Manipulating Political Decentralisation

Africa's Inclusive Autocrats

Lovise Aalen, Ragnhild L. Muriaas

• Publisher: Taylor & Francis
• Year: 2017
• eBook ISBN: 9781315472393
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: African Politics
• Index: Politics & International Relations

Can autocrats establish representative subnational governments? And which strategies of manipulation are available if they would like to reduce the uncertainty caused by introducing political decentralisation? In the wake of local government reforms, several states across the world have introduced legislation that provides for subnational elections. This does not mean that representative subnational governments in these countries are all of a certain standard. Political decentralisation should not be confused with democratisation, as the process is likely to be manipulated in ways that do not produce meaningful avenues for political participation and contestation locally.

Using examples from Africa, Lovise Aalen and Ragnhild L. Muriaas propose five requirements for representative subnational governments and four strategies that national governments might use to manipulate the outcome of political decentralisation. The case studies of Ethiopia, Malawi, South Africa, and Uganda illustrate why autocrats sometimes are more open to competition at the subnational level than democrats.

Manipulating Political Decentralisation provides a new conceptual tool to assess representative subnational governments' quality, aiding us in building theories on the consequences of political decentralisation on democratisation.

1
Introduction

Under the rule of the Ethiopian People’s Revolutionary Democratic Front (EPRDF) (1991–), an altered definition of a verb has been introduced in Ethiopia. The verb ‘to participate’ normally means that a person actively takes part in or shares an activity with others based on that person’s initiative. According to anecdotes, the altered Ethiopian definition of the verb relegates the person who participates as a passive receiver of an order to take part. Instead of saying that ‘we must be open for participation’, the new saying among party cadres and government officials is rather telling: ‘we must participate them!’ This linguistic invention is a conspicuous example of how participation is understood in an autocratic regime, which uses mass mobilisation as an instrument for extending and sustaining its powers. In Ethiopia today, the population is participating in government affairs on a scale never before seen in the country’s history. However, the regime is still known as one of the most autocratic in Africa today and has a low score on both national and local democracy ratings. From the national level all the way down to neighbourhoods, citizens are invited to participate by voting for candidates in elections. From the EPRDF’s point of view, a high voter turnout is a better measure of success in elections than multiparty competition. At the grassroots and household levels, people are ‘participated’ into community groups for local development work, where a failure to participate is seen as a sign of resistance. Therefore, the sheer number of participating citizens becomes an indicator of the regime’s capacity and will to organise instead of an opportunity for citizens to influence government affairs.

In Malawi, the situation is a stark contrast to the massive participation that takes place at a local level in Ethiopia. Local government offices are not bustling. On the contrary, they appear more like empty shells in comparison. The local government structures that were established in districts and towns in 2000 were already dissolved in 2005 when the government issued a radio message saying that the local assemblies had been suspended. At an old assembly house in Mangochi, a town in Malawi’s southern region, someone in the early 2000s stuck a sticker on the wall stating: ‘Register now to vote in Malawi’s 1^st Tripartite Elections (May 2004).’ This turned out to be an ironic message, as tripartite elections or local elections were cancelled or postponed seven times until they were finally held in 2014. In the meantime, no local councillors were elected, and there were no subnational arenas where ordinary Malawians could participate or compete. When elections were finally held, the number of seats over which candidates could compete was reduced to nearly half – a total of 462 seats across the country. In comparison, there are 257 Traditional Authorities in Malawi, which is the highest level of a three-tiered chieftaincy system. Traditional Authorities are non-voting members of local assemblies. Members of the Malawian Parliament gained voting rights in local assemblies after an amendment to the local government act in 2010. In addition, five non-elected local stakeholders are also supposed to meet – but not vote – in the assemblies of each 28 districts in the country. In conclusion, there are 128 fewer members elected through local elections in Malawian assemblies than other kinds of representatives or appointees. Opportunities for participation and contestation at the subnational level, to say the least, are not overwhelming for the ordinary Malawian. Yet, the country is commonly classified as a minimalist democracy with competitive elections held on a regular basis at the national level.

When comparing the meagre decentralisation efforts in democratic Malawi with the apparent mass participation in autocratic Ethiopia, it is evident that there is no clear relationship between the level of democratisation and decentralisation. This is further underlined by the massive system of subnational elections in another authoritarian state like Uganda. Data presented in this book shows that both democratic and autocratic regimes may have high local government scores, and similarly, other autocracies and democracies may score low on the same index. This fuzzy relationship between regime and decentralisation contradicts with a common assumption behind decentralisation reform – that decentralisation enhances democracy – and motivates us to look deeper into the mechanisms behind local governance.

The book aims therefore at unpacking what happens in the wake of introducing political decentralisation reforms in terms of contestation and participation at a local level. We develop a set of tools to analyse these processes: five requirements that are needed for a system to be classified as a representative subnational government, and four strategies that regimes may use to manipulate the formal institutions of Subnational Governments (SNGs). In this way, we are better equipped to understand how political decentralisation is implemented or avoided across regime types. The findings from our four case studies, representing variation in both local government score and regime type, underline the main message of the book: that decentralisation cannot be confused with democratisation and that autocrats can be inclusive and sometimes even open for competition at the subnational level.

