Four decades after Sidney Tarrow’s call for comparativists to examine “how politics is fought out across territory,” the publication of five important new books has substantially widened and deepened our understanding of territorial politics in Latin America.Footnote 1 Increasingly, political scientists who study the region are leaving its national capitals behind, or at least not confining their research to these sites, and embarking instead on sustained programs of research in subnational spaces. This deployment of what Richard Snyder identified as the “subnational research design” faces a number of special challenges, including additional field research costs, logistical obstacles in more remote sites, and more extreme forms of data scarcity—which make the many achievements of these new books all the more impressive.Footnote 2
While not all significant research questions in contemporary Latin America require that scholars move beyond the capital, the books reviewed in this essay demonstrate the salience of territorial politics for our understanding of two central issue areas: democratic governance and public policy. Three of these books (Boundary Control, Democrats and Autocrats, and Illiberal Practices) examine one of the most intriguing features of democracy in Latin America nearly four decades after national political regimes began to (re)democratize, which is the persistence of authoritarian or undemocratic subnational governments within these regimes. In addition to Guillermo O’Donnell and Jonathan Fox, Edward Gibson is the scholar whose 2005 theory of “boundary control” published in World Politics did the most to galvanize and shape this new literature; his 2013 book builds on that earlier article by offering a more comprehensive conceptual framework and by comparing the cases of Argentina and Mexico with the US. Agustina Giraudy focuses on the same two Latin American cases, Argentina and Mexico, for which she provides comprehensive regime measurements of all provinces and states, and from which she derives a powerful new theory of subnational regime continuity. The third book in this set is a volume edited by Jacqueline Behrend and Laurence Whitehead, who argue that we should broaden our focus beyond regime type to include a wider range of practices and structures, and who bring together chapters by leaders in the study of subnational Latin America, including Carlos Gervasoni, Julián Durazo Herrmann, Celina Souza, and Behrend herself.Footnote 3
Turning from regime type to public policy, the remaining two books focus on the adoption and implementation across territory of innovative new social policies in the areas of health and education. In Diffusion of Good Government, Natasha Borges Sugiyama asks what led municipal governments across Brazil to emulate two kinds of social policies: conditional cash transfers (CCTs) and community-oriented health programs. In Avoiding Governors, Tracy Beck Fenwick asks a related question, which is why national governments have been so much more successful in Brazil than in Argentina when they sought to implement national CCT programs in subnational units. Sugiyama and Fenwick thus have a coincident interest in the territorial spread of CCT programs in Brazil, about which they offer strikingly different accounts. Whereas Sugiyama questions the importance of institutional and political incentives and emphasizes the role played by ideas and norms, Fenwick dismisses ideology and highlights the importance of institutional differences, including constitutional rules that either grant or deny autonomy to municipal governments. Both argue that important new social rights have been extended across territory in Brazil, but for very different reasons.Footnote 4
Common Issues in the Study of Territorial Politics
The five books paint very different pictures of the balance of power between national and subnational governments and of the relative leverage that each wields over the other. Whether national authorities can get what they want from subnational governments is a question that generates varied responses. Nowhere does the national government appear more powerless than in the account by Sugiyama, whose statistical analysis shows that federal fiscal transfers played no role in encouraging municipalities to adopt policy reforms, and whose interviews suggest that these transfers were too irregular and insufficient to shape municipal decision-making.Footnote 5 Fenwick would mostly agree in concluding that national governments are dependent on subnational actors and not the other way around, although another unavoidable conclusion from her book is that municipal autonomy liberates mayors from governors only to transform them into mere agents of the federal government.
National authorities in contrast appear quite powerful in the works of Gibson and Giraudy. For Gibson, national democratic authorities tolerate subnational authoritarianism when it serves their interests and can and do intervene to end the practice when it does not; the analytical focus is on understanding why the center might decide to intervene, with the assumption that center-led transitions end subnational authoritarianism once the decision to breach provincial boundaries is taken. Giraudy’s claim is even more straightforward: presidential power over subnational undemocratic regimes suffices as an explanation for their continuity. Behrend in her own work on the Argentine provinces comes to a very different conclusion; one is struck by how little changed in the aftermath of federal interventions in Corrientes, Catamarca, and Santiago del Estero (one of Gibson’s main cases), and by the resilience of illiberal practices in these provinces despite the alternation in power that followed each intervention.
