Regionalism,Ethnic Diversity,and Variation in Public Good Provision by National States

Abstract

This article argues that variation in public good provision is determined by the salience of demographic and economic regionalism, conceptualized as the presence of multiple large population centers and distinct economic units within a single country’s borders. Two mechanisms—divergent public good preferences and regional self-sufficiency—underpin this relationship. Evidence of these mechanisms as central to low public good provision is found in case studies of Ecuador and Colombia, two countries that score high on standard measures of ethnic diversity and that saw little public good provision during the first century after independence. A region-wide examination of post-independence South America and an investigation of change over time in Venezuela provide further support for the claim that regionalism affects public good provision. A plausible case for this aspect of geography as a cause of low public good provision poses a challenge to the scholarship attributing it to ethnic diversity.

Keywords

Latin American politics subnational politics political economy state building race ethnicity politics

Latin America is a particularly interesting setting to re-examine the ethnic diversity hypothesis for several reasons. There is great variation across countries in the region in levels of public good provision, even among the smaller set of relatively large countries that were ruled by the same colonial power and gained independence at the same time, that is, Spanish South America plus Mexico. Moreover, this variation is patterned in ways predicted by the ethnic diversity hypothesis; high diversity is correlated with lower levels of public good provision. Yet this article finds little evidence of ethnic diversity’s dampening effect on public good provision in Latin America. This challenges not only the claims about these cases but also the broader cross-national correlational evidence that includes them.

Instead of ethnic diversity, this article shows that variation in public good provision is determined by the salience of demographic and economic regionalism—conceptualized as the presence of multiple large population centers and distinct economic units within a single country’s borders. Thus, like Stasavage (2011), I explore the purely spatial roots of political development, highlighting that existing scholarship on ethnic diversity’s effects draws a false dichotomy between it and other, often overlapping, political cleavages. I show that even in the absence of cultural fractionalization, the very existence of urban concentrations outside the capital can shape public good provision. I identify two mechanisms—divergent public good preferences and regional self-sufficiency—that underpin this relationship, and trace them through case studies of Ecuador and Colombia, two countries that score high on standard measures of ethnic diversity and that saw little public good provision during the first century after independence. In both cases, I show that regionalism, rather than ethnic diversity, explains the low levels of public good provision. I then turn to region-wide evidence from Spanish South America to show that more public good provision resulted when national borders contained a single dominant urban center than when they contained multiple large cities of similar size. Further support for the relationship I propose derives from a brief cross-time examination of Venezuela, in which an increase in state efforts to provide public goods emerged only after the demographic rise of Caracas as the national center.

A plausible case for this aspect of geography as a cause of low public good provision poses a challenge to the scholarship attributing it to ethnic diversity. A weak version of this challenge would hold that regionalism is an omitted variable in the ethnic diversity scholarship that should be taken into account. A stronger version would go further to endogenize diversity, tracing the politicization of identity-based cleavages to economic and political conflict over public good provision between various regions. In this article, I mostly limit myself to the former claim; I do show some suggestive evidence from the case of Ecuador to provide some plausibility for the latter version.

Definition and Mechanisms

For the purposes of this article, I use the term “regionalism” to refer to the existence of multiple urban concentrations with associated economies within a single country. Regionalism refers, then, to the opposite of urban primacy, which describes countries that have a disproportionate share of the population living in or near a single major city. It refers to a demographic and economic phenomenon that generates political implications, rather than to political mobilization or identity formation. In cases with high degrees of regionalism, there is no single dominant center, but instead multiple urban concentrations that sit astride distinct regional economies reaching across sizable portions of the national territory and incorporating significant shares of population and economic activity.

My definition of regionalism draws on the concept of sectionalism discussed by Bensel (1984), which traces across history the political consequences of the presence of multiple economic “sections” within the United States. Bensel defines sections by the boundaries of “trade areas” composed of urban centers and their surrounding hinterlands; I use the same logic to define regions. The politics generated by this distribution of population and economic activity cannot, of course, be reduced to economic determinants alone. They are constructed by political entrepreneurs (Agnew, 2002; Appelbaum, 2003) and shaped not only by objective economic conditions but also by what Herrera (2005) calls “differences in the imagination of economic interests” (p. 11). Those politics can take the form of regional lobbies in national-level institutions or of separatism—sovereignty or autonomy claims. Indeed, it is plausible that the economic issues at stake in regional conflict over public good provision may foster the formation of the identity cleavages along regional lines that underpin regionalist and separatist mobilization.

Nevertheless, the argument developed in this article draws on the purely demographic and economic component of regionalism, showing that the distribution of population and economic activity, independent of identity-based mechanisms, is sufficient to reduce public good provision by national states.¹ Two mechanisms (public good preferences and economic self-sufficiency of regions) lead the salience of regionalism to undermine public good provision by the national state. After elaborating the logic of each mechanism, the next section traces each at work in a separate case study.

Locational Determinants of Public Good Preferences

Two currents of existing scholarship trace low public good provision to political conflict between actors with divergent interests: This causal logic characterizes the ethnic diversity arguments to which this special issue responds, as well as the region-specific accounts of public good provision in Latin America, which tend to emphasize socioeconomic inequality as the determinant of provision (Engerman & Sokoloff, 2012; Kurtz, 2013; Mahoney, 2010). Notably, these region-specific accounts are in tension with the ethnic diversity arguments because of the mismatch between the reified nature of ethnic categories in those theories and the more deeply constructed, plastic, and fundamentally socioeconomic nature of ethnic categories highlighted in accounts of ethnic diversity in Latin America. This can be seen in the fact that both Mahoney (2010) and Engerman and Sokoloff (2012) blur the distinction between ethnic and socioeconomic cleavages in describing Latin American inequality.² Whereas existing scholarship overwhelmingly emphasizes social or economic inequality and its role in shaping political institutions as the cause of low public good provision, my first mechanism instead emphasizes regional tensions in tracing low public good provision to political conflict between actors with divergent interests.

I argue that distinct regions, defined and characterized by the organization of economic activity around distinct urban centers, are likely to have divergent preferences about the location and type of public goods provided by the state.³ As conflict between regions over public good provision priorities intensifies, provision is diluted. Conflict over the location of public goods emerges because willingness to contribute to public goods is likely to be correlated with proximity to them, which facilitates access to the benefits they generate.⁴ Given the scarcity of government resources, choices about where to locate state services will create high-stakes conflict over basic state-building priorities. Distinct preferences about the types of public goods the state is to provide derive from the fact that the multiple urban centers are often each associated with a distinct regional economy. As the case studies below illustrate, this may in some cases imply that regions have distinct linkages to the international economy.

Tensions between multiple salient regions over public good priorities contribute to political instability. Resolution of this conflict, whether through compromises that attempt to satisfy actors from all regions (logrolling) and the resulting dilution of policy priorities, or through regionalist conflict, which may escalate to secession, is likely to take concerted attempts to increase public good provision off the table.⁵

Though its logical structure is parallel to the argument (made, for example, by Alesina et al., 2003) that ethnic diversity is associated with disagreement over the types of public goods to be provided, this locational mechanism has two advantages. First, because public goods are by their nature fundamentally spatial, the claim that people in different places have different preferences over them requires fewer assumptions about the origins of those preferences than does the claim that preferences about public good provision are shaped by ethnicity. Whereas it is possible to imagine important ethnic differences in preferences over types of education, the effects of identity on preferences over many other kinds of public goods, like transportation, are harder to see.⁶ By contrast, the vast literatures on NIMBY (Not In My Backyard) politics, and on the geographic clustering of political preferences (Rodden, 2010), both lend support to a locational account. Alesina and Spolaore (2005) rely centrally on the locational determinants of public good preferences to account for the development of state capacity.

