Varieties of State-Building: Ecology, Clientelism, and Bureaucratic Rule in Chile

Abstract

Current research suggests that all states share a perpetual appetite for extraction and standardization. However, this research overlooks the fact that subnational regions present different appeals and challenges to ruling coalitions. While states seek to extend bureaucratic rule over peripheries with valuable assets and favorable geography, they might instead seek to preserve local patrimonial bastions when those areas offer substantial electoral support. In turn, these strategies lead to broad subnational heterogeneity in the reach of the state. This paper focuses on regions’ ecological, military, and clientelistic features to explain local trajectories of bureaucratic rule and country-level state capacity. Empirically, I examine Chile, a successful case of capacity-building in Latin America. Prompted by a fiscal crisis in the mid-1850s, Chile’s central government launched state-building projects to offset its budgetary deficit. Using GIS and original data from censuses, budgets, and other primary sources, I show that Chile’s ruling coalition paradoxically modernized the country’s peripheries while deepening its own traditionalism. These results challenge prevailing narratives about the projection of political authority and Chile’s territorial uniformity.

Introduction

State-building—the extension of a ruler’s authority over distance—is essential for political development. The effort to project authority enhances the government’s presence across the territory, compels the formation of professional bureaucracies, and promotes an efficient and equitable production of public goods. Successful state-building also helps to prevent the emergence of stateless spaces, which can become the foundation for nonstate actors (e.g., rebel groups, criminal organizations) to challenge the state’s monopoly of violence and thereby threaten its existence.

Contrary to the dominant approach, which focuses on the strength of a state-building core, this paper focuses on subnational regions’ socioeconomic and physical features. ¹ Three types of factors, ecological, military, and clientelistic, strongly shape state-building projects, strategies, and outcomes. Regions with favorable ecologies—valuable commodities and advantageous geography—are targeted for state-building. However, the outcome—cooperative or imposed bureaucratization—depends on local elites’ ability to extract concessions through a military threat. When regions present unfavorable ecological conditions, leaders are unlikely to pursue state-building, settling instead for indirect rule. Paradoxically, ruling coalitions’ need to survive in office may compel them to form subnational enclaves to favor potential clientelistic allies, helping them to secure de facto autonomy and preferential access to public goods. In turn, the effort to disrupt ecologies and rule distant regions necessitates investments in administrative and technocratic reforms, contributing to national-level state capacity.

Empirically, this paper focuses on the dual effects of the global trade expansion of the mid-nineteenth century. As prices for raw materials increased rapidly, subnational regions previously considered insignificant gained newfound value. Advancements in transportation altered the perception of such peripheries, rendering them accessible and exploitable. However, the overall impact of expanded trade was a weakening of state authority, as ruling coalitions engaged in compromises with local elites, leading to revenue-sharing arrangements under the banner of indirect rule (Mazzuca 2021). Nonetheless, this period also witnessed a potent force driving state-building: fiscal shocks driven by commodity busts. Given a country’s typical reliance on a single commodity, sudden drops in international prices often triggered acute financial crises. In response, ruling coalitions strategically turned either to international credit or to state-building to produce or tax commodities they believed would gain value in the near future.

I assess this theory by studying nineteenth-century Chile. In a region known for endemic state weakness, Chile is a successful case of capacity-building in the absence of strong geopolitical pressures. Prompted by a fiscal crisis in the mid-1850s, Chile’s ruling coalition led several state-building efforts to offset fiscal vulnerability. The state negotiated a transition to bureaucratic rule in Atacama to the north and Concepción to the south, both peripheral areas with favorable ecological conditions that coalesced to make a military threat. Then, the state imposed bureaucratization in Araucanía, an area of dense forests and scattered geography in the far south dominated by the native Mapuche. Finally, to shore up political support, the government allied with landed elites in the capital’s hinterland, the Central Valley. Although Chile is often considered the poster child of evenly projected state power, this paper suggests that state-building was instead highly selective and uneven, with the very center being the most important deviation to uniform, bureaucratic rule.

The empirical section measures the state’s territorial reach using a geographic information system (GIS) with a combination of census and budget data. Censuses show individuals who identify as bureaucrats or members of the military and police, while budgets reflect the central government’s spending on wages and various public goods, including public works, the judiciary, and the police. The data is provided at the department level—the smallest administrative unit at the time. Unlike infrastructure-based indicators, this approach enables state-building and public goods to be analyzed separately.

My findings contribute to the study of state-building, democratization, and public goods distribution. First, the role of regional ecology can be employed to think about how states negotiate and regulate the contemporary expansion of illicit economies, including drug production, deforestation, and mining in hard-to-reach areas. Against conventional wisdom in democratization theory, this paper suggests that landowners can promote enfranchisement if they can secure a base of clientelistic support among peasants. Finally, my analysis suggests that there is a conceptual and empirical difference between public goods distribution and state capacity, as an increased flow of public goods to a region—commonly confounded with state capacity in the literature—can coexist and even thrive under patrimonialism.

In what follows, I discuss major works on state-building and Latin American politics, and chart a theory-building exercise that suggests how fiscal and coalitional needs combine with regions’ ecologies to shape subnational outcomes. Then, I provide a sketch of center–periphery relations in Chile and the events that led to state-building projects. Finally, I test the theory by showing the cross-sectional and temporal variation in the reach of Chile’s state and country-level state capacity, and by conducting case studies for each subnational outcome.

