Why do autocratic governments hold rigged elections that carry little or no credibility? How do such elections affect political stability and the survival of autocratic leaders? Although scholarly attention to autocratic elections has been steadily increasing, the existing literature does not provide sufficient theoretical leverage to explain why such elections are held nor to account for variations among them. The central theoretical contention of the proposed research project is that autocratic elections are held to shape beliefs about regime support and thereby reduce the likelihood of opposition revolt. In our theory, we identify the conditions under which autocrats are willing to call fraudulent elections and derive a number of hypotheses regarding variations in the competitiveness of elections and likelihood of post-election protest. This research is designed to test these predictions with data generated in a controlled laboratory experiment with human subjects, statistical analysis of a cross-national dataset, and in-depth analysis of regional executive elections in Russia during the 1995-2001 period.

This research will contribute to our understanding of the widespread phenomenon of unfree and unfair elections. First, the research contributes to the still nascent theoretical literature on autocratic elections by showing how such elections may affect public beliefs about the legitimacy of autocratic regimes and prolong (or shorten) their survival. Second, the research constitutes, to our knowledge, the first attempt to conduct experimental research on the informational effects of rigged elections. This research makes it possible to place further theorizing about autocratic regimes on firmer micro-foundations. Third, the research advances our understanding of democratization by showing how autocrats may insulate themselves from revolutionary pressures by instituting rigged elections instead of resorting to fully democratic rule. Fourth, the dataset on fairness of executive elections in Russia's regions will be instrumental for future scholarship on this topic.

Elections are frequently seen as a panacea for problems of political instability, bad governance, and lack of accountability in the developing world. The fact that U.S. Agency for International Development (USAID) alone spends more than $700 million a year to promote democracy and elections speaks to the positive impact policy-makers expect of elections. The research helps to identify to what extent this view is warranted. At the theoretical level, the proposed research suggests a mechanism whereby rigged elections may in fact prolong the survival of badly governed and unaccountable autocratic governments. In addition, the research better identifies the conditions under which post-election conflicts and protests are likely to occur. Learning how rigged elections affect public beliefs and post-election conflicts is important not only for political scientists but also for policy-makers concerned with the political development of non- democratic regimes.

Project Report

Many countries around the globe hold elections that are neither free nor fair by any standards. While this happens very often, neither researchers nor policy makers have a good understanding of how such elections work and how they contribute to the longevity of non-democratic governments. Many governments and internationalorganizations spend great resources in promoting elections around theworld; thus, the issue has great practical significance. Most importantly, we do not know whether unfair elections help non-democratic governments to survive or whether they present them a challenge. The goal of the research project was to collect data to facilitate the research of un-free and unfair elections. In particular, the award was used to (1) run a laboratory experiment on voting in unfair electoral systems; and, (2) conduct field-work research in the Russian Federation, a country with a history of unfree and unfair elections. In the laboratory experiment, the objective was to determine how people update their beliefs about the popularity of elected leaders if elections are not free and fair. The experiment involved 130participants from the Duke University student body and local residents of Durham, North Carolina. The data generated in the experiment indicate that people's incentives to protest against the government can be reduced by elections even if the elections are known to be un-free and unfair. The experiment provides novel evidence of how elections can trick citizens into believing that the government elected in unfair elections is popular. Another outcome of the project is an original dataset on the quality of gubernatorial elections in the Russian Federation from 1991 to 2001. Using information from local newspapers, we assembled data on 17 features of the quality of elections in 85 regions of Russia. These features include the number of candidates who attempted registration, the number of candidates that were not allowed to run, reports of candidate and voter intimidation, post-election protests, legal challenging of election outcomes, the number of complaints to the electoral commission, the number of signatures required to register a candidate. The final data set describes 167 elections comprising approximately 2,900 data points (cells). The data show remarkable variation in the quality of elections across regions of Russian federation. The data will enable researchers to study why some polities hold free and fair elections while others hold highly repressive ones. The data collected in this project are being used in the co-PI’sdoctoral dissertation, and they will be made available to the public in Spring 2012.

Agency
National Science Foundation (NSF)
Institute
Division of Social and Economic Sciences (SES)
Type
Standard Grant (Standard)
Application #
1023867
Program Officer
Brian Humes
Project Start
Project End
Budget Start
2010-08-01
Budget End
2011-12-31
Support Year
Fiscal Year
2010
Total Cost
$11,770
Indirect Cost
Name
Duke University
Department
Type
DUNS #
City
Durham
State
NC
Country
United States
Zip Code
27705