In October 2018, five days before the Brazilian presidential election, then candidate Jair Bolsonaro livestreamed a meeting with the leader of the Bancada Ruralista (Rural Bench, Brazil’s Agrarian Caucus), Representative Tereza Cristina. In the video, Tereza Cristina expressed the caucus’s support for Bolsonaro’s candidacy.

The Ruralistas’ endorsement was key in signaling Bolsonaro’s capacity to govern if elected. At the time, Bolsonaro’s party had almost no representation in Congress, while the Agrarian Caucus controlled around a quarter of the Lower Chamber seats, making the Ruralistas’ support crucial to advance Bolsonaro’s conservative policy agenda. The alliance was sealed when, shortly after winning the presidential runoff, Bolsonaro appointed Tereza Cristina as minister of agriculture. However, the rightwing president has not been the only one to seek the support of the Ruralistas in Congress to guarantee governability. In fact, after defeating Bolsonaro in 2022 to win a third term in office, center-left president Lula da Silva of the Workers Party (PT) also appointed a Ruralista to the ministry of agriculture.Footnote ¹

In neighboring Argentina, agrarian elites’ political standing could not be more different. Despite the sector’s economic importance as the main source of foreign currency for the country, agrarian elites have no meaningful representation in Congress or ties to political parties. Agrarian elites’ political weakness manifested itself most clearly in March 2008 when agricultural producers took to the streets to protest a hike in export taxes. After launching a four-month-long production and commercialization stoppage, accompanied by numerous mass protests and roadblocks across the country, landowners were able to repeal the most recent increase but not the export tax altogether, which remained at a whopping 35 percent.

Scholars of Latin America have widely studied the role of landed elites as obstacles to democratic consolidation throughout the region’s history.Footnote ² However, we know comparatively much less about how landowners have adapted to the new democratic context after the transitions of the mid 1980s. This book addresses this gap in our knowledge by studying the strategies agrarian elites employ to make democracy work to their advantage. As the two opening vignettes illustrate, there is important variation in how agrarian elites organize to influence policymaking. In countries such as Brazil, Chile, and El Salvador, landowners have organized in the electoral arena by building parties, running for office themselves, or supporting likeminded candidates. In other countries, such as Argentina and Bolivia, they have shunned the electoral arena, influencing politics through nonelectoral channels such as lobbying or – when this has failed to block unwanted policies – protests.

Understanding this variation in Latin American agrarian elites’ political strategies matters because it affects landowners’ ability to influence policy realms of great regional and global significance. In the region with the highest income inequality in the world,Footnote ³ and where almost half of the rural population still lives in poverty,Footnote ⁴ agrarian elites have been much more successful in blocking redistributive policies where they have organized in the electoral arena than where they have relied on nonelectoral means of policy influence. Figure 1.1 illustrates the contrasting capacity of governments to redistribute resources away from agricultural producers in four Latin American countries by displaying the evolution of the total support for agriculture as a share of each country’s GDP in the last two decades.Footnote ⁵ Positive values indicate net transfers from the rest of society to agriculture. Negative values indicate redistribution from the agricultural sector to the rest of society. As we can see, in the two countries where agrarian elites are organized in the electoral arena (Brazil and Chile), they have been able to secure net transfers towards their sector throughout the period. In the two countries where agrarian elites lack electoral representation (Argentina and Bolivia), in contrast, governments have extracted resources from them. Nowhere is this difference clearer than when comparing Argentina and Brazil. In Argentina, where agrarian elites have no electoral representation, every government since 2001 has extracted bountiful resources from them, reaching, on average, the equivalent of 3 percent of the country’s GDP annually during the administrations of the center-left Frente para la Victoria (Front for Victory, FPV) (2003–2015). In contrast, in Brazil, where agrarian elites are organized in a powerful multiparty congressional caucus, governments have consistently and generously subsidized agribusiness. This includes the leftwing administrations of the PT (2003–2016), which transferred annually, on average, resources equivalent to 0.65 percent of the country’s GDP from the rest of society to agriculture. This equals to billions of dollars in subsidies every year to some of the wealthiest people in Brazil, one of the most unequal countries in the world.Footnote ⁶

Figure 1.1 Total Support Estimate for agriculture as percentage of country GDP: Argentina, Bolivia, Brazil, and Chile, 2000–2020.

