American Political Science Review Vol. 100, No. 2 May 2006Electoral Institutions and the Politics of Coalitions: Why SomeDemocracies Redistribute More Than OthersTORBEN IVERSEN Harvard UniversityDAVID SOSKICE Duke University, London School of Economics and WissenschaftszentrumBerlinStandard political economy models of redistribution, notably that of Meltzer and Richard (1981),fail to account for the remarkable variance in government redistribution across democracies. Wedevelop a general model of redistribution that explains why some democratic governments aremore prone to redistribute than others. We show that the electoral system plays a key role because itshapes the nature of political parties and the composition of governing coalitions, hence redistribution.Our argument implies (1) that center-left governments dominate under PR systems, whereas center-rightgovernments dominate under majoritarian systems; and (2) that PR systems redistribute more thanmajoritarian systems. We test our argument on panel data for redistribution, government partisanship,and electoral system in advanced democracies.Why do some countries redistribute more thanothers? Most work on the politics of redis-tribution starts from the premise that demo-cratic institutions empower those who stand to benefitfrom redistribution. The basic logic is succinctly cap-tured in the Meltzer–Richard (1981) model, where thevoter with the median income is also the decisive voter.With a typical right-skewed distribution of income, themedian voter will push for redistributive spending upto the point where the benefit of such spending to themedian voter is outweighed by the efficiency costs ofdistortionary taxation.This argument implies that redistibution is muchgreater in democracies than in nondemocracies (atleast of the right-authoritarian variety), and that,among the latter, inegalitarian societies redistributemore than egalitarian ones. There is some evidence tosupport the first implication, although it is disputed (seeRoss 2005), but most of the variance in redistributionis probably within the same regime type. According todata from the Luxembourg Income Study, for example,the reduction in the poverty rate in United States asa result of taxation and transfers was 13% in 1994,whereas the comparable figure for Sweden was 82%(the poverty rate is the percentage of households below50% of the median income). To explain this variance,we have to look at political and economic differencesTorben Iversen is Professor, Department of Government, HarvardUniversity, Cambridge, MA 02138.David Soskice is Research Professor, Department of Political Sci-ence, Duke University, Durham, NC 27708.An early version of this paper was presented at the 2002 AnnualMeetings of the American Political Science Association, Sheratonand Marriott Hotels, Boston, August 29–September 2. We thankJim Alt, Klaus Armingeon, Neal Beck, David Brady, GeoffreyBrennan, Gary Cox, Thomas Cusack, Jeff Frieden, RobertGoodin, Peter Hall, Peter Katzenstein, Daniel Kselman RobertKeohane, Herbert Kitschelt, Peter Lange, Philipp Rehm, GerardRoland, Frances Rosenbluth, Fritz Scharpf, Ken Shepsle, MichaelWallerstein, anonymous reviewers, and the participants in the Work-shop on the economic consequences of democratic institutions,Department of Political Science, Duke University, April 1–2, 2005for their many helpful comments on a previous version of this paper.among democracies, but the second implication—–thatinegalitarian societies redistribute more—–turns out tobe of little help. In fact the empirical relationshipbetween inequality and redistribution is the oppositeof the predicted one (see B´enabou 1996; Moene andWallerstein 2001; Perotti 1996). Sweden not only redis-tributes more than the United States, but also is a muchmore egalitarian society. So the explanation for whysome democracies redistribute more than others wouldseem to lie more or less wholly outside the standardframework in political economy to explain democraticredistribution.One possibility is that the power of the working classand left political parties varies across countries (see,e.g., Korpi 1983, 1989; Hicks and Swank 1992; Huberand Stephens 2001). Because it is plausible that redis-tribution is a function of government policies, and suchpolicies reflect the preferences of those who govern,looking for differences in government partisanship isa promising avenue. Furthermore, if left governmentsnot only redistribute more but also reduce inequality ofearnings by, say, investing heavily in public education,partisanship may also explain why equality and redis-tribution tend to co-vary. Indeed, there is much evi-dence to the effect that government partisanship helpsexplain cross-national differences in redistribution(Boix 1998; Bradley et al. 2003; Kwon and Pontusson2003), and our finding corroborate this evidence.But i t raises another puzzle: why are some democra-cies dominated by left governments, whereas others aredominated by right governments?Although government partisanship is often assumedto reflect the level of working-class mobilization, weargue that it is in fact mainly determined by differencesin coalitional dynamics associated with particular elec-toral systems. Table 1 shows the strong empirical rela-tionship using a new dataset on parties and legislatures(see Cusack and Engelhardt 2002; Cusack and Fuchs2002). The figures are the total number of years withright and left governments in 17 advanced democraciesbetween 1945 and 1998, organized by type of electoralsystem. Mirroring a similar finding by Powell (2002),165Electoral Institutions and the Politics of Coalitions May 2006TABLE 1. Electoral System and the Number of Years With Left andRight Governments (1945–98)GovernmentPartisanshipProportion ofLeft Right Right GovernmentsElectoral Proportional 342 120 0.26system (8) (1)Majoritarian 86 256 0.75(0) (8)Note: Excludes centrist governments (see text below for details).about three fourths of governments in majoritariansystems were center-right, whereas three fourths ofgovernments under PR were center-left (excludinghere “pure” center governments). The numbers inparentheses convey a sense of the evidence at the levelof countries, classifying countries according to whetherthey have an overweight (more than 50%) of center-left or center-right governments during the 1945–98 period. We discuss the data (and the one outlier)in detail next.Our explanation for the association in