In the post-Cold War era, multiparty elections have become routine in both autocracies and democracies. The mere conduct of multiparty elections are thus no longer seen as an indicator upon which to decide whether a regime can be classified as authoritarian or democratic. As noted by Gandhi (2015: 446), elections appear less useful in distinguishing between democracies and autocracies but are more useful in characterising different types of autocracy. We contribute to this line of thinking by arguing that the scholarly literature on regimes should also incorporate the characteristics of variation in regimes’ subnational inclusiveness. Our point of departure is that subnational elections must give the national government something more than just legitimacy. Subnational elections must contribute to sustaining their power. Yet, although we know that subnational elections increase uncertainty as they multiply opportunities for participation and contestation in a country and that few regimes can rely on suppression as a meaningful way of controlling the population over a longer period of time, what is it that representative SNGs are able to deliver to certain autocrats? And which kinds of manipulation strategies are available on the one hand to reduce uncertainty, and on the other hand, multiply arenas for participation and contestation?

The motivation for writing Manipulating Political Decentralisation is the observation that autocracies with representative SNGs have not been discussed sufficiently in the mainstream literature on autocracy and democracy. We compare democratic and autocratic cases that have all introduced political decentralisation in the 1990s but have since developed different levels of political decentralisation and systems of representative SNGs. We look specifically at authoritarian regimes that are characterised by giving the electorate the opportunity to participate in elections at multiple levels of government through policies of political decentralisation. In the process of studying representative SNGs in an authoritarian setting, we noted a key challenge. We saw that it was possible for a regime to implement all institutional requirements for having representative institutions locally but that the element of oppression made it difficult to tell when such institutions are meaningful and when they are not. Hence, it was possible for two different authoritarian regimes to fulfil the requirements of having representative institutions at the subnational level, but the distortions of these institutions would vary so much between autocracies that researchers would find the one case as having quite strong locally empowered subnational units, whereas the other does not. We were thus curious about finding out what kind of dynamics were operating in these contexts that put them apart from each other.

The first step in exploring these differences would be to assess whether such apparently similar sets of institutions were indeed so similar when they were tested and compared along a common standard. Hence, we had to develop a set of institutional requirements based on theoretical literature on democratisation and political decentralisation. In this way, we could justify our claims that two cases look alike and actually have established representative SNGs. A second step would be to examine whether and how different national governments manipulated the institutions of SNGs. We assumed that through manipulation, the uncertainty of subnational elections is likely to be reduced and even help to sustain the ruling party. Still, we wanted to identify different ways of manipulating political decentralisation reforms and be open to the possibility that this list could go beyond just avoiding implementing institutional requirements. It is one thing to establish mock institutions, and another thing to manipulate representative institutions – and both opportunities should be explored as they might matter for democratisation. What we found is that the space for meaningful contestation and participation is determined by the level and kind of manipulation in each case. This indicates that although having representative SNGs in an authoritarian regime can be conducive for democratisation, surely it does not have to.

Furthermore, we have found it relevant to compare the processes of introducing – and manipulating – political decentralisation in autocracies with similar examples from democracies. Within the category of democratic regimes that have introduced political decentralisation, there is also a variation in terms of how SNGs open up for meaningful contestation and participation. In some instances, decentralisation reforms have remained unimplemented, and the process of political decentralisation should be considered flawed; whereas in other cases, reforms have been implemented and the requirements for representative SNGs are fulfilled. Thus, elements of manipulation can also be found in democracies. Dominant party regimes in democratic countries can even exhibit similar strategies of manipulation to those found in autocracies. It is, however, likely that in democratic regimes, compared to autocracies, there will be more institutional checks and balances present, limiting governments’ opportunities for manipulation.

The key questions addressed in Manipulating Political Decentralisaton are: Why do authoritarian, as well as democratic governments, multiply arenas for electoral competition by establishing representative SNGs? How can such institutions, although in theory open to the local citizenry, operate in an authoritarian setting? What repertoire of strategies is available for manipulating representation in SNGs across regimes?

To answer these questions, we first identify a set of strategies that national governments can deploy for manipulating SNGs. This list is supposed to expand the scope of manipulation opportunities available for national governments to illustrate the shortcomings of reducing the list of alternatives to only include dominance and mock institutions. Depending on the national government’s own characteristics and the political topography of the country, we single out four ways to manipulate political decentralisation. Although none of these strategies proves efficient over time and in all contexts, they can all be used by some governments to reduce the challenges from opposition after decentralisation reforms have been introduced. These four strategies are 1) creating institutional gaps, 2) national dominance, 3) overwhelming the opposition, and 4) regionalised opposition enclaves. The first of these strategies manipulates the process to such an extent that the process of political decentralisation must be identified as flawed. The last three strategies are ways to strengthen the position of already strong national governments. Hence, we assume that the first strategy is most likely to be deployed where the opposition is relatively strong, whereas the three last strategies are more likely to be used where the national government is already dominating national politics.

