English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Cited by Google 
Similars in
SciELO 
Similars in Google 
Permalink
Introduction
In the late 1980s and early 1990s, few Colombian municipalities had competitive party systems. Most were either bipartisan or had a hegemonic political party. Our research shows that between 1997 and 2015, the majority of Colombian municipal party systems experienced stable levels of competitiveness, and a very small portion of them evolved towards greater levels of competition among candidates and parties. This article focuses on this shift and the factors that appear to have driven it.
What explains the movement from restrictive local party systems to those that encourage higher levels of electoral competition? This study argues that the transition towards more competitive party systems takes place when a change in structural conditions occurs and, at the same time, opposition leaders understand that the costs for entering the electoral race are lower than before and that they might now be successful in challenging the previously dominant political elites.
In scenarios where the revenue produced by a specific industry is the main and practically sole economic source for the region or municipality, such dependency shapes social and political relations. The result is rentier subnational departments with characteristics distinct to those of municipalities with a more diversified economy. In this article, we state that emerald mining in the Western Province (Provincia de Occidente) of the Boyacá department, Colombia (hereafter, Western Boyacá), allowed local elites to develop powerful networks that resulted in, for a considerable period and in certain municipalities, less competitive elections. By the end of the 1990s, the emerald industry suffered a decrease in profits, with emerald revenues no longer as determinant for the economy of the region. This affected the elites’ capacity to control the electoral process, resulting in higher levels of competitiveness and a more open party system. In sum, in contexts of enclave mining economies, electoral competition is affected by the changing conditions that characterize economic structures, and by the strategies used to build and maintain power. Our analysis is based on both quantitative and qualitative information: electoral data, interviews with key informants, observation and analysis of local and regional newspapers.
Following this introduction, the article develops as follows: The second section describes the characteristics of subnational electoral competition in Colombia. The third reviews the literature on the relationship between subnational electoral competition and local rentier economies, with a focus on Latin America. The fourth presents our argument, the context, and the actors, and the fifth describes electoral trajectories in Colombian municipalities from 1997 until 2015. After that we present our case study and discuss our findings, and in the last section we state our conclusions.
2. Subnational Electoral Competition in Colombia
Recent literature has shown that electoral competition does not spread in a uniform way throughout a territory (Gibson and Suárez Cao 2010). On the contrary, in several cases, highly competitive national party systems coexist with a wide spectrum of subnational systems that do not demonstrate this dynamic. In some extreme cases, democratic national contexts concur with what has been conceptualized as subnational authoritarianism (Fox 1994; Gibson 2005, 2010; Higuera, Hoyos, Uribe, and Wills-Otero, 2014; Hoyos 2005, 2009; Giraudy 2011; Gervasoni 2011; Pino, 2013).
Colombia is a unitary country. Therefore, unlike a federal structure, one would expect a more homogenous competitive context throughout its territory. However, heterogeneity in competition is one of the most deeply rooted characteristics of Colombia’s political system. While some municipalities are highly competitive, others are not competitive at all and, as a multilevel party system (Swenden and Maddens 2008), national and subnational levels do not necessarily present the same dynamics of competition. Beyond this heterogeneity, the literature has also revealed the way in which the system has transitioned away from local authoritarian dynamics towards further openness, especially after the proclamation of the 1991 political constitution (Gutiérrez 2001). With this context in mind, we explore the differences in electoral competition in mayoral contests across the country over the last two decades (1997-2015), and we attempt to explain the shift towards more competitive elections at the municipal level in one specific context: municipalities in Western Boyacá, a region with a fragile state presence, high levels of violence and poverty, and income economies based on emeralds.1 The structural characteristics of the local context promote the activation of certain strategies by political actors, or at least make them more feasible.
The municipality of Pauna and, more generally, the emerald subregion of Western Boyacá, served as a laboratory to test theories related to the construction of subnational authoritarianism and provide some alternatives to it, similar to the study by Basset et. al. (2017) on the municipality of Riosucio in the Colombian department of Chocó. Specifically, and as will be developed later, we find that in scenarios of rentier economies that are not efficiently regulated by the central government, as in the case of emeralds, local elites depend less on their relationships with national politicians to maintain their hegemonic control, as Gibson highlights (2005, 2010). We also find that, in contrast to many cases studied in this literature, those who are in control of authoritarian dynamics are not necessarily politicians; given the sometimes illegal nature of the resource and/or its management, it is often too risky for those who hold the economic power to assume public office directly.
In this article, we use a subnational approach within a unitary system to test a body of literature that has focused almost exclusively on federal countries. Furthermore, we show the importance of building bridges between the literature on subnational authoritarianism and that of rentier economies to analyze the Colombian case. Specifically, presenting different electoral trajectories we show how, in contexts of enclave mining economies, electoral competition is affected by strategies used by local elites as well as by changing conditions that characterize economic structures.
