Joshua Tucker: The following is a guest post from Columbia University political scientist Timothy Fyre.
President Petro Poroshenko recently announced that Ukraine will elect a new Parliament on Oct. 26. One would expect voters to place a high priority on a candidate’s ethnicity and language. Both have been enduring themes in Ukrainian politics, and the tragic military conflict in the east of the country highlights these cleavages.
In a recent paper, however, I found that while a voter’s ethnicity and language influenced a hypothetical vote choice, a candidate’s language and ethnicity were far less relevant. Russian and Ukrainian voters were not much moved by learning that a candidate was Russian or Ukrainian or was a native speaker of Russian or Ukrainian. Far more important was whether a candidate favored an economic policy orientation toward Russia or Europe.
These findings are based on a national survey (see the paper for details) conducted in late June in which I created eight fictional candidates for a seat in the Ukrainian parliament who varied along three features: 1) ethnicity as revealed by either a distinctly Russian or Ukrainian name 2) native language of Russian or Ukrainian and 3) whether they supported closer economic ties with Russia or with Europe.More specifically, interviewers asked:
Let’s say that there were elections to the Supreme Rada. A candidate with the following features took part in the race. About how willing would you be to vote for this candidate? [Ivan Egorovich Filinov/Boris Bogdanovich Tkachenko] is a 40 year-old businessman who speaks [Russian/Ukrainian] as his native language. He is promising to reduce corruption, increase spending on education, and build tighter economic ties with [Russia/Europe.]
One of the eight versions of the question was then randomly assigned to each respondent. Caveats up front. This vignette does not capture the nuances of language use, ethnicity, or policy orientation. Economic policy orientation toward Russia and Europe is freighted with deep cultural and political connotations; ethnicity is more subtle than a name; and native language does not include the possibility of being bilingual. Yet comparing how small changes in a candidate’s profile shape vote preferences can help identify the independent impact of these factors that are often highly correlated.
Despite the candidates’ distinctive ethnicities, native languages, and an ongoing conflict laden with ethnic and linguistic overtones, there is little difference in the average level support for each of these candidates. The differences in the average support for 7 of the 8 candidates are statistically indistinguishable from zero. Surprisingly, the “average” respondent does not appear to be strongly swayed by candidate language, ethnicity, or policy orientation.
These “average” levels of support mask, however, vast differences in the preferences of voters of different ethnicities and native languages. Breaking down the responses according to the language and ethnicity of the respondents reveals a far different pattern. Table 1 reports the hypothetical vote preferences of three groups of respondents: ethnic Russians whose native language is Ukrainian (23 percent of the sample), ethnic Ukrainians whose native language is Ukrainian (59 percent), and ethnic Russians whose native language is Russian (16 percent).
For example, consider Candidate 3. Filinov is an ethnic Russian who speaks Ukrainian and favors closer ties with Russia. Among native Russophone-Ukrainians, this hypothetical candidate is quite popular and receives a score of 3.90; among native Ukrainophone-Ukrainians, however, the score is just 2.28. Among the relatively smaller number of native Russophone-Russians the score is 3.29. Looking across all candidates, we find significant differences in the responses of Ukrainian speakers who are ethnic Russian and who are ethnic Ukrainian in five of the eight candidates.
Three candidates (4, 6 and 8) all of whom favored closer economic ties with Europe drew roughly equal levels of support from all three groups of respondents. That voters of different ethnicities and language backgrounds express roughly similar support for these three candidates suggests that voting in Ukraine has not yet been reduced to an ethnic or linguistic census despite the ongoing violent conflict in Eastern Ukraine.
Most interesting, among all respondents candidate policy orientation toward Russia or Europe is a powerful mover of vote choice, even more so than candidate language or ethnicity. Neither native Russian speaking nor native Ukrainian speaking respondents were much moved in their vote choice by changing the candidate’s language or ethnicity. However, as shown in Table 2, native Ukrainian speaking respondents were significantly more likely to support candidates who favored an economic policy orientation toward Europe. The differences in responses in each of these four paired comparisons that hold candidate ethnicity and language constant, but vary policy orientation are statistically significant at the .10 level.
Among Russian speakers, the magnitude of the change in support for these four candidates is similar, but in the opposite direction as Russian speakers are far less likely to support a candidate who backs closer economic ties with Europe. The importance of policy orientation is even found among the subset of respondents from the four eastern regions of Ukraine. Of course, economic policy orientation here should be broadly conceived as ties to Europe and Russia area loaded with political and cultural meaning, as well.
In sum, native Ukrainian and native Russian speakers have different preferences over their candidates in many, but not all, cases, suggesting that voters consider factors other than ethnicity and language in the ballot box. Most interesting, a candidate’s policy orientation toward Russia or Europe drives vote choice far more than about a candidate’s ethnicity or language. That policy orientation, broadly understood, matters so prominently gives some hope that the parliamentary election in the fall will not be simply an ethnic or linguistic census.