Abstract
This article offers an empirical assessment of a number of aspects of the reduction of market income inequality as a result of government taxes and transfers, and of the distributive effect of wage-bargaining institutions and minimum wages, in thirteen developed countries over the last twenty five years. It considers five broad themes in the literature: the median voter approach, which argues that government redistribution is associated with inequality of pre-government income; the power resources approach, which emphasizes partisan political contestation and electoral participation; the institutional approach, which focuses on political institutions such as the electoral system; the labor organization approach, which argues that labor unions play a key role in government efforts to achieve a more egalitarian distribution of market income; and the economic globalization approach, which argues that integration into the global economy has undermined public social protection efforts. The article finds consistent positive relationships between direct government redistribution and four variables: the extent of pre-government inequality; the level of electoral turnout; the share of the labor force that is unionized; and the presence of proportional representation electoral systems. As to wage-bargaining institutions, the article confirms that there is a positive relationship between the degree of coordination of wage bargaining and a relatively egalitarian distribution of earnings across various income groups, and (more weakly) between the level of statutory minimum wages in a country and the earnings share of the lowest-income group.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Examples include Lane Kenworthy, Jobs with Equality (New York: Oxford University Press, 2008); Torben Iversen and David Soskice, “Electoral Institutions and the Politics of Coalitions: Why Some Democracies Redistribute More Than Others,” American Political Science Review 100 (May 2006): 165–81; Lane Kenworthy and Jonas Pontusson “Rising Inequality and the Politics of Redistribution in Affluent Countries,” Perspectives on Politics 3 (September 2005): 449–71; Jonas Pontusson, Inequality and Prosperity: Social Europe vs. Liberal America (Ithaca, NY: Cornell University Press, 2005); Torben Iversen, Capitalism, Democracy and Welfare (New York: Cambridge University Press, 2005); David Bradley, Evelyne Huber, Stephanie Moller, François Nielsen, and John D. Stephens, “Distribution and Redistribution in Postindustrial Democracies,” World Politics 55 (January 2003): 193–228; and Markus M. L. Crepaz, “Global, Constitutional, and Partisan Determinants of Redistribution in Fifteen OECD Countries,” Comparative Politics 34 (January 2002): 169–88.
As Morillas has recently pointed out, comparing inequality before and after public sector transfers are added and taxes deducted does not capture redistribution at the level of households but rather net redistribution at the level of entire countries; see Juan Morillas, “Redistribution as an Income Mobility Process: The Identification and Measurement of Redistribution,” Luxembourg City: Luxembourg Income Study Working Paper 513, 2009. While there are certainly many advantages to a household-level analysis, there are also advantages to the conventional approach if one's goal is to evaluate national-level redistributive policies; clearly, the aim of governments seeking to redistribute income is not to generate movement of incomes up and down but rather to achieve a more egalitarian distribution of post-government income. In an effort to make clear that I am focusing on net redistribution at the level of countries, I generally use the more precise term “government inequality reduction” rather than the conventional, looser, term “government redistribution”—although I do occasionally employ the latter term, which is used in most of the literature I cite.
The LIS surveys considered are Australia, 1981, 1985, 1989, 1995, 2001, 2003; Belgium, 1992, 1997; Canada, 1981, 1987, 1991, 1994, 1997, 1998, 2000, 2004; Denmark, 1987, 1992, 1995, 2000, 2004; Finland, 1987, 1991; 1995, 2000, 2004; France, 1979, 1984, 1989, 1994; Germany, 1981, 1983, 1984, 1989, 1994, 2000; Netherlands, 1983, 1987, 1991, 1994, 1999; Norway, 1979, 1986, 1991, 1995, 2000, 2004; Sweden, 1981, 1987, 1992, 1995, 2000, 2005; Switzerland, 1982, 1992, 2000, 2002; U.K., 1979, 1986, 1991, 1994, 1995, 1999, 2004; and U.S.A., 1979, 1986, 1991, 1994, 1997, 2000, 2004. (A 1987 survey is available for Ireland, but is not included because it represents a single, early point in time.) A number of LIS surveys for other developed countries are available, but are not suitable for analysis of government inequality reduction because their starting point is income net of taxes; a complete list of “net income” data sets is available on the LIS website. In 2007, the LIS released revised (“release 2”) data sets for Wave V, which includes income surveys conducted in about the year 2000. The main change was a reclassification of a number of pension systems that straddled the boundary of public and private in an effort to better reflect the actual operation of these systems. The data used here reflect these changes.