Source: Author, based on IDB-Agrimonitor.

Beyond taxes, subsidies, and land reform, agrarian elites’ capacity to organize electoral representation has been crucial in shaping another policy realm with great redistributive impact: environmental regulations. Environmental policies regulate how producers use natural resources within their properties (e.g., land, water, and forests), as well as the kind of inputs they can employ in their production (e.g., fertilizers, seeds, and pesticides), impacting their profits and constraining their property rights. The literature analysing the relationship between inequality and democracy typically looks at taxes and land reform as the quintessential policy realms of redistributive struggle between governments and agrarian elites. This book also analyses environmental policies as these have become a highly contested issue in the region in the context of the current climate crisis, while land reform has lost relevance in the public agenda, due to urbanization and changes in agricultural production.

Despite their ties to environmental groups and indigenous organizations, leftwing administrations in Latin America have encountered significant obstacles to pass and enforce regulations over the exploitation of natural resources such as forests, minerals, or water.Footnote ⁷ This book explains how agrarian elites have been able to use democratic institutions to block or significantly dilute environmental regulations, a critical issue given Latin America’s crucial role in mitigating climate change. For instance, forest protection legislation stalled for years in the countries’ legislatures, despite widespread and accelerating deforestation, and passed only after significant accommodations to agribusiness interests.Footnote ⁸

This book presents a framework to understand the variation in agrarian elites’ strategies of political influence. I explain when agrarian elites organize in the electoral arena to protect their interests and how they adjudicate between the different electoral strategies available to them. My work contributes to the ongoing debate on the relationship among economic elites’ representation, democracy, and redistribution by specifying the key mechanisms through which agrarian elites can use democratic institutions to protect their economic interests.

1.1 The Argument

The main argument of this book is that agrarian elites’ strategies of political influence are shaped by two factors: the level of threat they perceive and their level of intra-group fragmentation. The perception of an existential threat – defined as a policy that jeopardizes the continuity of agrarian elites’ business – is a necessary condition for electoral investment. Land reform, confiscatory taxes, or stringent environmental regulations are examples of existential threats. Nonelectoral strategies, such as lobbying or personal contact with policymakers, are ill-suited to deal with existential threats because they depend on a group’s ability to access an administration, but threatening policies are usually implemented by political rivals. By contrast, electoral strategies – such as party-building or sponsoring like-minded candidates – are a more reliable source of influence because they entail electing politicians to key policymaking positions that already share the group’s preferences and thus do not need persuading. Therefore, when confronting an existential threat, landed elites will be willing to pay the extra cost of organizing in the electoral arena. Absent this kind of threat, agrarian elites will prefer cheaper, informal means of exerting influence.

The perception of an existential threat at the time of democratic transition, when parties were being (re)built and looking for new constituencies, was particularly crucial for the development of electoral strategies. Organizing in the electoral arena to respond to threats after this foundational moment, when linkages between interest groups and political parties had already consolidated, was harder for groups that had not built these linkages during the transition. Thus, in countries where agrarian elites invested in electoral representation during the democratic transition, they were better positioned to neutralize new threats down the road.

The type of electoral strategy landowners will pursue is conditioned by their degree of intra-group fragmentation. Fragmentation hinders party-building because it increases coordination costs. Agrarian elites may be divided along regional, economic, political, or religious cleavages. When these divisions are significant, all the agreements and negotiations that developing a partisan organization entails – for example, selecting candidates and party leaders, developing a territorial organization, and designing a party platform – will be harder to bring about. For instance, where agrarian elites in different regions of the country have competing economic interests, it will be difficult for them to agree on a national policy platform for their party, while in cases where political divides exist among the rural elites, they may not be able to coordinate a common campaign strategy. Therefore, in cases of high fragmentation, landowners will deploy a nonpartisan, candidate-centered strategy that does not require those kinds of compromises. They will support like-minded candidates individually, across partisan lines. Figure 1.2 summarizes the argument.