Theories of State-Building

Studies on the emergence of modern states have emphasized the role of warfare. In this approach, geopolitical pressures between rival feudal lords after the end of the Carolingian era in the ninth century eventually led to the concentration of coercive and extractive resources in the hands of the state. This “bellicist” hypothesis was first proposed by Hintze (1975), who suggested that geopolitical pressures began in the early sixteenth century once France interpreted the Habsburgs’ territories in Spain and the southern Netherlands as a security threat. The French crown responded by “eliminating provincial particularism, centralizing administration, and creating a standing army to boost French military effectiveness” (Ertman 2017, 54). These events created escalatory pressures that led Spain and Austria to react similarly. From then on, the creation of standing armies and tax collection became the exclusive authority of the king (Hintze 1975, 194–95).

Several works have expanded the bellicist thesis by adding new dimensions, mainly the availability of taxable resources. Wherever rulers had to tax individuals instead of imposing trade duties—a far more significant effort—the state became sturdier (Tilly 1985; 1992). The timing and scope of the military revolution have also been associated with variations in state capacity (Ertman 1997). Where state-building took place before the military revolution (ca. 1450), technologies of state expansion were obsolete and became hard to replace (Ertman 1997, 27). ²

A second school emphasizes states’ enduring interest in making nature and populations legible in order to extract revenue (Scott 1998; 2009). To achieve such legibility, states develop official measurements, censuses, statistical yearbooks, scientific forestry, and cartography to make their surroundings easier to read and tax. Recent empirical works have suggested that legibility is associated with centralized governance (Lee and Zhang 2017), suffrage expansion (Brambor et al. 2020), and fiscal capacity (Vom Hau, Peres-Cajías, and Soifer 2023).

A third group of works counter the idea of states as control- and revenue-maximizers by instead emphasizing instances of subnational heterogeneity within states’ reach. Some of these works focus on the projection of authority as a cost-benefit trade-off that includes a region’s distance to the center, population density, and geographic ruggedness (Alesina and Spolaore 2003; Herbst 2014). Related works that focus on postcolonial states suggest that ruling coalitions tend to allow for decentralized governance with regional elites or even leave areas wholly ungoverned in an effort to minimize political conflict (e.g., Boone 2003; Giraudy and Luna 2017; O’Donnell 1993; Sánchez Talanquer 2017; Slater and Kim 2015).

State-Building in Latin America

A growing consensus in the study of Latin American politics suggests that the aftermath of the independence wars of the early nineteenth century had long-term repercussions on state development. Those who share this view agree on two conclusions: (1) state weakness is common in the region, and (2) cross-national differences are consistent. The first wave of scholars to arrive at these conclusions evaluated the bellicist hypothesis. Centeno (2002) claimed that although Latin American countries did experience wars in this period, these conflicts did not increase states’ capacity since they were financed with foreign loans instead of direct taxation. Unlike the effort to tax individuals directly, loans do not encourage bureaucratic development. Centeno’s hypothesis was revisited by Thies (2005), Schenoni (2021), and Queralt (2022). Discussing Chile, they suggest that its divergence lies in its disputes with Peru and Bolivia throughout the nineteenth century, especially during the War of the Pacific (1879–84). For Schenoni, the results of wars—rather than preparations for them—put countries on different trajectories of state development. Queralt (2022, 273–89) argues that since Chile could not access credit during the War of the Pacific, ruling elites had to impose taxation on elites, enhancing institution-building in the long run.

Four more scholars have evaluated sources of Chile’s state capacity. Saylor (2014) argued that, in Chile, an export-oriented coalition demanded new public goods during the double boom of copper and wheat, which led to capacity-building. For Kurtz (2013), state-building was unlikely where rural elites depended on a labor-repressive agrarian economy. Echoing Moore’s (1966) thesis that landlords are the elite faction most hostile to democratization, Kurtz argued that landowners oppose state-building too, since a strong state is more likely to take away their control over coerced laborers. Soifer (2015) identified ideational motivations behind state-building, as ruling elites opt for the expansion of the state when they see it as a means for development (2015, 4, 24). He suggested that the identities of administrators explain territorial unevenness, as state-building goals are more successful when the state deploys outsiders in communities than when they are appointed by local elites. Mazzuca (2021) argued that trade expansion produces weak states, since ruling coalitions and peripheral elites share incentives for indirect rule. While he does not study Chile in depth, his argument suggests that the origins of Chile’s comparatively high-capacity state lie in the lack of patrimonial peripheries (28).

Varieties of State-Building

Despite the progress made by the theories outlined above, fundamental challenges remain. Geopolitical theories have limited application outside Western Europe and do not explain successful episodes of capacity-building in the absence of war. The fact that ruling elites are usually reluctant to commit to costly projects of simplification and standardization undermines the legibility school’s most relevant claim. While theories that focus on the uneven nature of states have modified their assumptions accordingly, they usually suggest that states reduce their scope and appetite for taxation only when local challengers push back. While scholars of Latin American politics have pushed the field’s theoretical and empirical boundaries forward, the uneven nature of Chile’s projection of authority remains unexplored. The following subsections explain how economic incentives brought by the trade expansion of the late nineteenth century combine with regions’ attributes to explain subnational trajectories of bureaucratic rule and country-level state capacity.