For our case study we selected a region that, with some exceptions (Barón and Gutiérrez 2008; Leiteritz and Riaño 2018; Uribe 1992; Téllez 1993;), has been practically absent from previous research on politics and elections in Colombia. This case study allows us to analyze the mining context, and the behavior of actors within this context regarding specific electoral competition dynamics. By doing so, we take into account the dynamics of violence associated with the mining business resulting from emeralds not being effectively regulated at the national level and their commerce not being completely legal. Our analysis is based on both quantitative and qualitative information: electoral data, interviews with key informants, observation and analysis of local and regional newspapers.
3. Electoral Competition in Contexts of Extractive Economies
Although recent literature shows the Colombian party system has been changing towards more competitive elections, these evolving processes do not follow the same paths, nor do they share the same characteristics across the country. Our study focuses on an uncommon, insufficiently-studied setting as a vehicle for understanding the way structure and strategy interact in specific electoral contexts: mining-dependent local economies in Colombia.
Strategies are enabled by certain local conditions. Here we explore the relationship between political actors and the variations over time in levels of electoral competitiveness in mayoral contests in Colombia. We argue that the characteristics of the local context produce incentives for political actors to develop certain strategies that, in turn, influence the levels of competition in local elections. In particular, we contend that in order to understand the changes towards more open competition - or more competitive systems - two types of processes occur: a structural one, related to the situation of the natural resource-led economy (be it stable, rising, or declining), and an agency one, associated with the strategies followed by actors involved in the fight for power in these resource-rich regions. We find that strategies differ depending on the context.
Focusing on a key aspect of the democratic game, electoral competition, we seek to identify the different strategies actors use to control and build political power (Gibson 2007, 2010). These processes can seek to maintain a system with low levels of competitiveness or, on the contrary, to overthrow it. We intend to identify the differences in the strategies by observation at the subnational level through fieldwork.
This will also provide inputs on electoral competition relating to the extraction of natural resources and the violence generated in the communities where these resources are located and exploited. If the most important business structuring the local economy is the exploitation of natural resources, the strategies used by political actors will certainly be affected by the singularity and complexity of such a context. Some authors have shown how the fiscal dependence on the political center of subnational governments explains low levels of political competition in local democratic regimes (Gervasoni 2010).
Some of these territories are characterized by patrimonialist political and economic structures in which citizens’ economic well-being depends on existing social networks. Those who take office will be able to control access to goods and services, and to allocate positions in the civil service in exchange for electoral support. This situation reduces the possibilities of real alternation in political power. In turn, it affects the levels of competitiveness and the quality of the democratic regime (Calvo and Murillo 2005; Fox 1994; Hoffman and Gibson 2005). Additionally, it has been proven that territories with high levels of dependency on non-fiscal revenues - such as extractive industries, including oil, gas and mining - produce scenarios prone to less competitive political systems in which local governments are less responsive to voters (Goldberg, Wibbels, and Mvukiyehe 2008).
According to Gervasoni, in rentier provinces “public revenue originates from an external source that does not depend on the local economy’s taxation capacity and that is not necessarily proportional to its size. The rents obtained from these resources must, therefore, be placed in the same analytical category as the rents obtained from unconditional foreign aid or fiscal federalism” (2011, 583). Generating a considerable portion of its income from external rents allows the state to stand distant, or maintain distance, and be less accountable (Ross 2001, 329). The control of political and economic power follows the same path, shaping power relationships based on patronage, clientelism and, in countries like Colombia, violence. There are, thus, fewer incentives for electoral competition given that political and economic power fall into the same groups’ hands, as we will see in the emerald case study presented here.
As mentioned above, rural territories where rents extracted from natural resources become a significant portion of government income - at the municipal or departmental level - tend to comprise areas prone to the proliferation of authoritarian practices. This produces low levels of electoral competitiveness. Although this paper is not about fiscal federalism, Gervasoni’s concept is useful to explain how the emerald business affects the different strategies used by political actors to build and control power, affecting electoral competition. In our case study, we find that strategies developed by national and subnational players are conditioned by the dynamics of a rentier territory (Gervasoni 2011), in this case the emerald mines in Western Boyacá.
As Gervasoni states, the situation produced by this kind of rent gives more freedom to politicians, and powerful subnational players, to spend their resources and therefore find ways to build and nurture clientelistic and patronage networks. In that sense, these territories experience “the least favorable circumstances for political competition” (Gervasoni 2011, 584). Unlike federal and decentralized contexts, unitary centralized ones give subnational players less leeway. Therefore, analyzing what happens in a resource-rich territory where extraction contributes to the forging of political power by giving political players the means needed to develop their plans of power control, is relevant to exploring the relation between structure and agency.
