Scholarly article on topic 'Seeing Political Settlements through the City: A Framework for Comparative Analysis of Urban Transformation'

Seeing Political Settlements through the City: A Framework for Comparative Analysis of Urban Transformation Academic research paper on "Social and economic geography"

CC BY
0
0
Share paper
Academic journal
Development and Change
OECD Field of science
Keywords
{""}

Academic research paper on topic "Seeing Political Settlements through the City: A Framework for Comparative Analysis of Urban Transformation"

Development

and Change Published on behalf of the Institute of Social Studies, The Hague

Seeing Political Settlements through the City: A Framework for Comparative Analysis of Urban Transformation

Tom Goodfellow

ABSTRACT

This article presents a new analytical framework for studying cities in the developing world based on the 'political settlements' approach. This has a dual purpose: to enrich comparative urban research by bringing new theoretical ideas to bear on this field, but also to use capital cities as a lens to better understand national political settlements. The central argument is that urban built environments and their transformations in situations of late development reflect the workings of different varieties of clientelism, and by analysing the former we can better understand the latter. Specifically, issues such as the nature of urban land use and land allocation, the pace and form of construction, the effectiveness of environmental regulation and the provision of housing for different income groups are all revealing of political settlements and their broader development implications. The potential of this approach is explored through three narrative 'sketches' of contemporary urban development in Eastern Africa: the 'city as marketplace' (Kampala), the 'city as expo' (Kigali) and the 'city as construction site' (Addis Ababa). In presenting this framework, the article seeks to advance debate on epistemo-logical and analytical approaches to the study of both power relations and differential patterns of urban development.

INTRODUCTION

This article proposes a new analytical approach to comparative urban development research, building both on the renaissance in comparative urban thinking and growing interest in 'political settlement' analysis in development studies and political economy. It argues that the political settlements approach — which despite its growing influence has devoted little attention thus far to space, scale and subnational governance — holds potential to

The author would like to thank two anonymous reviewers and Ezana Haddis Weldeghebrael for their very helpful and constructive comments on earlier drafts of the text. All errors are entirely my own.

Development and Change 49(1): 199-222. DOI: 10.1111/dech.12361

© 2017 The Authors. Development and Change published by John Wiley & Sons Ltd on behalf of Institute of Social Studies.

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

illuminate how varying relationships between political power and institutional structure shape cities across the global South. In so doing, it aims to bring new interdisciplinary perspectives to bear on the field of comparative urbanism, and to enrich the political settlements approach itself. Urban political science has been slow to embrace comparison (Kantor and Savitch, 2005; Pierre, 2005), especially beyond the confines of high-income countries. This is despite the flourishing of comparative urban work within geography, but also despite the proliferation of comparative national-level approaches to the political economy of development, including political settlement analysis. Bringing together these trends, this article advances an interpretation of political settlement analysis that aims to generate new insights into the drivers of diverse urban transformations occurring across the developing world.

In recent years, scholars such as Ward (2008), McFarlane (2010) and Robinson (2011,2016) have highlighted a resurgence of interest in comparative urbanism with particular attention to developing countries, drawing on post-colonial theory and the literature on policy mobilities (McCann, 2011). This work, which I refer to as the 'new comparative urbanism', emphasizes that all cities are starting points for theory building, eschewing ideas of 'paradigmatic urbanism' and rejecting the notion that everything should be measured against how things work in US or European cities (McFarlane, 2010; Robinson, 2006, 2011). This marks an exciting turn in urban scholarship, exploding conventional global North/South divides and promoting 'new geographies of theory' (Roy, 2009). Its greatest contribution thus far has been conceptualizing comparison in new ways; although empirical comparisons inspired by this work are now growing in number, for the most part comparative empirical analyses of specific cities have so far been limited, with the comparative element often remaining implicit. For example, of the 30 articles included in a 2014 Virtual Issue of the International Journal of Urban and Regional Research on 'comparative urbanism', few explicitly compare specific cities, and only three involve direct comparisons where one or more of the cities is outside the global North.1 There is thus still relatively little by way of research on cities in developing countries predicated on systematic comparison.

Comparative research is also thriving in the study of the political economy of development, providing further opportunity for interdisciplinary cross-fertilization of urban research — if we are prepared to overcome scalar conventions. Indeed, Tilly's (1984) exhortation to understand the interactions of political processes at different scales is as important as ever. Separating the city from other scales can obscure the importance of 'national infrastructures' for urban analysis (Sellers, 2005) — especially in parts of the global South where the autonomy of municipal governments and mayors

1. This Virtual Issue was edited by Jennifer Robinson; see: http://onlinelibrary.wiley.com/ doi/10.1111/1468-2427.12171/full

tends to be much more constrained than in the Northern contexts that gave rise to most existing urban theory (Resnick, 2014). Thus while Robinson (2006) rightly argues that we should not allow urban studies to be colonized by ideas of 'cityness' from the global North, we should equally not allow urban studies to become too insular, depending only on ideas developed at the city level. This is particularly true of capital cities, where national and municipal 'infrastructures' of governance are often intertwined — and it is capital cities that are the primary focus of this article.

The political settlements approach, associated particularly with the work of Mushtaq Khan (1995, 2010) and explicated in more detail below, offers one way forward. This approach focuses on the tripartite relationship between formal institutions, informal institutions and the distribution of power in a given setting, and how economic structure underpins this relationship in contexts of late capitalist transition (Gray, 2016). It offers complementarities with existing urban theory while also holding the potential to advance understanding of different formal-informal institutional configurations in cities of the global South. While extant urban scholarship offers many insights into institutional and power structures, when it comes to cities of the developing world there is a risk that generic treatments of informality and clientelism homogenize the 'view from the South'. Significant differences in the way contemporary cities are evolving suggest that despite the insistent, neoliberal march of globalization, nothing is inevitable about 21st century urban development in the South. Such variation demands exploration. What the political settlements framework offers urban analysis is therefore a way to comparatively analyse the many varieties of clientelism and formal-informal interface in developing countries, and their concrete manifestations in situations of rapid urban growth and transformation. Meanwhile, an urban focus also offers something to the political settlements approach: a fruitful new empirical domain. For example, through studying changes to (and bargaining over) land use and the built environment, we can learn a great deal about the underpinning political-institutional situation.

Within a broader potential research agenda for the application of political settlements analysis to cities (outlined in the conclusion), this article is primarily concerned with explaining striking differences in cities' evolving built environment through a political settlements lens. In the first section, the article outlines the main tenets of the political settlements framework. This is followed by an exposition of how it can be applied to the evolving physical form of cities, and how the approach relates to some other relevant bodies of urban theory. Following this, the analytical potential of the approach is briefly explored through three short narratives of rapidly growing cities in Eastern Africa: the 'city as marketplace' (Kampala), the 'city as expo' (Kigali) and the 'city as construction site' (Addis Ababa). Finally, a conclusion offers some tentative thoughts on how a broader research agenda might be taken forward.

THE POLITICAL SETTLEMENTS APPROACH TO THE POLITICAL ECONOMY OF DEVELOPMENT

The political settlements framework developed by Khan (1995, 2010) is rapidly gaining traction in debates on the political economy of development (Di John and Putzel, 2009; Gray, 2015, 2016; Sen, 2013; Whitfield et al., 2015). The approach was developed in explicit contradistinction to the 'new institutional economics' (NIE) associated with scholars such as North (1990) and Acemoglu and Robinson (2012), which continues to characterize the mainstream accounts of economic transformation in the developing world. Central to the critique is the failure of NIE to engage fully with informal institutions and how these relate to formal ones (such as legally enshrined property rights), as well as how the specific nature of late capitalism shapes the power relations that underpin institutional arrangements.