Technically, “pre-government” income does include income from the state in the form of wages and benefits of public sector employees. The definition employed here is, however, the conventional use of the term “pre-government” in this context.
In accordance with most work using LIS micro-data, I have equivalized income by dividing household income by the square root of the number of household members, weighting households by the number of members they include. I thus compare income at the level of individuals, but in a way that accounts for the structure of the households in which they live. Households that report zero pre-government income (i.e., all of their income is derived from the state) are included but the small number of households that report zero post-government income are not, on the assumption that they must receive at least some income from unreported sources. I follow the standard LIS conventions by top-coding income at ten times the median of non-equivalized income and bottom-coding at 1 percent of equivalized mean income.
Kenworthy and Pontusson, “Rising Inequality and the Politics of Redistribution in Affluent Countries,” 467. As it happens, my absolute and relative Gini reduction measures are strongly positively correlated across our seventy one cases: r=+0.95.
An effort to offer a rough estimate of the redistributive effect of indirect taxes is Pablo Beramendi and David Rueda, “Social Democracy Constrained: Indirect Taxation in Industrialized Democracies,” British Journal of Political Science 37 (October 2007): 619–41.
Brian Burgoon, “Globalization and Welfare Compensation: Disentangling the Ties that Bind,” International Organization 55 (Summer 2001): 509–51; Karl Ove Moene and Michael Wallerstein “Earnings Inequality and Welfare Spending: A Disaggregated Analysis,” World Politics 55 (July 2003): 485–516.
Lars Osberg, Timothy M. Smeeding, and Jonathan Schwabisch “Income Distribution and Social Expenditures: A Cross-National Perspective,” Luxembourg City: Luxembourg Income Study Working Papers, Number 350, 2004.
I have also examined other percentile ratios, including 95/25, 90/50, 75/50, and75/25, with results broadly similar to those reported. Percentile ratios that include the tenth percentile, which have often been used in previous research, are not employed here. The reason is that in some countries more than 10 percent of households receive no pre-government income at all, since all of their income derives from the state. In these cases, percentile ratios are misleading, since they are based on a bottom-coded figure (and are often very large).
David Brady, “Rethinking the Sociological Measurement of Poverty,” Social Forces 81 (March 2003): 715–52. See also David Brady, Rich Democracies, Poor People: How Politics Explain Poverty (New York: Oxford University Press, 2009).
Obviously, wage bargaining directly affects only wages, not self-employment earnings. However, it is useful to also include the latter since it gives an idea of the share of the labor force that is not directly included in wage agreements but may be indirectly affected by them. Self-employment income generally constitutes between 5 and 10 percent of total earnings in the countries examined.
Exceptions include Jonas Pontusson, David Rueda, and Christopher Way, “Comparative Political Economy of Wage Distribution: The Role of Partisanship and Market Institutions,” British Journal of Political Science 32 (April 2002): 281–308, and David Rueda, “Left Government, Policy and Corporatism: Explaining the Influence of Partisanship on Inequality,” World Politics 60 (April 2008): 349–89.
The classic statement is Alan H. Meltzer and Scott F. Richard, “A Rational Theory of the Size of Government,” Journal of Political Economy 89 (October 1981): 914–27.
Branko Milanovic, “The Median Voter Hypothesis, Income Inequality, and Income Redistribution: An Empirical Test with the Required Data,” European Journal of Political Economy 16 (2000): 368–69.
Walter Korpi and Joakim Palme, “New Politics and Class Politics in the Context of Austerity and Globalization: Welfare State Regress in 18 Countries, 1975–95,” American Political Science Review 97 (August 2003): 425. See also Walter Korpi, “Power Resources and Employer-Centered Approaches in Explanations of Welfare States and Varieties of Capitalism,” World Politics 58 (January 2006): 167–206.
Major examples include Evelyne Huber and John D. Stephens, Development and Crisis of the Welfare State: Parties and Policies in Global Markets (Chicago: University of Chicago Press, 2001); and Duane Swank, Global Capital, Political Institutions and Policy Change in Developed Welfare States (New York: Cambridge University Press, 2002).
