: Concepts And Debates
At the Third High Level Forum (HLF3) on Aid Effectiveness in Accra in September 2008, the issue of governance was the subject of intense debate. The Accra Agenda for Action adopted on that occasion by governments from the North and the South brims over with statements on the importance of political frameworks for development and development cooperation (Accra Agenda 2008). However, it is not only the international donors who, for years, have shown great interest in good governance: civil society actors in sub-Saharan Africa, who grew more and more politically articulate in the 1990s, are now increasingly voicing their demands for more democratically oriented policies and governance from their leaders (Eberlei, 2008). In the Poverty Reduction Strategy (PRS) processes which have played an important role in many African countries since 2000, the governance debate has often been driven by civil society actors as well. In contrast to donors, however, it is not only the efficiency and effectiveness of the government apparatus that have generally come under fire but also, and above all, the lack of transparency in governmental politics, weak rights of participation, and a lack of accountability and other democratic elements of governance. These debates have now left their mark on African governments as well: recent declarations by the African Union spell out the democratic governance concept in detail.

Good Governance: The African Union's Perspective
The stage reached in the official debate about governance in Africa is reflected, first and foremost, in key declarations and decisions adopted by the African Union (AU). The 2007 African Charter on Democracy, Elections and Governance is of particular importance in this context. The Charter makes a number of statements about the issue of governance. However, the text does not clarify the precise relationship between the triad of concepts contained in the Charter's title. The term (good) governance is not defined. Nonetheless, various statements suggest a close proximity to the concept of democracy. For example, Article 2 (6) of the Charter's stipulates: "(...) nurture, support and consolidate good governance by promoting democratic culture and practice, building and strengthening governance institutions and inculcating political pluralism and tolerance." (African Charter 1990) The formulation of this objective itself makes clear that good governance, as the African Union understands it, is not only or primarily a matter of good bureaucratic administration, but is determined by a democratic culture, however that is defined. Chapter 9 of the Charter enumerates various strategies in order to advance political, economic and social governance, in particular, Article 27, 1 and 2 calls for strengthening the capacity of parliaments and political parties, and fostering popular participation and partnership with civil society organisations.
Admittedly, this is followed by various approaches, such as improving public sector management; improving efficiency and effectiveness of public services and combating corruption; and promoting the development of the private sector through an enabling legislative and regulatory framework: in other words, approaches found in donor documentations are often subsumed within the term good governance. Nonetheless, the priority with reference to strengthening parliaments and fostering popular participation clearly reveals a fundamental understanding of democratic governance.
For civil society actors, the Charter can become an important reference document. It contains frequent references to popular participation in general and partnership with civil society in particular as being key elements of democratic societies (See Art. 2,10;3,7;4,2;12,3;27,2;27,8;28;29,2;44,2 Bb). Article 36 of the Charter makes explicit reference to the Declaration on Democracy, Political, Economic and Corporate Governance adopted as part of the NEPAD process in 2001, and urges AU members to promote and deepen democratic governance in accordance with the Declaration's principles and core values, and, where applicable, by implementing the African Peer Review Mechanism (APRM). Summing up, then, the first point to note is that good governance, as the African Union understands it, goes far beyond technocratic or administrative dimensions. The key issue, according to the AU's understanding of governance, is the embedding of governance in a democratic culture and democratic institutions. This reflects not just a technical understanding of optimizing governmental procedures. It also aims at democratic governance (the road from rhetoric to reality is discussed in more detail below).