Figure 1.2 The argument.

1.2 Contributions to the Literature on Agrarian Elites and Democracy

From the classical works of Gerschenkron (Reference Gerschenkron1943) and Moore (Reference Moore1966), who focused on landowners’ reliance on labor-repressive institutions, to more recent analyses looking at the relationship between democratization and inequality (Boix Reference Boix2003; Acemoglu and Robinson Reference Acemoglu and Robinson2006; Ansell and Samuels Reference Ansell and Samuels2010), the consensus has been that agrarian elites’ economic interests are incompatible with democracy. Even when scholars disagree on whether democracy will lead to more (Boix Reference Boix2003; Acemoglu and Robinson Reference Acemoglu and Robinson2006) or less (Ansell and Samuels Reference Ansell and Samuels2014) redistribution from the rich to the poor, they agree on one thing: Agrarian elites will fare better under autocracy, especially in highly unequal societies such as those in Latin America. While Acemoglu and Robinson (Reference Acemoglu and Robinson2006) and Boix (Reference Boix2003) argue that landowners are easy targets for democratic governments’ redistributive ambitions due to the fixed nature of their assets, Ansell and Samuels claim that under autocracy agrarian elites’ have greater capacity to restrict labor mobility and influence agrarian policy (Ansell and Samuels Reference Ansell and Samuels2014, 38). While the causal mechanisms vary, the prediction is the same: “Big landowners oppose democracy of necessity” (Boix Reference Boix2003, 37) and, therefore, will always be on the side of autocracy (Ansell and Samuels Reference Ansell and Samuels2014, 12). Considering these theories, it is not surprising that Latin American countries, where land inequality has been extremely high since colonial times, struggled to consolidate democracy before the third wave.

Starting in the 1980s, however, supporting military coups to protect their interests was no longer an option for Latin American agrarian elites. On the one hand, military governments’ disastrous management of the economy, combined with their terrible human rights abuses during the dictatorships of the 1970s and 1980s, led economic elites in many countries to question the capacity of the military to govern effectively and protect elites’ interests (Bartell and Payne Reference Bartell and Payne1995). At the same time, due to changes within the military and in civil–military relations, the armed forces were no longer available as allies for economic elites looking to destabilize democracy (Pion-Berlin Reference Pion-Berlin2001). On the other hand, changes in the international community during the 1980s and 1990s increased the costs of coups. The promotion of democracy became a high priority of US foreign policy, and regional organizations such as the Organization of American States (OAS) and the Southern Common Market (MERCOSUR) made democracy a membership requirement (Hagopian and Mainwaring Reference Hagopian and Mainwaring2005). These domestic and international changes meant that agrarian elites were compelled to look for ways of protecting their interests within democracy.

When explaining how agrarian elites may attenuate the distributive effects of democratization, most of the existing literature has emphasized noninstitutional, often violent, mechanisms such as fraud, clientelism, or the use of paramilitaries (Ziblatt Reference Ziblatt2008; Acemoglu and Robinson Reference Acemoglu and Robinson2008).Footnote ⁹ Much less attention has been paid to democratic and institutional channels of elite protection. Following the pioneering work of Albertus and coauthors (Albertus Reference Albertus2015, Reference Albertus2017; Albertus and Gay Reference Albertus and Gay2017) and Ziblatt (Reference Ziblatt2017), this book makes an important contribution to the comparative politics literature by studying how landed elites may protect their interests by playing the democratic game.

Albertus and coauthors have shown that landowners can better protect their interests in democracies than in autocracies because democracies offer landowners institutional veto points to block redistribution that are absent in authoritarian regimes (Albertus Reference Albertus2015, Reference Albertus2017; Albertus and Gay Reference Albertus and Gay2017). These scholars have focused their attention on explaining how expropriation is easier in autocracies and why, consequently, landowners should support democratization. However, how landowners organize to use the institutional mechanisms of democracy to their advantage or why they sometimes fail to do so remain under-explained. This book addresses precisely these issues by explaining when and how landowners will organize in the electoral arena. Contrary to Albertus (Reference Albertus2015, Reference Albertus and Gay2017), my research shows that democratization is indeed threatening to landowners. However, contrary to the predictions of redistributivist theories of democratization, this perception of threat will not necessarily lead landowners to hamper democracy. Rather, it will motivate them to organize in the electoral arena. This book advances the literature by specifying the various mechanisms through which agrarian elites have been able to protect their interests democratically in contemporary Latin America. Put differently, while the existing literature has focused on understanding why economic elites will concede democracy,Footnote ¹⁰ this book investigates how they adapt to a democratic regime that they did not choose.