Second, the region-based account of preferences over public good provision improves on the similar ethnic-based argument because it does not depend on the collective action of groups as the path by which public good allocation is determined. In Latin America, as elsewhere in the world, exclusionary political arenas resulted in policies being determined by a small group of elites not necessarily representative (descriptively or substantively) of the population as a whole. Thus, there is a disconnect between levels of ethnic diversity in the population at large and the composition of policy makers within the political arena. The regionalist argument, by contrast, simply requires that regional elites (whatever their ethnic background) have some influence over policy either through formal channels or through the ability to mobilize challenges outside those formal channels. Even where elites are ethnically homogeneous, then, I argue that they may pursue distinct interests because they are based in different locations.

Regional Self-Sufficiency and Anti-Statist Politics

The first mechanism developed above builds on existing literature to identify a new cleavage that shapes conflict over the distribution of the costs and benefits of public goods. Yet even with this extension, this approach to the determinants of public good provision—whether it emphasizes region or inequality as the key determinant of preferences—remains overly narrow in its focus on how these costs and benefits are distributed. A complete account of the determinants of public good provision requires a step back to consider why the provision of public goods is sought in the first place. Public goods are, after all, not an end in and of themselves. Instead, they are provided when they are a means to an end sought by political actors. Explaining variation in public good provision, then, requires identifying when these goods are seen as the appropriate means to the ends pursued by political leaders, thus shifting our focus from regional interests to considering the possibility that the central state may operate as an independent causal force in determining policies relating to public good provision (Olson, 2000). This mechanism centers on regionalism’s role in shaping the normative commitments of the state to public good provision; the place of public goods in the ideology of development.

Here lies the second mechanism linking regionalism to low state provision of public goods: The absence of urban primacy is often accompanied by the presence of salient regional economies, which operate largely independent of one another and are separately integrated into the world market. Where intraregional trade and economic exchange is limited, and there is little potential for it to grow, there are fewer gains to be reaped from public goods that could knit the country together into a single national unit (Spruyt, 1994). Overcoming regional tensions and knitting the country together will produce less fruit for economic development and political stability where the country’s regions have self-enclosed and self-sufficient economies, especially if those regions either produce similar goods for export or goods for which there is no domestic market. Where the gains from national economic integration are expected to be minimal, given existing technology, I argue that public good provision comes to be seen by central state leaders as less propitious for economic development, political stability, and social peace. Under these conditions, states will more likely pursue development through policies that allow already salient regional economies to flourish. These policies will emphasize limited state intervention, rather than public good provision, as a means to economic development. Low levels of public good provision by the national state can therefore result from the disinclination of state leaders to pursue it, which is likely to emerge in a context of salient and economically independent regions.⁷ This laissez-faire model of state development is the antithesis of the “order and progress” pursued by state leaders in more centralized contexts like Mexico. Thus, the spatial distribution of population and economic activity shaped variation in the context of development projects in Latin America by influencing the consensus among Liberal and Conservative political elites about their normative commitment to public good provision.

Case Studies

I now turn to exploring these two mechanisms in a pair of case studies of post-independence Ecuador and Colombia. These are not intended to be compared with one another; instead, each case traces one of the causal mechanisms detailed above. The logic for selecting these two cases for further examination can be seen in Figure 1. First, by regional standards, both countries had low levels of public goods provided by the national government, as revealed in their poor performance in literacy rates and railroad development in the first century of independence in all panels of the figure.⁸ Second, on the Alesina, Devleeschauwer, Easterly, Kurlat, and Wacziarg (2003) fractionalization index, shown in the top row of Figure 1, they both score at similarly high levels, meaning that they are both cases for which we should expect that ethnic heterogeneity is the reason for low public good provision.⁹ Thus, they are useful cases in which to examine whether the proposed argument really is at work. Against the ethnic heterogeneity argument, I show that in each case, the low levels of public good provision can be traced to the salience of regionalism; both cases (as seen in the bottom row of Figure 1) have highly salient regionalism, which makes them a good site for observing the mechanisms that I developed above, and for showing that those mechanisms, rather than those linked to ethnic diversity, explain the low levels of public good provision.¹⁰

Figure 1.

Ethnic Diversity, Urban Primacy, and Public Good Provision.

The study of long-term patterns of public good provision in Latin America is appropriately focused on the 19th century for two reasons. First, as shown in the cross-national works of Mahoney (2001, 2010), Kurtz (2013), Saylor (2014), and Soifer (2015), as well as many single-country studies, the origins of contemporary divergence in state capacity and public good provision can be traced to the 19th century. A satisfying account of contemporary variation must emphasize the period in which this variation emerged. Second, there are good reasons to be concerned about the endogeneity of regionalism to state capacity to provide public goods (one might say the same about ethnic diversity, as noted by Wimmer in his article and by the editors of the special issue in theirs). This threat to inference can be addressed by studying the effects of regionalism on the initial construction of public good provision, granting that reciprocal causation operates in subsequent time periods. By showing that the two mechanisms elaborated above account for the underprovision of public goods in Ecuador and Colombia, and that regionalism is correlated with public good provision across Latin America, this article highlights its role in accounting for the long-term variation in public good provision that is often attributed to ethnic diversity.

Regional Conflict and Low Public Good Provision in 19th-Century Ecuador

Ecuador saw strikingly low levels of public good provision by the national government during the first century after independence, as seen in Figure 1. Compared with many of its neighbors, the post-independence Ecuadorean state performed poorly in the provision of public goods: It had the lowest railroad density of any country in the region and a below-median literacy rate. This is a country that might appear at first glance to have deep identity-based divisions; it has one of the most ethnically diverse populations in Latin America, as seen in its relatively high scores on both the ethno-linguistic fractionalization (EFL) and the Alesina et al. (2003) fractionalization measures of ethnic diversity. Yet I show that low levels of public good provision in this case are not explained by disagreement across ethnic groups about public good preferences, but instead by longstanding patterns of interregional disagreement among elites from the mountain and coastal regions. I further show that the deep and persisting regionalist divide in Ecuador has little or no ethnic component.

The Correlation With Ethnic Diversity

A cursory glance at Ecuador in cross-national perspective would suggest that the country’s ethnic diversity could plausibly account for its low public good provision. Not only does the country score 0.655 on the Alesina et al. (2003) ethnic fractionalization measure, which places it third highest among South American cases, but also a politicized ethnic cleavage certainly exists in contemporary politics and society with the emergence in recent decades of one of Latin America’s most powerful indigenous movements (Chartock, 2013; Yashar, 2005).

The distinct populations of the coast and highlands mingled little, and had separate regional histories to draw on. The highlands populace centered on Quito was heavily indigenous, while the importation of African slaves characterized the coast. There was little overlap of elite families across regional borders in the 19th century (Clark, 1998, p. 27). The coastal center of Guayaquil was actually removed from the rest of Ecuador and made part of the Peruvian viceroyalty during the very end of colonial rule (in 1804), and it gained independence separately from the rest of the country, enjoying a brief existence as an independent city-state. This distinct history is often referenced as a “repeated affirmation of its preference for autonomy” (Pineo, 1994, p. 276). Evidence also points to the existence of distinct identity characteristics across the regions. The most important was traditionalist Catholicism, and the church wielded “more spiritual and material power” in the highlands than on the coast (Kasza, 1981, p. 19). This difference was reflected in 19th-century party politics, which saw Liberals gain more support on the coast and Conservatives in the highlands.