Taking a lead from the NIE they critique, institutions are defined in this approach as 'the rules and norms that govern behaviour' (Di John and Putzel, 2009: 6). Yet unlike authors such as Acemoglu and Robinson (2012) whose treatment of power is largely decontextualized from economic structure, the political settlements approach is rooted in historical materialism. It thus draws our attention to the fact that formal institutions characteristic of developed democratic capitalism — for example the rule of law, secure property rights and transparent judicial processes — are a reflection of the evolution of capitalist societies. They exist to support capitalism, and 'work' in such societies because they reflect a power structure shaped to a significant degree by capitalist profits (Gray, 2016). Khan argues that in developing countries, although formal capitalist institutions are usually in place (often promoted by donors), the productive sector and capitalist foundations of the state are much weaker. Many powerful groups instead draw their power from sources such as control over land, coercive capabilities or traditional authority, and cannot maintain this power through formal capitalist rules and rights. This leads to a structural 'mismatch between the scale and productivity of activities protected by formal institutions and the distribution of power' (Khan, 2010: 30). In other words, formal institutions in most developing countries often do not serve the interests of the powerful very well. In such situations, informal institutions play a critical role in promoting political stability by addressing this mismatch and ensuring that benefits accrue to powerful groups who would otherwise mobilize to (violently) contest the institutional structure (Gray and Whitfield, 2014).

This focus on how context-specific power relations are maintained through particular configurations of formal and informal institutions is one of the distinctive contributions of the political settlements approach. For Khan, a political settlement involves 'an institutional structure that creates benefits for different classes and groups in line with their relative power' (Khan 2010: 20). Any situation in which there is not large-scale conflict (which would indicate the breakdown of a pre-existing political settlement) reflects

a political settlement of some kind; yet these vary in relation to a number of factors. One is the degree to which formal relative to informal institutions do the 'work' of maintaining stability by allocating benefits to the powerful. Rather than the material foundations of power distribution, it is this focus on the balance of informal versus formal institutions to allocate benefits that this article explores through the lens of capital cities.

If formal institutions refer to laws, regulatory frameworks and government policies, informal institutions are rules that are 'systematic enough to be identified', but not officially written down or implemented by the state (Khan, 2010: 10). Informal institutions, particularly in the form of clien-telistic relations through which economic rents are exchanged for political support, tend to be viewed in NIE as a hindrance to development. For political settlement theorists, however, such institutions are critical for ensuring the political stability on which sustained economic development is predicated. In his more recent collaborative work, North has acknowledged the importance of informal clientelistic institutions for generating stability in this way (North et al., 2009). However, the 'access order' framework developed by North and his colleagues fails to acknowledge how structural economic conditions produce historically contingent power relations, which in turn give rise to different configurations of formal and informal institutions in situations of late development (Gray, 2016).

Informal institutions exist in developed capitalist economies too, but here it is the formal ones — which have evolved over time to reflect the profit-making interests of capitalists — that do the primary work of maintaining the political settlement. By contrast, in most developing countries, even if the same formal institutions are in place, they are frequently subverted by clientelist informal institutions that serve better to allocate benefits to the powerful. This does not mean formal institutions are irrelevant, because they still play a role in shaping the broader institutional environment: indeed informal institutions often comprise 'adaptations to the ways in which particular formal institutions work' (Khan, 2010: 1).

Khan's overarching four-way typology of political settlements involves two varieties in which formal institutions are well aligned with existing power relations: capitalist political settlements (which broadly include most developed economies) and pre-capitalist political settlements (encompassing various forms of feudalism). Then there are two varieties of political settlement in which formal institutions are poorly aligned with significant elements in the distribution of power: 'clientelist' political settlements (in which formal capitalist productive rights exist but are not well enforced, and powerful groups can influence outcomes regardless of formal rights) and 'political settlements in crisis'. The latter refers to situations where formal institutions have virtually collapsed and the settlement is being violently contested, while the former encapsulates a wide range of developing countries not in large-scale conflict. The category of clientelist political settlements is thus extremely broad, and can be broken down further by exploring how

particular institutional configurations are supported by a context-specific balance of power.

In the political settlements framework a distinction is commonly made between the ruling elite — the top political leaders in an incumbent regime — and the ruling coalition—the factions that broadly support this ruling elite (Kj^r, 2015). Differences in the distribution of power are conceptualized by Khan (2010: 64) both horizontally (the relative strength of elite factions outside the ruling coalition) and vertically (the relative strength of factions at lower levels within the ruling coalition). To this, Whitfield et al. (2015: 98) have added a focus on the degree of cohesion within the ruling elite.

Khan subdivides the coalitions underpinning clientelistic political settlements again into four types, which reflect the differential capacity of factions to command resources 'off-budget' in the form of patronage and clientelistic favours. The categories (discussed at length in Khan, 2010: 64-69) are 'potential developmental', '(vulnerable) authoritarian', '(weak) dominant party' and 'competitive clientelist'. Khan also further categorizes clientelis-tic political settlements in terms of the power and capabilities of emerging productive entrepreneurs relative to other social groups (ibid.: 69-75). The power distribution among political factions, combined with the relative power of productive entrepreneurs in a given setting, is central to Khan's explanation of differential economic growth outcomes.

As Khan acknowledges, the dimensions of clientelistic political settlements outlined above 'by no means exhaust the different aspects of the distribution of power that can affect institutional performance' (ibid.: 75). The intention here is not to rigidly apply these categorizations at the level of the city, but to explore the potential for using the broader framework to understand the relationship between power distribution and urban outcomes other than economic growth. What the framework facilitates is an exploration of the institutional basis for different varieties of clientelism — paralleling in some respects the 'varieties of capitalism' literature focused on the global North (see Hall and Soskice, 2001). Before exploring how urban analysis and the political settlements framework intersect, however, we need to consider further the concept of power which is so central to the framework.

Conceptualizing, observing and measuring power is notoriously difficult. Proponents of the political settlements approach use Khan's concept of relative 'holding power', which has been conceptualized as the ability of individuals and groups to assert or maintain claims to the ownership of property and income flows (Gray and Whitfield, 2014: 11). 'Holding power' is therefore a specific conception of power that links it to the capacity to acquire and maintain material benefits.2 This relative holding power can be exercised through 'violence or the threat of violence, ideas, sociological and

2. Note here that 'holding power' is a compound noun, rather than a verb followed by a noun; i.e. an individual or group has 'relative holding power', rather than is holding power.

institutional hierarchies and traditional authority as well as through access to economic resources' (ibid.).

Significant holding power in developing countries can be based on organizational capabilities, for instance of 'intermediate class' elites who organize clients and use them in political confrontations (Khan, 2010: 54). Sources of holding power are inevitably context-specific and determined by historical patterns of conflict and organization, and are often very difficult to observe. However, utilizing a political settlements approach does not require an assessment of the sources of power ex ante, but rather an understanding of some of the outcomes of the distribution of power and the formal-informal institutional configuration through which these are realized. Knight's (1992) point that in any society it is unlikely that the overall institutional structure can produce a distribution of benefits significantly out of line with the distribution of power is central to the political settlements approach, and to op-erationalizing any analysis based on it. Following this logic, we can broadly 'read' the distribution of power by identifying the de facto beneficiaries of institutional arrangements.3 Building on this, the present article is concerned specifically with benefits that manifest in the form of urban land and property, and concomitantly how we can 'read' urban form to better understand the relative power of different groups and how this is institutionally organized. It is to this urban application of the approach that we now turn.