4. Context, Actors, and Strategies
According to Gibson (2005), there are three main strategies that will result in the configuration of a subnational authoritarian regime. The first is the parochialization of power, whereby elites dominate local power and intervene in the assignation of economic resources. The second strategy is the nationalization of the political influence of local leaders. It is unusual for an undemocratic local regime to remain isolated from the political dynamics that operate at regional and national levels. And the third is the monopolization of national-subnational linkages. This strategy aims to reduce the latent threats posed by the local opposition and the possibility that they will disseminate information about the conflict or authoritarian situation beyond the area of hegemonic control, by strengthening the connections of the local authoritarian players with the political center.
These strategies do not occur in a vacuum; they are permeated by the specific circumstances of the local context. In the absence of certain conditions, some strategies will not be feasible. In fact, one of the arguments we seek to emphasize is that political agency is conditioned by a structure and a context that make employing certain strategies more or less viable.
While authors like Gibson (2005) do not ignore the relevance of the local contexts in which these strategies are used, they do assume that the most significant variable affecting such strategies is the connection between local and national actors. The explanatory advantages these studies assign to agency probably reside in the fact that they focus on federal countries, where subnational actors are more powerful and can freely alter key aspects related to multilevel relations within the national territory, such as redesigning the local rules of the political game.
In unitary structures, such as Colombia, local agents are less autonomous or have less room for maneuver to change or maintain authoritarian practices at the subnational level. Therefore, certain strategies are more feasible in some contexts than others. For example, less economically developed rural territories are more prone to less competitive regimes. Furthermore, in rural areas, traditional political parties and their leaders usually receive greater support than in more urban areas, where the electorate tends to vote less for traditional candidates and more for independents or new parties. Historically, in rural areas the vote was less volatile, with strong support for Liberals or Conservatives - depending on the region - instead of other parties.
Building on the existing literature on rentier economies and subnational authoritarianism, in this article we want to combine the strategic interaction approach (agential dimension) with the perspectives of the rentier economies (structural dimension), by modulating the effects that each perspective has on electoral competition in unitary scenarios, and where the economic resource is not regulated by the state. We found both approaches useful to understand the changes in electoral competition in a rentier territory, such as Western Boyacá. Focusing on the interaction among political actors, institutions and the mining context allowed us to fully depict some of the dynamics that emerged in this rather unique context.
Specifically, with this case study we want to show that subnational authoritarian practices in contexts of rentier economies do not necessarily go through some of the situations proposed by the literature. First, in certain contexts, connections between the local and national political elite are not strictly necessary to develop an authoritarian enclave; in fact, on many occasions, low electoral competitiveness may develop outside the national political establishment. Second, the rentier economy that serves as an input to keep the clientele structure afloat may be illegal - or partly illegal - and that factor transforms the way in which the actors structure local political dynamics. In the following section we explain how we plan to answer the research question presented in the introduction: What explains the movement from restrictive local party systems to those that encourage higher levels of electoral competition?
5. Electoral Trajectories in Colombian Municipalities
Before presenting our case study, and in order to explain how and why we selected it, it is useful to describe Colombian municipal electoral trajectories from the late 1990s until 2015. To establish these trajectories, we observed the level of competition in mayoral contests in Colombian municipalities. To measure this variable, we focused on a specific attribute of electoral competition: competitiveness. It is a continuous variable that is calculated using the margin of victory, the difference in percentages between the first and the second candidate in a given election. The greater the distance between them, the lower the level of electoral competition. More competitive elections indicate the presence of various actors in the contest, while less competitive elections imply the existence of hegemonic political actors.
In order to observe the variation in electoral competition, we carried out a trajectory analysis known as “semi-parametric group-based longitudinal analysis” (Nagin 2005). Trajectory analysis is a form of statistical modeling that allows the identification of patterns of behavior of sub-populations within a given sample of observations. To identify similar patterns of behavior among the observations making up the sample, the model applies the logic of estimations of maximum verisimilitude. This allows the observation of, in this case, municipalities that behave similarly over time to be grouped within a given cluster, while those that behave differently are placed in other groups.
This is how different territorial trajectories of electoral competition in Colombian mayoral contests were developed. We show that there are four different types of trajectories, and that stability is the most common in Colombia. Municipal data allowed us to observe the differentiated dynamics of local politics, and also generated further observations. Additionally, mayoral election results allowed us to control for institutional factors, such as the district magnitude.
We decided to focus our qualitative analysis on the strong openness trajectory since it allows us to look more deeply at the changes in subnational competition, even if it does not represent the majority of municipalities during this research period. This choice also makes sense if we consider that most of the recent literature on parties and elections assumes that the Colombian party system has transitioned towards openness, in spite of the fact that: i) regional and local electoral data do not resonate with this conclusion, and ii) that in regions where this openness has taken place, it has not occurred in the same time frame or with the same characteristics from one area to the next.