A major contribution each of these books makes is to demonstrate why exactly we need to move beyond the simple “national vs. subnational” binary and disaggregate the “subnational level” into its component layers by differentiating, at a minimum, between intermediate-level governments and local-level governments. Whereas the political science literature on decentralization often privileged the role of intermediate-level governments in federal systems, all five books illuminate the pivotal role that municipalities play in these systems. One of Fenwick’s main contributions is to compare two federations that are typically identified as “robust” (Argentina and Brazil) but that perform very differently because municipalities enjoy constitutional status in the latter but not the former. Gibson concurs in emphasizing the importance of whether a given federal system is “municipal-empowering” or “province-empowering,” and identifies “plural cities” as potential problem sites for provincial authoritarians. Like Gibson, Giraudy identifies municipal opposition to provincial autocrats as a factor that may impede governors’ ability to reproduce undemocratic regimes when they are under attack by the president. For Behrend and Whitehead, the control of “municipal machines” is one of the four key mechanisms that sustain undemocratic provincial behavior, and Celina Souza’s case study of Bahia in Illiberal Practices shows how the control of municipalities served as a bulwark for the hegemonic rule of Antônio Carlos Magalhães. Finally, Sugiyama’s book is the most municipalist of all in the sense that she focuses exclusively on policy adoption by municipal governments.
Whereas all five books differentiate municipalities from provinces and showcase the overlooked but critical roles played by the former, they differ in terms of the importance they place on horizontal interactions—conflictual or cooperative—between territorial units at the same subnational level. These are critical for Gibson, who argues that the subnational hegemonic parties that sustain subnational authoritarianism and enforce “boundary control” are linked horizontally to a network of co-partisan (and not necessarily authoritarian) state parties. As he notes in the Argentine case, “Peronist hegemonic provincial parties were … minority members of a chain of provincial parties dominated by parties from large competitive provincial party systems, [and] benefitted from the reflected legitimacy of their national coalition” (156). Interaction between municipal governments is even more salient in Sugiyama’s book, which describes how shared professional norms and social networks between bureaucrats in different cities facilitated the spread of common approaches. These interactions seem less important for Giraudy and Fenwick. For Giraudy, struggles between democratic presidents and autocratic governors play out in a mostly bilateral fashion. In Fenwick’s study, municipalities without constitutional or fiscal autonomy from provincial governments (as in Argentina) have limited ability to interact with each other in ways not sanctioned by governors, but even where they enjoy autonomy (as in Brazil) they do not appear to use it to engage in any sort of collective action with other municipalities.
Despite the fact that all five books are interested in government institutions, the authors actually give very different causal weight to the role played by institutions. For example, whereas Gibson emphasizes formal institutional logics, including the ability to build hegemonic subnational parties and exercise leverage in national legislatures, Behrend and Whitehead see this approach as too limiting and argue that we need to examine informal institutions and to look beyond institutional incentives altogether to illuminate how political families exercise control through economic domination and media practices. As noted above, Sugiyama and Fenwick come to very different conclusions about the role of institutions. Whereas Sugiyama casts doubt on the notion that electoral incentives are what lead mayors to adopt innovative social policies, Fenwick explains the adoption of these policies as the direct consequence of institutional incentives. Against this backdrop, Giraudy’s theoretical framework stands out as an attempt to balance attention to institutional factors like fiscal rules, partisan incentives, and state structures with noninstitutional factors like the cohesiveness of political elites and the support of mass organizations.
We also see in these books sometimes converging and sometimes diverging insights about the role played by specific institutions. Consider, for example, the question of how hard budget constraints impact subnational autonomy. Fenwick argues that they help make municipalities more autonomous from provinces by reducing the fiscal largesse with which governors can otherwise co-opt mayors, and Giraudy suggests that profligate governors who turn to the federal government for bailouts lose a great deal of autonomy as a result. We also learn from these books that, while an institution like coalitional presidentialism in Brazil might be positive for policy reform, as Fenwick argues, it is an obstacle for subnational democratization by increasing the likelihood that opponents of democracy in the states will have co-partisan protectors at the center, as André Borges argues in his insightful chapter in Illiberal Practices.
