2014 Theses Doctoral

Essays on the Economics of Statelessness and State Formation

Sanchez de la Sierra, Raul

This dissertation empirically examines economic exchange in the absence of the state. While economic analysis focuses traditionally on economic exchange within a state able to enforce property rights, functioning states are an anomalous environment in the development process. However, a fundamental problem is that in the absence of a functioning state, there is no data collection capacity.
In order to observe economic exchange in the absence of a state, I focus on a collapsed state in the current period: the Democratic Republic of the Congo (DRC). As a foundation for this dissertation, I conducted fieldwork in areas of the Eastern provinces of the Democratic Republic of the Congo that are removed from the central state. I managed a team of 12 surveyors, as well as a team of 20 traders, in order to observe economic exchange in such areas. The DRC is a well suited location to study the rules that govern economic exchange in the absence of the state. The central state is considered a failed state", the second weakest state in the world. Lacking a state to protect property rights and provide a judicial system, the economy has organized informally. Actors independent of the state regularly use coercion to define property rights. Contracts are enforced under the threat of social sanctions, or the threat of village armed men who administer disputes.
To analyze economic exchange without a state, I use standard economic analysis and standard econometrics but I am constrained to introduce different notions than the ones on which traditional economic analysis usually relies (Acemoglu and Wolitzky, 2011, Grossman, 1999, Hirshleifer, 1995). Instead of assuming that property rights are stable, I allow some agents to use coercion in order to dene property rights over goods and labor services. In anarchy, coercion can lead to large efficiency losses, since uncoordinated coercive actors do not internalize the distortions they generate on the incentives of those who produce (Grossman, 1999). This leads me to the question of Chapter 1: when will individuals with a comparative advantage in coercion organize violence in a particular location, and hold a "monopoly of violence"? Since a dominant view across disciplines views a "monopoly of violence" as a sufficient characterization of the state and traces back state origins to coercion and organized crime in medieval Europe (Tilly, 1985), my exercise uncovers the economic forces that lead to the formation of states - whether these monopolies of violence are recognized as states by the international community or not. However, in this exercise, I take the notion of the state as a "monopoly of violence" as given. But is this a useful view of the state? Why should any monopoly of violence, even those with purely criminal motives, act as states? Olson (1993) endows this view of the state with theoretical foundations: since criminals who hold a monopoly of violence are partial residual claimants through the taxes they collect, it is in their interest to protect economic activity, promote growth. Indeed, by doing so they can increase the tax revenues they collect. Therefore, in Chapter 2, I confront these theoretical foundations to the data. I examine in detail the behavior of village monopolies of violence held by armed men in Eastern Congo, and introduce economic analysis to predict state-like behavior: optimal taxation, public goods provision, and popular support. I go a step further to refine this view of the state, and show that the view of the state as a "monopoly of violence" is useful only if the time horizon is long and the monopoly of violence is uncontested. Once I have established the economic causes of state formation and the links between organized crime and states, I turn to the impact of the state on economic organization. In Chapter 3, I implement a field experiment that allows me to observe the impact of the penetration of the state legal system. I find that the state legal system allows for an increase in trade, and does not crowd-out pre-existing social mechanisms generating trust and trade. However, there are limits to the effectiveness of the formal system of contract enforcement. Indeed, I nd that only the groups which have captured the state benefit from the leverage provided by the legal system. This confirms the view of the state, as an organization that holds a monopoly of violence but which can be captured by a subset of the population. In what follows, I describe each chapter in more detail.
