The following are Working Projects by Dr.Seim. Click on the titles to read the abstract and learn more.
Citizens vote for candidates they hope have traits conducive to performing well in office. But, as epitomized in the classic question “Does power corrupt?” holding office itself may be the cause of politicians’ undesirable behavior. Explaining whether and how holding office changes a politician’s behavior is fundamental to understanding political outcomes. It is particularly relevant to the study of politicians in developing countries, where institutions aimed at regulating elected officials are often weak. We argue that to achieve preferred outcomes (whether legal or not) an elected official must establish cooperative relationships. Given this necessity, we hypothesize that holding office increases a politician’s reciprocity. Estimating the effect of holding elected office on reciprocity, however, confronts the problem of selection bias; individuals who engage in reciprocal behavior may be more likely to run for office in the first place. This article reports on a behavioral field experiment designed to address this selection bias and estimate the treatment effect of holding office on a politician’s reciprocity. Using a regression discontinuity design, we recruit local politicians in Zambia who narrowly won or lost an election. These politicians play games drawn from behavioral economics that measure reciprocity. We find that office-holding politicians display significantly higher levels of reciprocity than do politicians who are not in office. This study is the first to identify a causal effect of holding political office on individual behavior, as well as the first to recruit standing elected officials to play behavioral games.
Corruption is costly for developing nations’ economies, but scholars and policy makers do not fully understand who bears the cost of corruption. We conduct a field experiment in Malawi to determine the impacts of political connections, socioeconomic status, and shared ethnicity on the degree of corruption a citizen encounters across two contexts — police roadblocks and electricity service offices — that vary the risk borne by corrupt officials. By disentangling wealth and political power, we show that political connections reduce exposure to corruption, while wealth only insulates one from corruption when it is seen as a proxy for political power. Coethnicty with a corrupt official increases corruption, but only under high risk, since trust in such contexts is crucial. These findings indicate that officials make strategic decisions about when to engage in corruption, which leaves certain individuals, particularly the poor and politically powerless, disproportionately bearing the costs.
The political science literature suggests that campaigning on and enacting policies with widespread and long-term benefits is often a poor political strategy in many African democracies. African politicians frequently win elections by buying votes, intimidating voters, committing electoral fraud, appealing to ethnicity, or provisioning short-term goods such as clothing or food. Yet, because of Sub-Saharan Africans’ vulnerability to adverse climatic conditions, successful implementation of long-run climate change adaptation policies is vital to the well-being of African nations in the future. In order to implement these climate policies, African politicians must be appropriately incentivized by their electorates. We ask: Is discussing climate change an effective campaign strategy for African politicians? Or, instead, do voters punish politicians for emphasizing climate change platforms at the expense of more immediate payoffs? We show that party campaign messages about climate change adaptation policies actually decrease voter support for that party. These effects persist for both long-run and short-run adaptation policies and are pronounced among both farmers and non-farmers. By conducting a large-scale experiment, providing subjects with accurate climate change policy information, and performing this research on the day before an actual African election, we provide the first empirical evidence that African politicians may not be electorally incentivized to allocate political resources to climate change adaptation.
Corruption is a significant barrier to development. Despite robust anti-corruption interventions, corruption persists. I suggest that one reason for the limited return on anti-corruption efforts is that corrupt political officials respond to such efforts by changing the form, not the level, of their corruption. Specifically, I theorize that officials respond to more information about corruption (transparency) by choosing less detectable forms of corruption. When information is publicized among citizens (bottom-up transparency), officials respond by choosing low-impact forms of corruption that are less likely to anger citizens. I test these expectations among local government officials in Malawi, a country with active anti-corruption efforts yet persistent corruption. Field research for this project was completed one month before a large corruption scandal “Cashgate’” broke in Malawi, in which many of the subjects of this research project were likely involved. Using a survey experiment, I examine how these in-office government officials anticipate responding to a typical anti-corruption intervention that changes the level and type of transparency in their environment. I find evidence of corruption substitution in the hypothesized direction: increased transparency shifts corruption to less detectable forms and increased bottom-up transparency shifts corruption to low-impact forms. This research offers a theory of corruption substitution, and an accompanying measure of corruption substitution among in-office government officials. The findings have vast implications for future anti-corruption policy, especially in Malawi, where anti-corruption reforms in response to the scandal are currently under design.
Political research heavily relies on survey responses by academics and policy-makers. Researchers and politicians solicit evaluations of governments, parties, and politicians to document and respond to political conditions. Often, survey respondents are “regular citizens,’” but many rely on “experts” to code political information cross-nationally. Regardless of the type of respondent, such surveys fundamentally rely on the assumptions that respondents are consistent and focused, and able to accurately evaluate political information while ignoring information that is irrelevant to solicited judgments. In this paper, we contribute to the growing literature testing the validity of these assumptions. In the context of a survey experiment conducted on Amazon Turk, we provide respondents with information about the political conditions in pairs of countries. After baseline comparisons, we randomly assign respondents to receive additional information about the countries’ economic and religious compositions. We find that the additional information substantially reduces variance in the evaluations of democracy and significantly affects respondents’ evaluations of quality of democracy, especially for Muslim and impoverished countries. This project represents a first step in a broader research agenda to understand the behavior of survey respondents in diverse contexts.