Focusing on openness trajectories allowed us to grasp the way change takes place over time and the roles played by actors and structure. With the aim of testing our argument, we built on our findings about the rise in electoral competition, and we identified the causal mechanisms behind this increase in the context of a declining emerald mining business. To accomplish this, we decided on a case study focused on an extreme case: the municipality of Pauna, in Western Boyacá, one of 15 provinces into which the department of Boyacá is divided. Prior to selecting Pauna as our case study, we visited different municipalities and conducted semi-structured interviews with eighteen people of different backgrounds, such as current or retired politicians, public servants, candidates, and journalists. We carried out eighteen semi-structured interviews between November 2014 and June 2015: three in Briceño, three in Saboyá, two in Chiquinquirá, six in Pauna, one in Maripí and three in San Pablo de Borbur. The observations and interviews were accompanied by analyses of local and regional newspapers, which helped us to cross-check the information provided by the interviewees.
Western Boyacá has always been a violent scenario. The power of the patrones has contributed to creating a culture of silence and mistrust that made it significantly harder to conduct interviews and access the data and evidence we needed. This was of course an obstacle, but it also highlighted the fact that there was a story to be told. Being able to explain some of the underlying factors in the electoral competition opening process is a decisive step towards further knowledge of subnational politics in these specific economies.
6. Trajectories of Electoral Competition in Colombian Municipalities (1997-2015)
Exploring the Whole Country
In this section, we analyze the levels of competition in the last two decades for more than one thousand municipalities in Colombia. Figure 1 illustrates the evolution of the average electoral competition in Colombian mayoral contests between 1997 and 2015. According to the data, the levels of competition in this period - in mayoral elections - show a slight increase. This consistent rise in the levels of competition hides significant internal variation which suggests that, even if on average there is a trend towards further political openness - higher levels of competitiveness - this is not necessarily the picture that best describes the subnational reality throughout the country.
Source: Pachón and Sánchez; Registraduría Nacional del Estado Civil, 2015.
Figure 1: Evolution of Electoral Competition in Mayoral Contests, 1997-20152
Figure 2 presents the results of the trajectories found for the period between 1997 and 2015. The estimated model allows a general prediction to be made regarding the percentage of municipalities that make up each of the identified trajectories, along with an estimation of the probability that each individual municipality would be included in them. As can be observed in Figure 2, four trajectories were identified: 1) closure, 2) stability 3) closure/openness, and 4) strong openness.
This estimation was run using a model involving three trajectories with different functions: one with a polynomial function (Trajectory 4), another with a cubic function (Trajectory 3) and the first ones with a linear function (Trajectories 1 and 2). The probability of a municipality belonging to one of these trajectories was, on average, 85% for Trajectory 1, 97% for Trajectory 2, 88% for Trajectory 3, and 95% for Trajectory 4. The Bayesian indicator (BIC) for the model was -35885.24.
As shown in Figure 2, 86% of Colombian municipalities followed a pattern of stability in electoral competition (Trajectory 2). Nevertheless, we observed other trajectories representing 14% of the municipalities (154 in total), which did not follow the pattern of stability seen in the rest of the country. One of these trajectories, which we call closure/openness, occurs in 5.6% of the municipalities and is characterized by a decrease in the levels of electoral competitiveness between 2000 and 2003, followed by an increase in this indicator with similar values to those registered before 2000. Margins of victory were wide and became narrower in the more recent elections. The strong openness trajectory clustered 65 municipalities (5.9%) that in 1997 had very low levels of electoral competition but later initiated a sustained openness trajectory that continued until 2015 (with tight margin victory gaps since the local elections of 2000).
Finally, in the remaining 27 municipalities (2.4%) we found the closure trajectory, where the level of competition registered at the beginning of the period shrank systematically over time. In other words, the margin of victory has risen election after election. The four trajectories had different starting points, but their arrival point - at least for three of them - was the same. This means that almost all Colombian municipalities currently have very similar levels of electoral competition, but the paths they followed to get to this position have been heterogenous.
The Four Trajectories in Western Boyacá
We found all four trajectories in Western Boyacá. This province is made up of Chiquinquirá, the largest town with a population of approximately 80,000, and another 14 municipalities, resulting in an overall current population of some 200,000 (Acevedo and Bornacelly 2014). On one hand, this region has been electorally dominated by the Conservative Party (Partido Conservador, PC). In research on this region in the 1990s, Uribe (1992, 107) argued that, “As in other backward rural zones, party affiliation in Western Boyacá is a question of inheritance strengthened by primary loyalties associated with family ties and membership of a particular political client network. In general, Boyacá is considered to be a typically conservative region whose adherence to the PC dates back to the Thousand Days War [1899-1903].”
On the other hand, the region has been characterized by scant levels of infrastructural development and a precarious state presence, a situation accompanied by widespread, elevated levels of poverty. Paradoxically, this combination of poverty and weak institutional presence, along with the historical predominance of the PC, has occurred in a province that is very rich in natural resources. The region has been at the epicenter of world emerald production for more than fifty years. However, rather than bringing wealth to its people, the emerald trade has generated intense and prolonged periods of violence, known as the Guerra Verde, the Green War.