Perspectives of Civil Society Actors
According to Mkandawire (2004), the inspiration for the concept of good governance came from African scholars. In the late 1980s, some of them were commissioned by the World Bank to prepare background papers on the current political situation in their countries and identified "the failure of authoritarian regimes in Africa to ensure both human rights and development". Mkandawire writes: The general understanding within African intellectual circles then was that the main challenge of development was the establishment of state-society relations that are (a) developmental, in the sense that they allow the management of the economy in a manner that maximises economic growth, induces structural change, and uses all available resources in a responsible and sustainable manner in highly competitive global conditions; (b) democratic and respectful of citizens' rights; and (c) socially inclusive, providing all citizens with a decent living and full participation in national affairs. Good governance should therefore be judged by how well it sustains this triad.( www.inwent.org/E+Z/content/archive-eng/10-2004/tribune_art1. html) However, these views reflect a very different perspective on governance from that espoused by the World Bank, which shaped the concept in publications from 1989 onwards (World Bank 1989 as the first example). Over the past 20 years, African civil society organisations have frequently criticised donors' limited technocratic approach to the governance concept and called instead for a comprehensive understanding of democratic governance. There have been numerous calls, notably the African Charter for Popular Participation in Development and Transformation of 1990, for democratic participation in political processes, in particular. Civil society organisations have repeatedly voiced demands for the core dimensions of democratic governance most recently in their preparations toward and of the HFL3 (see www.betteraid.org for examples). These core dimensions include: • Institutionalised participation of civil society in political processes, i.e. properly established dialogue structures, legally protected scope for participation, adequate resources, also in order to build the participatory capacities of disadvantaged social groups, and, in particular, far greater involvement of women in politics; • Transparency through full and immediate public access to information and safeguards on press freedom; • Full government accountability (by strengthening parliamentary and civil society participation in the monitoring of governmental politics).
Moreover, through their political education activities in advance of elections and their role as election observers, civil society organisations in Africa have done much to promote the democratic legitimation of governmental politics through free and fair elections. Civil society organisations have also been involved in constitutional reform processes in numerous countries such as Kenya or Zambia.
For civil society organisations in Africa, good governance always means more than the efficient and effective performance of bureaucracies. The democratic governance concept, then, is deeply rooted in African civil society.

Donor Notions of Good Governance
The term good governance was first introduced into the international debate by the World Bank in the late 1980s. Here again, it is the World Bank which has crucially shaped the "good governance" discourse over the past two decades by modifying its own understanding of the concept several times during the process. 7 However, the World Bank's operational understanding of good governance is not truly reflected in the statements made in its numerous publications, particularly those targeted at the Western public. It is more informative to look at the criteria applied by the World Bank as the basis for its decisions on disbursements to countries. The World Bank's disbursements to the least developed countries are based, to a large extent, on the scores these countries achieve in a Country Policy and Institutional Assessment (CPIA), a diagnostic tool that is intended to capture the quality of a country's policies and institutional arrangements.
Here, the primary focus is on fostering conditions that are favourable to economic development. Aspects of democratic governance are only marginally reflected (Eberlei, 2007b). The World Bank's understanding of governance tends to have such a technocratic bias, writes Waeyenberge (2006: 15 -16), because the Bank's mandate restricts its involvement in recipient countries' political affairs to the "economic" dimensions of governance only. However, Waeyenberge continues, this is merely a superficial explanation. In reality, the Bank's focus on these dimensions of governance rests on a particular model that prescribes a limited role for the state in development processesa model which is ill-suited for the context of developing countries and hence misguided, and fails to deal with complex underlying politico-economic processes.
The pronouncements of another major multilateral donor, namely the European Commission, reveal a rather differently accentuated understanding of governance. The EU favours an interpretation which takes account of political processes in the country concerned as well. Hence, in a document on governance and development, the Commission defines governance as "the state's ability to serve the citizens (...). 7 The World Bank defines Good Governance currently as "(1) the process by which governments are selected, monitored and replaced, (2) the capacity of the government to effectively formulate and implement sound policies, and (3) the respect of citizens and the state for the institutions that govern economic and social interactions among them" (World Bank, no date).