Yet these identity-based differences did not drive conflict over public good provision. Conflict was instead driven by the purely economic dimension of regionalism: The coastal region around Guayaquil and the central-northern highlands around Quito had different modes of production and distinct trade orientations (Pineo, 1994, p. 276). These created regional political coherence that was not based on identity, and they spurred conflict between these regions over how the state should pursue development through the provision of public goods. The fact that some coastal regions identified with the highlands and vice versa (Kasza, 1981, pp. 6-7) shows that economic regionalism, rather than a purely spatial or identity-based division, is at the core of this cleavage.

“A Tale of Two Cities”

In Ecuador, as in most Latin American countries, a degree of broadly Liberal consensus about national priorities was found across the entire political and economic elite (Soifer, 2015, Chapter 1). Despite their differences, elites nation-wide coincided in their pursuit of national development, and saw state provision of transportation infrastructure, education, and other public goods as a means to that end. Yet elites from different regions were deeply split on how these priorities should be enacted, and each region’s representatives in the national government pursued policies designed to benefit their native region rather than the country as a whole (Pineo, 1994, p. 277).

Quito and Guayaquil each stood astride a coherent region that Van Aken (1989, p. 39) compares with the city-states of Renaissance Italy. The provinces in which they were located accounted for more than half of the country’s population, dominating the country to such an extent that the country’s evolution has been described as “A Tale of Two Cities” (Rodríguez, 1985, p. 29). Because the political economies of the coast and highlands were so different, each had distinct preferences about public goods, revenues, and spending priorities.¹¹ Although there were tensions within both the coastal and highlands regions, the primary cleavage of political conflict split them from one another.

Guayaquil surpassed Quito in size at the end of the 19th century, and its growth escalated further with the anti-malaria campaigns after 1918 (Rodríguez, 1985, p. 27). Guayaquil was also the site of major banking houses that provided loans to the government, giving its elites another platform for influence over national policy. Thus, in terms of population, economic power, and financial importance, it was at least as powerful as the national capital of Quito throughout the post-independence era. It was the center of a coastal regional economy that was overwhelmingly based on export of tropical agricultural products, of which cacao was the most dominant (Henderson, 1997). In maximizing exports, Guayaquil could not be self-sufficient economically in terms of food production. Yet trade with the interior was so expensive that it preferred to pursue policies that facilitated trade with Europe. Guayaquil’s economic ties to Europe were greater than its connections to the domestic economy. The highlands around Quito, by contrast, produced primarily for the domestic market, exporting fairly little. Neither cattle nor grain could be profitably shipped to the coast, and the highlands had a sufficient urban population to act as a sizable market for agricultural production.

Regional Conflict Over Public Good Provision

The sharp difference in regional economies prompted divergent preferences about public good provision. These manifested in Guayaquil’s displeasure over bearing the burden of most of the customs duties while infrastructure spending was concentrated in the highlands. City officials periodically refused to turn over tax revenue collected in the port, often referred to as “Guayaquil’s money” (Pineo, 1994, p. 277) to the national treasury, and used it instead for local public works priorities (Henderson, 2008, p. 87). Similar regionalist conflicts erupted in areas of public good provision, including education, water management, public building, and sanitation, but the most salient divergence in preferences revolved around transportation development.

Highlanders sought a railroad from Quito across the spine of the Andes to the coast. Coastal planters, who relied on river transport but were the principal contributors to the fisc through import taxation, did not need the railroad for their export trade to thrive (Henderson, 1997, p. 172). Guayaquil made significant efforts to develop river navigation and steamship commerce along the coast, but had little interest in extending infrastructure into the highlands (Deler, 1994, p. 312ff). Coastal resistance in Congress prevented progress on railroad development, and a rail link connecting Quito and Guayaquil was completed (as discussed below) only in 1908.

The debilitating effects of regionalism drove political instability and “seemingly endless coups, pronouncements, and insurrections” (Rodríguez, 1985, p. 32). Regional interests, pursued both through institutional and extralegal means, fostered a series of conflicts including the 1859 Civil War, which threatened the country with dismemberment, and several other major conflicts. Regional tensions flared again when economic crisis broke out in the mid-1920s, as highland planters “rebelled” against the political hegemony of the coast and its emphasis on cacao export, and demanded the modernization of highland agriculture through technological investment and the facilitation of imports (Saint-Geours, 1994, pp. 179-180).

Not surprisingly, national leaders sought to contain regionalism and find a compromise that would prevent subsequent rebellions. Compromise could only be generated through massive pork-barrel politics that undermined infrastructure priorities (Spindler, 1987, p. 134). Congress used vast earmarks to log-roll nearly every public works project proposal it received from regional administrators, and thus abdicated the ability to coordinate infrastructure at the national level. Inefficiency was so high that in 1905, only 55 of the 346 public works projects authorized were under construction, with the remainder either abandoned or never begun (Rodríguez, 1985, p. 95). The result was that despite the ample revenues available to the state from cacao exports, “the country’s leaders failed to build the infrastructure necessary to modernize Ecuador” (Rodríguez, 1985, p. 1).

A Narrow Cross-Regional Consensus and Railroad Development

As mentioned above, a railroad linking Quito and Guayaquil was completed in 1908. This represents a major exception to the broad pattern of low public good provision described in this section, and it provides an opportunity to confirm the power of the regionalist explanation I have proposed. Clark (1998) shows that railroad construction was a product of a confluence of coastal and highland economic interests that emerged at a particular historical moment. Notably, this consensus did not extend to a broader range of public good provision—even other branches of the railroad network were excluded, and their construction was further delayed by decades or permanently stymied. Thus, the regional explanation can account not only for the general pattern but also for this important exception, which is not associated with a decline in the salience of any identity-based or ethnic cleavage. It thus outperforms alternatives in its ability to match the contours of the historical record of public good provision in Ecuador.

The engineering challenges to railroad development in Ecuador were vast. But this was not the major obstacle to rail development; instead, political deadlock between Quito and Guayaquil affected the allocation of funding and moved it down the list of government priorities for much of the second half of the 19th century. Only when highland and coastal interests coincided on the need for railroad development did it ensue, and most construction was accomplished in the course of a single decade leading up to the 1908 inauguration of the line connecting the two cities.

Unlike railroads in many other developing countries, the main economic function of the Quito-Guayaquil train line was not related to an effort to generate exports of highland agricultural production, since these could not compete on the world market. Highland elites focused instead on its effects for the domestic market - they hoped that reduced transport costs would facilitate the replacement of imports with domestic agriculture and livestock production in coastal consumption. (Clark 1998, 54) The growing interest of highland elites in market expansion crystallized late in the 19th century with the onset of a series of elements of modernization of production, including new cultivation methods, the rise of credit and banking, and along with the railway, agriculturalists envisioned incipient industrialization of the highlands, including vestigial food-processing, leather tanning, and textile production. None of these “could have expanded without the railway.” (Clark, 1998, p.119).

Nor were increased customs revenues from highland agricultural export a consideration in coastal elites’ shift to supporting the railroad. Instead, this was part of a broader effort to increase labor supply in the cacao sector. Coastal elites came to pursue the railroad after regional labor shortages began to limit their agricultural production; they hoped it would boost labor migration from the highlands. The 1895 Revolution brought coastal elites to power, and allowed them to use state power to increase the labor supply on the cacao plantations of the coastal region, which became an ever larger concern as cacao production boomed. A series of measures they introduced after 1895 that sought to weaken the hold of highland elites over the indigenous labor supply had little effect. (Clark 1998, Chapter 4) In this context, the railroad, with the transfer of resources from coast to highlands that it implied, became a palatable, if sub-optimal, strategy to increase the labor supply in the cacao sector.