When looking at mayoral elections, all four of the trajectories modeled previously in this section are found in the region. The openness trajectory occurred in municipalities that are mainly located within the zone of the mining basin (Pauna, Otanche and Quípama), which produces the highest number of emeralds. However, two municipalities in the area of mining influence - Muzo and San Pablo de Borbur - represented a trajectory of stability. The other municipalities that displayed a stable trajectory (Chiquinquirá, Coper, La Victoria, Caldas, Buenavista and San Miguel de Sema) are found principally in the easternmost zone of the province, where the impact of the emerald business has been minimal or non-existent. Finally, the closure/openness trajectory was found in Maripí, located in the mining zone, and the closure trajectory was found in Tununguá, a municipality with no significant mining influence.
The variation in the electoral results found in this region made it a suitable “laboratory” for our case study. We were able to maintain uniformity in the contextual variables (poverty levels, institutional precariousness, and high incidence of rural areas, among others) and find a wide range of political results. One structural characteristic that was different among the municipalities, and that we were interested in exploring further, was the presence of the emerald economy. Bearing this in mind, we suggest that there is a connection between the prominence of the mining economy and the openness shown in their electoral trajectories, and between the absence of the emerald industry and the existence of trajectories of stability in electoral competition. figure 3 shows the competition trajectories in the municipalities of Western Boyacá.
Figure 3: Territorial Trajectories in Electoral Competition in Mayoral Contests in Western Boyacá, 1988-2011
What are the mechanisms that have pushed the transition from electorally hegemonic political systems to competitive ones in Western Boyacá? What role have political actors played in this political trajectory, and what was the influence of structural characteristics? We believe that analyzing the openness trajectory is the most valid approach to answering these questions because it enables us to illuminate the intense processes of political change in electoral competition that took place over time, and to test the hypotheses that we judge to underpin this pattern of change. This trajectory occurred in municipalities highly influenced by the emerald-based economy.
We identified that this change from lower competition to openness is explained by the decline suffered by the emerald business in the zone towards the end of the 1990s, and by the emergence of a group of leaders who had been building their own power bases over the years. A weakened mining economy meant that opposition leaders, that is, figures who were not part of the clientelistic political game that revolved around mining, had the opportunity to establish themselves as viable electoral alternatives.
In contrast, the trajectory of stability was evident in the municipalities that were not influenced by the emerald economy, with the sole exceptions of Muzo and San Pablo de Borbur. In these towns, local politicians lacked the capacity to coopt voters, as they did not have access to the economic resources provided by the emerald business in the other scenario. These municipalities were poor, highly rural and - like those with a powerful mining economy - had scant state presence. Their economies were dominated by small scale peasant farms (minifundios) in which families did not rely on an economic elite that demanded they vote for specific candidates in return for the material support they required to survive.
Finally, the closure/openness trajectory also occurred in some mining municipalities in Western Boyacá. The electoral competition results in these municipalities are somewhat unclear. In general, they follow a stable trend with high levels of political uncertainty. However, there are also specific moments in which extremely low levels of competitiveness occur. The case of Maripí, a mining town, illustrates this irregular behavior. Political competition in this municipality has been dominated by two opposing families who have alternated power and historically fought for access to the economic power provided by emerald mining.
7. Case Study: The Openness Trajectory in Pauna
“Before, politics was in the hands of miners. Not like today, when it is run by politicians.”
As previously stated, the aim of our fieldwork was to understand the nature and characteristics of the territory and gain insights into its particular dynamics in a considerably under-studied region. Additionally, we sought to explore long-term political and electoral trends, to decode the relationships between key actors and the social and economic contexts in which they operated, and to understand the impact of these characteristics on electoral outcomes. With this case study, we tested our argument on the importance of combining explanations that emerge from theories of strategic interaction with those that emphasize the structurally-imposed context in a mining region within the literature of subnational authoritarianism.
The structural and contextual conditions in Pauna provided incentives to local political actors to take power for themselves. This municipality located near the Coscuez mine, and others in Western Boyacá, such as Quípama and Otanche, were in the center of mining activities from the late 1960s until the end of the 1990s. During this period, their populations depended highly on mining to survive. According to the Bank of the Republic (Banco de la República), in 1966, 95% of the world emerald market depended on the Colombian black market.3 The black market would operate from Coscuez to Bogotá and then abroad (Gutiérrez and Baron 2008, 109).
The sociodemographic characteristics of Pauna are very similar to those of other municipalities in this region. It has an overwhelmingly rural and poor population, concentrated in settlements far from the small urban centers of the municipality, which themselves have a precarious state presence. Today, Pauna has around 10,000 inhabitants according to projections of the National Statistical Office, with no more than 20% residing in the urban nucleus.
Elections in Pauna - and in most of the region’s municipalities - repeatedly resulted in an uninterrupted electoral hegemony of the PC, not just in mayoral contests but also in departmental and national elections. Table 1 summarizes the tendencies of political competition in mayoral contests in Pauna.