Another important theoretical contribution of this book is to show that economic elites can organize effective electoral representation in the absence of strong conservative parties. The existing comparative politics literature sees conservative parties as the main vehicle for economic elites’ electoral representation in democracies. In this vein, scholars of Latin American politics have long emphasized the importance of conservative parties for democratic consolidation as economic elites will tolerate democratization only where they believe their interests are well-protected (Di Tella Reference Di Tella1971; Gibson Reference Gibson1996; Middlebrook Reference Middlebrook and Middlebrook2000). More recently, Ziblatt (Reference Ziblatt2017) has advanced these ideas by showing how strong conservative parties played a crucial role in helping democracy come about and endure in Western Europe. Where landed elites did not build well-organized mass parties, they feared democracy and tried to undermine it (Ziblatt Reference Ziblatt2017). At the same time, the literature has also highlighted how difficult and potentially risky party-building is,Footnote ¹¹ especially for economic elites who comprise but a small fraction of the electorate.Footnote ¹² However, as this book shows, the absence of institutionalized, electorally strong conservative parties does not necessarily mean democracy will perish due to the lack of support from the elites. Conservative parties are not the only vehicle for the electoral representation of economic elites in democracies. Rather, economic elites can organize effective electoral representation through nonpartisan means.

This book analyses agrarian elites’ choice to build a conservative party in relation to other nonpartisan electoral strategies available to economic elites in democracies. In particular, it discusses a novel strategy by agrarian elites in Brazil: a multiparty congressional caucus. Multiparty caucuses allow interest groups to coordinate legislative work – one of the main functions of political parties in democracy – without necessarily building a centralized electoral machine to select their representatives. This is relevant because party-building has become increasingly harder as political fragmentation, electoral volatility, and the dilution of partisan identities are on the rise across the developing world.Footnote ¹³ In this context, where many parties are little more than the electoral vehicles of ambitious politicians,Footnote ¹⁴ and candidates’ personal characteristics are more reliable indicators of their policy preferences than their partisan affiliation,Footnote ¹⁵ candidate-centered strategies may become a more effective option for interest group representation than party-building.

1.3 Agrarian Elites in Contemporary Latin America: Urbanization, the Commodity Boom, and the Left Turn

The paradigmatic “undemocratic landowners” that reigned over the Latin American countryside until the last decades of the twentieth century based their power not on the economic importance of agriculture – as most farms were inefficient or outright unproductive – but on the political control of the peasants living on their lands. The Latin American rural landscape, however, has changed dramatically since then. Over the last six decades, a series of structural and political transformations have undermined landowners’ old sources of power while increasing their wealth.

Over the last thirty years, agriculture has become one of the pillars of many Latin American countries’ economies, with the old unproductive latifundio giving way to highly mechanized farms producing for the international markets. The exhaustion of the import substitution industrialization model (ISI) that ended in the debt crisis of the early 1980s led policymakers to focus on alternative growth models based on the region’s comparative advantages. The surge in commodity prices during the boom of the 2000s, which more than doubled in less than a decade,Footnote ¹⁶ helped consolidate this new development paradigm based on the large-scale production of agricultural commodities for the international markets, China in particular. Between 1980 and 2010 the area occupied by cropland in the region grew by 35 percent (Figure 1.3), turning many Latin American countries into world leaders in the production and export of agricultural commodities. Between 1995 and 2015, the region’s share of global agricultural trade rose from 8 to 13 percent (Martel et al. Reference Nin-Pratt, Falconi, Ludeña and Martel2015, 1).

Figure 1.3 Evolution of the rural population and croplands: Latin America, 1965–2016.