Although elites in the two regions did not share a common vision of the broad project of national development, the railroad served the goals of each group, and a narrow consensus was forged that facilitated its development. As Clark (1998) writes, “Neither group could impose a project entirely in its own interest, but each could hope to gain something from the railway” (p. 55). Thus an “uneasy agreement” or “tenuous consensus” (Clark, 1998) between regions was reached that, while it held, allowed this project of public good provision to proceed. Conflict between regions continued to simmer over development priorities, as the coast was reluctant to pay for state investment in highland agricultural development, whereas the highland elites resisted the elimination of debt peonage that the coast sought to further stimulate labor migration. But on this issue, a narrow consensus held for long enough to ensure that the railroad was funded and made a national priority, and thus it represents an exception to the broad pattern of low public good provision in post-independence Ecuador. Indeed, the pattern of regional tensions generating political conflict over public good priorities has continued to characterize Ecuadorian politics.¹²

Regional Conflict and the Roots of Politicized Identities

This case study has argued that regionalist conflict, and not politicized ethnic diversity, underpinned the poor record of public good provision in post-independence Ecuador. Though Ecuador had high levels of ethnic diversity, and some degree of salience of regional identities, these did not stymie infrastructural development, which was traced instead to economic differences across regions, which shaped distinct preferences for public goods. One can perhaps go a step further yet: Some scattered evidence suggests economic conflict has fed the emergence of regional identities.¹³

Thus, Ecuador’s regionalized political economy may underpin not only its low level of public good provision but also the salience of the identities said to drive low public good provision. Kasza (1981) makes this argument most explicitly in the Ecuadorian context, writing that “the cultural cleavages dividing the coast and the sierra are not so radical as to provoke conflicts in and of themselves; they serve mainly to embitter disputes caused by competing economic and political interests” (p. 5). Regional identities that began as fundamentally economic (Saint-Geours, 1994, p. 145) came to cleave Ecuadorian political debates and shape political alignments at both the mass and the elite levels, as historical experience “reinforced resentment” about regional divides on political issues (Kasza, 1981, p. 30). Thus while a correlation can be found between ethnic diversity and low public good provision in Ecuador, one would be mistaken in drawing causal conclusions from it.

Self-Sufficient Regions and No Central Public Good Provision in 19th-Century Colombia

A review of Colombia’s political and economic history shows that regionalism caused low levels of public good provision, which are reflected both in its development of railroad infrastructure (second-lowest railroad density in the region) and its relatively low literacy rate, as well as in many other aspects of state activity (Deas, 1982). Moreover, regions in Colombia, unlike in Ecuador, do not overlap with ethnic divisions to any significant extent, meaning that the regionalist tensions that I show to cause low public good provision operate independent of the country’s level of ethnic diversity. Thus, although Alesina et al., (2003) would claim Colombia as a case fitting their model—its fractionalization score is one of the highest in Latin America—I show that correlation in this case does not imply causation.

How Regionalism Undermined Public Good Provision

Like Ecuador, 19th-century Colombia lacked a single predominant urban center. But unlike Ecuador, which had two salient poles of demographic and economic regions, the country’s population was dispersed widely: While Bogotá had just shy of 30,000 residents in 1851, 30 other municipalities (including regional centers like Popayán, Socorro, and Medellín) had between 8,000 and 15,000. Because these regional cities were large enough to absorb local economic production, trade within the country was limited. Palacios (2006) describes Colombia as a “mosaic of isolated regions” (p. 5). Regions were oriented around their central cities rather than linked to the national capital. Even plans to develop internal commerce as a means to economic development tended to focus on fairly local projects rather than ones that crossed regions (Rosenthal, 2012, p. 125). There was simply no vision of Colombia as a national economic unit centered on Bogotá and “there was little to draw [elites] automatically to the capital” (Christie, 1979, p. 44). As a result of its lack of urban primacy, Colombia was unique “in the deep imprint its regionalist sentiment has left on the pattern of national development” (Park, 1985, p. 7).

Yet as Posada-Carbó (1996, p. 3) argues, regional divisions in Colombia were fundamentally economic rather than about “subjective identity.” This is further support for a theory of regionalism and public good provision that operates independently of ethnic or cultural diversity. The Ecuadorian case, discussed above, saw regionalist conflict over public good provision overlap with ethnic differences across regions. There I showed that the association with ethnic diversity was spurious. In Colombia, the national state undertook no significant project of public good provision in a context marked by regionalism that did not align at all with an identity-based cleavage.¹⁴

Though regionalist conflict over public good provision unfolded here as in Ecuador, the more important mechanism by which regionalism undermined public good provision was that the economic self-sufficiency of Colombia’s regions reduced the gains perceived from provision of public goods by the national state. Because each of Colombia’s regions had a large enough population and a diverse production profile, they were largely self-sufficient, and to the extent that they traded, enjoyed independent links to the world market.¹⁵ As a result, the gains perceived from the potentially costly process of integration were quite limited, and there was no imperative to undertake an effort to overcome the regional tensions. The fact that Colombia was “economically invertebrate” (Safford, 1988, p. 53) made the payoffs from establishing a powerful state unclear and made the path to development through the fostering of regional self-sufficiency seem plausible. As a result, “the integration of a national market, and therefore transportation improvement, may have been less fundamentally important to Colombians” than it was to political elites in our other cases (Safford, 1988, p. 37). Regions were tied into the world economy “without the presence and involvement of the central state” (Palacios, 1980, p. ix).

Trade between Colombia’s regions was minimal throughout the 19th century (Ramírez & Salazar, 2010, p. 427). One exception was textiles from Santander, which did supply clothing in other states. But food and other basic commodities were rarely exchanged across regions (Safford, 2010, p. 530). Even as the coffee export boom took off, peasants continued to grow food for local consumption, and larger haciendas cushioned against fluctuating coffee prices by producing food, sugar, and beef for local markets (Palacios, 1980, pp. 93-94). Regions evolved largely independent of one another in economic terms, and pursued economic policies chosen with regional rather than national imperatives in mind. One clear instance of this regionalist emphasis can be seen when a prominent Antioquian politician spoke of the need to promote “industries producing goods which we unnecessarily bring from other states” (Park, 1985, p. 254). He was not referring to the need to protect Colombian markets against imports from abroad but Antioquian markets against imports from other Colombian states.

Colombia enjoyed a series of commodity booms in the 19th century, culminating with the coffee booms. Because booms were regionally concentrated, they drove a certain amount of infrastructure development in particular locations. Tobacco, for example, led to the improvement of Magdalena River steam navigation. But none of these resulted in attempts to “promote other sorts of exports or promote regional integration by trade” (McGreevey, 1971, p. 236). Even the huge coffee booms, which transformed Colombian politics, had limited effects on infrastructure: roads remained undeveloped, mules remained pivotal to getting coffee to export markets, and most of the coffee producing regions remained unconnected to the railroad network as late as 1920 (Palacios, 1980, p. 12). The limited transport development that did occur linked each region to the external market, rather than unifying the country and its economy. Perhaps the starkest evidence of this absence of unification was the failure to coordinate railroad track widths across regions. Even where multiple train lines intersected, goods could not easily pass from one region to another (Safford, 2010, p. 563).

Unlike their counterparts in many other Latin American countries, Colombian elites from across the political spectrum believed that development should be pursued by the elimination of state intervention in the economy. This strikingly laissez-faire ideology was “virtually a dogma” across the political spectrum for the first century after independence, and most Conservatives “were also economic liberals” (Safford, 1988, pp. 36, 58). At the height of this laissez-faire moment, under the 1863 Constitution, the national government oversaw nothing more than foreign relations, national defense, trade policy, and the regulation of weights, measures, and coinage (Delpar, 1981, p. 12). Beyond these basic functions, all power was delegated to the states, including some degree of control over the postal system and public education. Even the purportedly centralizing government of the Regeneration (1886-1899) chose not to pursue the central provision of public goods. Its main political figure, Rafael Núñez, believed that transportation infrastructure simply required the creation of an attractive investment climate (Bergquist, 1986, p. 222). A Conservative president of this period revealed the limited ambitions of the central state by claiming that improved river transport “was in the hands of the heavenly powers” (cited in Posada-Carbó, 1996, pp. 157-158). Another, speaking even more broadly about the limited ambitions of government, stated that he “should not be judged for the new bricks that he laid, but for the tremendous ruins that he avoided” (cited in Bushnell, 1993, p. 164).