Second, we develop a methodological tool to distinguish between systems with or without representative SNGs. This is done through the identification of the five requirements. With the realisation that both democracies and autocracies can be decentralised – and engage in the process of manipulating the institutions introduced to establish avenues for contestation and participation locally, our analytical tool for measuring representative SNGs can be used across regime types. Our tool consists of five requirements that a system needs to fulfil. First, elections to local-level legislatures should be based on universal suffrage. Second, these elections should be held on a regular basis. Third, there is at least minimal pluralism and some openness for opposition candidates to stand. Fourth, the election results have to be the cause of executive appointment. Fifth and last, the subnational assembly should be composed of only elected representatives, excluding non-elected members. However, it is important to note that governments in autocracies do not multiply their arenas of political participation and contestation to moderate the authoritarian characteristics of the regime. On the contrary, they are likely to see representative SNGs as a means to sustain and consolidate their power on the national political scene. Thus, political decentralisation can be understood as an instrument that can help mobilise support for the regime and control local elites. However, as discussed extensively below, not all authoritarian governments have the same incentives and/or are equally equipped to use representative subnational elections as a tool to sustain the power of national regimes. Similarly, democratic regimes may have related incentives for attempting to manipulate or simply boycott implementing decentralisation.

Third, we present in-depth case studies of four African regimes that have all introduced political decentralisation reforms in the 1990s but which vary in terms of regime type and local government. We have selected two with high scores on the Varieties of Democracy’s local government index – Uganda and South Africa – and two with low scores – Malawi and Ethiopia (VDem, 2017). The two high score countries represent different regime types, democratic South Africa and autocratic Uganda. Similarly, the two low score countries include one autocracy, Ethiopia, and one democracy, Malawi. These scores are surprising; as for instance the Ethiopian government, without thorough examination of requirements, has apparently taken quite solid steps in establishing representative SNGs. Hence it may fail due to the obvious authoritarian features of the regime. But if we accept that, why should autocratic Uganda still have such a high score? Consequently, Ethiopia may – despite having established SNGs – fail in passing the ‘five institutional requirements’ test – similarly to Malawi. Our case selection is motivated by this puzzle – why should an autocratic country, where there is massive opportunities for participation at the subnational level fall into the same category as a minimally democratic country with almost no avenues for such? And if there is a high score on the local government index in an authoritarian regime like Uganda, what does participation and contestation look like in this context compared to a democratic country with high scores on the same index like South Africa?

Elections in Authoritarian Regimes

The role of elections in authoritarian states has emerged as an important topic within the literature on regimes during the last decade. Manipulating Political Decentralisation contributes to the ongoing debate by focusing on why some authoritarian governments take a step further and institutionalise representative SNGs even if national elections would suffice to demonstrate the governments’ invincibility and create international leverage. Schedler (2013: 22–23) argued that autocrats are notoriously insecure, and elections always bring a certain element of uncertainty. Therefore, why would autocrats engage in such hazardous acts as multiplying the arenas for participation and contestation if they could avoid it? Which authoritarian governments would avoid holding elections and which would typically not? Within the scholarship on elections in authoritarian regimes, the idea that subnational elections would help produce democratisation seems overly optimistic. Thus, we have to develop alternative explanations for why some governments introduce representative SNGs and even make them persist whereas others do not. At the same time, we should not rule out that such a synergy between elections and democratisation may happen in certain contexts.

The emerging literature on regime stability takes as a prerequisite that all politicians would like to remain in power if they can and that they strategise accordingly. Elections in authoritarian regimes are a consequence of autocrats’ wishes to stay in power. Therefore, we must assume that elections are held because autocrats believe that they can help them retain power (Gandhi and Lust-Okar, 2009). The explanation provided by Gandhi and Lust-Okar (2009) is that elections can neutralise groups that could otherwise pose a threat to the regime. Through elections, the autocrats can co-opt potential challengers to the regime by offering spoils through promising legislative seats or boosting credibility by promising to share power. Furthermore, in addition to co-opting rivals, elections may be a source of legitimacy, a way to deter opposition and to learn about the political opposition’s strength and standing in the broader population.

However, the great downside of elections is that they may lead to loss of power at the polls to a coordinated opposition, or trigger protests and coups. As Gandhi and Lust-Okar (2009) highlight, elections are likely to be useful in sending and receiving information signals, not just about the strength or weakness of oppositions, but also on the governing party itself. If efficient, the mobilisation of the ruling party’s apparatus, its campaign machinery and display of nationwide support send signals of regime strength. If an incumbent party is able to show its numeric strength, armed opposition against the regime appears futile, and internal...

Table of contents

• Cover
• Title
• Copyright
• Contents
• List of Figures
• List of Tables
• Series Editors’ Foreword
• Acknowledgements
• List of Abbreviations
• 1. Introduction
• 2. Four Strategies of Manipulation
• 3. The Requirements of Political Decentralisation
• 4. Decentralisation Through Dominance in Ethiopia
• 5. Institutional Interruption in Malawi
• 6. Overwhelming the Opposition in Uganda
• 7. Regionalised Opposition in South Africa
• 8. Conclusions
• Index