Finally, while these scholars generally disagree on the relative importance of formal institutions and on the role of institutional incentives, one common insight that can be drawn from these books—and it is an ironic one since all five exclusively examine federations—is that territorial politics should be studied in unitary cases as well as federal ones. Broadening the focus to include subnational politics in unitary cases follows directly from Behrend and Whitehead’s framework, which downplays formal institutions in favor of informal institutions and calls for a focus on illiberal practices rather than subnational regimes (which may be harder to establish in unitary contexts). But even in the books that emphasize formal institutions, like Giraudy’s and Gibson’s, there is plenty of space for the consideration of unitarism. Giraudy explicitly claims that her theory should be able to travel to nonfederal cases, and Gibson concludes his book by noting that “more important than the federal-unitary dichotomy are the specific powers and prerogatives granted to national and subnational governments” (149). This holds true for the books focused on public policy as well. Nothing about the conditions that either facilitate or interrupt policy diffusion across municipalities requires federalism in Sugiyama’s study. While Fenwick’s study is perhaps the most tied to a federal context, given her focus on interactions between intermediate and local governments, in recent years many unitary countries have introduced and/or strengthened the former, which means that the set of countries to which her arguments could apply has likely increased.
Democracy across Territory: Concepts and Theories
The literature on what could be called subnational departures from democracy is characterized by common substantive concerns and yet great heterogeneity vis-à-vis the terms scholars prefer to use. What is in dispute is both the noun and the adjective, so to speak; whereas some authors focus on “regimes” as the object of analysis, others question whether this is the appropriate concept, and even among those who focus on regimes there is disagreement about which labels to use, including “authoritarian,” “hybrid,” or “undemocratic.”
Gibson and Giraudy both use the term “regime” but define it differently. For Gibson it refers broadly to “the set of norms, rules, and practices that govern the selection and behavior of state leaders” (5). Whereas Gibson’s definition thus combines the two dimensions of “access to power” and “exercise of power” that Sebastián Mazzuca has argued should be kept distinct,Footnote 6 Giraudy considers only the first dimension in determining whether regimes are (un)democratic. While Gibson and Giraudy (along with several chapter authors in Illiberal Practices) conceptualize their object of study as “regimes,” Behrend reserves the term “regime” for the federal level (91) and suggests, with Whitehead, that fully fledged political regimes are not really possible at the subnational level (5). They argue instead that we should be focusing on “illiberal structures and practices” in the provinces, which is a more capacious but also more amorphous category that perhaps casts the net too broadly, and that has the additional problem of seeming to equate democracy with liberalism. One also notes that “practices” are indeed included in Gibson’s regime definition. Despite this fundamental disagreement over whether subnational regimes are even possible, the authors seem to agree that “enclave” is a noun best left behind since it suggests the absence of connectivity with national politics (though Julián Durazo Herrmann in Illiberal Practices does use the term).
Among those scholars who believe they are studying “regimes,” we see just as much heterogeneity at the adjectival level in terms of the labels they apply to these regimes, whether “authoritarian,” “hybrid,” or “undemocratic.” Here, however, the semantic variation seems to overstate the extent of actual disagreement since all agree that the location of a subnational regime within a national regime that is democratic puts real constraints on how blatantly authoritarian that subnational regime can be (which is what leads Behrend and Whitehead to adopt the more extreme stance of questioning the possibility of subnational regimes altogether). In Carlos Gervasoni’s evocative phrase, within national democratic regimes there can be no “subnational North Koreas” (157). In response, and following Kelly McMann, he and other authors in Illiberal Practices use the term “hybrid” to stress that the regimes they study are neither fully democratic nor fully authoritarian.Footnote 7 But neither Gibson nor Giraudy use the term “hybrid”; Gibson sees “hybrid” regimes as a subset of “authoritarian” regimes (14), his preferred adjective, and Giraudy prefers the term “undemocratic” to “hybrid” because the former “denotes cases that only fare poorly on the access to office dimension” while the latter “cannot tell us what specific aspect of the access or the exercise dimensions they are missing” (37). Nevertheless, there does not seem to be much daylight between any of these adjectives; all underscore the point that within national democracies, subnational deviations take the form of subverting rather than abolishing democratic institutions, and that they do so through the systematic violation of political rights.