In this context, we can see that the striking fragmentation of Colombia underpinned the absence of a state-building project that would seek to pursue development through the provision of public goods. This included not only transportation but also schooling—Colombia was an outlier in the absence of universal schooling throughout its first century after independence, making no attempt to standardize school curricula and promote national education like many of its neighbors (Soifer, 2015, pp. 134-137). It also made no effort to undertake other aspects of state building, failing to build a national army or systematize conscription practices and allowing individual states to undertake the independent importation of weapons and war materials (Soifer, 2015, pp. 229-230).

The striking hegemony of Colombian anti-statist views about development derived from the self-sufficiency of the country’s regions, which led policy makers to see “order and progress”—the sorts of goals that drove projects of public good provision elsewhere in the region—as goals to be pursued at the regional, rather than national, level. Regionalism, in other words, shaped the content of Latin American liberalism, which varied sharply across countries. Where a single national center existed and state building could unfold as the extension of its authority outward over the national territory, liberalism was marked by a more positivist tenor. By contrast, where a single national center did not exist, as in cases like Colombia, the extension of central state authority and the national provision of public goods were less likely to be seen as propitious strategies for economic and political development. The mechanism by which regionalism shaped the underprovision of public goods in Colombia operated through its effect on what Singh and vom Hau (2016) call normative commitments of state leaders to public good provision.

Urban Primacy and State Capacity: Region-Wide Evidence

Having shown that the proposed mechanisms can be found in various cases, I now turn to broader correlational evidence that salient regionalism is associated with low public good provision. Spanish South America is an appropriate locus for studying the effects of regionalism because its countries share some core properties in common, and thus one can rule these out as sources of variation. These include a common colonial background and a common independence process, broadly similar political economies centered on primary product exports through the first century after independence, and the common absence of the sorts of wars and territorial conflicts that promoted state building and public good provision elsewhere (Centeno, 2002).¹⁶ Moreover, countries in the region display striking long-term continuities in social and economic development (Mahoney, 2010), raising the importance of the study of the formative eras in which these long-term trajectories first crystallized and justifying the turn to the 19th century, when variation emerged (Soifer, 2015).

In this cross-national analysis, I use urban primacy as a proxy for regionalism, using population share as a crude but telling indicator of political and economic importance.¹⁷ A country has a primate urban distribution, and is dominated by a single center, when a disproportionate share of its population is found in a single city (Ades & Glaeser, 1995; Zipf, 1941). The standard measure of urban primacy calculates a ratio of the size of the largest city in a country to the sum of the four largest cities. Ordered scores for each country are shown in Table 1 below, which also includes the four largest cities by population for the year used.¹⁸ Because the standard measure of primacy divides the population of the largest city by the sum of that of the four largest, the primacy score can vary from 0.25 (if the four cities are equal in size) to 1 (if the largest city contains the entire population of the country). In Latin America, urban primacy scores vary from Uruguay at 0.84 to Venezuela and Paraguay at 0.39. The regional average is 0.54, with a standard deviation of 0.15.

Table 1.

Urban Primacy in South America.

Country	Year	Primacy score	Total population	Largest city	Population	Next three	Population
Uruguay	1908	0.84	1,042,700	Montevideo	309,400	Paysandú	21,000
						Salto	19,800
						Mercedes	15,700
Brazil	1866	0.73	9,221,000	Rio de Janeiro	420,000	Recife	95,000
						São Luis	30,000
						Belem	25,000
Argentina	1847	0.66	1,006,600	Buenos Aires	82,000	Paraná	15,000
						Córdoba	14,000
						Mendoza	12,000
Peru	1876	0.6	2,469,000	Lima	100,100	Callao	34,500
						Huaraz	17,000
						Cerro de Pasco	13,000
Mexico	1867	0.55	7,485,200	Mexico City	242,000	Guadalajara	78,600
						Puebla	75,500
						Queretaro	47,600
Chile	1865	0.53	1,819,200	Santiago	115,400	Valparaíso	70,400
						Talca	17,900
						Concepción	14,000
Ecuador	1885	0.47	1,004,700	Quito	80,000	Guayaquil	40,000
						Cuenca	30,000
						Riobamba	18,000
Bolivia	1858	0.46	1,742,400	La Paz	76,400	Cochabama	40,700
						Sucre	24,000
						Potosí	22,900
Colombia	1851	0.44	2,223,800	Bogotá	43,000	Popayán	20,000
						Socorro	20,000
						Medellín	14,000
Paraguay	1879	0.39	476,000	Asunción	22,000	Villarica	13,000
						Concepción	11,000
						San Pedro	10,000
Venezuela	1874	0.39	1,784,200	Caracas	48,900	Valencia	28,600
						Barquisimeto	25,700
						Maracaibo	22,000

The correlations of urban primacy with proxies for levels of public good provision are very strong, as seen in Figure 1. Across the 10 countries in the sample, urban primacy is correlated with railroad density (kilometers of track per square kilometer of territory) at .742 and with the literacy rate at .696. This suggests the existence of a region-wide relationship that operates through the mechanisms hypothesized above. In each case, higher urban primacy scores are associated with stronger states. These correlations remain sizable (in results not shown here) even with the exclusion of Uruguay, which is an outlier in its high primacy score. Confidence about the direction of the proposed relationship is increased by the fact that primacy is measured for the earliest post-independence date possible and at an earlier point in time than the literacy rate or railroad density. Thus, we can rule out the possibility that public good provision could affect urban primacy—for example, by drawing population to the capital city.¹⁹

A skeptical reader might think that this strong correlation between urban primacy and public good provision is a consequence of an underlying geographic factor that affects one or both of these. Yet other aspects of geography cited as affecting state development—mountainousness (Fearon & Laitin, 2003), ruggedness (Nunn & Puga, 2012), and size (Herbst, 2000)—are poorly correlated with these measures of public good provision. As Table 2 shows, none of these aspects of geography has a strong correlation with either literacy or railroad density. Moreover, the coefficients are inconsistent with the predictions from this scholarship. The signs on the correlation coefficients are backward in some cases; for example, more mountainousness is associated with greater railroad development, and (against the predictions made by Herbst) greater size is associated with more public good provision. Their relative size is also puzzling: Ruggedness has a stronger relationship with literacy than with railroad density. Thus, there is reason to cast doubt on these geographic explanations as the underlying cause of the correlation between urban primacy and public good provision in Latin America.

Table 2.

Correlations of Proposed Causes of Public Good Provision.

	Literacy	Railroad density
Urban primacy	.696	.742
Mountainousness (Fearon & Laitin, 2003)	−.136	.161
Ruggedness (Nunn & Puga, 2012)	−.304	−.167
Size (sq km; CIA World Factbook)	.156	.215
Ethnic Fractionalization (Alesina et al., 2003)	−.533	−.795
ELF (Fearon & Laitin, 2003)	−.465	−.239

Note. Country scores for geographic characteristics are calculated based on contemporary national boundaries. Sources for literacy and railroad density data are listed in Note 8. CIA = Central Intelligence Agency.

