Skip to main content
Support
Event

Fraud or Fairytales: Russia and Ukraine's Electoral Experiences

Peter Ordeshook, Professor of Political Science, California Institute of Technology; Mikhail Myagkov, Associate Professor of Political Science, University of Oregon

Date & Time

Thursday
May. 12, 2005
3:30pm – 5:30pm ET

Overview

At a recent Kennan Institute talk, Peter Ordeshook, Professor of Political Science, California Institute of Technology; and Mikhail Myagkov, Associate Professor of Political Science, University of Oregon, argued that regional and precinct level electoral data from Russia and Ukraine can be used to demonstrate that extensive electoral irregularities occurred in the 2004 presidential elections of both countries. Ordeshook explained that no single statistical test can conclusively prove the presence of irregularities. However, he contended that the evidence of several different analyses of electoral data, together with significant anecdotal evidence of ballot-stuffing and the use of administrative resources, combine to paint a convincing picture of questionable elections in Russia and Ukraine. Ordeshook hopes that statistical evidence of electoral irregularities can be used to back of the claims of election observers, who are frequently accused of having political agendas.

Myagkov explained that he and Ordeshook examined data on voter turnout and on number of votes for each candidate from national, regional, and local election commissions throughout Russia. A comparison of the data from the presidential election in 2004 with data from presidential elections in 1996 and 2000, he argued, clearly demonstrates that patterns of voter turnout were quite different in 2004. According to Myagkov, a distinct division appeared between regions with normal voter turnout, and regions with high voter turnout. Myagkov contended that increased voter turnout usually results in higher total numbers of votes for all candidates, but this was not the case in Russia's 2004 elections, where higher levels of turnout resulted in more votes for Putin and fewer for his opponents. This data, Myagkov argued, points to the conclusion that officials in some regions falsely recorded votes for Putin. He estimated that 10-14 million of Putin's total votes can be considered suspect.

Ordeshook explained that the second and third rounds of Ukraine's 2004 presidential elections provided an ideal environment to test the effectiveness of their methods of identifying electoral fraud. The second and third rounds of elections were held between the same two candidates with the same electorate, leaving level of fraud as the only independent variable. Ordeshook noted that turnout data from the second round of elections, which were widely reported to involve fraud, exhibit striking similarities to data from Russia's 2004 election. In districts where opposition candidate Viktor Yushchenko won the majority of the vote, the data fits the expected pattern of turnout, but in districts where prime minister Viktor Yanukovych won the most votes, turnout appeared artificially high. Also worth noting, Ordeshook argued, is the fact that an inordinately large percentage of votes cast during the last few hours of the election went to Yanukovych, suggesting that officials engaged in ballot-stuffing once it became clear that Yushchenko was in the lead. Yanukovych won some 1.5 to 3.1 million suspect votes, he said.

The data from the third round of elections, Ordeshook argued, is significantly different. The districts dominated by Yanukovych demonstrated a distribution of turnout much closer to normal. However, Ordeshook noted that there is evidence of abnormally high turnout in some pro-Yushchenko districts. Yushchenko may have gained some 200,000-700,000 suspect votes. This would not have been enough votes to influence the outcome of the election, Ordeshook said.

Ordeshook concluded with a warning that the statistical tests developed by himself and Myagkov cannot be used in the same way in all political systems. He argued that researchers wishing to use these methods must have a very deep knowledge of the political system and electorate that they are studying. Researchers must understand what the typical distribution of votes looks like in a given system in order to determine whether or not the distribution was skewed by fraud. As an example, Ordeshook noted that in Hamilton county, Ohio, in each of the past three presidential elections, higher levels of turnout have been correlated with a much higher number of votes for the Republican candidate and a slightly lower number for the Democratic candidate. At first glance, this appears to be evidence of suspicious practices similar to what happened in Russia and Ukraine, but Ordeshook argued that this is not the case. Hamilton county has a large number of precincts that are divided between middle- and upper-class white communities, on one hand, and poor black communities on the other. When turnout increases, it increases primarily among the white communities, who tend to vote for Republican candidates. Ordeshook emphasized that this type of knowledge about a region is vital in order to understand electoral data and accurately identify irregularities.

Tagged

Hosted By

Kennan Institute

The Kennan Institute is the premier US center for advanced research on Russia and Eurasia and the oldest and largest regional program at the Woodrow Wilson International Center for Scholars. The Kennan Institute is committed to improving American understanding of Russia, Ukraine, Central Asia, the Caucasus, and the surrounding region though research and exchange.  Read more

Thank you for your interest in this event. Please send any feedback or questions to our Events staff.