Turning from conceptual challenges to theory building, scholars working in this literature focus mainly on the factors that explain the persistence or transformation of subnational regimes, with greater attention to what enables authoritarian/undemocratic regimes to survive (continuity) than what causes them to give way to democracy (change). These books point to a large set of causal variables, including fiscal dependence (Gervasoni); social structures (Durazo), intra-elite cohesion or division (Behrend and Giraudy); party nationalization (Borges); the provision of private and public goods (Souza); legal changes and jurisprudence (Gibson and Desmond King in Illiberal Practices); and presidential powers and bureaucratic structures (Giraudy). Reconciling disparate, even contradictory, theoretical accounts is always challenging, but especially so considering the diversity of subnational departures from democracy. As Giraudy argues, it’s not just that subnational undemocratic regimes may differ from country to country in ways that might require different causal explanations, but that they differ within the same country. Or, as Gervasoni shows in his study of just one country, Argentina, different provinces are characterized by different illiberal practices. One is reminded of Tolstoy: “Happy families are all alike, every unhappy family is unhappy in its own way.” Though this heterogeneity poses special challenges for theory building, each of these three books nevertheless offers powerful theoretical insights that transcend specific cases. Behrend’s comparison of Argentine provinces demonstrates that it is far easier to eliminate baldy repressive features than other illiberal practices. Giraudy provides convincing evidence that autocratic governors can indeed resist national attempts to democratize them if they can count on cohesive elites and mass support. Gibson shows that decentralized federations will trigger center-led as opposed to party-led transitions away from subnational authoritarianism.
In addition to the diversity of subnational regimes, the role of the national government also poses real challenges for theory building. We see significant disagreement about whether the (democratic) national government plays a positive or negative role vis-à-vis subnational authoritarianism, and different assumptions about the content of national preferences vis-à-vis subnational regime type. For example, whereas Gibson conceptualizes boundary openings as democratizing attempts by national officials to support local oppositions against provincial authoritarians, Behrend argues that interventions do not always have democratizing goals, and Giraudy goes further to explicitly assume that presidents prefer undemocratic regimes at the subnational level. This leads to a stark disagreement; for Gibson “whenever a provincial authoritarian government exists, boundary control is taking place” (150), whereas Giraudy argues that presidents, if sufficiently powerful, will breach these boundaries to reproduce rather than overthrow undemocratic regimes.
Reconciling the various theoretical approaches and explanations in these books can also be difficult because of the different time frames the authors privilege. Gibson notes that his study of successful transitions from authoritarian rule in Argentina and Mexico “ended when power was transferred from incumbent parties to opposition parties … and therefore provides little empirical evidence about the long-term consequences for provincial democratization of center-led or party-led transitions” (170). As Maya Tudor and Adam Ziegfeld argue in their excellent chapter in Illiberal Practices on subnational democratization in India, there can be good country-specific reasons to focus on alternation in power as a meaningful measure of subnational democratization. And yet research by others who adopt longer time frames are loaded with less sanguine conclusions. Behrend, for example, picks up where Gibson leaves off, and the landscape she paints is a rather bleak one. In Catamarca, the end of the rule of the Peronist Saadi family in 1991 led to two decades of hegemonic Radical Party rule by Arnoldo Castillo; in Corrientes federal intervention was followed by the emergence of a newly dominant political family (the Colombi); and in Santiago del Estero, Governor Gerardo Zamora largely mimicked the practices of the Juárez regime after its displacement. From other chapters in Illiberal Practices we see similar stories; Durazo shows that non-PRI governors Gabino Cué in Oaxaca and Mario Marín in Puebla have reproduced the PRI’s authoritarianism in Mexico, and Borges’s case study of Rio Grande do Norte likewise shows that Natal mayor Vilma Faria defeated the Maia clan only to favor her own family as governor. Even Souza, whose sees deeper changes underfoot in post-Magalhães Bahia, acknowledges that Worker’s Party (PT) governor Jacques Wagner has had to bring into his coalition “politicians close to the previous political group” (224).