Table 2 also shows the correlations of country scores on railroad density and literacy for two indicators of ethnic diversity—the Alesina et al. (2003) scores and the ELF scores (drawn from Fearon & Laitin, 2003). As can be seen at a glance in Figure 1, correlations of the Alesina et al. (2003) measure are equally strong as those for urban primacy, while those using the ELF score with levels of public good provision are far weaker. Yet one cannot conclude from this that fractionalization causes low levels of public good provision, given that the fractionalization measure, which is calculated for the 1980s or 1990s in the Alesina et al. (2003) data set, is measured after the proposed outcome of public good provision. Moreover, the fractionalization scores for all Latin American cases in the data set are based on the Encylopedia Brittanica, which does not describe the sources for its discussion of the ethnic composition of national populations. Most likely its estimates are drawn from national census data. Yet Latin American countries assess ethnicity in their censuses in part based on questions about language (Loveman, 2014, pp. 254-263). As national language is acquired via primary schooling, a core public good, the correlation between homogeneity and public good provision may be accounted for, as Wimmer argues in his article for this special issue, by a common underlying variable of state capacity. Unlike this measure of ethnic diversity, the urban primacy measure of regionalism uses data from an earlier date than the public good provision indicators and avoids this problem. This fact, combined with the historiographical evidence presented above and the fact that the correlational evidence is as strong as fractionalization, suggests that the account I have proposed outperforms the ethnic diversity argument in accounting for variation in development in Latin America.

Change Over Time: Evidence From Venezuela

Additional evidence for the regionalism explanation I have advanced can be seen in a cross-time examination of trends in regionalism and public good provision in Venezuela, where a sharp increase in the primacy of Caracas in Venezuela, without a concomittant decrease in ethnic diversity, was associated with an increase in public good provision. Venezuela’s over-time trajectory thus confirms the power of the regionalist argument and shows that it operates independent of the ethnic diversity accounts.

In the early decades after independence, Venezuela was deeply fragmented. Though this was in part a function of the fragmented mobilization during the independence war (López-Alves, 2000, p. 201), each of Venezuela’s Atlantic ports “had close links to its own hinterland” and a salient regional economy (Ewell, 1984, p. 14). Maracaibo and the gold-mining Guayana region “openly challenged the authority of Caracas on several occasions” (Tarver & Frederick, 2006, p. 69) as they saw little benefit from national integration. The political solution to this conflict mirrored that seen in Colombia: The 1864 Constitution empowered local political actors at the expense of the national state. Though political elites envisioned a project of development through immigration and transportation development, this was expected to unfold through private initiative rather than the state provision of public good provision (Ewell, 1984, p. 21). This is exactly the pattern of low public good provision that the regionalism argument would predict based on Venezuela’s low urban primacy score shown in Table 1 above.

But after the 1870s, the country saw an increasing concentration of population and economic activity in Caracas and on the nearby north coast (Lombardi, 1982, pp. 194-196) as the population of the capital doubled in succeeding decades. This went along with a decline in the salience of regional divisions, and a rise in economic ties between the capital and many formerly self-sufficient regions (Ewell, 1984, p. 9). This shift away from regional fragmentation in demographic and economic terms coincided with increased public good provision by the national government. Under President Cipriano Castro (1899-1908) the country saw “the timid beginnings of the state commitment to a responsibility for economic development” and state involvement in public good provision, including the development of a national telegraph and rail network (Ewell, 1984, p. 29). This escalated under the dictatorship of Juan Vicente Gómez (1908-1935), which saw concerted efforts at public good provision, including the construction of the country’s national highway system, as part of a project of political, administrative, and economic centralization (Yarrington, 2003, p. 14). Public good provision in the pursuit of economic development in Venezuela was, of course, distorted with the arrival of the oil boom that began in the mid-1920s (Coronil, 1997; Salas, 2009). But the increased efforts by the central state to develop infrastructure and reshape society during the late 19th and early 20th century had a significant impact (Yarrington, 2014).

The case of Venezuela thus shows a shift in public good provision that coincides with a decline in regionalism and the rise of Caracas as a single national center. This shift over time provides further support for the importance of regionalism as a heretofore unexplored factor in accounting for variation in public good provision. Moreover, the increased provision of public goods cannot be traced to a decline in ethnic diversity: As shown by Wright (1990), the country continued to be characterized by a “multi-racial majority” (p. 79), and elite discourse continued to view race as a fluid, largely socioeconomic category. Along with public good provision, state leaders sought to increase European immigration as a means of “whitening” the country’s population, but these efforts bore little fruit and resulted in no changes in the demographic composition of the country. Thus, the Venezuelan case provides additional evidence that increased urban primacy, but not declining ethnic diversity, accounts for increased public good provision. This over-time, within-case evidence, when combined with the correlational evidence presented above and the process-tracing evidence of the two mechanisms in Colombia and Ecuador, makes a strong case for the regionalism argument.

Conclusion

This article began by first developing a set of proposed mechanisms that link regionalism—entirely independent of ethnic or cultural heterogeneity—to low levels of public good provision. Second, I have shown each of these mechanisms at work in a historical case study. The Ecuador case showed that regions have distinct public good preferences, which generates political deadlock, conflict, and low levels of public good provision. In the Colombian case, I showed that self-sufficient regions shaped an elite consensus that the national state should not prioritize public good provision and should instead leave its development to the regions. Third, I have shown that the absence of urban primacy (a proxy for regionalism) was associated with low levels of public good provision across Spanish South America during the first century after independence, with correlation coefficients as strong as or stronger than those found for ethnic heterogeneity in this set of cases, and that declining regionalism in the case of Venezuela was associated with increased public good provision.

Thus, we have strong evidence that regionalism is associated—independent of ethnic diversity—with lower public good provision. Yet these findings do not support the possibility that regionalism acts as an intervening variable or causal mechanism linking ethnic diversity to low levels of public good provision. For one thing, primacy is correlated with the fractionalization index only at −.336, and with the ELF scores only at .070; this strongly suggests that regionalism operates through a distinct causal channel. I have also presented ample evidence in both case studies showing that decisions about public good provision were shaped by divides across region, and not by ethnic diversity.

The level of regionalism in a country, one should note, is not sufficient to explain the level of public goods provided to its residents. The theory developed in this article is incomplete in two important ways. First, although it can account for national government policy in regards to public good provision, it cannot account for the differential provision by subnational and local authorities, which is sizable in some cases. Why public goods are provided to divergent extents by subnational administrative units, however, goes beyond the scope of the explanation I develop. I have shown that regionalist conflict is one reason why provision might be devolved below the national state, though there are of course many other explanations for this phenomenon. But under these circumstances, the salience of regionalism—a national-level variable—cannot explain the levels of public good provision by subnational units.²⁰ Second, my focus has been on the policies determined by political institutions and not their implementation: To the extent that policies are incompletely implemented, the causal framework developed in this article will over-predict levels of public good provision in cases where regionalism is not salient.²¹ Importantly, neither of these limitations is unique to the regionalism account. State-centered accounts of public good provision in general, though they improve in important ways on ethnic diversity-collective action frameworks, cannot on their own account for the full distribution of levels of public good provision.