POLITICAL SETTLEMENTS AND URBAN ANALYSIS

Cities constitute institutionally dense and complex environments. Multiple layers of government regulation shape urban life, yet these frequently operate alongside a range of 'twilight' economic and political institutions (Lund, 2006). There is already a rich literature on how engrained and more ephemeral informal institutions affect urban life in cities of the developing world, including with respect to land (see, for example, Bayat, 2007; Rakodi, 2006; Roy, 2011; Simone, 2004; Waibel and McFarlane, 2012; Watson, 2009). Pervasive informality means that many city dwellers' homes

3. This perspective is subject to important criticisms. First, the distribution of benefits may not always map neatly onto the distribution of power, broadly conceived. This is one of the reasons for using a narrower concept of 'holding power', which relates specifically to the degree to which groups can acquire and maintain benefits, enabling a focus on the institutional channels through which this link between power and benefits is realized in different contexts. The capacity of different groups to acquire benefits can also shift over time, both gradually and in more sudden ways. Moreover, not all benefits that accrue to the powerful are visible; like sources of power, outcomes can be difficult to observe. Despite these problems, given that power is largely absent in dominant 'new institutional' approaches, a framework that gets us closer to being able to identify power through a focus on associated outcomes and how they are institutionally distributed is worthy of further exploration.

and livelihoods exist in 'gray spaces' that contravene formal institutions but are not clearly illegal either, resulting in a precarious state of permanent vulnerability (Yiftachel, 2009).

Some of this literature is also explicitly comparative; even before the 'new comparative urbanism' gained momentum, Rakodi's (2006) Special Issue was a notable example of comparative work on formal and informal mechanisms of urban land delivery. However, coming primarily from an urban planning perspective, such work pays little attention to national political dynamics and how struggles for resources among social groups play into formal-informal institutional dynamics in different urban contexts. Urban land and real estate are highly prized, and thus constitute key instruments of benefit allocation for maintaining political coalitions. By applying a political settlements lens to cities, we can explore further how built environments reflect what Simone (2004) calls a 'functional tandem' between different institutional forms which, by allowing benefits to flow to particular groups, remains stable as well as being fluid.

All three components of the basic political settlements framework — formal institutions, informal institutions and distributions of holding power — have particular applications when considered in relation to urban built environments. Relevant formal institutions include national-level legislation on property rights and land tenure, as well as decentralized laws and municipal regulations concerning building control, land use and taxation. Urban master plans and zoning rules also constitute formal institutions, as do government tendering procedures and investment incentives. For the remainder of this article, the term 'formal institutions' is used as a shorthand for all these legal and regulatory instruments and planning frameworks. On the informal side, along with 'traditional' forms of land delivery that are not endorsed by the state, there are clientelistic norms, including the waiving of official rules or systematic non-enforcement of tax payment in exchange for political support, or the regularized use of bribes for access to government contracts or permission to use urban land in prohibited ways.

This leaves the question of the distribution of holding power, as reflected in a particular distribution of benefits. The specific benefits we are interested in here concern access to urban land, and rights to use and acquire income from land and the built environment. Many of the physical transformations that occur in cities are concrete manifestations of the allocation of such benefits, through which we can learn more about prevailing power structures. In capital cities especially, particularly where these are also economically core, nationally significant groups tend to make demands for benefits rooted in the urban terrain, which they are successful in realizing if they have sufficient holding power. Indeed, as Whitfield et al. (2015: 95) note, 'land is still the dominant mode of accumulation in African countries', and urban land in core cities is particularly lucrative, especially when used in highly intensive ways. The physicality of cities thus provides a window — albeit not a fully transparent one — onto how and to whom material benefits are accruing.

Benefits relating to land and the built environment may not be entirely visible, but they are more visible than many other types of benefits such as more obscure financial assets. This empirical focus therefore constitutes an epistemological lens onto 'holding power'.

To operationalize this framework implies gathering data on physical urban transformations such as major construction projects, expropriations, real estate and housing ventures, including with respect to how these are financed and who the beneficiaries are. We can then examine the degree to which such developments are taking place in accordance with the formal rules, in direct contravention of them but supported by recognizable informal norms, or in 'gray spaces' where formal institutions are unclear. This helps to map the contours of the national political settlement in a particular urban context, elucidating who benefits from existing institutions relating to the built environment, and whether these benefits are primarily allocated formally or informally. Analysing how the distribution of material benefits to support this power structure is organized through a particular configuration, formal and informal institutions help to explain social, environmental and land use outcomes that matter significantly for human well-being.

This approach shares elements with existing perspectives in urban studies across different disciplines, but is tailored to explaining the significant differences in formal-informal institutional interface in cities of the developing world and how they affect cities' physical evolution. Like contemporary Marxist urban scholars such as Peck, Theodore and Brenner, it is concerned with how global political economy in a marketized world 'touches down' in differentiated (and path dependent) ways in different urban contexts (Peck et al., 2009). Unlike these approaches, however, it calls attention to the partial and highly circumscribed capitalist development that characterizes many cities in developing countries (Parnell and Robinson, 2012), and the consequent importance of informal institutions. The political settlements approach also offers some complementarities with more agency-centred sociological theories of urban change, including Molotch's (1976, 1993) 'urban growth machine'. The role of'competing land-interest groups' (Molotch, 1976:311) and urban 'growth entrepreneurs' (Molotch, 1993: 32) in pushing for their interests is clearly important in the global South today. However, there is relatively little research on how this plays out in situations where the nascent property development industry is relatively uninstitutionalized.

The political settlements approach also offers some parallels with important strands of Urban Political Science — particularly Urban Regime Theory (URT). Like Khan, Stone is concerned with the 'informal arrangements by which public bodies and private interests function together' in order to make and implement governing decisions (Stone, 1989: 6). Urban regime analysis also emphasizes the 'iron law' that to be effective at governing, a coalition must have the capacity to make 'side payments' to groups negatively affected by particular developments (Stone, 2005: 317). While using different terminology, this corresponds with the attention in the

political settlements framework to the distribution of 'informal incomes like off-budget resources, land, and other types of rents that are created through the exercise of informal political power' (Khan, 2010: 54). In the context of urban planning and land development, such informally allocated 'side payments' include preferential access to land, the waiving of land use and construction regulations, and non-enforcement of local taxation.

Despite these important complementarities, URT rests on a clear divide between 'main street' and 'city hall' as two poles of power that must find ways to collaborate (Stone, 2005: 326). In many less-developed countries, this assumed division barely exists to begin with; a post-colonial fusion of power challenges conventional thinking on public/private divides (Gray and Whitfield, 2014; Migdal, 1988). While URT performed a crucial role inmov-ing urban political science beyond the elitism-pluralism debate (Mossberger and Stoker, 2001), its assumptions about particular forms of developed capitalism do not travel very well beyond the US and some European contexts (ibid.). It has also been criticized for its excessively 'localist' focus (Ward, 1996), assuming that city politics can be somehow 'sealed off' from broader national and global processes (Lauria, 1997: 75). This is a particularly pertinent criticism in relation to the global South (beyond a limited number of cases mostly in Latin America), given the prevalence of central government interference and the blurring of boundaries between national and urban institutions (Goodfellow, 2015; Resnick, 2014). In these respects the valuable attention to coalition formation, informality and 'side payments' in URT could usefully be supplemented by the political settlements framework. Specifically, the attention of the latter to national power dynamics and the variable institutional forms of clientelism is necessary for understanding urban outcomes in situations of late capitalist transition.

