Main

This article critically reviews the literature on the governments and politics of cities produced by different disciplinary traditions. The importance of the city government is paramount, as almost 60% of the world’s population live in cities nowadays. Therefore, the daily lives of most people worldwide are ruled, regulated and served by urban governments, with increasing diversity and inequality. This figure will only grow in the near future, with an estimate of 1.1 billion new urban inhabitants between 2015 and 20301,2. As this growth will mostly happen in the Global South, the urban world will also become increasingly different from the cities in the Global North that pioneered large-scale urbanization processes and inspired the dominant theories and models.

Cities are also central to democratic politics, especially considering the crisis of traditional systems of political representation worldwide. This scenario demands the urgent reduction of social inequalities, the promotion of social inclusion, and the widening of democratic opportunities and practices—all conditions for the creation of urban resilience and sustainability. Although these are my normative starting points, I am not interested here in discussing normative principles nor concepts charged with desired features of cities such as ‘intelligent’, ‘healthy’ or ‘smart’ cities. This article discusses the literature about the governments and politics of ‘already existing’ cities with the certainty that the more we understand them, the easier it gets to change them, promoting better, more efficient and more just cities worldwide.

To proceed in this direction, it is key to better explain city governments and their politics not merely as a ‘spatial scale of operation’ but as a political arena in itself associated with concrete ‘political jurisdictions’3. In this sense, to what extent are city governments different from national ones in terms of the way they operate, and also in terms of the politics they host or induce in relation to democratic practices? And how can we fully incorporate city and political diversity, analyzing the existing processes and structures, and not highlighting what we suppose they lack4,5? The rich analytical tradition that has studied city governments and politics since the early twentieth century provides a good starting point.

Two main intellectual preoccupations lie behind the many analytical models and thematic trends of this tradition. The first preoccupation is concerned with the degrees of political autonomy experienced by city governments in policy production, locally from economic actors and vertically from central states. The second preoccupation explores the relationships between cities and democracy, in light of their substantial and durable social and political inequalities. These two issues are still crucial for contemporary cities. Local autonomy has substantially expanded with recent trends of city empowerment, and the connections between cities and democracy are at the center of the multi-actor politics of local governments that inspire discussions about multi-level democratic politics6, urban governance7, civil society mobilization and participation8,9, and invitations to ‘see like a city’ or consider the ‘urban logic of political action’10,11. Therefore, this article critically synthesizes the literature following these two underlying threads.

Some preliminary remarks are relevant. This article focuses on explanations of local political institutions, governments, the politics that surrounds them and the policies they produce. To do so, the choice of authors discussed inevitably involves discretion. In this review, I started from my own reading of the field, from important major overviews of the literature12,13 and from a systematic, but non-automatic, search of relevant journals since 2015 including International Journal of Urban and Regional Research, Urban Studies, Urban Affairs Review, Journal of Urban Affairs, European Urban and Regional Studies, and Environment and Planning. Readers will notice a concentration on North Atlantic authors. As I will discuss, more than my preference, this expresses the concentration of international debates in English and on cities of the north, creating analytical limitations to the field14.

In addition, I must highlight the multidisciplinary nature of the field, with contributions from sociology, geography, urbanism, political science and policy studies (and even public administration in some cases). The first three disciplines have dialogued more intensely within a subfield usually called ‘urban studies’, while political science and policy studies have stayed distant from them during most of the period. This lack of dialogue has created important negative consequences in terms of knowledge accumulation and theory production already mapped in different countries15,16,17,18.

The reasons for that gap seem multiple, but are at least in part caused by theoretical divides. Urban studies are mostly dominated by political economy, with recent dislocations into the study of the geography of global flows, everyday urban life, urban ecology, and new epistemologies and ontologies of the urban (planetary urbanization, feminist, queer, postcolonial and more-than-human analyses)19,20,21,22. Political science and policy studies, on their part, were dominated until the 1980s by pluralism and public choice/rational choice theory. However, since the 1990s, historical neoinstitutionalism and the sociology of public action have become increasingly present, expanding possible contact points between disciplines. The analysis of politics was always present but looked from quite different angles. While urban studies focused on politics in (and around) society, political science and policy studies were interested in politics in (and around) the state, institutions and governments. Without denying the importance of the former preoccupation, this article focuses on the latter because of space constrains.