The purpose of this article has been simply to account for the part played by regionalism in accounting for variation and to question the conventional wisdom about the part played by ethnic diversity. The positive finding about the relationship between regionalism and public good provision is worthy of broader examination in its own right. Most broadly, one can see echoes of the account I have sketched in studies of how inter-sectoral linkages shape economic geography and economic development. Like Hirschman (1977), Saylor (2014), and Cardoso and Faletto (1979), my theory traces variation in public good provision to the nature of a state’s incorporation into the global economy: One similarity among all the cases discussed in this article is their reliance on commodity exports. Indeed, both of the mechanisms I have elaborated are underpinned by the presence of limited linkages among distinct regions; the salience of economic regionalism is defined by low levels of linkages. Yet while this scholarship has focused on economic development as the outcome of interest, I seek to explain how levels of demographic and economic regionalism shape public good provision.²²

Similar dynamics to those described in the cases in this article have been explored by scholars in cases far beyond Latin America. Perhaps the most prominent example is Bensel (1984), which traces conflict between regions with distinct economic interests over fundamental elements of public good provision throughout the post-Civil War era in the United States. Much like my account of Ecuador above, Bensel (1984) shows that American political development was shaped by “continuous political rivalry” (p. xix) between regions with different economic profiles and different orientations toward the world market but little “religious or ethnic rivalry” (p. 4). One important result of sectionalism was that the national state in the United States was a striking outlier in its low level of public good provision.

In a very different context, the most striking instance of state failure in early modern Europe, Poland, was also marked by salient regionalism. In 1600, six Polish cities were at least as large as the capital of Cracow, each with its own, economically distinct hinterland (Wyrobisz, 1989). This regionalism had no significant ethnic or religious dimension (Romaniszyn, 2005). But the inability of the central authority in Cracow to overcome “the centrifugal tendencies of its many regions” (Davies, 2005, p. 26) long delayed the creation of a single national currency (Davies, 2005, p. 103) and stymied other efforts to knit the country together economically through public good provision. It also played a crucial role in the creation of Poland’s famously dysfunctional political institutions, including the liberum veto that allowed any individual to block proceedings of the council of nobles (Downing, 1993).

Thus demographic and economic regionalism, though heretofore little studied, may extend its explanatory power to variation in state development and public good provision far beyond the cases studied in this article. To the extent that it remains unincorporated into scholarship on these questions, it represents an omitted variable in the ethnic diversity scholarship that should be taken into account. Yet one might go further to endogenize diversity by exploring whether interregional economic and political conflict over spending priorities and public good provision cause the identity-based cleavages to which underprovision is traced to become politicized in the first place, as I have suggested that it might characterize the case of Ecuador. To the extent that this claim can be made plausible and more general, arguments that link ethnic diversity to the underprovision of public goods are further called into question. Thus, this article provides a caution for existing scholarship as well as directions for further investigation of the limits of the diversity-development deficit nexus.

Footnotes

Acknowledgements

Earlier versions of this paper were presented at the annual meetings of the American Political Science Association and the Latin American Studies Association, and in seminars at Stanford University and the Catholic University of Chile. I received helpful comments and suggestions from audience members, discussants, and fellow panelists on each of these occasions. I also thank Kent Eaton, Ryan Saylor, Michael Bernhard, and the editors of the special issue for suggestions and guidance, and Travis Blemings and Cassandra Emmons for excellent research assistance.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Hillel David Soifer is an Assistant Professor of Political Science at Temple University. His book State Building in Latin America was published by Cambridge University Press in 2015. One plank of his research agenda centers on the origins, development, and effects of state strength and weakness in Latin America.

References

Ades

A. F.

Glaeser

E. L.

(1995). Trade and circuses: Explaining urban giants. Quarterly Journal of Economics, 110, 195-227.

Agnew

J. A.

(2002). Place and politics in modern Italy. Chicago, IL: University of Chicago Press.

Alesina

Devleeschauwer

Easterly

Kurlat

Wacziarg

(2003). Fractionalization. Journal of Economic Growth, 8, 155-194.

Alesina

Spolaore

(2005). The size of nations. Cambridge, MA: MIT Press.

Appelbaum

N. P.

(2003). Muddied waters: Race, region, and local history in Colombia, 1846–1948. Durham, NC: Duke University Press.

Baldwin

Huber

J. D.

(2010). Economic versus cultural differences: Forms of ethnic diversity and public goods provision. American Political Science Review, 104, 644-662.

Bensel

R. F.

(1984). Sectionalism and American political development, 1880-1980. Madison, WI: University of Wisconsin Press.

Bergquist

(1986). Coffee and conflict in Colombia 1886-1910. Durham, NC: Duke University Press.

Bushnell

(1993). The making of modern Colombia: A nation in spite of itself. Berkeley, CA: University of California Press.

10.

Campante

F. R.

Q. A.

(2014). Isolated capital cities, accountability, and corruption: Evidence from US State. The American Economic Review, 104, 2456-2481.

11.

Campante

F. R.

Q. A.

Guimaraes

B. V.

(2013, May). Isolated capital cities and misgovernance: Theory and evidence (NBER Working Paper No. 19028). Cambridge, MA: National Bureau of Economic Research.

12.

Carbó

E. P.

(1996). The Colombian Caribbean: A regional history, 1870-1950. Oxford, UK: Oxford University Press.

13.

Cardoso

F. H.

Faletto

(1979). Dependency and development in Latin America. Berkeley, CA: University of California Press.

14.

Centeno

(2002). Blood and debt: War and the nation-state in Latin America. University Park, PA: Penn State University Press.

15.

Chartock

(2013). “Corporatism with adjectives”? Conceptualizing civil society incorporation and indigenous participation in Latin America. Latin American Politics and Society, 55, 52-76.

16.

Christie

K. H.

(1979). Gamonalismo in Colombia: An historical overview. North-South Canadian Journal of Latin American Studies, 4, 42-59.

17.

Clark

A. K.

(1998). The redemptive work: Railway and nation in Ecuador, 1895-1930. Wilmington, DE: SR Books.

18.

Coronil

(1997). The magical state: Nature, money, and modernity in Venezuela. Chicago, IL: University of Chicago Press.

19.

Clark

A. K.

(1998). The redemptive work: Railway and nation in Ecuador, 1895-1930. Rowman & Littlefield.

20.

Davies

(2005). God’s playground a history of Poland: Volume I: The Origins to 1795. Oxford, UK: Oxford University Press.

21.

Deas

(1982). The fiscal problems of nineteenth-century Colombia. Journal of Latin American Studies, 14, 287-328.

22.

Deler

(1994). Transformaciones regionales y organización del espacio nacional ecuatoriano entre 1830 y 1930. In Maiguashca

(Ed.), Historia y región en el Ecuador: 1830-1930 (pp. 295-353). Quito, Ecuador: Corporación Editora Nacional.

23.

Delpar

(1981). Red against blue: The liberal party in Colombian politics, 1863-1899. Tuscaloosa, AL: University of Alabama Press.

24.

Downing

B. M.

(1993). The military revolution and political change: Origins of democracy and autocracy in early modern Europe. Princeton, NJ: Princeton University Press.

25.

Eaton

(2007). Backlash in Bolivia: Regional autonomy as a reaction against indigenous mobilization. Politics & Society, 35, 71-102.

26.

Eaton

(2011). Conservative autonomy movements: Territorial dimensions of ideological conflict in Bolivia and Ecuador. Comparative Politics, 43, 291-310.

27.

Engerman

S. L.

Sokoloff

K. L.

(2012). Economic development in the Americas since 1500: Endowments and institutions. New York, NY: Cambridge University Press.

28.

Ewell

(1984). Venezuela: A century of change. Stanford, CA: Stanford University Press.

29.

Fearon

J. D.

Laitin

D. D.

(2003). Ethnicity, insurgency, and civil war. American Political Science Review, 97, 75-90.

30.

Gootenberg

P. E.

(1989). Between silver and guano: Commercial policy and the state in postindependence Peru. Princeton, NJ: Princeton University Press.

31.

Henderson

(1997). Cocoa, finance and the state in Ecuador, 1895–1925. Bulletin of Latin American Research, 16, 169-186.

32.

Henderson

P. V.

(2008). Gabriel García Moreno and conservative state formation in the Andes. Austin: University of Texas Press.

33.