Trade Expansion and Fiscal Shocks

The literature broadly considers the global expansion of capitalism as inimical to state-building for two reasons. First, the wide availability of international credit limited the development of extractive institutions and, second, the possibility of sharing the benefits of trade incentivized decentralized governance. ³ On the contrary, this paper focuses on a less explored aspect of trade expansion: the fiscal crises prompted by commodity busts. ⁴ These episodes were among the first exogenous economic shocks experienced by the young republics of the Americas. Under such circumstances, fiscal needs offset the costs of disrupting friendly center–periphery relations and taking over newly valuable commodities becomes a plausible strategy to balance the deficit. Even though international finance became an important source of revenue for many governments under these circumstances, Chile could not balance its budget only with this mechanism, opting for state-building as a complementary strategy.

Regions’ Attributes

Ecological Suitability

State-building is more likely to succeed in ecologically suitable regions. Suitability varies according to two key factors: geographic accessibility and economic appeal. Scholars broadly agree that state development is probable in regions characterized by social and physical enclosure. Michael Mann (2012a, 42) explained the emergence of early states through the metaphor of a social cage, a “fixed settlement [that] traps people into living with each other, cooperating, and devising more complex forms of social organization.” These spaces limit the avenues of escape from the state’s ambitions. Olson (1993, 575) further argued that the advantages of being a “stationary bandit”—a ruler who establishes political order and facilitates economic development—are larger in contexts of geographic boundedness.

Political hierarchies first emerged in fertile agricultural regions with available connections (usually rivers) that were bounded by mountains, seas, or deserts. In such geography, individuals find it difficult to escape states’ attempts to tax them (Carneiro 1970; Dal Bó, Hernández-Lagos, and Mazzuca 2021; Fernández-Villaverde et al. 2023). Hintze (1975) linked geography and European state-building by suggesting that England escaped the escalatory pressures identified in the bellicist tradition due to its insular condition. But geographic features can also prevent state-building. In areas that Scott (2009) calls “zones of refuge,” rugged terrain hinders state control over local populations. Similarly, sparsely populated areas make it difficult for the state to access taxable populations, which limits its territorial scope (Herbst 2014).

A second feature of ecological suitability is a region’s economic endowments. The economic value of peripheries can increase when international markets drive up the price of commodities. Regions may contain newly valuable resources such as minerals, or have the appropriate climatic and geographic conditions for agricultural development. In contrast, regions without economic value are less likely to be targeted for state-building under these circumstances.

Figure 1 charts a theory-building exercise suggesting a relationship between geographic accessibility and economic appeal, leading to four ideal types: (1) cages, (2) marginal, (3) refuge, and (4) autonomous. Cages are ideal for state-building, as their geographic and economic features facilitate it; they are the only ecologically suitable region. Marginal regions have appropriate geographic conditions for state-building but lack economic appeal (i.e., before or after commodity booms). Refuge areas are inaccessible to the state, and, consequently, have no financial appeal. These are usually hills, tropical jungles, dense forests, and deserts. Finally, autonomous regions refer to hard-to-reach yet productive areas. These include, for instance, Colombia’s coffee-producing intermontane valleys, which stand in stark contrast to Chile’s Central Valley in terms of geographic accessibility (see empirical analysis below).

Figure 1 A Typology of Ecological Suitability for State-Building

Regional classifications are not fixed. Rather, fiscal shocks or commodity booms change regions’ economic outlook. Similarly, technological advancements in transportation, engineering, or cartography can change accessibility. The phrase “ecologically suitable” does not encompass every type of region. Rather, it describes regions whose ecological features most interact with state-building. Even though cages are the only regional type defined as ecologically suitable, ruling coalitions may still attempt to extend their rule into less suitable areas. As this paper focuses on state-building, the empirical section evaluates ecologically suitable regions only.

Ruling coalitions’ efforts to turn regions into cages can prompt tangible gains in state capacity. However, this too depends on regions’ features. Areas suitable for grazing, like the southwestern US and the central Atlantic coast of South America, may require little to no investment by the state yet still bring revenue to it. Conversely, if challenging “zones of refuge” hold highly valuable resources, states may, in the event of land shortages, spend significant energy and resources to make them legible, leading to investments in technology, infrastructure, and administrative reforms and enhancing capacity-building in the long run. As the empirical section below shows, the Chilean state invested many resources to turn Araucanía, an area of dense forests where native peoples escaped colonial ambitions for centuries, into a wheat-growing area. As a side effect, this project shaped the formation of technocratic cadres and prompted important bureaucratic reforms.

Credible Military Threats

A second attribute that impacts state-building outcomes is the ability of peripheral elites to threaten a ruling coalition. Regions that pool coercive resources and create a credible threat to the control of the ruling coalition are more likely to obtain concessions than those that do not. In particular, what matters is the perceived threat of ruling elites, not regions’ baseline defense capabilities. This difference is well illustrated by the threats created by colonial-era peripheral elites and first nations in the Americas during the nineteenth century. Empowered by the benefits of trade expansion, many peripheral elites were able to resist state-building efforts by raising armies. In Latin America, these were the well-known caudillos. On the other hand, indigenous peoples such as the Mapuche in present-day Chile (analyzed below) were able to secure fewer territorial concessions than wealthier, politically relevant colonial-era regional elites. The former often launched rebellions aimed at limiting state-building and upending the national ruling coalition, while the latter’s uprising attempted to stop internal colonization projects within their own regions.