The article is structured in the following two sections and a conclusion. The next section discusses the formation of the field from classical contributions of the early twentieth century to some of the most used analytical models for urban politics analysis derived from political economy in the 1980s and 1990s, passing through urban Marxism and public choice theory in the 1970s and 1980s. The second section presents the debates since the 1990s, including the focus on urban governance and the renewed interest of political science in cities since the 2000s. The conclusion summarizes the main arguments and discusses the field’s main challenges.

From classical contributions until the 1990s

Debates about local governments started in the United States, focusing on the political machines of the late nineteenth century, and the later urban reform movement. Political machines were city-centered party organizations that captured institutions by controlling votes, usually of poor immigrants, in major US cities. They were ruled by bosses who extracted and monopolized public sector resources and were deeply involved in patronage, corruption and even political violence. They were considered inefficient, corrupt and a danger to democracy itself23, and therefore the local autonomy of the political was perceived as opposed to democracy. A wide urban reform movement mobilized public opinion in the early nineteenth century, leading to institutional changes in the format of local governments of many cities, explaining their large variation until today (https://www.nlc.org/partisan-vs-nonpartisan-elections and https://www.nlc.org/forms-of-municipal-government). Later research suggested that machines were in fact central for the inclusion of the poor and tended to survive, at least in the logic of massive political mobilization24, while others showed that in terms of political strategy, bosses and reformers were not so different25.

By contrast, early theorizations in Europe focused on the role of cities as loci of citizenship and political autonomy during the formation of modern nation states26. Historically, it was mainly the presence of important local economic actors (merchants and guild masters) that explained the capacity of some cities to resist the concentration of power in the hands of absolute monarchs, and later states themselves, suggesting that the vertical autonomy of local governments came at the expense of their horizontal autonomy from local economic elites. The later consolidation of national institutions reduced cities’ political autonomy, although some authors consider the legacies of this history still relevant to understand politics and collective action in European cities27.

The relationship between democracy and power in the city came to the forefront in the community power debate of the 1950s and early 1960s. On one side, the so-called community power structure was inspired by elite theory and sustained that the structure of inequality of North American cities—and later in the whole US polity—created the conditions for an ample and durable undemocratic control of politics and policies by a unified elite28. The arguments of this tradition were eloquent, although the concrete mechanisms of political capture were not specified. In this formulation, social inequalities would endanger political equality and democracy itself, leaving very low political autonomy for cities, considering their elites. On the other side, scholars of pluralism formulated the key question of ‘Who governs?’ cities29, answering that regardless of its large inequalities, the US polity was not based on rank, and inequalities were not perfectly cumulative. Sociological changes during the nineteenth century, such as migration and industrialization, impeded political capture by specific elites by pluralizing society (and elites themselves). Elections, suffrage expansion and political competition generated power alternation in government control. For pluralism, therefore, democracy would be at the center, with interest groups as the main units of political action. Cities would exert large degrees of political autonomy, although the political relevance of non-elite groups would be minor, and inequality patterns would be of lesser importance. While elitist analyses tended to downplay political autonomy and agency, pluralist studies overstated this autonomy and the effects of inequalities in access to power resources. Both debates downplayed the role of institutions and state actors in politics and policies.

In contrast to this debate, Marxist urban sociology and critical geography of the 1970s discussed urban policies with structuralist arguments that were very sensitive to inequalities but granted very little autonomy to cities and politics itself. Marxist urban sociology discussed the role of cities in capitalist societies and their relations with the state30 and produced detailed analyses of actors and processes such as urban developers31 and public equipment distribution32. Critical urban geography started by discussing land values and space production, to later expand its arguments towards the role of cities in processes of capital accumulation33. This literature strongly impacted the field and is still very influential in recent discussions of neoliberalism, financialization and social movements, among others. However, their assumption of a structural capture of the state implied that eventually states would always promote the interests of capital, making politics epiphenomenal, underplaying city autonomy, diminishing the relevance of democracy and reducing the researcher’s capacity to explain variability worldwide.

The so-called growth-machine analytical model of the late 1970s was also based on political economy arguments and resonated interpretations from elite theory, but was better anchored in political mechanisms to explain political capture by elites34. This interpretation sustained that US federalism left local governments with very few financial resources to implement policies and provide services. Mayors and local political actors would have become prisoners of economic actors that control resources in cities, such as local developers and landowners. Promoting urban redevelopment would be one of the only ways to mobilize private resources for urban projects (which in fact are matched by public resources, in the end), pushing cities to become growth machines that disproportionally benefit private interests. Local autonomy would be low and democracy would be reduced as local agendas would always converge towards redevelopment. Although very elegant and accurate, this model underestimated other possible local power configurations of actors and electorates, even in US cities. It did not apply very well to other regions where fiscal federalism is not so restrictive to cities, where land is less concentrated in private hands and where governments are relatively more powerful. It also wrongly assumes that the local agenda always converges around one single topic, while we know that politics in cities frequently involves multiple disputes and agendas.

