Following New Zealand’s electoral system change from first-past-the-post to a mixed member proportional, the representation of the country’s indigenous people has improved considerably. More Maori gained seats in the parliament – outside the special constituencies reserved for the group – the Maori Party was founded and has even come to exercise governmental responsibilities by supporting the National Party’s minority government (Summersby 2009). It would be convenient to interpret this development as a typical example for a relationship that appears to be common knowledge in the field of electoral system change: Proportional electoral systems, in contrast to majoritarian formulas, are auspicious to the representation of minorities like indigenous people, national minorities, or immigrants (Lijphart 1991: 75; Phillips 1995: 14). Nevertheless, developments in a number of countries defy this understanding: For example, Canada, a country using single-member districts, implements extensive multicultural policies the French-speakers, its indigenous minorities, and citizens of immigrant origin (Banting and Kymlicka 2012). Are, therefore, proportional electoral systems truly superior to majoritarian electoral systems for the initiation of minority-supported legislation?
Earlier contributions in the field of minority representation argue that the more accurate translation of votes into seats under proportional representation increases the likelihood for minority members and minority ethnic parties to enter parliaments. These actors are most likely to promote group interests and their presence in parliament thus enhances the chances for successful policy-making in the interests of minorities (see e.g. Phillips 1995; Lijphart 2004; Chandra 2011; Ruedin 2012). On the flipside, some researchers stress the fact that the presence of minority-friendly representatives does not guarantee majority support for legislation in the interest of minorities. Legislators who are minority members or belong to ethnic parties might fail to occupy the powerful positions necessary to influence policies (Eisinger 1983; Heath et al. 2005). Further, majorities of legislators might refuse to vote for minority-supported legislation, especially if the power relation changes due to increasing influence of minorities (Blalock 1967; Yoder 1991; Lublin 2004).
In this blog contribution, I analyze the relationship between minority-supported legislation and the proportionality of the electoral system for 106 ethnic, linguistic, religious, or national minorities in 34 member countries of the European Union and the OECD. To identify minority-supported policies, I build on the World Directory of Minorities and Indigenous People (Minority Rights Group International 2013), a text-based description of important developments concerning minorities. The most basic information in these texts indicates whether a country recently adopted any new laws improving the protection of a specific minority (between the two editions of the directory in 1997 and 2013). I created a dummy variable that takes the value ‘1’ if a minority profited from the introduction of new minority rights, and ‘0’ if not. The key explanatory variable of the analyses is district magnitude – the main determinant of electoral systems for the proportional translation of votes into seats (Gallagher 1992: 494-495). The weighted average of the seats per district informs about district magnitude at the country level (Database on Political Institutions by Beck et al. 2001 as updated January 2013). Control variables include minority characteristics such as the ethnic composition of districts, being an immigrant or Roma minority, and group size as well as country-level factors such as the quality of democracy, the level of human development, the existence of formal electoral thresholds or federal power-sharing, and the strength of left-wing parties.
Before starting my analyses, I check to what degree the electoral system indeed provides a reasonable approximation of the permissiveness of the electoral system for minority-friendly actors. This is necessary because in the research design the seat share of ethnic parties and minority members are not included as control variables (since they constitute the causal mechanisms through which the proportionality of the electoral system influences minority-supported legislation). Table 1 provides correlations between the mean district magnitude, and the seat share of ethnic parties, minority members, as well as minority districts. The figures reveal that the mean district magnitude explains circa half of the variation of ethnic parties’ electoral success, and fourteen percent in the share of minority members in parliaments. Other factors explaining of the presence of legislators belonging to minorities are ethnic parties (with a determination coefficient of 0.41) and minority districts (with a determination coefficient of 0.66). This implies that, when controlling for the ethnic composition of districts, the permissiveness of the electoral system is a key determinant for ethnic parties and minority members’ presence in parliaments.
Table 1: Correlation coefficients of the mean district magnitude, the proportion of minority party seats, minority members, and legislators elected in majority minority districts, including levels of statistical significance and numbers of observations.Annotations: All figures display pairwise correlations. District magnitude refers to the average of the mean district magnitude between 2000 and 2010, but using the average from 1997 to 2013 instead the figures remain very similar. Given the challenges researchers face when trying to identify minority belonging, the data for this variable is incomplete and includes only thirty-two observations.