Source: Author, based on FAOSTAT Land Use domain.

However, this prodigious growth in the economic weight of agriculture brought with it the loss of some of the landowners’ old sources of political power. The rural population has been in steady decline in the region since the 1960s (Figure 1.3) and the expansion of agriculture did not reverse this trend, as new production techniques required less labor and replaced tenants with seasonal workers (Fearnside Reference Fearnside2001; J.T. Roberts and Thanos Reference J. Timmons and Thanos2003; Vergara-Camus and Kay Reference Vergara-Camus and Kay2017a).Footnote ¹⁷ While in the mid-1960s half of the Latin American population was rural, by 2008, in the heyday of the commodity boom, only 22 percent of the region’s inhabitants lived in rural areas (Figure 1.3). The decline of the peasant population together with the consolidation of democracy and the expansion of cash transfers to the rural poor undermined the ability of rural elites to mobilize votes through coercion or clientelism,Footnote ¹⁸ thus leaving landowners without an electoral safeguard against the preferences of the urban majorities. As a consequence, agrarian elites had to find new ways to influence policymaking in the region as they lost their old source of electoral power.

The beginning of the twenty-first century saw another important change in the region, the resurgence of the political left.Footnote ¹⁹ Many countries elected leftwing presidents who governed until the mid-2010s, when the trend started to reverse. The rise to power of leftwing governments in the region with the highest land inequality in the world (Vollrath Reference Vollrath2007) raised alarm among the agrarian elites, especially when in some cases, such as Lula da Silva in Brazil, Evo Morales in Bolivia, and Fernando Lugo in Paraguay, leftwing presidents had promised to implement land reform during their campaigns and were supported by peasant movements. However, once in power these leftist presidents not only failed to change the unequal structure of land ownership in their countries but also ended up implementing policies that helped the expansion of agribusiness.Footnote ²⁰

There are structural explanations for why the leftwing administrations that governed Latin America during the commodity boom did not challenge agrarian elites’ interests. First, the economic transformations described earlier meant that agrarian elites were no longer the backward, inefficient actor that leftist parties impugned in the past. In the Latin America of the commodity boom, agricultural production was an important driver of economic growth and one of the main sources of foreign exchange, which created a tension between leftwing governments’ redistributive agenda and their fiscal needs. Second, high levels of urbanization in the region meant there was no longer a large constituency calling for land reform. Consequently, governments prioritized redistributive polices benefiting the urban poor.

Although structural changes may have played a role in explaining why most leftwing administrations in Latin America did not plunder the agricultural sector to finance their redistributive agenda, they cannot explain why agrarian elites’ have not been able to block redistributive politics everywhere. A central claim of this book is that agrarian elites’ capacity to organize politically to protect their interests is crucial in explaining this variation. As discussed earlier, both Brazil and Argentina, two of the world’s biggest exporters of agricultural commodities, were governed by leftist parties with redistributive agendas during the commodity boom of the 2000s. However, while Brazilian agricultural producers managed not only to avoid taxation but also to secure massive subsidies thanks to their representation in Congress, Argentine elites’ lack of political connections left them defenseless to fight increasingly high taxation. This quick comparison suggests that even if landowners’ structural power has increased throughout the region as agricultural exports have become the pillars of many Latin American economies, agrarian elites’ capacity to translate this economic power into political influence has varied. This book proposes a theoretical framework to understand this variation.

1.4 Research Design and Case Selection

The main purpose of this study is to explain the variation in agrarian elites’ strategies of political influence across contemporary Latin America. Specifically, this study aims to understand when landowners will invest in electoral representation and why they deploy different electoral strategies. To this end, I compare three country cases since their most recent democratic transitions: Argentina since 1983, Brazil since 1985, and Chile since 1990. Cross-country comparisons are combined with within-case longitudinal analyses of agrarian elites’ strategies over the last three decades.Footnote ²¹ A comparative, small-N research design is the best suited for the study of economic elites’ strategies as it allows me to focus on the causal mechanisms connecting each explanatory factor to landowners’ strategic decisions.Footnote ²² Moreover, it is not clear that the research question motivating this book can be adequately addressed through a large-N quantitative analysis since measuring agrarian elites’ perceptions and strategic calculations requires access to their thinking. Thus, employing the typical tools of process tracing,Footnote ²³ I present an in-depth analysis of agrarian elites’ incentives to pursue a given strategy of political influence, as well as of how each strategy is deployed and adapted over time.