A radically different interpretation of city politics was inspired by public choice assumptions35. In this case, cities would have low political autonomy and limited control over their choices. Policies would be developmental, allocational or redistributive considering their expected effects in the local economy and social groups, in decreasing order of desirability. The author sustained that efficiency would be opposed to equality, and local leaders should prioritize developmental instead of redistributive actions. The primary interest of cities would be to improve their economic productivity, defined as the “interest of the city as a whole” (page 4 in ref. 35). These seem normative choices of the author and not logical consequences of the analysis. The model does not acknowledge the role of politics, the composition of the local electorate, the strategies and actions of political parties, or the role of ideology, to name just some of the elements that impact the local political scene.

The most influential model for analyzing urban politics, however, was formulated in the late 1980s and early 1990s36. It criticized both elite theory and pluralism and sustained that the study of urban politics should be centered on the coalitions constructed by local political leaders with local economic elites, the so-called urban regimes. As economic resources in market economies are in the hands of the private sector, the main power that local political elites could exercise would be the ‘power to’ govern interests, instead of the ‘power over’ citizens. This power would originate from the ballot box but in fact would be anchored in the capacity to bring into cooperation coalitions of actors of several kinds. Depending on local conditions, urban regimes would vary from—in increasing order of difficulty and resources—maintenance, development (similar to what was described by growth machines), progressive (middle-class regimes focused on the complex regulation of improving urban quality) and redistributive (oriented to the expansion of lower-class opportunities). This model granted large autonomy to cities and their leaders and recognized variation in local contexts, differently from the growth machine. Although regimes are permeable to different power configurations, their explanatory capacity tends to be reduced in situations where local governments have larger fiscal capacities and are relatively more powerful than private actors, as in cities of the south. However, it still suggested the existence of unitary regimes in each city, assuming one single, or at least strongly hegemonic, agenda.

The model of ‘dominant political coalitions’37 also brought political and economic actors to the center, but resulted from two criticisms. The first involved pluralism and its voluntarism and blindness to inequalities, and the second focused on structuralist (including Marxist) explanations for economic determinism and their difficulty in accepting contingency. Instead, the analysis considered the relationships between the local state, its economic and social actors, local institutions, and the ways in which political competition and economic and social change mobilized the interests of the city’s social groups. A key element in the formation of coalitions would be the actors of the local economy and city’s electorate and their class, race and immigrant composition. The model sustained that the key was creating coalitions that amalgamated different interests and that would win elections and secure the necessary cooperation from other actors to govern. Considering local conditions, some of these could be durable until some moment of intense social and economic transformation. This model incorporated local autonomy, democracy and inequalities in quite an interesting way, but received less attention than growth machines and urban regimes.

Political science: comparisons and governance since 2000

The new millennium brought at least three important changes to the tradition we analyze here. First, although some of the models discussed earlier continued to be relevant and influential, there was a certain dislocation from the search for broad theoretical answers to more localized debates around specific issues. This sometimes led to a certain fragmentation of the field, but it also allowed more analytical and theoretical diversity, even though almost always focused on larger cities38. Second, this happened in parallel with a renewed interest in political science in urban politics, incorporating neoinstitutionalism in dialogues with the American Political Development tradition39,40,41, and development studies3,5, while policy studies incorporated elements of the sociology of public action42. A third transformation follows recent trends in political decentralization, or more broadly the ‘rearticulation of scale’ of governments and policies43. Among these scales is the intraurban, associated with the propinquity44—the fact that political phenomena, actors and interests localize in space, demanding sometimes detailed spatial analysis19. A fourth and final element concerns the expansion of the scope for comparative international studies in urban studies and political science. This was partially an effect of what became called the ‘comparative gesture’ in urban studies45,46, but was also certainly due to the globalization of knowledge production and academic circles themselves. Some may see a contradiction between the trends of city empowerment within their nation states, which could lead to more parochialism, and the globalization of cities and of knowledge, which would be associated with globalism. These poles, however, represent combined facets of the recent worldwide transformations of scales and flows of the urban and the knowledge produced about it.