Policy across Territory: Horizontal vs. Vertical Dynamics
Understanding public policy in Latin America today also requires that we look beyond the national arenas, where almost all important policy struggles took place in the past. Decentralization deserves much of the credit or blame for this altered landscape, both because, as Sugiyama demonstrates, substantive policy-making prerogatives have been transferred to subnational governments, and because, as Fenwick shows us, national policy making now often depends on the buy-in of subnational governments that are no longer controlled politically by the national government. Sugiyama and Fenwick are fundamentally asking the same question (which is what makes comparing the two books especially fascinating): Why do elected politicians in control of subnational governments decide to adopt or support CCTs?
At first blush, it would be hard to imagine answers to this question more at odds than the ones Sugiyama and Fenwick provide in their meticulously researched and carefully executed books. As noted above, Sugiyama finds no quantitative or qualitative evidence for the importance of electoral incentives and instead traces the adoption of social policy reforms to leftist ideologies prevalent among mayors who belong to the Worker’s Party, and to health and education bureaucrats who are socialized into the norms of their professions. Fenwick, in contrast, rejects the hypothesis that we could attribute the greater territorial diffusion of CCT programs in Brazil than in Argentina to ideological differences between the Worker’s Party and the Peronist Party, and emphasizes instead the degree to which mayors respond to three kinds of institutional incentives: constitutional rules, fiscal rules, and the degree of majoritarianism in the political system.
While it would appear to be impossible to reconcile such contradictory theoretical explanations of CCT adoption in Brazil, on closer inspection three factors may help partially account for the discrepancy of these two accounts. First, both scholars focus on municipal decision makers, but with a key difference: Sugiyama asks why Brazilian mayors adopt municipally designed and funded CCT programs, whereas Fenwick asks why they agreed to cooperate with the national government in the implementation of a nationally designed and funded CCT program. Perhaps normative commitments matter more when mayors are considering the kinds of municipal policies they want to define their tenure, as opposed to the more strategic kinds of calculations that dominate when they interact and bargain with higher-level authorities. Second, though Sugiyama and Fenwick are studying the same policy revolution, the former is focusing on Brazil’s “municipal period” between 1995 and 2001, whereas the latter is focusing on Brazil after Lula’s rise to the presidency in 2002. While Fenwick argues that “the nationalization of the Bolsa Escola led to the eventual extinction of most subnational CCTs” (63), this does not mean that norms and ideas did not play, as Sugiyama argues, the critical role in diffusing the CCT approach in the earlier “municipal era.” A third factor has to do with the size of the municipalities on which Sugiyama and Fenwick are basing their accounts. While Fenwick argues that financial impoverishment is what lead many municipalities in Brazil to agree to implement the Bolsa Família, which supports her argument about fiscal rules, Sugiyama focuses on larger municipalities, where the lessening of fiscal constraints may have allowed other factors like professional norms to loom larger.
Another key difference follows from research design. While Fenwick is comparing countries (Argentina and Brazil) and not policies, Sugiyama is comparing policies (education and health) and not countries; hence the variables Fenwick identifies as explaining the difference between Brazil and Argentina are constant across all the municipalities in Brazil that Sugiyama studies. This difference in research design leads them to make different kinds of contributions to different kinds of literatures. Sugiyama’s research design generates significant insights for the policy literature; she finds that the simplicity of policy ideas matters (which helped the diffusion of the CCT approach), but flexibility also matters: that both health and education policies could be adopted in the main but then tinkered with and added onto as befitted local contexts rendered them more popular. Though it might not make sense to equate policy emulation with innovation as Sugiyama does (I would argue that innovation could more accurately be equated with the decision not to merely copy other municipalities’ approaches), she has deepened our understanding of policy stability (which can be achieved when a certain policy becomes “the professional standard”) and of the role that policy entrepreneurs can play in transforming the ideological valence of certain policies (as public health expert David Capistrano did in convincing fellow PT members they could support the new family-based health approach).
At the same time, while it is laudable that Sugiyama wants to go beyond “a simple tale of vertical diffusion” (126), the role of the federal government in municipal policy adoption strikes me as somewhat undertheorized, important in the empirical material but absent from the theory. This is especially the case for health reforms, where Sugiyama’s statistical results show that the introduction of federal funding for the Programa Saúde da Família (PSF) after 1998 did increase the likelihood of municipal adoption between 1998 and 2003 (70, 73). One can’t help but think that the greater role of the federal Health Ministry was indeed a salient factor in explaining the greater diffusion of health (as opposed to education) policy reforms in Brazil, and not just the more robust civil society of the health sector.