Foreign direct investments have been shown to facilitate growth and development, generate employment, increase technological transfer, and play a role in improving host country governance, so why are foreign investments sometimes resisted by the very communities they are supposed to benefit? Even within the same sector, why are some investors welcome while others encounter vehement protests? Long surrounded by controversy, in recent years Zambia’s mining sector has galvanized even greater attention with the influx of Chinese investments and several brutal attacks on Chinese investors. Other international investors on the other hand have largely avoided public violence.This paper takes a multi-level approach to investigate the determinants of public perception of investor behaviors in Zambia’s mining industry. Using data from original surveys of the local communities and foreign businesspeople in Lusaka and the Copperbelt, we investigate economic vs. non-economic determinants of public opinion on FDI in Zambia. To supplement large-N analyses, we conduct interviews of foreign company managers and workers to assess their understanding of their own public image. We go behind the “anti-China sentiment” and offer a nuanced explanation with structural and cultural elements. In addition to contributing to the international political economy scholarship on FDI, we add to a nascent but growing literature on China’s role in Africa’s development.
A scholar conducting research with human subjects must go to great lengths to avoid biasing her subjects’ behavior by unintentionally signaling that a particular behavior is expected of them. Such bias is often called “experimenter demand effects.” Experimenter demand effects are especially likely to bias subject behavior in studies conducted by non-native researchers in developing countries, where many citizens expect foreigners to distribute scarce resources on the basis of certain preferred behaviors or traits. Researchers in the developing world mitigate experimenter demand by translating surveys and protocols, hiring co-ethnic enumerators and crafting culturally relevant questions. However, it is possible that researchers bias subject behavior merely by their presence, physically or implicitly, in the research environment. We use a survey conducted in Zambia to illustrate the bias introduced by the presence of foreign researchers, showing that this specific type of experimenter demand is surprisingly strong. Through “as if” random assignment of foreign researchers to survey location, we find that the presence of foreign researchers significantly changed the distribution of responses to survey questions and changed behavior recorded in an unobtrusive measure. We conclude by discussing the types of research questions that are particularly vulnerable to this type of experimenter demand effects.
A core assumption of positive political theory is that politicians respond rationally to their strategic environments. Game theorists develop models to explain or predict strategic behavior, and experimentalists seek to test these models in the lab. However, game theorists frequently object to the use of undergraduates in laboratory experiments that purport to test such models. They argue that politicians may be more strategically sophisticated than college students are, and more likely to adopt equilibrium strategies. To test this assumption of many political science models, we ran identical experiments, involving a wide variety of classic economic games, on American undergraduates and on a large group of elected government officials, to see whether the latter are more strategic than the former. We find that game-theoretic predictions are no more likely to match the behavior of political elites than they are to match the behavior of typical undergraduate subjects. The findings suggest that, if politicians are more strategically sophisticated than undergraduates, they do not use that sophistication to employ equilibrium strategies in the types of games political scientists use to model institutional environments.
Scholars of American politics have widely documented incumbency advantage provided to incumbents at all levels of government. However, few have empirically studied incumbency advantage in other countries around the world. In this paper, we consider incumbency advantage in Sub-Saharan Africa. Building on the work in the American electoral politics literature, we measure incumbency advantage with two different methodological approaches, highlighting their advantages and disadvantages. First, drawing on a large original dataset of ten elections in five African countries and two levels of government, we use a Bayesian hierarchical model with limited covariates to show no evidence of incumbency advantage or disadvantage. These results provide a benchmark of no finding of incumbency effect in Africa, challenging an assumption made by many scholars. Second, using data from the 2006 and 2011 local government elections in Zambia, we employ a regression discontinuity design with electoral margin as the forcing variable. We show that incumbency only provides a 4% electoral advantage for those Zambian local government incumbents who very narrowly won their first term in office and who therefore were as-if randomly assigned to the treatment of incumbency in the 2011 second election. We conclude by discussing why incumbency advantage might not be as widespread of an issue as assumed in the American politics literature.
If you are interested in working as a Research Assistant on my research projects, please send a CV and cover letter via email. Please be sure you discuss particular research skills and past research experience that you have. If you are interested in pursuing a Ph.D. in Public Policy at UNC-Chapel Hill, please send me an email and find general information about the program and application process here. If you are interested in pursuing a M.A. in Global Studies at UNC-Chapel Hill, please send me an email and find general information about the program and application process here.