Clientelistic Reserves

Successful state-building can paradoxically reinforce local patrimonial strongholds. As ruling coalitions disrupt friendly center–periphery relations to seek more revenue, they must forge new alliances to stay in power. Landed elites, for instance, employed patron–client relationships with the state to resist democratization pressures in the late nineteenth century. In agriculture-based regions reliant on labor, landowners could secure votes through clientelistic ties with a large pool of peasants. In contrast, rural political machines did not become prevalent in pastoralist economies due to low labor requirements, which prevent landlords from creating such clientelistic networks. This center–periphery dynamic—where local lords traded autonomy for political backing—became known as caciquismo in Spain, coronelismo in Brazil, and gamonalismo in the Andes. ⁵

Outcomes: Subnational Bureaucratic and Patrimonial Rule

The outcomes serve as answers to the following question: “Who rules a periphery?” A central authority enforces regulations under bureaucratic rule. This category is also characterized by division of labor, defined hierarchies, regular salaries, and free contracts (Mann 2012b, 444; Weber [1921] 1978, 1:220–21, 2:952). While common definitions include a meritocratic dimension, ⁶ my definition aims only at identifying who holds political and administrative authority within a region. Bureaucratic rule can be classified into two subtypes: imposed and cooperative. Imposition refers to replacing a regional patrimonial administration through force without granting concessions to local notables. It usually takes the form of internal colonization and involves a large deployment of coercive forces and technocrats—engineers and surveyors—to make new areas legible to the state. Cooperative bureaucratization, on the other hand, involves the extension of state authority into a region while also granting rights to local elites, typically through political incorporation and public goods. ⁷

Patrimonial rule is defined by the discretionary power wielded by local private actors, such as landowners, warlords, local parties and politicians, or the clergy. In place of state institutions, they rely on private patronage networks to enforce regulations. Under cooperative patrimonialism, allies secure high degrees of de facto autonomy—the right to use coercion, settle conflicts, and collect taxes—and preferential access to public goods. ⁸ Under patrimonial reinforcement, the national ruling coalition also bolsters legal and political boundaries to safeguard a subnational enclave where patrimonialism can endure and thrive. ⁹

Figure 2 illustrates the theory, including independent variables and outcomes. During fiscal crises, some regional attributes become valuable to the center. Ruling coalitions attempt state-building in ecologically suitable regions while targeting regions with clientelistic advantages for patrimonial reinforcement. Regions that lack both attributes are ignored for state-building by ruling coalitions, who settle instead for patrimonial cooperation. If a periphery is economically valuable but ecologically unsuitable—autonomous regions as defined in figure 1—ruling elites opt for patrimonial cooperation too. These areas can be made productive by the private sector, without the state’s intervention. Finally, in regions where ruling coalitions do attempt state-building, the type of bureaucratization depends on the extent to which local notables pose a credible threat to the incumbent’s role in the ruling coalition.

Figure 2 Regions’ Attributes and State-Building Outcomes

Note: Figure assumes that money borrowing is no longer an option and assumes the presence of regions that can be made productive.

Bureaucratic Rule and State Capacity

Geopolitical theories suggest that the effort to raise an army to wage war prompts state capacity. The greater the effort, the larger the “organizational residues” (Tilly 1985, 181). I suggest a slightly different proposition. The effort to disrupt local ecologies, govern distant regions, and make them productive creates residues in the form of investments in technocratic cadres and infrastructure. Such efforts can lead to capacity-building at the central level, as the availability of such expertise and institutional templates enhances the state’s capacity to deploy public goods across the territory.

The Case: Chile

Chile is a positive divergent case of state development in Latin America. Figure 3 illustrates the distinctiveness of Chile’s evolution through the development of railroads and telegraph lines per square mile—a commonly used indicator of the state’s territorial reach. The Chilean state’s territorial control was higher than the regional mean as early as the mid-1860s, a pattern that continued over time.

Figure 3 States’ Territorial Reach in South America

Source: Banks and Wilson (2021).

The Chilean central government’s budget expanded steadily from the late 1850s (Humud Tleel 1969; López Taverne 2014; 2017, 66–69). According to figure 4, the Treasury and Interior budgets saw the most significant growth. ¹⁰ These ministries allocated resources for tax collection and bureaucrats’ wages, respectively. Customs witnessed a 111% increase in employees, rising from 276 people in 1845 to 581 in 1880, in stark contrast to the rest of the Treasury’s agencies. ¹¹ This change suggests a deliberate effort to enhance the state’s ability to extract resources from Chilean society.

Figure 4 Chile’s Annual Budget, 1845–78

Source: Interior Ministry budgets.

The Chilean case is ideal for studying capacity-building in the absence of the strong geopolitical pressures identified in the bellicist tradition. Scholars have noted that the animosity between Chile, Peru, and Bolivia may have driven the Chilean divergence (Schenoni 2021; Thies 2005). However, Chile’s successful state-building can be traced back to the early 1860s, 20 years after the war against the Peru–Bolivian Confederation (1836–39), and 15 years before the War of the Pacific (1879–84). Moreover, the 1860s were a period of amity between Chile and Peru, as both countries allied against Spain in the Spanish–South American War (1865–79). ¹² The state-building projects of the 1860s were not a legacy of the war against the Peru–Bolivian Confederation either. Chilean elites perceived the confederation as a threat since Peru’s main port could surpass Chile’s Valparaíso and control trade in the South Pacific. Chile allied with the Peruvian opposition in the Ejército Restaurador, a coalition aimed at breaking down the confederation and restoring the Peruvian state. As such, Chile’s role in the conflict was not characterized by territorial conflict and did not include the annexation of regions. Internal conflict is another plausible hypothesis. Indeed, Chile had two civil wars, in 1851 and 1859. In my interpretation, the first conflict was mainly over presidential succession, while the latter was a peripheral backlash to state-building.