Data are from Klaus Armingeon, Philip Leimgruber, Michelle Beyeler, and Sarah Menegale, Comparative Political Data Set, 1960–2005 (Berne, Switzerland: Institute of Political Science, University of Berne, 2008), www.ipw.unibe.ch/mitarbeiter/armingeon.CPD_Set_en.asp.
Arend Lijphart. “Unequal Participation: Democracy's Unsolved Dilemma,” American Political Science Review 91 (March 1997): 4.
If two elections are equally distant from a LIS survey, I use the earlier. Figures are for votes as a share of registered persons, except in the two countries in which registration is not automatic, France and the U.S.A., where votes are expressed relative to the voting-aged population. Data are from International Institute for Democracy and Electoral Assistance [IDEA], Voter Turnout: A Global Survey (2008), www.idea.int/voter_turnout, except for the U.S.A. where they are from Michael P. McDonald, United States Election Project (2008), elections.gmu.edu/voter_turnout.htm. The denominator of the U.S. figure excludes non-citizens but has been adjusted to include those not permitted to vote because they are in prison, on probation, on parole, or are former felons, since these exclusions are a matter of choice—one that is not exercised in any other developed democracy.
Iversen and Soskice, “Electoral.” See also Torben Iversen and David Soskice, “Distribution and Redistribution: The Shadow of the Nineteenth Century,” World Politics 61 (July 2009): 438–86.
The prototypical majoritarian system employs single member districts with a first-past-the-post electoral rule, while the prototypical PR system awards parties seats commensurate with their share of the vote on multiple-candidate lists, which are usually drawn up at the regional level and sometimes include a minimum threshold. There are a few nuances. Although Germany's system includes both party list and single-member-district components, the number of seats a party receives in the Bundestag is, except in unusual circumstances, determined by the party list result, so it is coded as PR. France's system is majoritarian but its presidential elections involve two rounds. If a party does not receive a majority in the first round, there is a second round in which only the top two candidates compete. Australia's House of Representatives employs a majoritarian system in which voters express first, second, third, etc. preferences for candidates. If a candidate receives a majority of first-preference votes, he or she is elected. If not, the candidates with the smallest number of first-preference votes are eliminated, one by one, and their lower preferences reallocated to other candidates until one receives a majority and is declared elected.
Jelle Visser, ICTWSS: Database on Institutional Characteristics of Trade Unions, Wage Setting, State Intervention and Social Pacts in 34 Countries between 1960 and 2007. Amsterdam: University of Amsterdam Institute for Advanced Labour Studies, Version 2, January, 2009.
Visser, Database.
David Rueda, “Insider-Outsider Politics in Industrialized Democracies: The Challenge to Social Democratic Parties,” American Political Science Review 99 (February 2005): 61–74.
David Neumark and William L. Wascher, Minimum Wages (Cambridge: The MIT Press, 2008), 138–39, 190.
OECD StatExtracts, Minimum Relative to Average Wages of Full-time Workers, http://stats.oecd.org/Index.aspx?DatasetCode=MIN2AVE, 2009 and earlier versions. In some countries certain groups, particularly young workers, are excluded or receive a lower minimum wage.
Ramesh Mishra, Globalization and the Welfare State (Northhampton, MA: Edward Elgar, 1999).
Dani Rodrik, Has Globalization Gone Too Far? (Washington, D.C.: Institute for International Economics, 1997), 54–55.
Dennis Quinn and Carla Inclán, “The Origins of Financial Openness: A Study of Current and Capital Account Liberalization,” American Journal of Political Science 41 (July 1997): 771–814. The authors have updated twelve of the fourteen points of their scale (capital and current account openness) through 1999 and made them available to researchers; to maximize the number of years covered, values based on these twelve points are used for the entire period. Years from 2000 on are coded at their 1999 level.
Data on investment and trade are from United Nations Conference on Trade and Development, UNCTAD Handbook of Statistics On-Line (2008), www.unctad.org, Tables 2.1 and 7.3. They are divided by GDP figures from United Nations, Statistics Division National Accounts, Main Aggregates Database, 2008, mdgs.un.org Investment. Trade figures for Belgium cover Belgium and Luxembourg.
I did, however, experiment with using them individually, with similar (non-)results.
Armingeon et al., Comparative Political Data Set.
