One of the enduring tensions in a capitalist democracy is that between the foundational commitment to universal rights of citizenship and the existence of large inequalities of wealth and income. Given the tension between equal rights and unequal rewards, one of the most important questions for those who study the capitalist democracies is the extent to which states can and do reduce the degree of inequality and, in particular, why some do more than others to reduce inequalities. That is the question Vicki Birchfield addresses in Income Inequality in Capitalist Democracies.

The book builds upon the author's work with Markus Crepaz (“The Impact of Constitutional Structures and Collective and Competitive Veto Points on Income Inequality in Industrialized Democracies,” European Journal of Political Research 34 [October 1998]: 175–200, which found that the degree of distributional inequality varied inversely across 18 countries with the extent to which the country had consensual rather than majoritarian institutions. Believing that Arend Lijphart's formulation of that distinction (Democracies: Patterns of Majoritarian and Consensus Government in Twenty-One Countries, 1984, revised and updated in Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries, 1999) is “undertheorized” in terms of democratic governance, Birchfield argues that consensual institutions are more conducive to redistribution because they are more inclusive and, thus, more likely to transfer the preferences of society into policy.

The most original aspect of Birchfield's analysis is her emphasis on the role of societal values. She argues that institutions alone cannot explain the variation across the capitalist democracies in the extent of inequality and that societal values, as mediated by institutions, must be incorporated into the explanation. Drawing on Robert Lane's distinction between market justice and political justice (“Market Justice, Political Justice,” American Political Science Review 80 [June 1986]: 383–402), she focuses in particular on the extent to which a society supports or opposes policy to lessen market-generated inequalities. She uses survey data to measure those preferences.

Birchfield uses both cross-national quantitative analysis and qualitative analysis of a few critical cases to test the hypothesis. The quantitative analysis uses measures of inequality for 16 countries at several time points obtained from the World Income Inequality database of the United Nations University–World Institute for Development Economics Research. She reports a correlation of −.4 between the extent of consensus institutions and the extent of inequality and −.2 between the extent of “political justice” and extent of inequality (p. 149). Regression analysis (p. 152) reveals a significant inverse relationship between consensus institutions and inequality, although neither the measure of political justice nor the interaction term is significant. But when other variables—most notably, the extent to which parties of the Left were represented in the cabinet but also measures of trade openness, economic growth, and social spending—are introduced, the statistically significant association between consensus institutions and inequality disappears. However, the extent to which a society values political justice has a significant indirect effect in reducing inequality through its impact on social spending (pp. 155–58). The effect is not overwhelming and is exceeded by the direct effects of Left control of government and trade openness (p. 157); nevertheless, it is enough to allow her to conclude that societal values about redistribution influence the degree of inequality.

There is some reason to question these findings. For one thing, the measure of inequality overstates the degree of inequality in several of the countries included in the analysis. A. B. Atkinson (“The Luxembourg Income Study (LIS): Past, Present and Future,” Socio-EconomicReview 2 [May 2004]: 165–90) suggests that the Birchfield and Crepaz (1998) finding that consensual institutions are associated with lesser degrees of inequality is largely the result of using measures of inequality that are not comparable across the countries. Their measures exaggerate the shares of income received by the top quintiles of recipients in the United States, United Kingdom, New Zealand, Canada, and Australia; when cross-nationally comparable data are used, the relationship between institutions and inequality is not statistically significant. Although Birchfield recognizes their noncomparability (p. 135), she uses data from the same source in the book.

The problem of noncomparability extends beyond the Anglo-Saxon countries. If one compares Birchfield's measure of the average level of inequality in the 16 countries (p. 141) with an average of the Gini coefficients reported in the first five waves of the LIS (a roughly comparable period) or an average of the middecade coefficients reported in the recent study of inequality and poverty by the Organization for Economic Co-operation and Development (Growing Unequal: Income Distribution and Poverty in OECD Countries, 2008), one finds that her measure overstates the degree of inequality not only in the United States, Ireland, and Australia but also in a number of other countries—Norway, Finland, Denmark, Sweden, the Netherlands, and Switzerland—as well.

If one replicates her analysis with those data, one finds—contrary to Atkinson—a stronger inverse relationship between the existence of consensus institutions and the degree of inequality (in the range of −.6) than the one she reports. Moreover, if one disaggregates the original Lijphart and Crepaz measure of institutions (“Corporatism and Consensus Democracy in Eighteen Countries: Conceptual and Empirical Linkages,” British Journal of Political Science 21 [April 1991]: 235–46) and considers only the corporatist component, one finds an even stronger inverse relationship with the degree of inequality (in the range of −.8). One also finds a stronger inverse relationship between a measure of the extent of Left presence in government and the degree of inequality (in the range of −.7) than the one she reports. In contrast, there is no relationship at all between the extent of political justice and the degree of inequality. In short, institutions—in particular, corporatist arrangements—and the partisan composition of government may be more consequential, and societal values less consequential, than Birchfield's analysis suggests.