These considerations bring us back to the 'new comparative urbanism', and its call for theory generation predicated on existing conditions in the global South. The approach proposed here benefits from the above trends in urban analysis, but also from the literature on clientelism and the formal-informal institutional interface in cities of the global South (Roy, 2011; Waibel and McFarlane, 2012; Watson, 2009). By exploring national political settlements at the city level, this article complements and extends some of these approaches by systematically comparing varieties of clien-telism and analysing the physical manifestation of different formal-informal institutional configurations.

The following sections further elaborate the approach through three 'sketches' from Eastern Africa. The primary material is drawn from semi-structured interviews with government officials, politicians, property owners, community members and architects during research on urban planning, land and other property-related issues in the three countries over a six-year period from 2009 to 2015. A range of literature on the recent evolution of the three cities is also consulted, as well as relevant policy documentation. Building on this empirical base, what follows constitutes an exploration of

how the political settlements framework might be used to analyse significant divergences in the built evolution of three cities. The narratives are necessarily brief and heuristic, intended to indicate the framework's value for generating comparative insights and to spur further systematic research along these lines. Greater detail on each has been sacrificed for the breadth of a discussion encompassing three distinct cases, to challenge the common tendency to collapse difference into binary opposition, and thereby better explore the complex variety of clientelist political settlements. In each case, I outline key characteristics of the national political settlement before exploring how the urban built environment can enhance our understanding of the deeper workings and institutional organization of this settlement.

KAMPALA: THE CITY-AS-MARKETPLACE

The political settlement in Uganda has to accommodate a relatively diffuse distribution of power. While some of the main productive entrepreneurs in the economy consist of Asian business people formerly expelled by Idi Amin and invited back by President Museveni in the 1990s, these are relatively small in number and generally not considered politically threatening due to their broad support of the National Resistance Movement (NRM) regime (Rubongoya, 2007). Much more significant numerically and politically are various factions excluded 'horizontally' from the ruling coalition who have relatively low entrepreneurial capabilities and thus do not benefit significantly from Uganda's (colonial and donor-driven) formal institutional framework of capitalist property rights, yet by virtue of their relative holding power, command benefits through informal channels. These include elites from traditional authorities whose support was crucial to NRM success in the civil war but who increasingly feel excluded, and ethnic groups that formerly dominated government and are associated with the capacity to mobilize violence (Kj«r, 2015; Lindemann, 2011). In addition, some 'lower-level' factions (including former soldiers and highly politicized informal economic groups) possess a degree of holding power due to their numerical significance as a voting bloc, evidenced by their capacity to subvert government regulations in their interests (Goodfellow and Titeca, 2012). Uganda's long history of rebellion and complex ethnic mix render any governing coalition fundamentally fragile in this regard (Tripp, 2010). The strategy of the ruling NRM, in power since 1986, consequently involves 'side payments' to a relatively large number of important factions in order to hold together the ruling coalition (Kj«r, 2015: 232).4

In the capital city, the NRM ruling elite has to accommodate older established interests such as leading figures in the Buganda Kingdom, within

4. For a detailed discussion of the evolution of the political settlement in Uganda, see Whitfield etal. (2015: 161-76).

which the city is geographically located (Gore and Muwanga, 2014). Moreover, the economy rapidly informalized in the 1970s under Idi Amin and the subsequent regime of Milton Obote (Kasfir, 1983), and although a small returnee Asian capitalist class exists, the majority of the urban workforce is employed in the informal economy — around 70 per cent in 2010 (ILO, 2013). In the context of a low (and declining) capacity for the regime to effect structural economic transformation (Kj«r, 2015), the need for continued patronage renders investing in productive capitalist enterprise even less appealing to the elite because the payoffs are uncertain and may take a long time to materialize (Tripp, 2010). The recourse to informal side payments linked to land use and distribution thus became highly significant over successive decades under the NRM. Against this backdrop, Kampala's physical evolution, particularly in the period 2000-10,5 can be understood as reflecting the dominance of informal institutions which directly subverted formal rules in the process of allocating benefits to individuals and groups with relatively substantial holding power.

Exacerbating existing trends from previous decades, this period was characterized by the degradation of infrastructure, encroachment of illegal construction onto environmental resources and informalization of wealthy neighbourhoods (Goodfellow, 2013; Gore and Muwanga, 2014). This occurred alongside the virtual abandonment of existing plans and regulatory frameworks. The transformation of the urban landscape over this period is aptly illustrated by the mushrooming of high-end hotels that contravened basic regulations, the virtual eradication of the city's green spaces (apart from the golf course) and the large number of informal marketplaces occupying land throughout the city (Goodfellow, 2013; Lindell and Appelblad, 2009). In some respects the city itself thus resembles a huge marketplace, in which virtually all urban land is exploited for private profit, very often without formal approval.6 The only substantial park in the city was turned into a cluster of restaurants targeted at the elite and middle class, and the only children's playground in the city centre was likewise sold off to a group of former military personnel for development, in both cases without planning permission (Goodfellow, 2013).

The Ugandan government in this period thus systematically allowed certain groups of people to build and work in ways that conflicted with formal institutions, in exchange for political or economic support.7 There was, however, another important political dimension to this, concerning decentralization. In part because of the extensive number of factions that require accommodating in Uganda, the government enthusiastically adopted the

5. Significant changes to urban governance in Kampala took place in 2011. While I allude to the significance of these below, there is no space here to discuss the post-2011 situation in more detail (see Gore and Muwanga, 2014).

6. Various interviews with planning officials in Kampala, 2009-10.

7. See Goodfellow (2013) for further detail.

decentralization reforms being promoted by donors from the 1990s, as they provided a new route to dispersed patronage (Lindemann, 2011). This enabled opposition figures to be elected to power in Kampala City Council, bringing the struggle between the NRM and opposition forces to the city level, which was reflected in the way that politicians on both sides sought to buy the support of the city population through urban land-based side payments, as a 2006 report into land leasing and sales in Kampala starkly illustrated (RoU, 2006).

The situation in Kampala was thus a product of the NRM's effort to maintain a political settlement in ways that were necessarily informal, because they violated land-related regulations and required meddling in affairs that had officially been decentralized to the city government. Yet it was also a product of how factions excluded from the ruling coalition, but with significant holding power for historical or ethnic reasons, themselves attempted to build and maintain support. In this respect, a national political settlement that might in Khan's terms be characterized as a '(weak) dominant party' settlement starts to look more 'competitive clientelist' at the capital city level. When bureaucrats or local politicians attempted to implement formal institutions, other (usually national) politicians would routinely intervene to 'protect' urban groups from these efforts — 'babysitting' their various client groups, in one interviewee's words.8

Exploiting the city to maintain a political settlement in this way had some negative consequences for the ruling NRM elite, and these became increasingly evident over time. The widespread allocation of benefits through informal channels, reflected in the virtual free-for-all in physical development, not only resulted in a city so dysfunctional that even those in power felt the negative effects, but also strengthened the opposition factions threatening the regime's dominance (Goodfellow and Lindemann, 2013; Gore and Muwanga, 2014). This led to a move in 2011 to radically reverse decentralization and disempower powerful urban opposition factions by introducing a new capital city authority directly accountable to central government (ibid.).

KIGALI: THE CITY-AS-EXPO

The basic contours of the national political settlement in Rwanda differ from those in Uganda in some fundamental ways. A first important difference is that what Khan terms 'lower level factions' within the Rwandan Patriotic Front (RPF)-led ruling coalition are weaker, and factions excluded from the ruling coalition are also relatively weak (though this has started to change in recent years). The trauma of genocide and outflow of certain opposition forces, alongside legal restrictions on ethnic discourse and political activism,

8. Interview, 11 June 2014.

all contribute to the relatively low level of holding power of groups outside the ruling coalition (Reyntjens, 2013). As such, there is amuch smaller range of clients to whom it is necessary to allocate 'side payments'. Moreover, the dominance of the RPF is not built on a painstakingly constructed political coalition of diverse groups, as was the case with the NRM when it achieved power (Rubongoya, 2007), but rather on its military supremacy, delivery of certain developmental results and suppression of opposition in a context of just two major ethnic groups (Reyntjens, 2013).