Historical Background

Chile’s territory consisted of four regions at the time of independence (1810): Atacama, Santiago, Concepción, and Araucanía. Atacama was a mining enclave near the Atacama Desert. Santiago included the capital, Valparaíso (the main port), and the surrounding wheat-producing area in the Central Valley. ¹³ Concepción was originally a Spanish military outpost created to fight wars of expansion against indigenous people. Over time, it developed into a large city with a frontier economy and culture like neighboring Araucanía, home of the native Mapuche. Figure 5 shows these areas and the territories that were annexed later on. ¹⁴ Due to their geographic features—bounded by the Andes, the Chilean Coastal Range, the Atacama Desert, and the western region of Patagonia—Chilean regions were cages, except for Araucanía. Due to its dense forests and scattered geography, that region can be defined as a refuge. ¹⁵

Figure 5 Chile’s Regions

A conservative ruling coalition sanctioned a unitary constitution in 1833, bestowing preeminence on the capital Santiago. Each province would be governed by an intendente (intendant), and each department by a gobernador (governor). However, given that the central government did not have a real presence across regions and local notables retained significant agency, the centralist nature of the constitution was nominal rather than real. ¹⁶ The supposed ascendancy of the capital rested on brittle political alignments within the ruling coalition. While named by the president, intendants were usually chosen in agreement with local elites, and “official” electoral lists were drafted in close alignment with, and in recognition of, local elites’ sensibilities. ¹⁷

Each region hosted numerous nonstate actors. These local elites had different origins. ¹⁸ In Atacama, mining businessmen filled administrative posts and named the intendente (Fernández Abara and Jerez Leiva 2020; Godoy Orellana 2018). Private actors built and operated ports, which led to inefficiencies and smuggling (Godoy Orellana 2018). Since rapid urbanization followed wherever a new mineral deposit was discovered, mine owners soon had to finance essential public goods like infrastructure and policing. A mine-owners’ guild organized the police; appointed administrative, political, judicial, and religious authorities; collected taxes; and distributed public goods (Fernández Abara 2016; Venegas Valdebenito 2008).

Concepción saw the development of a local export economy of wheat and wine exports to Lima and the emergence of trans-Andean commercial routes (Pinto Rodríguez 2020). Yet independence brought a period of decline, as banditry and civil conflict brought expropriation and forced conscription. A new foreign elite arrived in the 1830s, helping to redevelop the wheat and milling sectors (Campos Harriet 1979; Mazzei de Grazia 2015). Mapuche elites participated actively in Concepción’s local economy by exchanging meat, salt, and textiles (Pinto Rodríguez 2003). Having also settled across the Central Valley, they stopped the advancement of the Spanish army at the Biobío River (Bengoa 2015), and developed diplomatic relations with the Spanish Empire.

The origins of Central Valley’s haciendas lay in the need to provide food for the military garrison in Concepción in the early colonial period. After the Spanish crown gave land grants to a select number of people, and given the need for cheap labor, landlords created a service-tenancy system. They began incorporating a small number of inquilinos—permanent laborers—and their families into haciendas, where they were given a small plot of land to grow food. Inquilinos and their families were tied to the land by tradition rather than law (McBride 1936). Beyond their work in the fields, inquilinos also provided domestic services (Carrière 1981, 20). The process of mestizaje—the racial mixing of Spanish and native Chileans—created a large mass of seminomadic people in the area. In such a context, being included in the hacienda structure could be seen as an important benefit by peasants. Central Valley elites enjoyed preferential access to public goods, as the ruling coalition directed state-owned railroads, roads, and irrigation canals to this area to the detriment of Concepción’s farmers. ¹⁹

Fiscal Shocks and State-Building

The fiscal crisis of the mid-1850s triggered economic uncertainty, leading actors to reassess the advantages of state-building compared to the prevalence of indirect rule. First, silver prices plummeted due to the Panic of 1857 (Encina 1949, 589). Stagnating silver deposits limited the currency supply, since this metal was used to mint coins (Edwards 1932, 147; Humud Tleel 1974, 78–79). Revenue from customs duties declined and the trade balance became negative for the first time ever (Edwards 1932, 146; Humud Tleel 1974, 22, 74). Added to a few poor harvests in previous years, Manuel Montt’s government (1851–61) was now facing an existential economic downturn.

At the same time, the opening of international markets created new opportunities. First, a wheat boom began with the California gold rush (1848–55). Elites foresaw a promising future for grain given improvements in navigation through the Strait of Magellan that could open up the European, Argentine, Brazilian, and US East Coast markets. ²⁰ A copper boom due to the spread of electricity in emerging industrial economies also provided relief. Unlike silver, copper production survived thanks to smelting technologies that reduced the need for labor (Fernández Abara and Jerez Leiva 2020, 95) and the discovery of coal near Concepción that lowered fuel costs (L. Valenzuela 1992, 507). Together, these changes catalyzed state-building as an attempt to halt the looming fiscal crisis. ²¹

Chile: The Argument

Figure 6 illustrates the main argument, combining the theory outlined above and the Chilean process. After the emergence of a ruling coalition (1830), center–periphery relations were patrimonial and cooperative. After the price shocks in the 1850s, the central government changed its approach to governance in the periphery, opting for state-building. In Concepción and Atacama, the outcome was bureaucratic cooperation. Because these regions had gone through steady economic growth since the 1830s, local elites were able to pose a credible threat in the civil wars of 1851 and 1859. The central government defeated the rebels in a pyrrhic victory, as peripheral elites were able to extract concessions in the form of political incorporation and public goods.