Table 1: Mayors Elected in Pauna, 1988-2015.
| Year | Pauna | Margin of victory |
|---|---|---|
| 1988 | Joselín Casallas Lancheros (PC) | 88 |
| 1990 | Luis Villamil Lancheros (PC) | 100 |
| 1992 | Gilberto Ortiz (PC) | 55 |
| 1994 | Luis Gonzalo Núñez Peña (PC) | 54 |
| 1997 | Joselín Casallas Lancheros (PC) | 100 |
| 2000 | Gustavo Heladio Torres Sánchez (Independent) | 32 |
| 2003 | Jorge Alirio Murcia Chaparro (PC) | 25 |
| 2007 | Gustavo Heladio Torres Sánchez (Alas-Equipo Colombia) | 12 |
| 2011 | Omar Casallas Sánchez (Partido Social de Unidad National-PSUN) | 2 |
| 2015 | Eulices Augusto Caro (PC) | 1 |
Source: Pachón and Sánchez (2014). Registraduría Nacional del Estado Civil, 2015.
The PC dominated the first mayoral elections held in Pauna in 1988 and 1990, and in most of the subsequent elections the successful candidate won by a wide margin. These initial low levels of electoral competition, accompanied by the predominance of the PC in power, contrast with the election results of the first decade of the 21st century. During this latter period, not only did control of the municipality begin to alternate but, of greater importance for our argument, competition for political power began to intensify. How can this transition from political hegemony towards openness be explained? The following subsections provide detail on this case study and exemplify the openness trajectory.
Emerald Bonanza and the Patrones’ Hegemony: Strategies to Win and Control Political power
As shown, Western Boyacá is a clear example of how a certain combination of structural conditions and strategic decisions made by the patrones converged to facilitate a situation of low electoral competition. In Pauna, this situation changed when the emerald boom ended, and other local political actors saw the opportunity to use different institutional and contextual elements in their favor to contest the patrones’ power.
Between the first popular mayoral election in 1988 and the 2000 elections in Pauna, the electoral scene was dominated by Luis Murcia, aka El Pequinés. Murcia was a mine owner (patrón in local jargon) who was responsible for most of the political and economic decisions made in the zone, even around more peculiar issues, such as picking the new town priest and controlling public and private police, and other local security forces with their own armies. In electoral terms, decisions on mayoral candidacies were made exclusively by patrones such as Murcia, and their inner circles. In the words of a Pauna politician: “It was just a matter of him (Murcia) giving the order. It should be whatshisname! That was it.” (Local politician interviewed in San Pablo de Borbur, June 2015). Different strategies were successfully used by the patrón to maintain power and remain the most important figure, the uncontested authority in the municipality.
Voters were generally passive subjects who received and obeyed orders. This power rooted in the patrones’ political authority was the result of their capacity to parochialize the productive and economic sectors of communities highly dependent on the emerald business, which provided direct employment for some 25,000 people during the emerald bonanza. Mining seemed to be the only option. As one of the interviewees stated: “Every child’s dream was to have the patrón invite him to his car, give (him) them a gun and tell (him) them: son, come with me to the mine!” (Local businessman interviewed in Pauna, June 2015).
The emerald business in Western Boyacá produced paradoxical scenarios involving the informal flow of vast amounts of money in circumstances that were, overall, characterized by immense poverty, institutional weakness, and security issues. The patrón was seen not only as a leader who responded to the fundamental needs of the population, but was also feared and not to be disobeyed. He was protected by private armed forces, frequently associated with paramilitary groups, which gave a symbolic dimension to his electoral and military power. The phrase “If you don’t vote for X, you’re not coming back to the mine” sums up the pressure exercised by patrones over the population. People did what they were told in order to reduce the possibility of putting their job stability and survival at risk. This phrase was repeated multiple times during our fieldwork by different local actors that tried to depict the way the emerald patrones operated.
The mining patrones rarely occupied positions of power themselves. As long as their strategies for parochializing power guaranteed electoral success for their candidates, the costs of public exposure outweighed the advantages. To resolve this problem, their practice was to nominate someone they could trust to represent their interests in the municipal government. Given the fact that Pauna was fiscally weak, the patrones did not necessarily expect mayors to contribute to their financial interests. They preferred instead to be treated in a reverential manner by them as a way of exercising symbolic power. For example, the patrones expected to be received with honors during local festivities.
In addition, the patrones also controlled links with the political center through a chain of intermediaries that began locally with the mayor (the first link), involved deputies in the Departmental Assembly (the second link), and extended to members of Congress at the national level (the third link). The cases of former Congress members of the PC who enjoyed high levels of electoral support during the hegemonic period illustrate this connection. This is explained by a local politician: “Patrones were able to provide support for candidates running for the office of departmental deputy or member of Congress. However, many times, national politicians would end up betraying them after the election.” (Local politician interviewed in Pauna, June 2015).