The country ranks even lower than Uganda on questions of economic transformation and productivity (ACET, 2014), and domestic capitalists are generally lacking in capabilities, with the RPF itself constituting the main entrepreneurial actor (Behuria, 2015; Booth and Golooba-Mutebi, 2012). As in Uganda, the vast majority of the urban population work in the informal economy. However, they are not engaged in the same kinds of competitive clientelistic relations with local government as in Kampala, in part because there is no effective opposition and limited political decentralization (Chemouni, 2014).

The relative absence of interests with substantial holding power outside the ruling coalition, and the weakness of lower-level factions in the coalition, has enabled the government to focus more thoroughly on economic growth and the transformation of its economy (Behuria, 2015; Booth and Golooba Mutebi, 2012). In Khan's terms, the ruling coalition can therefore be characterized as a 'potential developmental' one, because the weakness of excluded factions minimizes the need for side payments and also enhances the government's capacity to implement policy (Whitfield et al., 2015). However, it still faces a situation where there is some demand for land-based side payments because powerful elites within the coalition cannot easily enrich themselves through formal capitalist rights (which underpin many of Rwanda's donor-driven legal and policy frameworks) as there is little by way of a capitalist economy through which to profit. Some of the ways in which this dilemma is institutionally managed are illuminated quite vividly by studying Kigali.

Kigali's development since the turn of the millennium has been characterized by remarkable strictness in regulatory enforcement. Developers encroaching on green spaces often find their buildings demolished and permission to build in such spaces is routinely refused; officials claim they have not heard of anyone who succeeded in being granted permission to build 'by appealing to high office'.9 Similarly, the situation with hotels is a direct contrast to that in Kampala, with a spate of demolitions in the late 2000s affecting hoteliers who had contravened planning regulations, or who simply could not prove that they had complied with them.10 The comment made by

9. Interview, 10 February 2010.

10. Interview with District Mayor, 15 December 2009.

one land policy advisor that 'we are all impaled on the master plan'11 aptly sums up the situation with respect to land use, zoning and other aspects of urban development control. Commercial and residential real estate developments have been advancing, with entirely new areas of the city opened up for suburban development and a series of towers in the old city centre transforming its skyline. Very noticeably, however, there are certain central areas that have been controversially cleared of slums since 2007 but without much being put in their place, as they await the construction of a proposed new central business district envisaged in the Singaporean-designed plans (Goodfellow, 2014); unlike in Kampala, such plots have not been surreptitiously sold off and developed for profit. Another feature of the evolving urban landscape has therefore been a number of 'blank spaces' in central areas, on which nothing can currently be built due to the stringent zoning regulations.

On the one hand, this apparent absence of widespread side payments in the form of non-implementation of land laws and regulations comes as little surprise, given the relatively few groups with relative power who require informal material benefits. Yet our understanding of the political settlement is deepened if we consider why even powerful elites are apparently unable to easily circumvent land use regulations in their interests. Indeed, even the wealthy and well-connected are sometimes forced to sell their land because they cannot get around the regulations preventing them from using it as they wish (interviews with investors, 2009 and 2014). Why do formal institutions have greater force than in Kampala if they are not always closely aligned to the existing distribution of power?

To understand this requires attention not only to power distribution but to image management and the role that discipline and 'zero tolerance' to corruption plays in RPF elite ideology and strategy (Behuria, 2016). Visible enforcement of rules is part of a broader 'performance' (Rollason, 2013) to present Rwanda as a developmental success, which the government considers crucial to the RPF mission. The city is consciously produced as a site of secure, orderly and forward-looking development — the antithesis of the 1994 genocide — with a particular eye on international investment (Gatsinzi and Donaldson, 2010; Goodfellow and Smith, 2013). Kigali's landscape illustrates that visible adherence to formal institutions is in its own right an important aspect of the political settlement, irrespective of the fact that many such institutions do not directly allocate benefits to the powerful. This display of rule-bound development is both inward facing, due to the potential threats to power that might emerge if perceptions of corruption and disorder among a regime based on an ethnic minority were to spread, and outward facing. Kigali is, by the government's own admission, where Rwanda is most obviously on show to the world (interviews with government officials,

11. Interview, 19 February 2009.

2010-11); the city is a sort of large-scale 'expo' for Rwanda's development. The perceived need to be seen doing things in accordance with formal processes can also help explain the prevalence of 'blank spaces' noted previously. Architects working in the city note that rigid interpretation of planning regulations causes long delays during which nothing can be built.12 The blank spaces reflect a government striving to minimize the degree to which formal institutions are overridden with informal ones, a concern rooted in its obsession with image and made possible by the relative weakness of excluded groups.

There is, however, another side to the institutional management of Rwanda's political settlement, which again manifests in Kigali. This is that alongside some formal rules and regulations that constrain the allocation of benefits to the powerful, there are others that clearly do help to allocate benefits to powerful groups — or have the flexibility to do so. These play an important role in holding the political settlement together. Strong legal powers of expropriation, for example, allow for a very expansive definition of 'public interest' through which land can legally be expropriated for profit under certain conditions. Tax reforms have also been successfully resisted by elites to facilitate maximum profits from real estate (Goodfellow, 2014). Relatedly, the vesting of authority in a master plan that privileges high-rise business development is generally well aligned with the interests of domestic elites seeking profit; many lower-income groups are displaced as a result (Manirakiza and Ansoms, 2014).

Moreover, some laws and regulations are easily changed without much consultation (such as the property tax rate), and many are also unclear or actively in flux, providing further opportunities for elites to profit without actively breaking formal rules.13 For example, in 2012-15, zoning regulations for a certain (currently undeveloped) part of the city were changed back and forth five times over three years with little explanation, causing ongoing delays for landowners wanting to build there. Some speculated that this was an attempt to prevent this potentially lucrative area from being developed, so that certain powerful actors could profit from it later on.14 This capacity to change formal laws relatively easily can itself be seen as a form of 'side payment', echoing Khan's point that '[i]nformal power can also be used to change formal laws to benefit groups who would otherwise not have benefited' (Khan, 2010: 54). In all, the political settlement thus involves balance of formal institutions that directly serve the goal of rent accumulation by elites and formal institutions that do not, privileging instead some other development goal — including image management and the aversion to visible corruption that are perceived as central to the governing elite's survival.