W. Eberlei: Democratic Governance in SSA
Governance refers to the rules, processes, and behaviour by which interests are articulated, resources are managed, and power is exercised in society" (EC, 2003: 3). As a logical consequence, the EU makes explicit use of the term democratic governance in various declarations. Bilateral donors, too, have since the early 1990s integrated good governance issues into their development policy strategies. In all cases, these are very broadly formulated strategies, covering almost every aspect of political development in one way or another and, broadly in line with the approaches pursued by the World Bank or the EU.

Emerging meanings
The governance debate has produced many different understandings of the governance concept. In simple terms, three separate approaches can be defined: • a more technocratic understanding of governance which focuses on the modernisation of government administrations; • an understanding of governance that is primarily economic in orientation and mainly considers how economic development can be supported through government action; • a politically oriented understanding of democratic governance.
Depending on their perspective and interests, the various actors accentuate different aspects of governance. The concept of democratic governance would appear to be a farreaching approach. This presupposes, firstly, that governments come to power as a result of free and fair elections and can be voted out of office again, i.e. legitimation as a prerequisite for governmental power.
It also implies, secondly, the legitimation of governmental action per se (and the principles of transparency, accountability and participation deriving from it). Both dimensions of democratic legitimation require an appropriate constitutional framework and functioning separation of powers.
Thirdly, implicit in this understanding is that democratically based governance must have a well-performing bureaucracy which is capable of implementing a government's pro-poor policies efficiently and effectively, for example. The governance concept itself does not determine which policies a government should pursue per se. Instead, it opens up a political arena in which social negotiation processes can take place. It does, however, prescribe who should be the ultimate beneficiaries of governmental politics: legitimation of governmental work through results in the interests of citizens.
The democratic governance concept is now well-established on a broad basis. First, the United Nations Millennium Declaration describes democratic governance as being an essential prerequisite for development. The Accra Agenda for Action, adopted in 2008 by governments from the North and the South with the involvement of civil society actors, reinforces this message: in its preamble, democracy (along with economic growth, social progress, and care for the environment) is stated to be one of the "prime engines" of development (Accra Agenda 2008). Frequent mention is made of the key elements of democratic governance throughout the document: of particular note are the repeated and explicit references to the role of parliaments and the explicit acknowledgement of civil society organisations as independent actors.
The fact that the African Union uses the term explicitly in the African Charter on Democracy, Elections and Governance as well as the spirit of the Charter in general confer upon the concept of democratic governance the greatest political legitimation on debates about governance in Africa. At the same time, the AU has established therein a benchmark against which to measure the actions of leaders.

Theory And Reality: Better Governance?
From a long-term perspective, the fact that state and non-state actors in Africa and in the Northern hemisphere alike are formulating and accepting notions of democratic governance as a normative criterion is an important development. But how will these lofty declarations impact on the reality of political rule in Africa?

A growing public discourse about governance
First and foremost, a dynamically intensifying public discourse about the requirements of governance can be observed in African societies. Public debates about the quality of governmental politics and the content of political decisions have increased considerably in recent years. As a result, the notion of democratic governance is steadily being wrested away from the control of those in power; instead, it is evolving into a benchmark of the demand for good governance and is becoming firmly established in this role.
The public debates about the annual government budgets are a significant example of this trend. Only a decade ago, government budgets were prepared behind closed doors at the ministries of finance. No member of parliament and no one in the public ever had the chance to discuss a draft of the budget before the finance minister unveiled what was, in effect, a finished document in parliament on budget day. The budget draft is more secret than the colour of Her Majesty's underpants was a popular quip among Kenyans. And yet in less than ten years, this situation has changed dramatically. Budget policy is now the subject of public and often heated debate in a wide range of African countries. Anyone following the media coverage of the budget debates in the parliaments of Zambia, Uganda, Ghana or Kenya, for example, will find a wealth of reports, analyses and opinion. Scholars and civil society think tanks analyse the draft budget, and in countries like Zambia, civil society actors are invited to attend parliamentary hearings. As these examples show, government spending policy in particular is now subjected to far more intense scrutiny in a number of African countries than was the case only a few years ago. This surge in public debate has much to do with the fact that as with the development discourse generally, the good governance discourse is no longer confined to the level of declarations of intent. Instead, the actual implementation of reform programmes and the outcome and impacts of these reforms are now the subject of comprehensive monitoring. Governance is observed, analysed, discussed and evaluated. Due to the substantial improvements in access to information over the past two decades -politically desired, but also driven by the Internet -this monitoring now makes a significant contribution to the public discourse, as the following section explains.