In contrast to what was expected from the mayors, these higher-level positions were to provide much more than merely symbolic representation. Patrones would require them to represent more tangible interests. Members of Congress and, to an extent, departmental deputies, were responsible for identifying resources to improve local infrastructure, such as maintaining the roads that connected the urban center with the rural hamlets and settlements known as veredas, where the mines were located. This shows, very in tune with the strategies proposed by Gibson, the importance of these links for the maintenance of subnational political control. However, in this case the political ties from the local to the national level were not necessarily led by the mayor, but by the patrón.
Even more importantly, they were expected to keep the illegal nature of the business out of the national public sphere. Patrones needed to retain their low profile and conceal the fact that their power was based on coercion of the population, at times armed. In this regard, the same local politician asserts: “Patrones would mostly demand that politicians with national power preserve the low profile of the crap they pulled at the local level.” (Local politician interviewed in Pauna, June 2015). In this context, and in exchange for the vote of national and regional politicians, the heads of the emerald business remained locally unbeatable and almighty.
Finally, the patrones were also very successful at maintaining influence beyond the borders of their territories by monopolizing and exerting control over local institutions linked to the national level. As suggested in the theoretical argument presented above, this made it difficult for any actor to share information on the local conflict in broader contexts or media, impeding the dismantling of the authoritarian nature of power.
For example, an inhabitant of Pauna described how, during electoral periods, luxury cars appeared, distributing astonishing quantities of food and alcohol. These were provided by the patrón of the mine that operated in the community and were intended to capture votes for the candidate he had imposed. On some occasions, when the police tried to impose order, the patrón’s private military would disarm them and lock them up until the electoral fiesta was over. This ability to exert control over the police at local levels, and to prevent the population from informing the authorities, demonstrates the capacity of the patrones to determine the result of an election and to maintain strict control over what can and cannot be done in their territories.
But why would a politician with national influence protect local elites who operated in the gray area spanning legal and illegal activities? The answer lies in the fact that the emerald patrones had the power to provide regional and national leaders with a highly disciplined clientele, who depended on the emerald business and, hence, could be depended upon in electoral periods. This meant that regional and national politicians had to exert less effort to win votes than would have been the case in a competitive environment. Thus, had a local police officer decided to denounce what had happened higher up the chain of authority, some departmental deputy, member of Congress or powerful politician would have intervened, putting an end to such a complaint.
As Gutiérrez and Barón describe, these strategies took place in a highly violent context where constant clashes between the patrones and their clans shaped the so-called Guerra Verde. This war had different stages. First, in the late 1960s as a reaction to the state’s intention to start controlling mining production in the region, violence was the result of disagreements on the best way to articulate the patrones’ interests with those of the state. The second phase started in 1975 with the murder of two brothers of the patrón “Pacho” Vargas by another patrón, Daniel Bustos Ortega, aka El Chamizo. The third stage of the Guerra Verde was the most violent one and took place in the 1980s, resulting from a disagreement regarding work shifts in the Coscuez mine (2008, 102-129). In the 1980s, between 800 and 2000 people died as a result of the Guerra Verde.4
The Emerald Decline and the Emergence of New Political Leaders
This electoral hegemony, which had been established by a powerful economic elite in the municipalities by parochializing political and economic power, and by monopolizing the links and channels of communication and representation with the departmental and national levels, began to collapse in the early 2000s. In electoral terms, the tipping point was reached in the 2000 local elections when the local opposition took advantage of the growing decline of the emerald business. At that time, the most important mines in the region - Peñas Blancas and Coscuez - were losing productivity.5
Under these circumstances, mining patrones began to lose their capacity to extort the population, since individuals were less and less dependent on the material opportunities associated with the business. This led them to focus more on the enormous structural problems of the communities, an aspect that was highlighted by the emerging opposition. At the same time, the national government decided to start formalizing the mining sector. Concessions began to be granted to national and international companies, and the rules governing the industry were tightened up and made more rigorous.6 The informality and clientelistic practices that had prospered under the old arrangements were seriously affected.
This transition is interestingly explained by a national/regional politician who describes the reduced support received by patrones as a consequence of the emergence of vote-buying dynamics in local politics, as opposed to a previous relationship characterized by trust and loyalty between voters and local mining elites: “People used to be more honest when they voted in the past. Nowadays, people just sell their vote” (National politician interviewed in Pauna, June 2015).
In this context, local opposition leaders, who had been strengthening their positions over a number of years in scenarios such as local council elections, understood that a window of opportunity had been opened. The time had come to enter the fray and stand for mayor, the highest local office, as a local politician describes it: “The struggle for political power from ‘beyond the mines’ started in the town council. The council was not as infiltrated by the patrones. It then became a place where there was room for other political alternatives” (Local politician interviewed in Pauna, June 2015).”