12. Interview, 5 June 2014.

13. Various interviews, January 2015.

14. Interview, 12 January 2015.

ADDIS ABABA: THE CITY-AS-CONSTRUCTION SITE

Ethiopia's political trajectory bears considerable similarity to Rwanda's in terms of relatively tight and authoritarian control by a dominant party (Mat-fess, 2015), but there are significant differences in the distribution of power. This relates both to Ethiopia's communist history and the complex ethnic calculus of the Ethiopian polity. On coming to power in 1991 after years of civil war, the Ethiopian People's Revolutionary Democratic Front (EPRDF) — dominated by a group from the Tigray region of Northern Ethiopia — was a minority group within the country and attempted to deal with the 'ethnic question' by instituting a system of ethnic federalism (Markakis, 2011). Addis Ababa is governed as a distinct multi-ethnic federal city, but the size and complexity of the Ethiopian population, as well as the historical dominance of the Amhara ethnic group, mean that the 'holding power' of a number of excluded factions in the country and city is relatively strong. Like the other two countries discussed here, Ethiopia falls in the category of political settlements in which capitalists are relatively politically weak and lacking in productive capabilities (Khan, 2010: 75), though this is arguably changing (Oqubay, 2015). However, very significant in Ethiopia is the role of diaspora, who comprise an important group of investors (Lavers, 2012; Lefort, 2015) but one which is excluded from the ruling coalition. They cannot invest in a number of economic sectors which are preserved for domestic or state actors, and they are often hostile towards the ruling party.15

The communist legacy of the 1974-91 Derg regime (including government ownership of all land, which the EPRDF has retained), alongside resistance to some of the donor-sponsored liberalization reforms adopted wholeheartedly in Uganda and Rwanda (Markakis, 2011; Lavers, 2012), provides the regime with key formal institutional instruments to allocate benefits to relatively powerful groups — particularly in relation to land. As will be explored below, the evolving landscape of Addis Ababa provides insights into how these instruments are used to maintain a political settlement, as well as when and how recourse to more informal institutions is apparent.

One of the most striking features about Addis Ababa's evolving form is the scale and extent of construction, which dwarfs anything taking place in Kampala or Kigali. Indeed, '[no] other capital city on the continent has known such fast-paced and extensive modernization over the last few years' (Lefort, 2015: 357). Notwithstanding considerable Chinese investment in infrastructure, the proliferation of commercial and residential buildings is primarily funded by Ethiopian investors — many of them diaspora, who arrived in large numbers after the 2008 global financial crisis led to a flight from investment in the global North — and by the state itself (Duroyaume, 2015). In neighbourhoods such as Bole, Olympia and Mexico, towering frames of

15. Interview, 1 October 2014.

half-built high-rise commercial structures line the (recently widened) main roads. Estimates of how much money has been invested by diaspora into real estate in the last few years range from US$ 1bn to over £ 3bn (Lefort, 2015: 368). In the eastern suburbs, meanwhile, newly-developed areas now gaining connectivity to the centre through the new light railway network are rapidly evolving residential real estate hubs. To the north and south of these (and dotted around the edges of the city) are hulking blocks of condominium apartments, part of the government's innovative affordable housing scheme which had produced 171,000 units in the city by early 2015 (allocated through a lottery and with some prioritized for low-income residents), while being both widely admired and criticized (Duroyaume, 2015; UN-HABITAT, 2011).

Unlike Kigali, this construction is not especially regimented and there is no current master plan in place, though a new one was being developed at the time of research.16 Unlike Kampala, however, the state has a strong hand in much construction, not only through the public housing programme — which can be seen as partly a political project to 'win over' the growing urban middle classes and aspiring lower classes17 — but through ownership and control over land. Although government has retained land ownership, a 1993 law introduced new rules allowing land leasing and rental, influenced by study trips in the 1990s to countries transitioning from socialism to capitalism, including China.18 This opened the door for commercial urban land development, leading to several modalities for allocating land leases. In some cases prime plots are auctioned for as much as £ 3,250 per square metre.19 Other modalities of allocation (for example through a modality of 'negotiation', which accounted for most land leases from 2002-11) can be opaque, as can the process by which land is made available for leasing (Yusufetal., 2009).

The leasing of land to commercial interests plays a particular role in accommodating the profit-making desires of the diaspora, who are often unable to invest in other key sectors.20 Indeed, this can be seen as part of a political effort to neutralize the potentially destabilizing role of groups hostile to the regime by allowing them to profit substantially from urban land and real estate development (which most find to be much more profitable than manufacturing) — especially as government exercises more restrictive rules in a number of other sectors (Duroyaume, 2015; Lefort, 2015).

The city's physical development sketched above illustrates how retaining land control and channelling resources into social housing provide the government with significant scope to allocate benefits. This is necessary given

16. Interview, 30 September 2014.

17. Various interviews, September-October 2014. See also Planel and Bridonneau (2017).

18. Interview, 4 October 2014.

19. Interview, 29 September 2014.

20. Interviews, 30 September 2014.

relatively strong excluded factions, and to a substantial degree facilitated by formal institutions for land and housing delivery which differ from the more heavily donor-influenced legal and policy frameworks evident in Uganda and Rwanda. The condominium housing programme itself is a potent indication of the relative holding power of the urban masses in Addis Ababa, who in 2005 resoundingly elected an opposition party to power in the city (illustrating their much greater capacity to exert political damage than the lower classes in Kigali), before the EPRDF tightened control and restricted political freedoms to recapture the capital in 2010 (Di Nunzio, 2014). The fear of urban upheaval is very apparent in Ethiopia (ibid.; Lavers, 2012), and the city's reshaping reflects the need for some material benefits to accrue to the city population as a whole. While some of this happens through formal channels and official government programmes, the city's evolving form also illustrates that excluded groups' relative holding power makes it more necessary than in Kigali to supplement formal mechanisms for allocating benefits with informal side payments. For example, the widespread belief that informal negotiation and corruption permeates the process of land leasing (even after the abolition of the 'official' modality of negotiation in 2011),21 and the relatively haphazard nature of construction in the absence of Rwandan-style strictness in planning and zoning (Duroyaume, 2015), are testament to this.

In Addis Ababa there is thus a scramble for land in some respects akin to that in Kampala, reflecting the wide range of interests that need accommodating. Yet the government retains a certain capacity to accommodate these interests through its formal institutions, largely due to the ideological legacy of public land ownership and restrictions on foreign investment. In this respect, the political settlement does not rest so thoroughly on informal competitive clientelism as in Uganda, yet it does require informal 'supplements' to enable benefits to flow to all the requisite groups. Both the formal and informal dimensions of this settlement have spatial effects, because in the context of weak productive capabilities they depend on maximizing the use of land to placate the range of interests vying for benefits in the city, affording very little room for the 'blank spaces' of Kigali. The city thus becomes a vast construction site where both formal and informal institutions fully exploit the political value of urban land.

CONCLUSIONS

These three narratives aim both to illustrate the utility of a political settlements approach for interpreting developments in the built environment, and to demonstrate the value of studying urban form for understanding political

21. Various interviews with government officials and investors, September-October 2014.

settlements. Capital cities such as those described above can be viewed as spatial-physical manifestations of the interaction between power relations, formal institutions and informal norms. The governing regimes in all three countries are clientelist, and it is clear that they all rely to a significant extent on systems of patronage and the inconsistent application of laws and rules. How institutions are used to maintain a political settlement varies significantly between them, however. If Kampala represents a sort of modal competitive clientelism, Kigali exhibits a structured, formalistic clientelism, in which official rules must be visibly respected (and where necessary sometimes changed). In Addis, meanwhile, state control of key resources provides additional formal instruments of clientelism distinct from those in the other two cases. Studying urban form is a window onto these differences in clien-telism because it is relatively easy to observe how land-based benefits are distributed and whether this happens in accordance with formal rules or not. Comparing transformations of urban space can therefore help to reveal not only underlying differences in the distribution of power, but how these combine with legacies of recent history (including ideological ones) to affect the degree to which formal relative to informal institutions can be the instruments through which the distribution of benefits takes place.

In linking transformations in the built environment (and sometimes strategically 'unbuilt' spaces) to benefit distribution through specific institutional configurations, this approach develops an aspect of the political settlements framework that is relatively underexplored in the literature. Most existing literature on the framework focuses on the distribution of power itself, and how this relates to the relative power of productive entrepreneurs and political factions. However, it is also important to consider how even among clientelist political settlements in situations of late capitalist transition, the degree of dependence on informal institutions — and the way formal ones are used — can vary widely. Moreover, particular combinations of formal and informal institutions, predicated in part on historically specific distributions of power, themselves have particular developmental side effects, including the spatial/physical ones that are the main focus here.