Figure 6 Chile: Outline of the Argument

The ruling coalition opted for patrimonial reinforcement in the Central Valley by creating legal and political boundaries around the traditional hacienda-based social order to offer a base of clientelistic support to the conservative side. Finally, the state chose bureaucratic imposition in Araucanía through a military campaign to make the area arable for wheat and silviculture. Patagonia’s ecological conditions did not present any advantages for state-building. Its arid, cold climate and scattered geography thwarted any productive activities during the colonial period and shortly thereafter. ²²

Empirical Analysis

Cross-Sectional and Temporal Variation

This section draws on original data from censuses and budgets to show the cross-sectional and temporal variation in the reach of the state (Véjares 2024). I use a population-adjusted measure of the presence of bureaucrats and members of the military and police, and of resources spent on bureaucrats’ wages, the judiciary, and public works. ²³ Recent works measure states’ projection of authority through the distribution of infrastructure—mainly railroads (e.g., Bignon, Esteves, and Herranz-Loncán 2015; Cermeño, Enflo, and Lindvall 2022). I choose an alternative approach that distinguishes between bureaucratic rule and public goods distribution. The analysis below suggests that an increased flow of public goods combines very well with the protection of local patrimonial bastions. As local elites enjoy preferential access to the state, they are more likely to receive resources for infrastructure projects while keeping the central bureaucracy at bay.

Figure 7 displays the presence of bureaucrats in each department’s capital between 1865 and 1920. ²⁴ The central bureaucracy expanded widely through Chilean territory in this period, with a distinctive increase in the northern and southern peripheries. The ruling coalition already had an important presence in Atacama by 1865. While we lack census data for the pre-1859 rebellion period, historiographic accounts agree that mining businessmen ruled the area through their private clientele and networks. ²⁵ Budget data in figure 9 corroborates this conclusion. Together with a large expansion in the number of bureaucrats, the central government increased the number of departments in Araucanía the most. The figure also shows the underdevelopment of the bureaucracy in the Central Valley, which is the only region that did not undergo bureaucratization. ²⁶ By 1920, the presence of the central government became even greater than in previous decades. While such a pattern denotes a special effort by the ruling coalition to govern specific regions in earlier years, it is also worth noting that the central bureaucracy was still weak in the Central Valley 60 years after state-building projects took place.

Figure 7 Territorial Reach of the State, 1865–1920

Notes: White and black dots represent one standard deviation—positive and negative, respectively—from the mean. Given that departments’ limits were often drawn decades after census data were taken, the figure shows each department’s capitals instead of administrative boundaries. See González, Compán, and Sagredo Baeza (2016, 311).

Figure 8 uses budget data to show the ruling coalition’s priorities over public goods in 1899. The pattern is similar to figure 7 with one relevant exception: the Central Valley received much more resources for productivity-enhancing public goods than for bureaucracy. Panels 2 and 3 show that the state prioritized funding for the judiciary and public works where landed elites were the strongest (particularly in provincial capitals like Talca and Curicó). Unlike high regulatory capacity, landed elites welcomed these public goods: the judiciary enforced property rights, while public works to improve irrigation and transportation were essential to increase production and exports. This pattern reflects a compliant relationship between the state and local notables, as state weakness was combined with infrastructural development and expanding state authority in other regions.

Figure 8 Spatial Distribution of Public Goods, 1899

Source: Dirección General de Contabilidad (1901).

Political elites saw state-building as a plausible strategy to increase revenue as early as 1865. Treasury ministers suggested that improving tax collection methods and creating unified administrative procedures could increase fiscal income without affecting trade. Until then, customs offices operated independently from one another. ²⁷ By the mid-1870s, the Treasury Ministry was able to create a unified accounting system. ²⁸ The plan included hiring guards and bureaucrats directly, whose wages would now be included in the central government’s budget. ²⁹ Intendants and governors also began writing formal reports in the annual Interior memorias (reports) by 1863, denoting higher vertical accountability.

Subnational Outcomes

Bureaucratic Cooperation

What began as a strong alliance between local conservative factions in Atacama, Concepción, and the ruling coalition in the early 1830s became fragile 20 years later. The rebellious regions shared similar grievances. The first move of Manuel Montt’s government to offset the deficit was to raise taxes in regions already affected by the Panic of 1857 and in a context of a highly uneven tax rate. The agricultural tax—paid mainly by Central Valley landowners—accounted for only 3% of fiscal income (Zeitlin 1984, 38). Most of the revenue came from exports from the peripheries.

In Concepción, liberalism gained traction in the late 1840s due in part to the rise of its new business-oriented elite. Local actors condemned the 1851 elections over fraud and, angered by the economic downturn, followed a local warlord. José María de la Cruz and his followers rebelled against Montt’s newly elected government. In Atacama, middle-class sectors allied with elite families after the government’s refusal to change the tax rate once the crisis hit. Amid a predatory lending system (habilitación) that had fueled a contentious relationship between both parties, the crisis instigated shifts in loyalties. As a result, the province united to oppose the ruling coalition, initially in elections and later on the battlefield. From the ruling coalition’s perspective, the frontier economy and Concepción’s hopes for autonomy were a threat to its overall control.

Both regions shared favorable geographic conditions for state-building. Bounded by the Atacama Desert, the Andes, and the Pacific Ocean, these areas were protected from foreign interference and could not ally with trans-Andean provinces and become part of Argentina’s complex territorial political game. ³⁰ These areas also had important economic resources the state could use. In Atacama, the ruling coalition needed to increase its presence to end tax evasion and take greater advantage of the copper boom. Concepción offered areas suitable for wheat growth, coal mining, and access to Araucanía.