Many of the leaders who emerged to compete against the power of patrones were also encouraged by institutional changes that increased political openness, ushered in by the 1991 political constitution. This new charter considerably reduced the requirements that candidates had to meet in order to stand for office, thus putting an end to the monopoly over electoral competition that the traditional political parties had enjoyed. These politicians aspired, initially, to seats on the municipal councils, and once they were successful, they became more popular and sought to pursue political careers.
The best-qualified in terms of political ability, and the most popular, later began to stand for the office of mayor and, though not successful at first, subsequently came to power. As a result, leaders from the traditional parties who had previously been coopted by the emerald patrones were defeated. The former mayor of Pauna (in 2000 and 2007) is a clear example of someone who, with the support of a political group other than the PC, took advantage of the critically important decline in the emerald business. He was able to capture the votes of small farmers, as well as others who were attracted by his promises to support agriculture and other sectors. With a novel, explicit and well-informed discourse that demonstrated the deficiencies of emerald mining as a sustainable development model for the region, this leader and his political group succeeded in Pauna. They created a precedent that, a decade later, in 2015, ended up with six candidates running for mayor.
The candidates who had been named by the patrones did not disappear completely from the electoral scene in this or other municipalities of Western Boyacá. On the contrary, they may still be found actively engaging in politics. And there are some groups in the community that still support mining families and believe they have a chance of winning. However, these clans now have to fight for each vote. Uncertainty is now a fact for those who compete for political power in this region.
Conclusions
In this article we aimed to shed light on heterogeneity in electoral competition across Colombia by examining the way structure and agency have interacted to produce change in a given context. In analyzing competition over time, we identified four trajectories: closure, stability, closure/openness, and strong openness. We also focused our case study analysis on the factors producing a shift across the trajectories. Our investigation also highlights the fact that subnational electoral competition in Colombia differs across the country, and that the assumption of a decisive increase in competition since the 1990s obscures the specificities of certain remote regions that have been absent from the research on elections and political representation.
Only 65 municipalities, out of more than 1100, fell into the strong openness category. Thus we decided to focus on one of these rare cases, constituting less than 6% of the total, to grasp the way change occurs and the role played by agency and structure in very distinctive contexts of natural resource extraction, and where violence and illegal practices are the rule.
Taking these factors into account, we analyzed the characteristics of these different trajectories in a specific and scarcely studied region of the country, Western Boyacá. This area was a world-class emerald mining region between the 1960s and the 1990s. When focusing on the electoral dynamics of this rentier territory, we argue that the characteristics of the local context affect the incentives of the actors to develop certain strategies to pursue and hold on to political power. These strategies would result in the development of low electoral competition and generally predictable elections. When emerald business revenues start to decrease, controlling electoral outcomes is much more difficult: uncertainty rises, as does competition.
For our case study, we focused on Pauna, a municipality where what we call strong openness was evident. In Pauna, patrones created a network of several stakeholders: local and national politicians, local elites, and even the Catholic Church. They became interconnected thanks to mining extraction and the rentier economy it created. Within these networks, patrones were able to successfully develop different strategies of parochialization of power, nationalization of political influence, and monopolization of political power (Gibson 2005). This phenomenon eventually led to low levels of electoral competition and high levels of certainty regarding who would win the elections.
Once the mining bonanza ended and emerald revenues were no longer the nearly singular income source for Pauna and the region, the patrones’ power began to decrease and different actors saw a window opening towards the possibility of contesting the political power built by the mining elite over three decades. At the end of the period, the mayoral elections in Pauna were more competitive, and both candidate selection processes and electoral outcomes became less predictable. The context had changed, and the actors knew they could take advantage of this new situation and develop successful strategies to challenge the patrones’ power.
Strong openness, therefore, became the result of the combination of the decline in the emerald business and the decision of new political leaders not linked to the traditional parties or emerald patrones to challenge the status quo. The new institutional design was now more favorable to independent candidates and played a key role in their ability to create more open political opportunities and participation.
The study of this specific region, and choosing Pauna as our case study, helped us begin to understand the political and electoral dynamics found in territories that depend almost exclusively on mining revenues, where profits are not regulated by the state or managed by politicians. However, the region in question constitutes just a small part of an extensive and varied country in which, despite the unitary structure, we can find a variety of situations and strategies in the struggles to gain and maintain political power.
Therefore, in considering a future research agenda, it would be informative to consider other regions of the country that are structurally different from our case study. This would make it possible to test again the theoretical argument advanced here, and to identify other contextual variables that might affect or influence the behavior of political actors and, in turn, electoral trajectories. We have identified other cases of non-extractive economies in which strong openness has also occurred. What, then, are the structural and agency factors that explain this particular result?
Finally, this study focused on mayoral contests. The questions of what happens, and of what trajectories exist, in other kinds of elections, such as municipal councils, departmental assemblies, Congress and the Presidency, remain to be answered. It would also be instructive to analyze the ways in which other political variables behave, such as electoral participation and/or political alternation.
