Poland Divided: Spatial Differences in the June 2003 EU Accession Referendum. Jean Monnet/Robert Schuman Paper Series, Vol. 4 No. 1, January 2004

Clem, Ralph S. and Chodakiewicz, Marek Jan. (2004) Poland Divided: Spatial Differences in the June 2003 EU Accession Referendum. Jean Monnet/Robert Schuman Paper Series, Vol. 4 No. 1, January 2004. [Working Paper]

Preview

PDF Download (270Kb) | Preview

Abstract

(From the introduction). Thus, our purpose here is to describe and analyze geographical patterns within Poland of approval, disapproval, and abstention from this crucial vote on joining the EU, and to link those outcomes to the social and economic situation obtaining in the regions. From these associations we can shed additional light on how Poland divided on this pivotal issue and posit some challenges for both Polish and EU policy makers in the years ahead. To guide us we refer to several studies of the emergence of electoral politics in former- Soviet states (notably Russia and Ukraine) and in Central and Eastern Europe that have pointed to the salience of geographical differences in voting outcomes and voter turnout in the postcommunist period. (7) Complementing research based on individual level, or survey, data, these geographic studies using aggregate data relate variations in the social, economic, and demographic traits of regions to party, candidate, and issue preferences across these same units. For example, it has been almost universally the case in the post-communist countries that rural, older, agricultural populations have voted mainly for parties of the left and against reform, while urban, better-educated, white collar areas have, for the most part, favored parties and candidates that have advocated reform and privatization. These outcomes match very closely the kinds of divisions within societies that we find in surveys. Clearly, however, we must recognize the limitations of aggregate data analysis, especially the need to avoid imputing individual action from collective figures. Surveys, of course, have advantages over one-time aggregate data, including their ability to probe attitudes and to conduct sampling over time. However, polling results have their own limitations especially that they tend to be a–spatial (and therefore cannot usually be used to illuminate important regional issues), and that respondents are not always truthful, particularly on sensitive subjects. But taken together, survey and aggregate data provide us with a higher degree of confidence in the analysis of the correlates of voting behavior. Following the methodology used in these other geographic studies of post-communist states, here we will test several propositions relating to the affinity for EU membership within different segments of Poland’s electorate. We will do this by cross-tabulating results of the June accession referendum with key social and economic variables among the 373 powiaty of Poland. According to the administrative reform of 1999, Poland is divided into 16 provinces (województwa, or voivodships), which are in turn divided into sub-regions (podregiony) and further into the powiaty. (8) The powiat scale of analysis is ideal for our purposes; powiaty are “county” level units or individual cities “with powiat status” (what we will call here “urban powiaty”) that provide an excellent degree of spatial resolution and, most importantly, for which the Polish government provides superb, detailed socioeconomic data and electoral results. (9) Powiaty typically range from 50,000 to 150,000 inhabitants and between 500-2,000 km2. Urban powiaty range from relatively small to medium-sized urban centers with populations around 100,000 people to the largest cities such as Poznań (572,000), Wrocław (624,000), Kraków (741,000) and Łódz (786,526). The Warsaw conurbation comprises the powiat of Warsaw (1.610 million). (10) As will be seen, there is a remarkable diversity within Poland—not unexpectedly in such a large and heterogeneous country—with regard both to social and economic conditions and to the demonstrated preferences of voters, with the two being related.

Actions (login required)