The impact of local and national economic conditions on legislative election results

Abstract

Using data for 278 Portuguese mainland municipalities, we estimate the impact of national and local economic conditions on legislative electoral outcomes over the period from the reestablishment of democracy in 1974 to the present. Empirical results indicate that the performance of the national economy is important but that the municipal situation also conditions electoral outcomes.

Acknowledgements

The authors wish to thank Henry Chappell for very helpful comments and the Portuguese Foundation for Science and Technology (FCT) for funding the project ‘Interactions between economics and politics in Portugal’ under research grant POCI/EGE/58 641/2004 (partially funded by FEDER).

Notes

¹ Among others, see Holbrook (1991), Strumpf and Phillippe (1999), Eisenberg and Ketcham (2004) for US presidential elections; and Johnston and Pattie (2001) for British general elections. For French legislative elections, refer to Auberger and Dubois (2005).

² A right-wing alliance was formed by the Social Democratic Party (PSD), the Democratic and Social Center/People's Party (CDS/PP) and the Monarchic People's Party (PPM).

³ The ‘Quadros de Pessoal’ is a yearly mandatory employment survey that covers virtually all privately owned firms employing paid labour in Portugal (public servants and own employment are not included). Although the most recent year for which data are available is 2003, unfortunately, there is no data on wages for 2001.

⁴ There were 11 legislative elections, since the restoration of democracy in 1974. But, the use of the results obtained by the principal government party in the previous election implies that the first one (1976) is not considered. The last one, which occurred in 2005, is also not considered because there is no data available for 2005 on most of the municipal level variables.

⁵ The income index takes into consideration the fiscal burden, electricity consumption, automobiles sales and the number of bank agencies and of retail stores in each municipality.

⁶ Since part of the grants transferred to municipalities is formula-related, non-formula grants are more easily manipulated by opportunistic governments than total grants.

⁷ Since most Portuguese governments did not complete their terms, we have different term lengths in our sample. Thus, in order to make the changes in economic variables comparable over terms, they are expressed as average annual changes.

⁸ Municipal dummy variables are globally statistically significant, and Hausman tests indicate that a fixed effects specification is always preferable to a random effects one.

⁹Veiga and Pinho (2007) showed that Portuguese governments have in fact opportunistically increased grants to municipalities in election years.

¹⁰ The average change in the income index over the entire term is not statistically significant.