![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
Party finance reformers often call for European parties to increase their financial reliance on small donors, but researchers have made few efforts to establish the feasibility of such strategies. This study examines partisan giving in Europe, investigating the potential for parties and policymakers to increase this type of political participation. It also asks whether there are national-level factors that make such efforts more likely to succeed in some countries. The research uses data from the European Social Survey to examine patterns of contributions to political parties in 16 European countries. It finds that the strength of partisanship predicts political giving independent of party membership, which suggests that European political parties may have some scope to increase their number of individual donors. On the other hand, existing tax policies to encourage political giving do not seem to be effective in increasing this type of partisan political participation.
Acknowledgements
We would like to thank Kenneth Janda, Brandon Rottinghaus and the anonymous reviewers for helpful comments on earlier versions of this paper.
Notes
1. The countries are: Austria, Belgium, Denmark, Finland, Germany, Greece, Italy, Luxembourg, Netherlands, Norway, Poland, Portugal, Slovenia, Spain, Sweden and the United Kingdom. Questions about donations to political parties were not asked in the Czech Republic and Switzerland. We could not include France, Hungary and Ireland in the model because these were missing other variables, most importantly household income, which is generally regarded as a key predictor of donations, but also marital status (France). We did, however, run the model without household income and including Hungary and Ireland. This produced only very slight changes in the individual-level results.
2. The question has not been asked in this form in subsequent rounds of the ESS.
3. Number of respondents for each category:
4. There is almost certainly some variation in the per capita level of these subsidies, but no data are available to determine this. However, given that political parties seem to be perennially short of funds even in countries with spending caps and high public subsidies, it is unlikely that parties in any of these countries were so richly funded from public sources as to be completely uninterested in attracting additional funds from private sources.
5. These measures may raise the question of how accurately surveys can assess donor behaviour. As with other types of participation, such as voting, it is likely that surveys provide an inflated estimate of actual financial participation. However, in contrast to voter turnout, there are no measures of actual donation rates against which to compare the results, so it is hard to gauge the magnitude of over-reporting. Given predominantly negative attitudes towards political parties (Dalton and Weldon Citation2005), it may be that social desirability does not affect reports of political giving as much as self-reported turnout, but we have no way of verifying this.
6. Alternatively, Gellman and Hill (2007: 246) state the advantages of employing a multilevel model as follows: ‘Accounting for individual- and group-level variation in estimating group-level regression coefficients … Modeling variation among individual-level regression coefficients. In classical regression, one can do this using indicator variables, but multilevel modeling is convenient when we want to model the variation of these coefficients across groups, make predictions for new groups, or account for group-level variation in the uncertainty for individual-level coefficients…Estimating regression coefficients for particular groups … With a multilevel model, we can get reasonable estimates even for counties with small sample sizes, which would be difficult using classical regression.’ In our analysis, we take into account these regression coefficients for particular European countries when reporting predicted probabilities.
7. Concerned with the effects that multi-colinearity might pose for the statistical significance of the country-level determinants, we ran several regressions with all possible combinations of this set of variables. In these combinations, we reduced the number of the country-level variables to verify the robustness of our results. None of these specifications found any country-level variable (excluding religious fragmentation) having an influence when explaining political giving.
8. We tested this in another way by asking whether tax policies made a difference for higher-income individuals, the group most likely to be attentive to such incentives. In a separate model we interacted tax incentives and income, but found no consistent impact of these variables across several specifications.
9. To corroborate this assertion, we ran several regressions to test whether party membership can predict donations to other types of giving (education/culture and churches). Even when employing several specifications as robustness checks, we did not find any explanatory relevance of party membership when explaining donations to other types of institutions.
10. In France, Germany and the United Kingdom all political parties with parliamentary representation solicited funds and included information on how to donate on their web pages. In Ireland and Portugal some parties did, and in Austria, the Czech Republic, Poland, Slovakia and Spain none of the party web pages included this information. This survey, while not contemporary with the individual level data analyzed above or covering precisely the same countries, nevertheless suggests that one reason that citizens do not contribute to politics may be because political parties are not doing a lot to solicit such contributions.