Abstract: When an outside power gains control of a piece of territory, such as in a colonial endeavor or a military occupation, it inherits the burden of managing internal conflict among the groups already residing there. Using a formal model, this paper develops a new theory of how the policies of an extractive outside force affect the occurrence of internal social conflict, and vice versa. When an outside ruler confronts a highly fractionalized society, the most profitable policy is one that maximizes internal conflict and minimizes resistance against external rule. However, the broader analysis does not lend support to the conventional wisdom of "divide and rule." All else equal, a ruler would rather govern a unified society than a divided one, as internal conflict is economically destructive.
Abstract: I provide a set of methods to estimate logistic regression coefficients when there is nonignorable missingness or measurement error in the outcome variable. Instead of requiring knowledge of the process by which the data went missing, these methods estimate bounds on the set of results that could be obtained under any assumption about the source of missingness. This approach is conservative by design, allowing analysts to discover exactly what kinds of results are—and are not—consistent with their data, given the uncertainty induced by missingness. Unlike existing approaches to nonignorable missing data, the estimators presented in this paper do not depend on untestable assumptions, nor do they require application-specific programming. I use simulations to illustrate which applied settings are most favorable to each of the three methods that I develop. I apply the methods to data from two previous studies, Oneal and Russett's (1997) analysis of the liberal peace and Lyall's (2010) analysis of counterinsurgency success, to examine the robustness of their results against potential miscoding of outcomes.
Abstract: I study the problems of diplomatic communication between states that have a common purpose, such as in a joint military effort. I analyze a formal model to characterize the conditions under which a state can credibly reveal private information to its ally. On one hand, the joint effort is more likely to be successful if the allies are fully informed about the international environment. On the other hand, each individual state would prefer to free-ride on its partners' efforts, all else equal. A state therefore may have an incentive to conceal information from its ally, if revealing such information would cause the ally to contribute less. Using a formal model, I characterize what kinds of topics states can (and cannot) credibly communicate about with their allies. No informative communication is possible when the source of uncertainty is a state's valuation of the issue at stake. In this case, a state prefers to pretend to be relatively unwilling to participate, so as to induce greater contributions by its partners. A similar result holds when one state has private knowledge of the common enemy's strength. However, when there are multiple ways to contribute, a state can credibly reveal information about its comparative advantage.
Co-authored with Mark Fey.
Abstract: How do flexibility and the other attributes of the bargaining process affect the resolution of international crises? We use Bayesian mechanism design to investigate the relationship between the wide variety of possible negotiation protocols and the equilibrium outcomes of crises. Our main conclusion is that although there is seemingly endless complexity in the possible forms that negotiations can take, there is a very simple structure to the possible outcomes that can emerge from negotiations. More precisely, we show that any outcome of a complex negotiation procedure could also be attained by a simple take-it-or-leave-it offer. In this sense, the great diversity of potential bargaining structures is an illusion. Our results cast doubt on the common supposition that a state is better off using a flexible bargaining strategy than an inflexible one. In an empirical analysis of international crises, we find that the issuance of an ultimatum has no discernible effect on the result of a crisis, in line with our theoretical predictions.
Abstract: I develop a model of electoral competition in which candidates' ability to raise money is related to their private information about the policy they will implement if elected. I use the model to analyze how politicians' fundraising decisions are affected by concerns about signaling their policy intentions, and to query whether campaign finance reform can reduce the possibility of electing well-funded candidates with policy intentions far from the median voter. I find that if centrist candidates can raise money more easily than others, they always exploit this advantage in equilibrium, using high spending as a signal of their proximity to the median voter. The reverse is not true when non-centrists have the fundraising advantage. In this case, candidates who spend highly are perceived as having extreme policy intentions, offsetting the otherwise positive electoral effect of spending. The electoral consequences of campaign finance reform are also asymmetric. When non-centrists can raise money more easily, a marginal decrease in the size of their advantage may increase the chance that a centrist is elected; however, in the reverse case, such measures have no effect on the chance of electing a non-centrist.
Abstract: We introduce a new statistical model to estimate the shape of potentially nonlinear, multivariate relationships. The method is built on basis regression, in which simple functions like polynomials are combined to approximate more complex relationships. To reduce the instability typically associated with basis regression, we use penalized regression techniques that perform automatic model selection, eliminating many terms from the final estimate. We focus on methods that satisfy the oracle property, which guarantees that terms with no true effect on the outcome are excluded from the estimated model in sufficiently large samples. Finally, we calculate standard errors and other estimates of variability via the bootstrap. In a series of simulations, we show that our method can accurately estimate nonlinear relationships, even if the exact functional form is not known in advance. We apply our method to Gartzke's (2007) data on the "capitalist peace" and find that joint democracy and trade dependence may increase the chance of conflict in some cases, contrary to the original results.
Abstract: The standard practice when estimating a treatment effect is to include all available pre-treatment variables in the propensity score, and we demonstrate that this approach is not always optimal when the goal is bias reduction. We characterize the conditions under which including an additional relevant variable in the propensity score increases the bias on the effect of interest across a variety of different implementations of the propensity score methodology. Moreover, we find that balance tests and sensitivity analysis provide limited protection against overadjustment.