I calculated multi-level logistic regression models with fixed effects and standard errors clustered at the country level for the likelihood that minorities receive new rights. Figure 1 below displays a marginal effects plot and the predicted probability plot, which show how district magnitude impacts minority-supported legislation. Higher numbers of seats per districts lead to lower predicted probabilities for new minority rights. Above a district magnitude of twenty (or a logged district magnitude of three) seats per district, the effect is statistically significantly different from zero. In consequence, the predicted probability for new minority rights is circa 0.5 in single-member districts, and decreases with every additional seat. At the other end of the scale, it is only half as likely that minorities reach new rights in the country with the largest districts included in the data set, which is Israel, where the whole country constitutes a single district with 120 seats (or a logged district magnitude of 4.7). It is thus only in case of large districts that district magnitude unfolds a statistically significant effect on the chances for new minority rights – and in these cases, it is a negative effect with more proportional electoral systems meaning less minority-supported legislation.
Figure 1: Marginal effects plot and predicted probability plot for the effect of district magnitude on the probability for new minority rights.
The findings presented demonstrate that the larger the district magnitude in parliamentary elections, the lower the likelihood that legislatures adopt policy drafts in the interest of minorities. Elections taking place in single-member districts create legislatures that most successfully initiate laws serving the same groups that the electoral system tends to exclude. By contrast, the increased seat share for minority-friendly actors in proportional electoral systems does not translate into relevant legislation. Powerful positions for minority legislators as well as the willingness of the majority to support relevant policy drafts are hence more important for policy outputs in the interest of minorities than the mere presence of minority members and parties during decision-making.
Corinna Kroeber in May 2018
 To ensure that it captures the same time horizon of the dependent variables, the variable contains the average of the mean district magnitude over several elections. Moreover, I use the logged district magnitude, since increases in district magnitude have different leverage depending on the size.
Banting, K., and Kymlicka, W. (2012) ‘Multiculturalism Policy Index’. Available at: http://www.queensu.ca/mcp/ [accessed 12 December 2014].
Beck, T., Clarke, G., Groff, A., Keefer, P., and Walsh, P. (2001) ‘New tools in comparative political economy: The database of political institutions’, World Bank Economic Review, 15 (1): 165-176.
Blalock, H. M. (1967) Toward a theory of minority-group relations (New York: John Wiley & Sons).
Chandra, K. (2011) ‘What is an ethnic party?’, Party Politics, 17 (2): 151-169.
Eisinger, P. K. (1983) Black employment in city government, 1973-1980 (Washington: Joint Center of Political Studies).
Gallagher, M. (1992) ‘Comparing proportional representation electoral systems: Quotas, thresholds, paradoxes and majorities’, British Journal of Political Science, 22 (4): 469-496.
Heath, R. M., Schwindt-Bayer, L. A., and Taylor-Robinson, M. M. (2005) ‘Women on the sidelines: Women’s representation on committees in Latin American legislatures’, American Journal of Political Science, 49 (2): 420-436.
Lijphart, A. (1991) ‘Constitutional choices for new democracies’, Journal of Democracy, 2 (1): 72-84.
Lijphart, A. (2004) ‘Constitutional design for divided societies’, Journal of Democracy, 15 (2): 96-109.
Lublin, D. (2004) The Republican South: Democratization and partisan change (Princeton: Princeton University Press).
Minority Rights Group International (2013) ‘World directory on minorities and indigenous people’. Available at: http://www.minorityrights.org/directory [accessed 2 December 2014].
Phillips, A. (1995) The politics of presence (Oxford: Oxford University Press).
Ruedin, D. (2012) ‘The representation of women in national parliaments: A cross-national comparison’, European Sociological Review, 28 (1): 96-109.
Summersby, K. M. (2009) ‘Does Māori representation matter? An analysis of the relationship between Māori descriptive and substantive political representation in parliament and local government’, University of Auckland. Available at: https://researchspace.auckland.ac.nz/bitstream/handle/2292/5484/02whole.pdf?sequence=2 [accessed 3 June 2013].
Yoder, J. D. (1991) ‘Rethinking tokenism: Looking beyond numbers’, Gender and Society, 5 (2): 178-192.