Following Slater and Ziblatt’s (Reference Slater and Ziblatt2013) recommendations, I selected cases that both represent the whole range of variation on the dependent variable and maximize control over alternative explanations to generate causal inferences that can extend to a broader set of cases. The three analysed cases are representative of the variation in agrarian elites’ strategies of political influence found in the larger case universe.Footnote ²⁴ Argentina before 2008 is a case where agrarian elites did not organize in the electoral arena, influencing politics instead through direct contact with public officials and protests. Both Brazil and Chile are cases of electoral participation by agrarian elites but through different channels. In Chile, landowners have built a conservative party to represent their interests. In Brazil, the electoral strategy has been nonpartisan and candidate-centered. Brazilian agrarian elites have run their congressional candidates in different parties, coordinating their work in a multipartisan caucus.

The country cases were chosen so as to discard alternative explanatory factors while holding certain background conditions constant. In terms of similarities, Argentina, Brazil, and Chile are among the most developed, industrialized, and urbanized countries in Latin America. Moreover, the three countries share a history of military governments that enjoyed the support of agrarian elites in the 1970s, transitions to democracy in the 1980s, and leftwing governments in the 2000s. Similarly, the three countries have presidential systems of government that grant the executive branch considerable legislative power (Mainwaring and Shugart Reference Mainwaring and Shugart2003; Samuels and Shugart Reference Samuels and Shugart2003).

Despite these similarities, the three countries present significant variation in other characteristics, but not in ways that predict the observed outcomes, which allows me to discard these features as alternative explanatory factors. First, the observed variation in agrarian elites’ capacity to influence policymaking rules out structural power as an explanation. In capitalist societies, business’ structural power emanates from their contribution to a country’s economy in terms of employment, investment, value added, or foreign exchange.Footnote ²⁵ The three countries are leading exporters of agricultural products, oilseed, and grain in the case of Argentina and Brazil, and fruit in the case of Chile. However, the economic weight of agriculture is quite different for each country. In 2019 agricultural products (foodstuffs, vegetable products, and animal products) comprised 56 percent of total exports in Argentina, 34 percent in Brazil, and 27 percent in Chile.Footnote ²⁶ This variation does not, however, map onto the variation in agrarian elites’ political influence or capacity to organize in the electoral arena. Argentina, the case where structural accounts of business power would predict that agrarian elites would wield the greatest political influence, is actually where governments have implemented policies most harmful to the interests of producers. Moreover, despite a considerable difference in the importance of agriculture for the economy of their countries, producers have been able to organize in the electoral arena and influence policymaking in both Brazil and Chile. While agrarian elites’ political organization – what the business politics literature calls instrumental power – compensates for their weaker structural power in Chile, it reinforces their economic prominence in Brazil.Footnote ²⁷

The three countries also differ in the degree of institutionalization and fragmentation of their party systems at the time of democratic transition and in their electoral rules. Before the 1973 military coup, Chile had had a stable democracy for forty-one years, while both Argentina and Brazil experienced multiple military coups during that period. Despite this, landowners were able to organize in the electoral arena in both Chile, where a strong democratic tradition existed, and Brazil, where it was much weaker. Party-system fragmentation at the time of transition varies across the three cases but does not correctly predict the observed outcomes. The effective number of parliamentary parties (ENPP) in Chile in 1989,Footnote ²⁸ the country where landowners invested in party-building, was considerably higher (4.9) than in Brazil (3.1) in 1985 where agrarian elites chose a multiparty strategy.Footnote ²⁹ Despite the common overrepresentation of rural areas (Snyder and Samuels Reference Snyder, Samuels and Gibson2004), the three countries have different rules to elect Congress. Argentina and Brazil use proportional representation, but with closed and open lists, respectively. Until 2015, Chile had a unique “binomial” system.Footnote ³⁰ Electoral rules cannot be the only factor explaining variation in agrarian elites’ strategies of political influence, as landowners organized in the electoral arena in both Brazil and Chile despite very different electoral rules. Moreover, the within-case analysis of Argentina, where agrarian elites changed strategies after the 2008 conflict, highlights the inadequacy of electoral rules as an explanatory factor, given that they remained unchanged through the analysed period.