Both civil wars followed similar trajectories, with local elites mobilizing armies through their local clientele and funding them with resources acquired during the bonanza period. They presented significant threats to the survival of the ruling coalition. Even though the central government emerged victorious both times, conservatives lost control of the ruling coalition and were forced to include liberals. ³¹

After the wars, both regions transitioned to bureaucratic rule. In Atacama, the government expanded the judiciary and coercive forces (Godoy Orellana 2018). Policing became part of the intendancy’s tasks, signaling greater control by the state (Fernandez Abara 2016, 25). Concepción’s elites lost their discretionary power to name public officials. Figure 9 illustrates these regions’ transition to bureaucratic rule, as increased expenses for bureaucrats suggest an effort to enforce regulations directly. ³² This change was disproportionately large compared to that in other regions, which signals a special priority to increase the state’s presence in these locations.

Figure 9 Bureaucrats’ Wages, 1845–78

Source: Interior Ministry budgets.

Notes: Budget data were produced at the province level. The average includes every province except for Atacama and Concepción. The number of observations varies as new departments were created over time.

The newly created fusionista coalition threatened the executive, first politically, by joining forces to develop alternative electoral lists, and then militarily. While fusionistas lost the 1859 civil war, the ruling coalition was weakened and President Montt did not have political support to name his close ally Antonio Varas as his successor—a tradition upheld by every previous president. As the new Liberal-Conservative coalition came to power, the losing side obtained concessions, including public goods and the incorporation of previously unrepresented groups. ³³

Peripheral elites also led the formation of new parties (Campos Harriet 1979, 280). Radicalism in particular played a key role in pushing for democratization reforms. Their first manifesto promoted strengthening Congress and decreasing the executive’s power (Snow 1972). Table 1 shows elections won by parties before and after 1861. This cutoff signifies the end of the conservative tenure. ³⁴ Radicals dominated elections in both peripheries compared to the rest of the country, indicating that these elites were indeed incorporated into the political system.

Table 1 Representatives by Party

Source: Valencia Avaria (1951).

Notes: Electoral districts changed over time, and did not fully align with other administrative boundaries like provinces or departments. Cells show the total number of elections, meaning there are some elections won by the same candidate.

Bureaucratic Imposition

The fiscal crisis made Araucanía highly attractive to the state. The area between the Biobío River and Patagonia was the last region suitable for wheat growing. Haciendas in central Chile had reached maximum productivity given technological and labor constraints, and many locations suffered soil erosion from deforestation to meet California’s demand for wheat (Schneider 1904). The government saw in Araucanía an opportunity to shift the country’s economic profile toward agriculture as early as 1867. ³⁵ Technological improvements in military equipment, transportation, and engineering allowed the state, in alliance with private companies, to log the area and make it productive. Unlike other regions, Araucanía could not be used as a potential clientelistic bastion because the Mapuche were not enfranchised.

The ruling coalition invaded once Arauco’s intendant and military leader, Cornelio Saavedra, attempted to regulate the de facto expansion of private colonizers in 1861. Mapuche elites’ support for the 1859 rebellion and Concepción’s defeat helped to legitimize the incursion. Although the original project aimed only at rebuilding towns destroyed in previous wars, the incursion slowly became more encompassing. The state perceived the forests in the area—an essential element of Mapuche cosmology, diet, and economy—to be unmanaged nature and displaced the native population to reservations, creating sedentary communities and dissolving indigenous networks of exchange, trade, and migration (Klubock 2014, 31–32).

The state added one million hectares as early as 1871, yet a Mapuche uprising led to a stalemate. ³⁶ Investments in railroads and telegraph lines allowed the central government to direct the war effort in real time. Together with improvements in weaponry, these advantages permitted the state to defeat outbreaks of Mapuche resistance. The effort to increase grain production drove the penetration of Araucanía. Figure 10 shows the expansion of the agricultural frontier. Provinces north of the Biobío River did not increase their production between 1875 and 1885. ³⁷ On the other hand, the provinces south of the river experienced outstanding growth in these years.

Figure 10 Wheat Production in Selected Provinces

Source: Chilean national censuses.

The central government claimed ownership over occupied land, clearing large portions (Solberg 1969). From then on, state emissaries developed infrastructure and enforced property rights (Bengoa 2002; Klubock 2014). ³⁸ Figure 11 shows a population-adjusted measure of individuals who identified as bureaucrats, members of the police or the military, or engineers. The military intervention came first, followed by bureaucrats and engineers. ³⁹ The military was in charge of defeating outbreaks of resistance and clearing land so engineers and surveyors could divide and measure plots (Bengoa 2002, 46). While the number of engineers per capita was much lower than the other two categories, by 1895 there were three times as many engineers in Araucanía than the national average. The deployment of engineers is associated with state-building’s legibility dimension, as they are commonly hired to measure, assign value to, and divide land.

Figure 11 The Transition to Bureaucratic Rule in Araucanía

Source: Chilean national censuses.