The combination of these transformations brought changes to the field that unfolded in several parallel lines of enquiry. A prolific debate continued a classical political science agenda of analyzing mayors, their characteristics and styles. Although local political autonomy is the landmark of these approaches, some authors have sustained that entire cities could be characterized as ungovernable47,48, or as having styles that derived from their social compositions as progressive49 or collaborative50, which obviously reduces the discretion for political choices and government decisions. In contrast, many other authors have been bringing political autonomy to the forefront by investigating the effects of the interaction of leadership, contexts, institutions and strategies imprinting different mayoral terms with certain characteristics51,52,53 and leading to diverse policy agendas influenced by political ideology, among other factors54,55.

However, local governments vary intensely in format52,56,57,58. In general terms, the different institutional configurations created a gradation of situations from strong mayors to weak mayors. Worldwide, mayors may be directly elected, appointed by elected councilors or even inexistent when professional managers are local chief executives. Councilors, on their part, may be elected by single-member districts or at large, and elections may be partisan or non-partisan, although parties seem to be important even when elections are non-partisan59.

To explore the consequences of that diversity, scholars have been investigating the effects of institutional variation on policy agendas, municipal budgets, representation of minority groups, political ideology, executive–legislative relations and program fragmentation54,55,60,61,62,63. Another important related dimension concerns the forms of institutional (and political) coordination or fragmentation between local governments in urban agglomerations and their effects on inequalities64, and the relations between central city and suburban governments in those conurbations. The picture that emerges from this debate is one not only of great heterogeneity but also of the joint relevance of both institutions and politics: institutional constraints, as well as individual styles, political ideology, choices and processes, are highly significant depending on local processes and context, highlighting both political autonomy and the democratic potential of urban politics.

Another intense (and classic) debate has focused on electoral mobilization and clientelism understood as personalized, contingent and monitored political exchanges. These practices were considered an obstacle to democratic politics. In fact, in the most traditional interpretation65, this would be the initial step of political development that would pass through political machines as a second phase, to reach the depersonalized democratic mass politics of mature democracies as the final stage. This evolutionist view is in fact very suspicious of the role of everyday relationships in democratic politics, and although it has been long abandoned, survives in more nuanced versions. Recent debates, however, have been trying to find precise mechanisms and actors involved in local political mobilization on the ground, especially brokers. While some studies frame these practices as clientelism66,67,68, others have been connecting it to constituency services, policy delivery and different types of brokerage, including organizational69,70,71,72,73. This is still an open agenda, but the more stimulating contributions show active political relations that are at the center of policy delivery both during elections and between them, articulate scales of the political system, and include civil society actors and political parties. The challenge seems to be understanding the precise role of locally grounded networks of political mobilization to coalition building, government formation and policy production in contexts where services are not universalized; party systems are not consolidated and local executives are relatively more central, without recurring to concepts that turn the politics of southern cities as an exception or a deviation.

Recent decades have brought large transformations to cities’ governments, directly impacting debates on urban politics and governments. These changes simultaneously involved the expansion of local executives’ political autonomy and promoted changes in the institutional formats of policy production and delivery. The first of these transformations included processes of devolution, decentralization and the creation of direct elections of local executive officers, among others, that empowered local governments43. In fact, in different parts of the world, the increasingly multilayer character of urban policies came to the forefront, not only by the reconstruction of federalisms or the decentralization of unitary states but also by the increasing relevance of supranational organizations74,75,76,77. It has been maintained that the governments of capital cities are strongly influenced by their relations with central states78. In all cities, however, it became impossible to understand urban politics and governments without taking into account their multi-scale features6,43, considering the simultaneous effects of processes, regulations, actions and finances from different layers of national and international levels.

The second transformation involved changes in the tasks performed by private and public actors, such as in several forms of public procurement, processes of privatization79, service outsourcing, special purpose authorities75, partnerships under different names and formats, and webs of contracts between companies80. Large urban projects became more and more frequent, both in the north75 and in the south81, although frequently suffered substantial local resistance82,83, and sometimes just represented an excuse for local elites to make it easier to create coalitions, in what the literature called ‘coalition magnets’84. All this obviously increased the presence of private actors in urban politics, both in a corporatist, Weberian sense, as organized development interests27, and in policy provision. This has been strongly criticized by several scholars77,85,86, while others have been dedicated to detailing the roles of different actors and to specifying the respective effects of diverse institutional formats87,88. In any case, we are really very distant from the centrality that public agents had in service provision during the 1970s and 1980s77.

The global dissemination of these changes itself has been a subject of enquiry, within a broader subfield of studies of policy mobility, circulation and transfer involving networks of international organizations, consultants and mayors. Institutions and policy solutions never arrived at the destination equal to how they left their location of origin89 but they certainly impacted local policy agendas, many times depoliticizing local policy processes and resulting in interventions out of context86,90.