Brazil and Chile have a history of electoral organization by agrarian elites prior to the last democratic transition that is absent in Argentina. Although this historical antecedent cannot be completely ruled out as a facilitator of landowners’ organization in the electoral arena in the first two countries, the social, economic, and political transformations the region experienced prior to the third wave limit this factor’s explanatory power. First, while the absence of a large peasantry in Argentina has been cited as one of the main causes of the failure to organize partisan representation by the country’s landed elites,Footnote ³¹ this difference has become less relevant in explaining agrarian elites’ capacity to organize in the electoral arena in the contemporary period due to the erosion of landowners’ clientelistic networks elsewhere in Latin America. As the Chilean case suggests, the political control of the rural poor is no longer a necessary condition for agrarian elites’ electoral organization. Landowners in this country have obtained electoral representation despite the dismantling of clientelistic networks in the countryside. Second, it is important to remember that in both Brazil and Chile, the military restructured the party system and changed the electoral rules.Footnote ³² Therefore, even when agrarian elites had previous experience with electoral involvement, by the time of the transition, the rules and players had changed since their last encounter with democratic competition. Lastly, the fact that Argentine agrarian elites attempted electoral representation after the 2008 conflict indicates that the variable “threat” explains when agrarian elites will organize in the electoral arena better than historical legacies of electoral involvement do. Table 1.1 summarizes the case selection criteria.

My multilevel, longitudinal research design involving cross-country and within-country comparisons across time allows me to discard the influence of alternative explanatory factors that could be driving the observed outcomes. First, I compare the three cases during the democratic transition to show how differences in the level of threat explain agrarian elites’ decisions about whether or not to enter the electoral arena. Whereas Brazilian and Chilean landowners feared they could be a target of the new governments’ redistributive policies, Argentine producers did not face an equivalent threat and consequently refrained from investing in electoral representation. Second, the within-case analysis of Argentina enables me to show – while controlling for national-level alternative explanations – how variations in level of threat explain changes in agrarian elites’ strategies. In that country, landowners entered the electoral arena in the aftermath of the 2008 agrarian conflict, when perceived levels of threat reached existential levels. Third, I compare the two cases of high threat during the transition, Chile and Brazil, to identify the factors that led landowners in these two countries to choose different electoral strategies. In Chile, agrarian elites were politically cohesive, which helped them (re)build a party to protect their interests in the new democracy. In Brazil, in contrast, agrarian elites’ political rivalry at the local level hampered the building of a national party to represent agrarian interests. Consequently, Brazilian landowners built their own multiparty congressional caucus instead.

1.5 Data Sources

The main source of data for this study are semi-structured, in-depth interviews with elite informants. In total, I conducted 158 interviews during fieldwork carried out over the course of ten months between 2013 and 2017 in the three countries.Footnote ³³ Interviewees included leaders from the main national and state-level producers’ associations, ministers of agriculture and other high-ranking officials in sectoral agencies, national and state-level legislators with an agricultural background or from districts where agribusiness is an important source of revenue, and party cadres with connections to the sector among other key informants.Footnote ³⁴ In addition, in Brazil, I conducted several participatory observations of the work of Bancada Ruralista members in the Agricultural Congressional Committee as well as during private meetings at their headquarters.

I triangulated interview evidence with archival analysis of newspaper articles, business associations’ publications, legislative debates, as well as statistical analysis of electoral and campaign contribution records. Through the analysis of historical primary sources,Footnote ³⁵ I gained access to what producers’ associations were doing and saying at the time of the democratic transition to then compare that information with what my interviewees remembered thirty years later. Statistical analysis of campaign contribution records in Argentina and Brazil allowed me to identify electoral investments by agribusiness.Footnote ³⁶ Lastly, secondary sources – historical, anthropological, sociological, and political science studies on agrarian elites by local and foreign scholars in the three countries – were key in operationalizing and measuring the explanatory factor “elite fragmentation” and in checking my assessment of the levels of perceived threat at the democratic transition.

