Learning the Economic Vote at Local Elections: Case of Lithuania, 1995-2011 Mažvydas Jastramskis Institute of International Relations and Political Science, Vilnius University mazvydas.jastramskis@gmail.com

Problem Economic voting theory: if the economy is doing all right, the citizen votes for the government (reward effect); in an opposite situation the vote is against the government (Lewis-Beck and Mary Stegmaier, 2000). Can economic voting theory explain voting in the local elections, or is it only limited to the first-order elections? Lithuania as an interesting case for the search of economy’s influence in local elections: weak party identification and unsteady cleavages.

Structure of the paper Theoretical possibilities of economic voting in local elections generally and particularly in Lithuania. Other relevant political-contextual factors. Statistical analysis of all periods between the particular municipal elections in Lithuania since the transition to democracy.

Economic voting in local elections? Role of the local governments in regional economy: local taxes, potential influence on unemployment and income (Vermeir and Heyndels, 2006). Even if local government has not enough power to shape the macroeconomic trends, they simply can be held responsible by voters. Referendum effect: local elections could be treated as an event where voters express their satisfaction or disdain with the policies conducted by the national government (Marsh, 1998).

Lithuanian context: municipalities and economy Lithuania’s municipalities are considered quite weak regarding both the budget independence and autonomy of the tax policies (Mačiulytė and Ragauskas, 2007; Čivinskas and Tolvaišis, 2006; Davulis, 2009) Though some authors argue that “municipal council is incapable of luring the capital, entrepreneurs” (Čivinskas and Tolvaišis, 2006: 87), electorate may form a different opinion: electoral promises of parties and candidates. Hyper-accountability (Roberts, 2008): being part of national government may negatively affect the votes of parties, but less in the times of rising economy and more in the periods of economic downturn.

Hypotheses Hypothesis 1: economic situation in region (municipality) is expected to be positively associated with the change in votes (in municipal council elections) of the party that is dominant in the municipality’s governing coalition: the greater positive economic change, the greater probability of reward (and vice versa). Hypothesis 2a: in periods when economic situation in the country is worsening, it is expected to find significant differences between the parties that are dominant in municipal councils and also belong to the national government on one hand, and parties that are dominant in the municipal councils, but at the national level are in opposition on other hand: parties that are governing at the national level should lose more. Hypothesis 2b: in periods when economic situation in the country is improving, it is not expected to find any significant differences between the parties that are dominant in municipal councils and also belong to the national government on one hand, and parties that are dominant in the municipal councils, but at the national level are in opposition on other hand. Hypothesis 3: influence of regional economic trends on voting in municipal elections should become more significant over time.

Independent variables Dependent variable: changes in votes casted for the dominant party in the municipal council (mayor party) between two municipal elections. Independent variables Unemployment change between the elections. Fragmentation change between the elections: change in the number of parties with at least one seat in the municipal council. Turnout change between the elections. Being part of the national government: dummy variable.

Empirical analysis: 1995-2002 Model nr. 1. Change in votes of the dominant party in the municipality, year 1995-1997 Model nr. 2. Change in votes of the dominant party in the municipality, year 1997-2000 Model nr. 3. Change in votes of the dominant party in the municipality, year 2000-2002 B coefficient Beta coefficient B coefficient Beta Unemployment 1.850* 0.358* 1.004 0.117 1.382 0.219 Fragmentation 0.710 0.175 -2.910* -0.235* -0.637 -0.071 Voter turnout 0.198 0.134 -0.942* -0.287* -0.314 -0.084 National government - -15.289*** -0.434*** -0.501 -0.022 Adjusted R² 0.104 0.413 0.007 Statistical significance level *p<0,05, **p<0,01, ***p<0,001 N 46 55 60

Empirical analysis: 2002-2011 Model nr. 4. Change in votes of the dominant party in the municipality, year 2002-2007 Model nr. 5. Change in votes of the dominant party in the municipality, year 2007-2011 B coefficient Beta coefficient Unemployment -0.730* -0.289* -2.041** -0.400** Fragmentation -3.338*** -0.533*** -1.912* -0.268* Voter turnout 0.015 0.011 0.005 0.002 National government -2.828 -0.169 -5.611* -0.284* Adjusted R² 0.323 0.256 Statistical significance level *p<0,05, **p<0,01, ***p<0,001 N 59 60

Conclusions Results of empirical analysis reveal that Lithuanians are learning the economic vote, unemployment being more significant in more recent period than in the first several elections: more the regional unemployment increases, the bigger is punishment for the dominant party in the municipal council (the one that holds mayor post). Effect of referendum is observed: Lithuanian electorate tends to clearly punish the governmental (national) parties in the municipal elections when economy is worsening (in comparison to other parties). Analysis presented in this paper suggests that economic voting theory is not limited to the analysis of first-order elections: it can also explain the patterns of voting in the local elections.