Just as the role of the federal government could be theorized more explicitly in Sugiyama’s book, so the role of subnational governments may deserve more extensive theoretical attention in Fenwick’s account of national policy successes. For Fenwick, Brazilian municipalities were more likely to sign on to national CCT policies than Argentine municipalities because the former enjoy a greater degree of constitutional, fiscal, and political autonomy. From the work of Sugiyama and others, however, we know that Brazilian municipalities like Campinas and Ribeirão Preto played a leading role in incubating this policy idea that was then nationalized. On the basis of Sugiyama’s analysis, one might speculate about the importance of a kind of “policy legacy” hypothesis for Fenwick, namely that a preexisting municipal CCT might be a factor in whether municipalities subsequently cooperate or not with federal CCTs. More generally, did the municipal origin of CCTs in Brazil (but not in Argentina), along with the municipal diffusion process in Brazil that Sugiyama documents, lessen the resistance to subsequent efforts by the national government to diffuse the policy vertically? Fenwick dismisses this argument as tautological (15), but subnational policy legacies in Brazil seem worthy of further consideration.
What makes Fenwick’s book quite unusual is that she pays sustained attention to the incentives facing elected officials at all three levels of government (federal, provincial, and municipal) without threatening her ability to tell a story that is coherent, plausible, and even elegant, and that generates the pithy recommendation summarized in the title of the book: “avoid governors.” Whether it is really necessary to avoid governors, however, may require further debate. According to Sara Niedzwiecki, for example, governors may or may not cooperate with the kind of national-level policy overtures that Fenwick examines, depending on their partisan orientation; she documents the extent to which governors from the president’s party cooperated extensively with the implementation of CCT programs in both Argentina and Brazil. Only governors from opposition parties refused to cooperate. At the same time, the terrific policy history that Fenwick provides for Argentina, as it lurched from Eduardo Duhalde’s Programa Jefes y Jefas de Hogares Desocupados (PJJHD) to Néstor Kirchner’s Programa Familias to Cristina Fernández’s Asignación Universal por Hijo (AUH), shows that federal policy designers themselves came to believe in the necessity of avoiding not just governors but mayors as well, resulting in the highly centralist AUH. This leads to a further comparative question: Why was the exclusion of municipalities fatal for the AUH in Argentina, but not for the Oportunidades program in Mexico, to which Fenwick’s framework should also be applied?
Conclusion
Each of these five books represents a significant response to Tarrow’s call in 1978 that we study territorial politics. Not only have these books shown that we can no longer understand democracy and policy in Latin America without taking subnational spheres seriously, but they have generated conceptual and theoretical debates that should be the focal point of future research. In closing this review essay, I’d also like to gesture toward factors that are de-emphasized in these books but that might be usefully brought into future work on territorial politics. Here I’d like to challenge Tarrow, who wrote that territorial politics “is not about territory, but is about how territory is fought out across territory.”Footnote 8 One way to enrich this already incredibly rich new literature might be to ask whether and how territorial politics may indeed be about territory, and to bring territory “back in” to the study of territorial politics.
One is struck, for example, by the fact that none of the following territorial factors seem to generate much causal leverage in these books: geographic location/remoteness, unevenness in the level of economic development, proximity to international borders, regionally specific cultural practices, or the possible overlap between territorial and ethnic boundaries. On this later point, if the focus of this literature shifts to unitary cases in places like the Andes, where territorial claims overlap with ethnic claims more frequently, ethnicity is likely to matter in much more central ways. Another useful direction would be to examine more explicitly the role that external and/or transnational actors and forces play in the outcome of these conflicts between national and subnational actors. While territorial struggles within countries over regime type and policy outputs might seem like quintessentially domestic conflicts, we may gain additional insights into these phenomena by considering how different external actors—governmental and nongovernmental—also seek to influence the outcome of these struggles.
Acknowledgements
I’m grateful to Fabrice Lehoucq, Sara Niedzwiecki, Juan Diego Prieto, and Matthew Sparke for their helpful comments on this review essay.
Open access