Monitoring of government work intensified
The critical monitoring and analysis of what Governments do take place through various institutions, primarily African parliaments and independent government review bodies, as well as civil society and donor organisations.
The parliaments have only recently been "discovered" in the international good governance discourse (mainly in the context of the debate about budget support) (see for example Accra Agenda 2008). At present, however, they are still neglected and often bypassed institutions. Although the constitutions adopted by the African countries in the 1990s make it very clear that parliaments have key political functions which include scrutiny of the government, a combination of factors has prevented these functions from being performed effectively. These include electoral systems that tend to strengthen the president/executive (Southall, 1999) and a lack of incentives for parliamentarians to adopt critical positions towards the government (Barkan, 2008). Not least, however, it is also due to pressure from international donors which, by demanding accountability from governments, has systematically weakened the parliaments. As the important example of the PRS processes shows, legislatures are still largely excluded from political decisionmaking processes or, at most, are only involved peripherally (Eberlei, 2007a).
Nonetheless, parliamentary control of the executive appears to be gaining momentum, as shown in reports from Kenya, Ghana, Uganda or Zambia, where parliaments are now far more articulate and critical of their governments (see reports on www. internationalbudget.org). Ghana is just one example, where in 2008, the parliament's Public Accounts Committee held public hearings for the first time, thereby unleashing a major public debate about the misuse of public resources and corruption.
One category of players generally overlooked in the debate about the monitoring of governance is government supervisory bodies, particularly the auditor generals' offices and/or courts of audit. For many years, they have been viewed as obedient servants of, and therefore loyal to, the government and, what's more, as generally being quite unequal to their tasks. The courts' audit reports for concluded fiscal years often took years to reach parliament, which meant that they were largely weak in political terms. This situation has now changed in a number of countries partly due to strong support for these institutions from donors, incidentally. The report on government spending policy by Zambia's Auditor General, for example, has become a nightmare for those in government year after year, for it documents mismanagement and misappropriation of public funds in detail and holds them up to public scrutiny. The report, which is published on the auditors' own website, is now released just one year after the end of the budget year, which means that those responsible are often still in office. It is regularly the subject of extensive media coverage and triggers heated debate in Parliament. However, in Zambia too, the Auditor General is dependent on his employer, the President, so the report cannot go too far. In 2004, a working group, controlled by the government, was set up in Zambia to track the use of funds generated from debt relief. When it voiced criticism of the government (by highlighting wastage/misuse of resources, etc.), it was promptly dissolved.
It was civil society actors whose critical engagement, in the 1980s, prepared the way for, and -once the external parameters after the end of the Cold War permittedbecame the driving forces behind the democratic transformation of numerous regimes (Ssemakariyanga, 2009, Apusigah, 2009a. Critical monitoring and public discussion of governance have therefore long been core elements of civil society activities in Africa too. These actors are still driving the public discourse about governance. However, the fact that this is now being recognised by political leaders is a relatively new development (see the Accra Agenda for Action 2008 as an example). Civil society actors have arrived at the macro level of political processes in Africa. They monitor and criticise violations of political rights and fundamental freedoms, and recently, have increasingly monitored violations of economic, social and cultural rights as well. Topics such as corruption and bad governance are taken up and driven by civil society actors. An increasingly diverse media landscape is creating the requisite transparency here. As mentioned above, civil society actors now have a greater presence in budget policy processes and in sectoral policy debate as well (Apusigah, 2009b). To sum up, they increasingly enter and capture the public political sphere in African countries which, for decades, was occupied solely by governments and their international donors.
Another politically exciting process which contributes to the public debate about governance in Africa is the African Peer Review Mechanism (APRM) mentioned above. This was developed in 2001 within the framework of the NEPAD initiative and was then adopted by the AU, founded in 2002. Today, 29 countries -i.e. more than half of the countries in sub-Saharan Africa -have formally joined the APRM, although the majority of these countries are still at the start of the multi-stage process. So far, only seven countries have completed all the stages of the process, with Burkina Faso and Nigeria being close to completion. Completion also marks the starting point for the implementation of the programme of action to address the identified weaknesses, which is subject to regular monitoring. Ghana, for example, has already submitted several reports on the implementation of its national programme of action, and its 2009 progress report is currently in preparation. The report will also contain the results of extensive consultations throughout the country, in which citizens were asked for their views on the governance situation.
The added value of the APRM approach can be identified at various levels: firstly, by adopting this process, the AU has broken away from the "Crow principle" as formulated by Grimm and Nawrath (2007): henceforth, "no crow will peck out the other ones' eyes". Discussion of governance issues now features regularly on the AU agenda, even though mutual criticism has been extremely muted so far. Secondly, the approach genuinely facilitates a mutually inspired learning process, focussing on the question of how good governance can become a reality in Africa. When seeking answers, the democratic processes in neighbouring countries -rather than the democracies of the North -can serve as a model for African countries' own policy action. Thirdly, and perhaps crucially, the APRM process at home, i.e., the dialogue between government and other institutions and actors (parliaments and political parties, traditional actors, civil society, the private sector) offer what is, to some extent, a regulated approach to the debate about governance issues in the country's public sphere. There is a risk however that the APRM will be misinterpreted as a one-off event rather than a process, and this is rightly identified as a major issue worthy of wary (Chikwanha, 2007: 6).
Finally, the public discourse about governance is spurred on by contributions from international organisations as well. With their greater focus on governance issues, the international donors have, over the past 10-15 years, developed various governance evaluation tools (OECD, 2008). Donor reports based on the results of these assessments also boost the governance debate -even though their methods and intentions often prompt well-founded criticism in some cases (Afrimap, 2007, for example). Besides the tools to analyse government policy that are utilised directly as a basis for donors' operational commitment, various international institutions are attempting to develop measurable and quantifiable indicators to score developments in the field of democratisation or quality of governance (see Arndt  Overall, then, external monitoring and review processes are an important source of information for domestic actors and provide points of reference for political discourse. They can also bring external pressure to bear on leaders to react more strongly to the call for good governance. However, they are no substitute for the domestic demand for a form of governance that is in line with the will of the people. Internal checks and balances play a crucial role in sustainable political reform processes, and it is exciting to see that these external and internal processes have increasingly converged and, in many cases, have intermeshed.