A third transformation brought an expansion of civil society actors in policy decision-making and provision. It is true that considering their density, scale, internal heterogeneity and political atmosphere, ‘cities breed contention’91. Therefore, urban social movements were always central to democratic politics, collective action and the construction of political identities. Although the first analyses about them tended to be quite localized30, studies gained different features with the recent new geographies of mobilization and identity formation92, and have certainly been impacted by the changes of what the urban itself means today21,22. However, my point here is more concentrated in government decision-making, as the recent decentralization of competences from central to subnational tiers of governments was combined in many places with improving social participation and citizen engagement in decision-making and policy delivery. Studies have accompanied this in local policies in both southern8 and northern9 cities. Others have been showing that public agencies and civil society actors are interconnected within local politics and policy communities, generating important consequences for government capacities, the policies themselves, and social movement formation and consolidation62,93,94. This suggests that civil society analyses and policy studies must dialogue more frequently with each other to better understand the production of contemporary urban policies. Although this debate is not unanimous, these changes have apparently enhanced democratic participation in cities, improved efficiency and made policies more redistributive, with important consequences for access to services.

Policy delivery and access to services have been other important themes in the literature, especially for cities of the Global South, although, in this case, urban politics is a driving force, more than the process to be explained. The issue has attracted academic interest as well as great attention from multilateral organizations, although with a more descriptive and normative focus. This subfield has been dialoguing with sustainability and development studies95,96, and with the informality and urban precarity literatures70, highlighting the difficult and slow (although varied) processes of service expansion in the contexts of southern cities. Service provision has been connected with the previously mentioned processes of political mobilization/clientelism, the different formats of policy delivery and the increasing presence of civil society actors in what has been called ‘hybrid systems’3,97.

Efforts to make all these changes fit into the concepts of urban regimes or growth machines have been strongly criticized, as these concepts travel with difficulty from the United States to other settings98. Some authors have sustained that these transformations just expressed the changing roles performed by the public and the private sectors in policy provision in contemporary market economies in what they called ‘regulatory capitalism’99. Others went in a similar direction and suggested that models for managing simultaneously public action and market economies must be defined by the combination of the degrees of autonomy given to local governments by multi-level institutions with the reorganized division of tasks between public and private actors76. Many others, however, have captured the changing role of the state in the management of the economy and territory under the label of neoliberalism100. However, the concept comes with intense a priori judgements and implies at least some core features in widely varying cases. It has been maybe overused in urban studies101, and it may be wiser to define these recent changes as multiple forms and arrangements of governance, and proceed with the analytical effort of specifying their exact institutional configurations, politics and consequences, including negatives.

Therefore, the broader concept (and debate) that summarizes these changes is urban governance. The concept has expanded intensely in the field since the late 1990s, providing an umbrella concept for these political and institutional changes around urban governments and politics. This becomes clear in Fig. 1, which shows the incidence of the expressions ‘urban politics’, ‘growth machines’, ‘urban regimes’ and ‘urban governance’ in the titles of books published in English until 2019, using the Google Books Ngram Viewer tool. As we can see, ‘urban politics’, a general formulation of the field, has maintained a quite high presence since the 1970s, much higher than ‘growth machine/s’ and ‘urban regime/s’, which have had a lower, but stable presence since the 1990s. However, it is ‘urban governance’ that comes to the forefront in the late 1990s with impressive growth, maybe pushed by the concept of good governance by multilateral institutions102 and its critique103, or by broader concerns about the recent blurring of the frontiers between the government, private sector and civil society mentioned earlier104.

Fig. 1: Presence of selected concepts in books published in English (1960–2019).
figure 1

Percentage of occurrences of selected bigrams in books published in English from the Google Books database in a given year (y axis) over time (x axis). Note that 2019 is the latest corpora available in the Google Books Ngram Viewer, elaborated in February 2020.