Herbst

(2000). States and power in Africa: Comparative lessons in authority and control: Comparative lessons in authority and control. Princeton, NJ: Princeton University Press.

34.

Herrera

Y. M.

(2005). Imagined economies: The sources of Russian regionalism. Cambridge, UK: Cambridge University Press.

35.

Hirschman

A. O.

(1977). A generalized linkage approach to development, with special reference to staples. Economic Development and Cultural Change, 25, 67-98.

36.

Kasza

G. J.

(1981). Regional conflict in Ecuador-Quito and Guayaquil. Inter-American Economic Affairs, 35, 3-41.

37.

Krugman

(1991). Increasing returns and economic geography. The Journal of Political Economy, 99, 483-499.

38.

Kurtz

M. J.

(2013). Latin American state building in comparative perspective: Social foundations of institutional order. Cambridge, UK: Cambridge University Press.

39.

Lombardi

J. V.

(1982). Venezuela: The search for order, the dream of progress. New York, NY: Oxford University Press.

40.

López-Alves

(2000). State formation and democracy in Latin America, 1810-1900. Durham, NC: Duke University Press.

41.

Loveman

(2014). National colors: Racial classification and the state in Latin America. Oxford, UK: Oxford University Press.

42.

Mahoney

(2001). The legacies of liberalism: Path dependence and political regimes in Central America. Baltimore, MD: Johns Hopkins University Press.

43.

Mahoney

(2010). Colonialism and postcolonial development: Spanish America in comparative perspective. New York, NY: Cambridge University Press.

44.

Maiguashca

(Ed.). (1994). Historia y región en el Ecuador: 1830-1930. Quito, Ecuador: Corporación Editora Nacional.

45.

McGreevey

W. P.

(1971). An economic history of Colombia 1845-1930. London, England: Cambridge University Press.

46.

Michalopoulos

(2012). The origins of ethnolinguistic diversity. The American Economic Review, 102, 1508.

47.

Nunn

Puga

(2012). Ruggedness: The blessing of bad geography in Africa. Review of Economics and Statistics, 94, 20-36.

48.

Olson

(2000). Power and prosperity: Outgrowing communist and capitalist dictatorships: Outgrowing communist and capitalist dictatorships. New York, NY: Basic Books.

49.

Palacios

(1980). Coffee in Colombia, 1850-1970: An economic, social, and political history. Cambridge: Cambridge University Press.

50.

Palacios

(2006). Between legitimacy and violence: A history of Colombia, 1875-2002. Durham, NC: Duke University Press.

51.

Park

J. W.

(1985). Rafael Núnez and the politics of Colombian regionalism, 1863-1886. Baton Rouge, LA: Louisiana State University Press.

52.

Pineo

(1994). Guayaquil y su región en el segundo boom cacaotero [Guayaquil and its region in the second cacao boom]. In Maiguashca

(Ed.), Historia y región en el Ecuador (pp. 251-294). Quito, Ecuador: Corporación Editora Nacional.

53.

Quintero

(1991). La Cuestión regional y el poder: Proyecto FLACSO-CERLAC, III [The Regional Question and Power: Project FLACSO-CERLAC III] (Vol. 29). Quito, Ecuador: Corporación Editora Nacional.

54.

Ramírez

M. T.

Salazar

(2010). El surgimiento de la educación en Colombia: En qué fallamos? [The growth of education in Colombia: What is missing?]. In Meisel Roca

Ramírez

M. T.

(Eds.), Economía Colombiana del Siglo XIX [The Colombian economy in the 19th century] (pp. 419-482). Bogotá, Colombia: Fondo de Cultura Económica.

55.

Rodden

(2010). The geographic distribution of political preferences. Annual Review of Political Science, 13, 321-340.

56.

Rodríguez

L. A.

(1985). Search for public policy: Regional politics and government finances in Ecuador, 1830-1940. Berkeley, CA: University of California Press.

57.

Romaniszyn

(2005). Europe and the formation of the Polish state, nation, and national identity. In Ichijo

Spohn

(2005). Entangled identities: Europe and the nation (pp. 154-171). London, England: Ashgate.

58.

Rosenthal

J. M.

(2012). Salt and the Colombian state: Local society and regional monopoly in Boyacá, 1821–1900. Pittsburgh, PA: University of Pittsburgh Press.

59.

Safford

(1988). The emergence of economic liberalism in Colombia. In Love

Jacobsen

(Eds.), Guiding the invisible hand: Economic liberalism and the state in Latin American history (pp. 35-62). New York, NY: Praeger.

60.

Safford

(2010). El problema de los transportes en Colombia en el siglo XIX [The problem of transport in 19th century Colombia]. In Meisel Roca

Ramírez

M. T.

(Eds.), Economía Colombiana del Siglo XIX (pp. 419-482). Bogotá, Colombia: Fondo de Cultura Económica.

61.

Saint-Geours

(1994). La Sierra Centro y Norte 1830-1925. In Maiguashca

(Ed.), Historia y región en el Ecuador: 1830-1930 (pp. 143-188). Quito, Ecuador: Corporación Editora Nacional.

62.

Salas

M. T.

(2009). The enduring legacy: Oil, culture, and society in Venezuela. Durham, NC: Duke University Press.

63.

Saylor

(2014). State building in boom times: Commodities and coalitions in Latin America and Africa. Oxford, UK: Oxford University Press.

64.

Singh

(2015). How solidarity works for welfare subnationalism and social development in India. New York, NY: Cambridge University Press.

65.

Singh

vom Hau

(2016). Ethnicity in time: Politics, history, and the diversity-development deficit hypothesis. Comparative Political Studies, 49(10), 1303-1340.

66.

Soifer

H. D.

(2013). State power and the economic origins of democracy. Studies in Comparative International Development, 48, 1-22.

67.

Soifer

H. D.

(2015). State building in Latin America. New York, NY: Cambridge University Press.

68.

Spindler

F. M.

(1987). Nineteenth century Ecuador: An historical introduction. Fairfax, VA: George Mason University Press.

69.

Spruyt

(1994). The sovereign state and its competitors: An analysis of systems change. Princeton, NJ: Princeton University Press.

70.

Stasavage

(2011). States of credit: Size, power, and the development of European polities: Size, power, and the development of European polities. Princeton, NJ: Princeton University Press.

71.

Stepan

A. C.

(1999). Federalism and democracy: Beyond the US model. Journal of Democracy, 10, 19-34.

72.

Tarver

H. M.

Frederick

J. C.

(2006). The history of Venezuela. New York, NY: Palgrave Macmillan.

73.

Van Aken

M. J.

(1989). King of the night: Juan José Flores and Ecuador, 1824-1864. Berkeley, CA: University of California Press.

74.

Weber

(1976). Peasants into Frenchmen: The modernization of rural France, 1870-1914. Stanford, CA: Stanford University Press.

75.

Wright

W. R.

(1990). Café con leche: Race, class, and national image in Venezuela. Austin: University of Texas Press.

76.

Wyrobisz

(1989). Power and towns in the Polish Gentry Commonwealth. Theory and Society, 18, 611-630.

77.

Yarrington

(2003). Cattle, corruption, and Venezuelan state formation during the regime of Juan Vicente Gomez, 1908-1935. Latin American Research Review, 38, 9-33.

78.

Yarrington

(2014). Tax farming, liquor, and the quest for fiscal modernity in Venezuela, 1908–1935. Hispanic American Historical Review, 94, 237-269.

79.

Yashar

D. J.

(2005). Contesting citizenship in Latin America: The rise of indigenous movements and the postliberal challenge. New York, NY: Cambridge University Press.

80.

Zipf

(1941). National unity and disunity. Bloomington, IN: Principia Press.