This article thus extends the political settlements framework into new terrain in several ways: first, by empirically focusing on urban land and its use as a site of accumulation and benefit allocation to maintain a political settlement; second, by drawing attention to the different ways in which clientelist settlements employ a balance of formal and informal institutions, which is conditioned not only by power relations but ideational factors such as developmental ideology and image management; and third, by exploring some of the spatial effects of this.

In presenting this approach, the hope is to spur further comparative urban research drawing on the political settlements approach. There are, however, many issues to consider when taking this forward. One key problem is that of scale. This article has aimed to situate national political settlements in relation to the physical development of capital cities, exploring questions

about how the structure of power between different groups at the national level affects institutional configurations at the city level. This is a deliberate attempt to challenge both the tendency in some urban theory to treat cities as sealed off from national dynamics, and the nation state fixation in much political science and political economy. Nevertheless, there is need for further theorization regarding how national and city-level institutions interrelate, and the scale at which informal institutions are most appropriately conceptualized and analysed. The question of whether and how such a framework might be useful in smaller cities less core to national and political life is important, as well as whether one can meaningfully talk of political settlements at the city level rather than just interrelation between cities and broader political settlements. The hope is that these ideas and the approach outlined above will feed into research and theorization to further this agenda.

There are a number of other ways in which this research agenda could be developed. While this article primarily explores the relationship between political settlements and city form in one direction (how the specific nature of the political settlement influences urban physical development), there is also scope for examining the converse: how changes to urban physical environment shape the political settlement, for example through creating new sources of power or affecting how institutions work. The potential of the political settlements approach for studying different urban phenomena from those discussed here — such as the nature of public goods provision or the prevalence of violence and crime — could also be explored. Another issue relates to the development of methodological tools for studying the relationship between political settlements and change. As argued here, the ongoing transformation of cities can be linked to relatively settled distributions of power. Yet political settlements are themselves dynamic, subject to both gradual change and critical junctures. There is therefore scope for exploring how trajectories of urban transformation are affected by (and thus reflective of) particular shifts and ruptures within an evolving political settlement.

REFERENCES

Acemoglu, D. and J.A. Robinson (2012) Why Nations Fail: The Origins of Power, Prosperity and Poverty. New York: Crown Publishing.

ACET (2014) African Transformation Report 2014: Growth with Depth. Accra and Washington, DC: African Centre for Economic Transformation.

Bayat, A. (2007) 'Radical Religion and the Habitus of the Dispossessed: Does Islamic Militancy Have an Urban Ecology?', International Journal of Urban and Regional Research 31(3): 579-90.

Behuria, P. (2015) 'Between Party Capitalism and Market Reforms - Understanding Sector Differences in Rwanda', The Journal of Modern African Studies 53(03): 415-50.

Behuria, P. (2016) 'Countering Threats, Stabilizing Politics and Selling Hope: Examining the Agaciro Concept as a Response to a Critical Juncture in Rwanda', Journal of Eastern African Studies 10(30): 434-45.

Booth, D. and F. Golooba-Mutebi (2012) 'Developmental Patrimonialism? The Case of Rwanda', African Affairs 111(444): 379-403.

Chemouni, B. (2014) 'Explaining the Design of the Rwandan Decentralization: Elite Vulnerability and the Territorial Repartition of Power', Journal of Eastern African Studies 8(2): 246-62.

Di John, J. and J. Putzel (2009) 'Political Settlements: Issues Paper'. Birmingham: Governance and Social Development Resource Centre.

Di Nunzio, M. (2014) '"Do Not Cross the Red Line": The 2010 General Elections, Dissent, and Political Mobilization in Urban Ethiopia', African Affairs 113(452): 409-30.

Duroyaume, P. (2015) 'Addis Ababa and the Urban Renewal in Ethiopia', in G. Prunier and E. Ficquet (eds) Understanding Contemporary Ethiopia: Monarchy, Revolution and the Legacy ofMeles Zenawi, pp. 395-413. London: Hurst.

Gatsinzi, J. and R. Donaldson (2010) 'Investment Challenges in the Hotel Industry in Kigali, Rwanda: Hotel Managers' Perspectives', Development Southern Africa 27(2): 225-40.

Goodfellow, T. (2013) 'Planning and Development Regulation amid Rapid Urban Growth: Explaining Divergent Trajectories in Africa', Geoforum 48: 83-93.

Goodfellow, T. (2014) 'Rwanda's Political Settlement and the Urban Transition: Expropriation, Construction and Taxation in Kigali', Journal of Eastern African Studies 8(2): 311-29.

Goodfellow, T. (2015) 'Taming the "Rogue" Sector: Studying State Effectiveness in Africa through Informal Transport Politics', Comparative Politics 47(2): 127-47.

Goodfellow, T. and S. Lindemann (2013) 'The Clash of Institutions: Traditional Authority, Conflict and the Failure of "Hybridity" in Buganda', Commonwealth & Comparative Politics 51(1): 3-26.

Goodfellow, T. and A. Smith (2013) 'From Urban Catastrophe to "Model" City? Politics, Security and Development in Post-conflict Kigali', Urban Studies 50(15): 3185-202.

Goodfellow, T. and K. Titeca (2012) 'Presidential Intervention and the Changing "Politics of Survival" in Kampala's Informal Economy', Cities 29(4): 264-70.

Gore, C.D. andN.K. Muwanga (2014) 'Decentralization is Dead, Long Live Decentralization! Capital City Reform and Political Rights in Kampala, Uganda', International Journal of Urban and Regional Research 38(6): 2201-16.

Gray, H. (2015) 'The Political Economy of Grand Corruption in Tanzania', African Affairs 114(456): 382-403.

Gray, H. (2016) 'Access Orders and the "New" New Institutional Economics of Development', Development and Change 47(1): 51-75.

Gray, H. and L. Whitfield (2014) 'Reframing African Political Economy: Clientelism, Rents and Accumulation as Drivers of Capitalist Transformation'. LSE International Development Working Paper No. 14-159. London: London School of Economics and Political Science.

Hall, P. and D. Soskice (eds) (2001) Varieties of Capitalism. The Institutional Foundations of Comparative Advantage. Oxford: Oxford University Press.

ILO (2013) Women and Men in the Informal Economy: A Statistical Picture (2nd edn). Geneva: International Labour Organization.

Kantor, P. and H.V. Savitch (2005) 'How to Study Comparative Urban Development Politics: A Research Note', International Journal of Urban and Regional Research 29(1): 135-51.

Kasfir, N. (1983) 'State, Magendo, and Class Formation in Uganda', Commonwealth and Comparative Politics 21(3): 84-103.

Khan, M. (1995) 'State Failure in Weak States: A Critique of New Institutionalist Explanations', in J. Harriss, J. Hunter and C. Lewis The New Institutional Economics and Third World Development, pp. 71-86. London: Routledge.

Khan, M. (2010) 'Political Settlements and the Governance of Growth-enhancing Institutions'. SOAS Working Paper. London: School of Oriental and African Studies.

Kjsr, A.M. (2015) 'Political Settlements and Productive Sector Policies: Understanding Sector Differences in Uganda', World Development 68: 230-41.

Knight, J. (1992) Institutions and Social Conflict. Cambridge: Cambridge University Press.

Lauria, M. (ed.) (1997) Reconstructing Urban Regime Theory: Regulating Urban Politics in a Global Economy. Thousand Oaks: Sage Publications.