Prototypical developments: Ethiopia, Ghana and Zambia
From the numerous analyses of the reality of governance in Africa, a clear picture emerges, despite some variance on points of detail: the quality of governance is evolving at very different rates in sub-Saharan Africa. Four groups of countries can be identified here (rationale and examples will follow below): • Countries with democratic governance, the quality of which has generally improved over the past ten years; • Countries with limited democratic governance whose development has fluctuated considerably over the past ten years; • Countries with authoritarian governance; • Countries with weak or failing governance (failing / failed states). This latter group is a special case in Africa, characterised by ongoing stagnation or setbacks, and will therefore only be touched on peripherally here. The other three groups represent different types of political transition processes in sub-Saharan Africa. Ghana, Zambia and Ethiopia can be regarded as prototypes for the three groups. A comparison of the results of the Worldwide Governance Indicators (WGI) for each of these three reference countries reveals marked differences between them.8 Since the start of the surveys in 1996 to the present, Ghana has performed best in the six WGI dimensions. Zambia has only been able to achieve similarly good scores in the political stability dimension, in 2006-2007. Ethiopia brings up the rear in four out of six dimensions, achieving broadly similar scores to Zambia only in relation to government effectiveness and control of corruption; however, both trail far behind Ghana. Zambia occupies the classic midfield, with fluctuating scores between the two extremes and with no discernable trends: instead, there are improvements, deteriorations or stagnation from year to year. Overall, it is clearly positioned between Ghana and Ethiopia. The three-way split is immediately apparent in the voice and accountability dimension, which is particularly relevant to the issue of democratic governance.9 Ghana's scores lie well above Zambia's and Ethiopia's and have continuously improved over the past decade. Zambia's scores lie mid-range and fluctuate without any discernable trend. Ethiopia, which has trailed behind Ghana and Zambia from the outset, actually shows a negative trend: its scores for voice and accountability have worsened in the years since 2000.
A comparative discussion from a quality perspective is necessary to enhance the quantitative approach. Using the above mentioned dimensions of democratic governance, the three case studies reveal the clear variance in these dimensions.10