In the qualitative analysis, the author examines two apparent exceptions to the rule that consensus institutions and a societal preference for political justice contribute to a lower degree of inequality. Switzerland's high degree of inequality, despite its consensus institutions, is attributed to the fact that those institutions accurately translate the country's strong commitment to market justice into policy. There can be no doubt that the institutions do indeed translate that preference into policy; data in the recent OECD study indicate that in no other country does fiscal policy do less to reduce market-generated inequality. However, the LIS and OECD data suggest that, notwithstanding its strong commitment to market justice and its minimally redistributive policy, the country is no more inegalitarian than other consensus democracies, such as Austria, Belgium, Finland, and the Netherlands.

The UK also appears to be an exception because of its high degree of inequality despite a moderately strong preference for political justice. The explanation, Birchfield suggests, lies in the majoritarian institutions that allow governments to come to power that do not subscribe to the society's values with respect to redistribution. There can be no doubt that the “aberration of Thatcher” (her words) resulted in an increase in inequality; the annual data in the recent OECD study indicate that the degree of inequality increased every year between 1979, when Thatcher took office, and 1990, when she left. But the data also suggest that a high degree of inequality is not an inevitable consequence of majoritarian institutions but, rather, the result of prolonged control of government by a party that is not committed to redistribution. If one goes back to the mid-1970s, when Labour was in office, the degree of inequality was considerably lower than what it became in the 1980s and 1990s. And however much one may doubt the redistributive ambitions of Tony Blair's Labour government, those data indicate that degree of inequality in the UK has decreased each year since 2000. The Swiss case suggests that institutions matter more than values, the British case which parties control the institutions.

If the data Birchfield uses to measure inequality raise questions about some of her findings, the most problematic aspect of the book is the less-than-complete elaboration of the conceptual underpinnings of the argument that the degree of inequality is influenced by societal values as mediated by institutions. If those values do indeed influence the degree of inequality and if, as Birchfield's survey data demonstrate, the capitalist democracies vary widely in the extent to which the citizens prefer market justice or political justice, the reader would expect some discussion concerning why the preferences of citizens with respect to redistribution vary across the countries. However, aside from a discussion in the qualitative chapter of the development of the preference for market justice in Switzerland and political justice in the UK, there is no discussion of the sources of societal values and possible reasons why the extent of support for or opposition to redistribution varies from country to country. Instead, the values are regarded as exogenous, and that variation is simply taken as a given. As a result, the reader is provided no explanation why, for example, citizens in France, Ireland, Italy, and Belgium are disproportionately supportive of political justice while those in Switzerland, Australia, Canada, the United States, and New Zealand are disproportionately supportive of market justice.

It may well be the case that societal values as mediated by institutions do influence the degree of inequality. But if that is the case, the reader expects some elaboration of precisely how that happens—that is, which institutions translate, accurately or otherwise, the values into redistributive policy. As noted previously, there is good reason to think that consensus institutions contribute to lesser degrees of inequality than majoritarian institutions. But that term aggregates a number of institutional attributes—a reliance on larger-than-minimum winning coalitions, a balance between executive and legislative powers that favors the latter, multiparty systems, multiple issue dimensions, proportional representation rather than first-past-the-post elections, and corporatist bargaining arrangements. Are all of those attributes equally important in translating societal values into redistributive policy, or are some—perhaps the use of proportional representation and the existence of corporatist arrangements—more important than others? And do they have a direct impact on redistributive policy or, as the link in Birchfield's path model (p. 158) between consensus institutions and the frequency of participation in government by Left parties suggests, an indirect effect—for example, by increasing the likelihood that particular groupings of parties, with particular preferences about redistribution, control government?

If those institutions do in fact transform societal preferences into policy, one would also like to know precisely how they do that—which domains of policy they influence and in what ways. One would like to know what the institutions themselves contribute to the creation, maintenance, and extension of taxation and spending policies that reduce the degree of inequality in pretax, pretransfer incomes. In particular, one would like to know to what extent they result in extensive reliance on taxes on income that have a progressive incidence, as well as in extensive spending on social assistance programs that are explicitly redistributive. In addition, one would like to know to what extent they contribute to the creation, maintenance, and extension of policies—for example, active labor market policies, extensive public employment, policies that encourage active participation in the labor force including by women, and policies that contribute to high rates of employment and low rates of unemployment—that reduce the degree of inequality in market incomes. It is certainly understandable, given her focus on the interplay of values and institutions and the voluminous literature that exists on the welfare state, that Birchfield would choose not to consider the impact of institutions on taxation, spending, and employment policy and the impact of those policies on inequality. Nevertheless, at the end of the day, the ultimate test of her argument about values and institutions depends on the extent to which societal values with respect to redistribution, as mediated by institutions, do in fact contribute to taxation, spending, and employment policies that have an impact on the degree of inequality.