The incursion into Araucanía demanded substantial investments and technical knowledge. The government helped to develop the civil engineering field, and included these experts in policy-making communities later on. The state-owned railway company (Empresa de Ferrocarriles del Estado, EFE), for instance, became a hub for the advancement of engineers. Slowly, they took over important roles in that company, helping them to acquire a high reputation among politicians and the broader public (Crowther 1973, 303; Guajardo 2007, 26). The deployment of infrastructure into Araucanía through EFE and a topographic commission in charge of measuring plots gave engineers the necessary legitimacy to take over more policy-making spaces (Ibáñez Santa María 2003, 118). Engineers developed an “antipolitics” policy-making style and obtained de jure bureaucratic insulation (Crowther 1973, 400), and subsequent reforms created a meritocratic mechanism for the incorporation of engineers into the public administration. Governments replicated the railway company’s organizational and technical structure in other areas. By the 1920s, the practice of engineers assuming roles as decision makers expanded into the private sector (Ibáñez Santa María 2003, 119).

Patrimonial Reinforcement

The state reinforced patrimonial rule in the Central Valley to benefit landowners, who secured political and economic survival by keeping a clientelistic base within haciendas. Landlords’ control over peasants worked as insurance against expropriation, decreasing the costs for the conservative side to survive in an increasingly competitive environment. Landed elites opted for this strategy after they lost their seniority within the ruling coalition. The plan created legal and political boundaries to keep the state’s potential regulatory capacity away from landlords’ traditional sources of power. ⁴⁰

Haciendas were highly authoritarian, self-contained systems. Inquilinos “lived and worked on the estates and depended on the patron for housing, for medical attention, for food, and the small plot of ground necessary for subsistence” (Kaufman 1972, 22). The patrón (owner) could arbitrarily change work requirements and land allotted to each family at any point (Petras and Zemelman Merino 1972, 54). Land inequality in the Central Valley was among the highest in the world, as fewer than three hundred people owned half of the agricultural land as late as 1935 (Carrière 1981, 30). While cities had policing institutions, the countryside “depended on a handful of ill-equipped constables, [and] ad hoc vigilante units led by the hacendados” (Bauer 1975, 166). Landlords also had de facto territorial sovereignty by controlling entry and exit into their properties (Loveman 1976, 34). By the late nineteenth century, these estates were not particularly productive. Instead, they became a source of status and prestige for the elite (Carrière 1981, 20).

Maintaining this institution (inquilinaje) was crucial for the viability of the landlords’ political goals. If inquilinos migrated to cities and became part of the urban poor, they could have become part of the Left’s political base. A transition to wage labor would have ended the system’s intrinsically paternalistic nature, hurting landlords’ chances of keeping peasants as an electoral base (Bauer 1995, 26–28). Keeping the hacienda-based social order intact would preserve the system of domination necessary to control peasants’ votes (30). ⁴¹

Two strategies were especially salient. First, Congress passed a law in 1874 that enfranchised literate males. While seemingly progressive, the effort aimed to create an electoral base across estates (Bauer 1995, 30). The bill was approved in the same months the conservative faction left the executive (Encina 1949). The number of peasants enfranchised increased disproportionally between 1872 and 1876 across the Central Valley (J. S. Valenzuela 1985, 119). The literacy requirement was only enforced in cities, which gave landowners an advantage vis-à-vis urban parties. At the same time, conservatives were able to reform electoral boards, taking them away from the executive’s control. Second, landowners lobbied the executive through the Sociedad Nacional de Agricultura (Agricultural National Society, SNA). The group was originally created to push for price regulations in the 1890s and soon became the most influential lobbying association. Between 1873 and 1928, 20% of congressional members were SNA affiliates (Carrière 1981, 36–37). Together, these strategies helped landowners to maintain the status quo in the long run.

Change started in the 1930s, when labor inspectors began visiting estates to fine landowners over labor and housing-code violations. At the same time, leftist parties began unionizing peasants (Loveman 1976, 31). Arguably, subnational enclaves ended after the introduction of the secret ballot in 1958 and land reforms in the 1960s, effectively ending landlords’ control over peasants. Table 2 shows the evolution of political representation in Central Valley districts. In stark contrast to Atacama and Concepción, the right wing (Conservatives, Liberals, and Nationals) dominated, and the Radicals did not make any gains. The Right kept a sizable majority in Congress well into the twentieth century, even after the election of the first Radical government in 1938 (Correa Sutil 2005, 71).

Table 2 Central Valley Representatives by Party

Source: Valencia Avaria (1951).

Conclusions

This article suggests that state-building prospects hinge on the socioeconomic and physical profiles of subnational peripheral regions. States attempt to extend bureaucratic rule onto regions that have valuable commodities and advantageous geography. At the same time, their desire to optimize political support compels rulers to ally with landed elites that offer a clientelistic bastion. I build this theory by studying Chile, a case that stands out as a successful instance of capacity-building in Latin America. The Chilean state negotiated a transition to bureaucratic rule in Atacama to the north and Concepción to the south. Then, the state imposed bureaucratization in Araucanía, the home of the Mapuche in the far south. Finally, the government solidified its alliance with agrarian elites in the capital’s hinterland to secure electoral support.

Although Chile is often considered the poster child of the even projection of state power in Latin America, my argument shows that state-building was instead highly selective and uneven, with the very center being the most important deviation to uniform, bureaucratic rule. These projects created durable legacies. First, the effort to rule peripheries produced long-term gains in state capacity. Second, the enduring alliance between landed elites and the state helped the hacienda-based social order to survive mostly intact for a hundred years. This strategy allowed the Conservative party to gain a sizable portion of Congress and maintain the status quo well into the twentieth century, shaping Chile’s competitive yet highly restrictive regime. These results challenge the dominant geopolitical and legibility traditions by emphasizing how the need of elites to ensure their political and economic survival interacts with regional characteristics to create broad variation in the extent of rulers’ territorial reach and state capacity.