1.6 Plan of the Book

This book is organized as follows. The next chapter introduces a new theory to explain the variation in agrarian elites’ strategies of political influence. It highlights the role of two independent variables – perception of an existential threat and intragroup fragmentation – to explain when and how agrarian elites will organize in the electoral arena. Chapters 3–5 develop the comparative historical analysis for the cases of Chile, Brazil, and Argentina, respectively.

Chapter 3 analyses a case of party-building by agrarian elites in Chile. It presents evidence of Chilean landowners’ financial support of the political right, their identification with rightwing legislators, and the programmatic convergence between agrarian elites’ preferences and the policy positions of rightwing parties, Renovación Nacional (National Renewal) in particular. The chapter argues that agrarian elites in Chile decided to invest in an electoral strategy of political influence at the time of the democratic transition because they feared a center-left government would endanger their property rights. It presents evidence of how this perceived threat was founded on landowners’ previous experience with democracy during the 1965–1973 period, when their farms were expropriated. The chapter also illustrates how low intragroup fragmentation facilitates party-building. Shared political and economic interests among the Chilean economic elite in general, and agrarian elites in particular, decreased the coordination costs associated with building a party to represent them.

Chapter 4 discusses a novel electoral strategy by which landowners have successfully influenced policymaking in democratic Brazil: a multiparty congressional caucus. It shows how agrarian elites finance the campaigns, encourage other producers to support, and subsidize the work of like-minded legislators independently of their partisan affiliation, as well as how legislators of agrarian origin collaborate across partisan lines. The chapter argues that Brazil’s Agrarian Caucus is the product of agrarian elites’ collective efforts to build a channel of electoral representation to protect their interests under democracy in a context of high political fragmentation. The threat of radical land reform during the democratic transition prompted landowners to engage in electoral politics. However, high political fragmentation among the agrarian elite rendered party-building unfeasible. The Brazilian case illustrates the advantages of an electoral, candidate-centered, multipartisan strategy over other strategies available to economic elites in democracies such as lobbying or party-building. First, self-representation has granted Brazilian agrarian elites direct access to key policymaking positions from which they have shaped sectoral policy according to their interests. Second, by being multipartisan, the Agrarian Caucus has multiplied the agenda setting positions it controls in Congress and increased its leverage vis-à-vis the executive.

Chapter 5 analyses the Argentine case, where, until the 2008 conflict, agrarian elites had historically shunned the electoral arena. Prior to 2009, Argentine landowners influenced politics through informal personal contacts with high-ranking government officials and, when lobbying failed, protests. This chapter shows how, in the absence of an existential threat, agricultural producers had no incentive to invest in electoral representation during the democratic transition or in the following elections. This lack of political organization, in turn, left them defenseless against the hostile policies of the center-left Peronist FPV administrations (2003–2015). In 2009, landowners switched strategies and entered the electoral arena to confront the confiscatory policies of the FPV. Given Argentine agrarian elites fragmentation, they deployed a candidate-centered strategy, sponsoring the candidacies of a dozen agricultural producers for Congress under diverse party affiliations. However, institutional features and ideological differences among producers’ associations blunted the effectiveness of the strategy and led to its abandonment.

Finally, in Chapter 6 I explore how the arguments developed in this book travel beyond the three analysed cases and discuss their broader implications for the field of comparative politics, in particular for the relationship among economic elites’ political representation, democracy, and inequality. First, I test the scope conditions of the argument by analysing agrarian elites’ strategies of political influence in a country where democracy is less consolidated: Paraguay during the Lugo administration (2008–2012). Next, I look at party-building by agrarian elites beyond South America, in a different historical context marked by civil war: post-1979 El Salvador. Finally, I extend the argument beyond agrarian elites, focusing on nonpartisan electoral representation by other interest groups in two contemporary cases: for-profit universities in Peru and conservative religious groups in Colombia.