Legitimation of governmental power
Regular elections take place in all three countries, albeit with great variations between them. In Ethiopia, the conditions for free and fair elections are judged to be very weak. There was general consensus among international observers that the 2005 parliamentary elections were manipulated to such a massive extent by the government that they could not be described as democratic (Bruene, 2007). Protests and demonstrations against election-rigging were brutally crushed by the government, resulting in several hundred deaths and more than 3000 arrests. More than 100 leading opposition politicians, including elected Members of Parliaments, were kept in prison for about two years. There is no semblance of democratic legitimation of governance through elections or a democratically elected parliament. In Zambia, the situation is infinitely better, although here too, electoral fraud and ballot-rigging have regularly occurred in the past. According to election observers from SADC, the AU and EU, the most recent presidential elections in 2008 were free and fair, although the main opposition party protested about alleged ballot-rigging. Ghana leads the field when it comes to the democratic quality of its elections: it is far removed from the electoral fraud and manipulation widespread in other African countries, although incumbent presidents are regularly criticised for exploiting their position in government for party-political purposes (an accusation which is familiar from many countries around the world as well). However, there is no direct ballot-rigging (the few exceptions prove the rule), and a free and highly critical media landscape and dynamic civil society facilitate open pre-election debate. Above all, however, legitimation through elections in Ghana is underscored by the fact that it is one of the few countries in Africa where governments have not only been voted out of office but have actually vacated the field for the new incumbents (in 2000, 2008). The prospect of a peaceful change of government through elections can be regarded as the quality mark of democratic legitimation of governmental power.

Legitimation of governmental politics through transparency
Comprehensive information about governmental politics is the prerequisite for popular participation in political processes but also for scrutiny of governance. Transparency creates a public sphere and a political public. Here, there are substantial disparities between the three countries. Ghana's political public is strong. The government has initiated the establishment of various legal frameworks to improve the transparency of government action (cf. APRM, 2005: 30 f.). A diverse and largely independent media landscape helps to create a critical political public. In the Reporters without Borders Press Freedom Index (recent years), Ghana is achieving very good scores. A very active civil society is now firmly established vis-à-vis the government, especially in urban areas, although less so in rural regions. Various civil society actors are creating more transparency in government spending policy, for example, through their monitoring activities such as SEND on debt relief spending and ISODEC on budgeting (Apusigah, 2009a,b). The APRM Report for Ghana emphasises the efforts being made to enhance transparency, but also identifies a number of existing weaknesses. Besides a lack of transparency in ministries, departments and public sector agencies, transparency of decentralised local government is still very weak and in need of a major overhaul (APRM, 2005: xv, 60f.). Nonetheless, the fact that Ghana's government was prepared to become the first African country to submit to the APRM process demonstrates the political will to subject its own conduct to scrutiny and thus create transparency. This in itself distinguishes Ghana from Ethiopia and Zambia, which in 2010 -five years after the publication of Ghana's APRM report -were still at the start of the review process. A lack of transparency in governmental politics is a particular point of criticism in Ethiopia, where important political decisions affecting the country are taken by a small group of party intimates and rubber-stamped by the cabinet. Admittedly, the government has increased reporting on its politics in recent years but