In Chapter 1, I focus on the following question: why do states arise, and when do they fail to arise? A dominant view across disciplines is that states arise when violent actors impose a "monopoly of violence" in order to extract taxes (Tilly, 1985). One key fact underlies all existing studies: statistics exist only after the state is formed. In this chapter, I provide the first econometric evidence on the determinants of state formation at their early stages. As a foundation for this chapter, I conducted fieldwork in areas of Eastern Congo removed from control by the central state, managing a team that collected village-level panel data on current armed groups. I develop a model that introduces optimal taxation theory to the decision of armed groups to form local "monopolies of violence", and argue that the returns to such decision hinge on their ability to tax the local population. A sharp, exogenous rise in the price of a bulky commodity used in the electronics industry, coltan, leads armed groups to impose a "monopoly of violence" in coltan villages. A later increase in the price of gold, easier to conceal and hence more dicult to tax, does not. Results from auxiliary tests provide additional support to the theory. The findings support the hypothesis that the expected Revenue from taxation, in particular tax base elasticity, is a determinant of state formation.
In Chapter 2, I go a step further, and challenge the theory associating monopolies of violence to state-like behavior. A dominant view across disciplines defines states as a monopoly of violence" and argues that states are a developed form of successful organized crime. But, is this a useful definition of the state? If current states are monopolies of violence and also engage in state-like behavior, this may be the result of a long historical process unrelated to the fact that they are a "monopoly of violence". In this chapter, I examine this view of the state empirically, and provide one of the first studies with econometric evidence linking criminal "monopolies of violence" before they develop into the states as most of us know them, to state-like behavior. To observe the behavior of criminal "monopolies of violence", I draw on the same data collection exercise as in Chapter 1. I apply optimal taxation theory to armed groups, and show that armed groups will tax and promote growth like states whenever their "monopoly of violence" is uncontested. Based on this data, I find that criminal "monopolies of violence" consistently provide public goods, develop popular support, and elaborate taxation schemes consistent with optimal taxation and European medieval states. In addition, using the timing of a peace agreement as an instrument for the presence of armed men in the village, I establish that the village "monopolies of violence" have a positive impact on economic activity. In 2003, a peace agreement (Sun-city agreement) led a large number of village "monopolies of violence" to relocate, in order to integrate the national government. This led to a security vacuum. I find that the vacuum caused by this peace agreement led to reductions in economic activity. I then use the timing of a large scale military operation in order to instrument for changes in the time horizon of existing village "monopolies of violence", which came under threat as a result of the military operation. I find that when their "monopoly of violence" comes under threat, armed groups turn from partial expropriation through stable taxation with full commitment, to full-scale violent expropriations. These findings support the view that the origins and workings of states can be obtained from criminal "monopolies of violence", but only if they have a stable time horizon.
In Chapter 3, I turn to the experimental component of the dissertation. Having established the causes that lead to the emergence of states - taxation and coercion - I investigate how to take advantage of the state in order to increase trade in contexts where distrust prevents trade to occur. Absent states to enforce contracts, societies may develop group-based mechanisms to generate trade when commitment problems prevail (Greif, 1993). However, groups are often fragmented. Can expanding access to the state legal system complement social mechanisms of contract enforcement sustained by groups? Or does legal intermediation by the state crowd them out? In this chapter, I provide evidence on the impact of state contracts on trade. As a foundation for this chapter, I created a home delivery business in East Congo. Traders sell a domestic good in households of different ethnic groups, and commit to deliver the good in the future. Shared ethnicity and formal contracts are equally effective at increasing trade by sustaining trust. Furthermore, contracts do not crowd-out ethnic group based mechanisms of trust production. Why do contracts increase trust? Results from a last experiment suggest that contracts are enforceable. However, they only protect claimants of ethnic groups that have captured the state administration. These findings suggest that even the state apparatus is embedded in the social structure, and expanding the reach of the state may have positive welfare effects, only for the groups that have vested control in the state institutions. Each of the three chapters is self-contained, so a reader interested in only one of these parts can focus on only one chapter. Each chapter contains an overview of the context relevant for that chapter. There are overlaps in the descriptions of the context. Chapters one and two are based on the same data collection exercise, but are the focus of different questions and different variables. Chapter 3 draws on a different field experiment.