Lavers, T. (2012) '"Land Grab" as Development Strategy? The Political Economy of Agricultural Investment in Ethiopia', Journal of Peasant Studies 39(1): 105-32.

Lefort, R. (2015) 'The Ethiopian Economy: The Developmental State vs the Free Market', in G. Prunier and E. Fiquet (eds) Understanding Contemporary Ethiopia, pp. 357-94. London: Hurst.

Lindell, I. and J. Appelblad (2009) 'Disabling Governance: Privatisation of City Markets and Implications for Vendors' Associations in Kampala, Uganda', Habitat International 33(4): 397-404.

Lindemann, S. (2011) 'Just Another Change of Guard? Broad-based Politics and Civil War in Museveni's Uganda', African Affairs 110(440): 387-416.

Lund, C. (2006) 'Twilight Institutions: Public Authority and Local Politics in Africa', Development and Change 37(4): 685-705.

Manirakiza, M. and A. Ansoms (2014) 'Modernizing Kigali: The Struggle for Space in the Rwandan Urban Context', in A. Ansoms and T. Hilhorst Everyday Forms of Land Grabbing in the Great Lakes Region of Africa, pp. 186-283. Oxford: James Currey.

Markakis, J. (2011) Ethiopia: The Last Two Frontiers. Rochester, NY: James Currey.

Matfess, H. (2015) 'Rwanda and Ethiopia: Developmental Authoritarianism and the New Politics of African Strong Men', African Studies Review 58(2): 181-204.

McCann, E. (2011) 'Urban Policy Mobilities and Global Circuits of Knowledge: Toward a Research Agenda', Annals of the Association of American Geographers 101(1): 107-30.

McFarlane, C. (2010) 'The Comparative City: Knowledge, Learning, Urbanism', International Journal of Urban and Regional Research 34(4): 725-42.

Migdal, J.S. (1988) Strong Societies and Weak States: State-Society Relations and State Capabilities in the Third World. Princeton, NJ: Princeton University Press.

Molotch, H. (1976) 'The City as a Growth Machine: Toward a Political Economy of Place', American Journal ofSociology 82(2): 309-32.

Molotch, H. (1993) 'The Political Economy of Growth Machines', Journal of Urban Affairs 15(1): 29-53.

Mossberger, K. and G. Stoker (2001) 'The Evolution of Urban Regime Theory: The Challenge of Conceptualization', Urban Affairs Review 36(6): 810-35.

North, D.C. (1990) Institutions, Institutional Change, and Economic Performance. Cambridge: Cambridge University Press.

North, D., J. Wallis and B. Weingast (2009) Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. Cambridge: Cambridge University Press.

Oqubay, A. (2015) Made in Africa: Industrial Policy in Ethiopia. Oxford: Oxford University Press.

Parnell, S. and J. Robinson (2012) '(Re)theorizing Cities from the Global South: Looking beyond Neoliberalism', Urban Geography 33(4): 593-617.

Peck, J., N. Theodore and N. Brenner (2009) 'Neoliberal Urbanism: Models, Moments, Mutations', SAIS Review of International Affairs 29(1): 49-66.

Pierre, J. (2005) 'Comparative Urban Governance: Uncovering Complex Causalities', Urban Affairs Review 40(4): 446-62.

Planel, S. and M. Bridonneau (2017) '(Re)making Politics in aNew Urban Ethiopia: An Empirical Reading of the Right to the City in Addis Ababa's Condominiums', Journal ofEastern African Studies 11(1): 24-45.

Rakodi, C. (2006) 'State-Society Relations in Land Delivery Processes in Five African Cities: An Editorial Introduction', International Development Planning Review 28(2): 127-36.

Resnick, D. (2014) 'Strategies of Subversion in Vertically-divided Contexts: Decentralisation and Urban Service Delivery in Senegal', Development Policy Review 32(s1): s61-s80.

Reyntjens, F. (2013) Political Governance in Post-genocide Rwanda. Cambridge: Cambridge University Press.

Robinson, J. (2006) Ordinary Cities: Between Modernity and Development. New York: Rout-ledge.

Robinson, J. (2011) 'Cities in a World of Cities: The Comparative Gesture', International Journal of Urban and Regional Research 35(1): 1-23.

Robinson, J. (ed.) (2014) Virtual Issue on Comparative Urbanism. Available at: http://www .ijurr.org/virtual-issues/comparative-urbanism/ (accessed 5 September 2017).

Robinson, J. (2016) 'Thinking Cities through Elsewhere: Comparative Tactics for a More Global Urban Studies', Progress in Human Geography 40(1): 3-29.

Rollason, W. (2013) 'Performance, Poverty and Urban Development: Kigali's Motari and the Spectacle City', Afrika Focus 26(2): 9-29.

RoU (2006) The Report of the Commission of Inquiry into the Sale, Lease and Purchase of Land by Kampala City Council. Kampala: Ministry of Local Government, Republic of Uganda.

Roy, A. (2009) 'The 21st-century Metropolis: New Geographies of Theory', Regional Studies 43(6): 819-30.

Roy, A. (2011) 'Slumdog Cities: Rethinking Subaltern Urbanism', International Journal of Urban and Regional Research 35(2): 223-38.

Rubongoya, J. (2007) Regime Hegemony in Museveni's Uganda: Pax Musevenica. New York: Palgrave Macmillan.

Sellers, J.M. (2005) 'Re-placing the Nation: An Agenda for Comparative Urban Politics', Urban Affairs Review 40(4): 419-45.

Sen, K. (2013) 'The Political Dynamics of Economic Growth', World Development 47: 71-86.

Simone, A. (2004) For the City yet to Come: Changing African Life in Four Cities. Durham, NC: Duke University Press.

Stone, C.N. (1989) Regime Politics. Lawrence, KS: University Press of Kansas.

Stone, C.N. (2005) 'Looking Back to Look Forward: Reflections on Urban Regime Analysis', Urban Affairs Review 40(3): 309-41.

Tilly, C. (1984) Big Structures, Large Processes, Huge Comparisons. New York: Russell Sage Foundation.

Tripp, A.M. (2010) Museveni's Uganda: Paradoxes of Power in a Hybrid Regime. Boulder, CO: Lynne Rienner.

UN-HABITAT (2011) Condominium Housing in Ethiopia. Nairobi: UN-HABITAT.

Waibel, M. and C. McFarlane (eds) (2012) Urban Informalities: Reflections on the Formal and Informal. London: Ashgate Publishing.

Ward, K. (1996) 'Rereading Urban Regime Theory: A Sympathetic Critique', Geoforum 27(4): 427-38.

Ward, K. (2008) 'Editorial -Toward a Comparative (Re)turn in Urban Studies? Some Reflections', Urban Geography 29(5): 405-10.

Watson, V. (2009) 'Seeing from the South: Refocusing Urban Planning on the Globe's Central Urban Issues', Urban Studies 46(11): 2259-75.

Whitfield, L., O. Therkildsen, L. Buur and A.M. Kjaer (2015) The Politics of African Industrial Policy: A Comparative Perspective. Cambridge: Cambridge University Press.

Yiftachel, O. (2009) 'Critical Theory and "Gray Space": Mobilization of the Colonized', City 13(2-3): 246-63.

Yusuf, B., S. Tefera and A. Zerihun (2009) 'Land Lease Policy in Addis Ababa'. Addis Ababa: Private Sector Development Hub/Addis Ababa Chamber of Commerce.

Tom Goodfellow (t.goodfellow@sheffield.ac.uk) is a Senior Lecturer in the Department of Urban Studies and Planning at the University of Sheffield. His research concerns the political economy of urban development, the politics of urban governance and planning, the informal economy, land and taxation in sub-Saharan Africa.