Legitimation of governmental politics through participation
There is a high level of participation of civil society actors in political processes in Ghana and Zambia. In both countries, civil society actors have played an active role in the preparation of the current national development strategies and in monitoring these strategies. In budgetary policy and sectoral political processes, too, civil society actors play a discernable role.
Nonetheless, there is clear scope for improving the opportunities for civil society participation in both countries, as civil society actors themselves, but also independent observers, have noted (for Ghana, cf. APRM, 2005: 52 and Apusigah 2009; for example; for Zambia: various articles in Eberlei, Meyns & Mutesa, 2005). In both countries, civil society participation at local level in particular remains poor, and the permanence of civil society participation even at national level is by no means secure, as the relevant institutional and legal frameworks are still too weak. Nonetheless, compared with the situation in these two countries ten years ago, for example, the opportunities for participation have greatly increased. The same cannot be said of Ethiopia. Admittedly, during a phase of political liberalisation (from 2000-2005 or thereabouts), the government created some political scope for participation (e.g. in Ethiopia's PRS process) but this has been substantially curtailed since the 2005 elections and the ensuing political unrest. Ethiopia's civil society currently has no opportunity for effective participation in political processes. On the contrary, following the introduction of an even more restrictive legal framework for NGOs in 2009, it must expect to encounter growing repression.
In all three countries, the participation of women in political processes is weak. Even Ghana -a model in many other respects -shows clear deficits here: women still play a very subordinate role in politics (Allah-Mensah, 2005). This applies to the national parliament and district assemblies, government office and the higher echelons of the civil service, such as the District Chief Executives (DCEs) (the most powerful figures in the administration of the districts) who are appointed by the President, and the political parties. With the parliamentary elections in 2008, the number of women in politics actually decreased further, with only 18 women -instead of the previous 25 -now sitting in the national parliament. Some progress on the gender front was achieved in January 2009, when a women was elected Speaker of the Parliament of Ghana for the first time (and hence to the third-highest public office in Ghana).

Legitimation of governmental politics through accountability
As shown in the previous paragraphs, in the case of Ethiopia -with its lack of transparency and opportunities for participation, as well as a parliament which serves as little more than a ceremonial institution -domestic accountability of the government cannot be said to exist. A different situation obtains in Ghana and Zambia. Although their parliaments show numerous weaknesses, their supervisory and control functions have been strengthened in recent years, notably -as mentioned -in the context of the annual budget debates. The pro-active work of civil society in both countries also forces the government to report regularly on its activities.
To sum up, then, a comparative debate about the quality of accountability in the three countries can do much to refine, differentiate and generally underpin the quantitative overview presented above. The three cases selected are emblematic of their respective groups of countries, which vary considerably in the development of governance.

Better Governance, Better Results?
Democratic governance can be measured in terms of compliance with principles. However, the ultimate test of governments must surely be the extent to which their work produces outcomes and results that benefit the large majority of their populations: legitimation through results. It was Amartya Sen (1999) who argued that democratic governance is the best prerequisite for governance that serves the majority of the population. Scholars have recently produced complex quantitative analyses to support this view. Drawing on data from the year 1975 to 2000, from a sample of more than 80 developing and industrialised countries, Faust (2007) provides empirical support for the assumption that increasing levels of democracy have a positive impact on economic performance. He calls this "democracy's dividend". Berg-Schlosser (2008) produces similar findings from an analysis of data for African countries in the 1990s -and concludes by explicitly endorsing Sen's position. These observations need further validation, however, they underpin the hypothesis that governance matters -or, more precisely, that democratic governance matters.