Full size image

Governance is a polysemic concept that captures several different processes, but intends to characterize the complex arrangements of multiple organizations—private and public—connected by self-organized networks recently involved in policy production104. Dialoguing with Dahl’s classic book title, we can say that governance tries to better understand ‘who governs what’, ‘who is governed’ (and by whom) and ‘who governs when nobody governs’7. This could result in different governance formats in cities88. These are highly heterogeneous and may vary internally to a single policy, even during the different phases of the same program105. Many new policy instruments have been developed, understood as condensed forms of knowledge that constitute tools and modes of policy operation that, once enacted, produce effects independently of social agency with profound consequences to policies81,106. The role of networks of actors in these new arrangements of policy production has also been analyzed, defined as the (formal and informal) relational patterns constructed through the historical formation of policy fields and communities57,107. Negatively, it may generate competition between institutions, giving rise to fragmented public authority in what has been called ‘twilight institutions’108. The major contribution of the literature about these changes, therefore, has been the creation of a concept that is plastic enough to include the different actors, processes and dynamics present in each specific case, and at the same time can travel between local configurations, allowing for broad comparisons that can help accumulate knowledge. Politics is at the center of these arrangements, implying autonomy and contingency.

Most of the field has been focused on cities of the Global North. This has generated a negative effect on the capacity to explain urban politics worldwide, as most existing concepts and explanations come from the study of a reduced (and concentrated) set of cities45. And to accumulate knowledge, theories must account for the complete variation of the phenomena they try to explain. Therefore, even if not necessarily recurring to postcolonial ontologies, expanding case studies and producing concepts that can grasp the increasing variation of cities, their polities and politics is an urgent matter. Developing comparisons that cross-analyze the south and the north is a way forward. Most importantly, concepts that can travel across the world need to be created and used, instead of insisting on pairs of concepts marked by dualities. Mid-range concepts tend to help more than broad theories in this endeavor109.

Several of these elements have been combined in recent years in a comparative agenda of the historically constructed governance patterns in cities of both the north and the south, understood as the sets of actors—public, private and associative—connected by formal and informal (legal and illegal) ties and surrounded by institutions and organizations. These configurations include networks, institutions and social agency and are behind policy production—decision-making and implementation—in cities worldwide. These governance patterns vary between policy sectors and cities, may superpose and even contradict each other, and incorporate influences and regulations from several levels of government6,43. They are also in constant change, although in a very path-dependent manner that builds different kinds of policy trajectories, including oscillating ones110. This agenda has been investigating cities of both the Global North and the Global South62,111,112,113 and has shown how the combination of political, urban and policy concepts enlarges our understanding of urban politics, especially when supposedly very different cities are analyzed with comparable lenses and concepts.

Summarizing the trajectory and its future challenges

The field of urban politics has had a rich trajectory since its early studies of political machines, passing through the community power debate, Marxist urban sociology, growth machines and urban regimes, among others. The political autonomy of cities (and their governments), vertically from actors and institutions of other scales, and horizontally from local economic actors and institutions, was always an underlying concern. Depending on the approach, local politics and institutions did or did not matter. A second underlying preoccupation was the relationship between urban politics and the pursuit (and possibility) of democracy, even in face of large economic inequalities. From these debates, we learned about the relevance of coalitions of actors and local contexts, and also about the importance of specific agents, mainly politicians and urban economic interests, especially developers, building companies and urban landowners. During this period, the field only peripherally considered institutions, and was strongly concentrated in cities of the Global North, generating difficulties for explanations of the politics of southern cities. These, which increasingly represent most of the urban world, were treated until recently as exceptions.

Since the new millennium, the field has been relying more on mid-range theories and models, better incorporating urban political institutions and their actors, and developing a denser comparative drive. This led to a certain fragmentation but also a better representation of emerging themes and cities of the Global South. In terms of themes, the characteristics of mayors and their governments, in interaction with institutions, brought back old preoccupations of agency versus structure in local affairs, while the role of clientelism highlighted the dilemmas of incorporating into political explanations cases where political mobilization has happened in contexts of weaker party systems and non-universalized service provision. In recent years, the new formats of urban policy production, with varied institutional arrangements and a larger presence of private interests and civil society actors came to the center of discussions. For the study of these complex configurations, the concept of governance became increasingly disseminated, due to the need of versatile concepts that might travel between settings populated by different actors and surrounded by diverse institutions and polities.

Substantively, urban resilience and sustainability came to the forefront of public debates, as did themes such as democracy, the environment and health, pushed by the challenges posed by several forms of liberalism, the risks of global warming, and the recent pandemic and other possible future events of the kind. These, however, joined a long list of lasting urban challenges such as segregation, inequalities, service provision, socio-political participation and urban precarity. The main theoretical and analytical challenges that lay ahead involve continuing the incorporation of political institutions without losing sight of actors and processes, and remaining sensitive (and flexible) enough to construct concepts that incorporate the variety of actors, institutions and configurations that characterize and explain the politics of the contemporary urban world.