Since my LIS surveys constitute an unbalanced pool (i.e., the years of LIS surveys vary slightly and not all countries have conducted surveys in all years), I have employed OLS regression with Huber White “sandwich” robust standard errors clustered by country; studies which employ this method include Bradley et al., “Distribution and Redistribution in Postindustrial Democracies”; Kenworthy and Pontusson, “Rising Inequality and the Politics of Redistribution in Affluent Countries”; and Clem Brooks and Jeff Manza, Why Welfare States Persist: The Importance of Public Opinion in Democracies (Chicago: University of Chicago Press, 2007). Regressions were conducted using Stata 10.0. An effort has been made in this study to offer a parsimonious explanatory model, and collinearity is not serious: the highest Variance Inflation Factor is a bit over two, well below the conventional criterion of ten, or even the more restrictive criterion of four that is sometimes employed. As a robustness test, I have also conducted the reported analyses using the following alternative methods, available in Stata: xtgls (CSTS feasible GLS); xtreg (CSTS random effects GLS); xtreg with mle (CSTS random effects GLS with a maximum likelihood estimator); and xtscc (CSTS regression with Driscoll–Kraay standard errors). The results are similar to (in fact, generally stronger than) those reported here.
My earlier measure differs in that it includes all income groups, not just prime-aged workers, and measures income from private sector sources other than work, such as property income, private transfers, and private sector pensions.
Pontusson, Inequality and Prosperity, 199; Geoffrey Garrett, Partisan Politics in the Global Economy (New York: Cambridge University Press, 1998); and David Brady, Jason Beckfield, and Martin Seeleib-Kaiser, “Economic Globalization and the Welfare State in Affluent Democracies, 1975–2001,” American Sociological Review 70 (December 2005): 921–48.
Alexander Hicks, Social Democracy and Welfare Capitalism: A Century of Income Security Politics (Ithaca, NY: Cornell University Press, 1999), 179; Swank, Global Capital, Political Institutions and Policy Change in Developed Welfare States, 118; Huber and Stephens, Development and Crisis of the Welfare State, 66–71; James P. Allan and Lyle Scruggs, “Political Partisanship and Welfare State Reform in Advanced Industrial Societies,” American Journal of Political Science 48 (July 2004): 496–512.
Data are from Armingeon et al., Comparative Political Data Set.
Evelyne Huber, Charles Ragin, John D. Stephens, David Brady, and Jason Beckford, Comparative Welfare States Data Set (2004), www.lisproject.org. Where necessary, I have updated figures through 2005.
Gøsta Esping-Andersen, The Three Worlds of Welfare Capitalism (Princeton, NJ: Princeton University Press, 1990), 56.
Karl Ove Moene and Michael Wallerstein, “Inequality, Social Insurance, and Redistribution,” American Political Science Review 95 (December 2001): 859–74.
Bradley et al., “Distribution and Redistribution in Postindustrial Democracies”; Iversen, Capitalism, Democracy and Welfare, 154.
Perversely, when a low-income worker is laid off, this inequality score actually improves, since he or she is no longer part of the full-time work force.
Lijphart, “Unequal Participation: Democracy's Unsolved Dilemma,” 2.
Vincent A. Mahler, “Electoral Turnout and Income Redistribution by the State: A Cross-National Analysis of the Developed Democracies,” European Journal of Political Research 47 (March 2008): 161–83.
Iversen and Soskice, “Electoral.”
Iversen and Soskice, “Electoral,” 171.
Iversen and Soskice, “Electoral,” 171.
Timothy M. Smeeding, “Public Policy, Economic Inequality and Poverty: The United States in Comparative Perspective,” Social Science Quarterly 86 (December 2005): 955–83.
A very interesting discussion of this approach is Michael Shalev, “Limits and Alternatives to Multiple Regression in Comparative Research,” in Capitalisms Compared, Comparative Social Research, Vol. 24, ed. Lars Mjøset and Tommy H. Clausen, (Amsterdam: Elsevier, 2007), 261–308. See also critiques of Shalev in the same volume, particularly Lyle Scruggs, “What's Multiple Regression Got to Do with It?” 309–24, and Jonas Pontusson, “Methods in Comparative Political Analysis,” 325–34.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Mahler, V. Government Inequality Reduction in Comparative Perspective: A Cross-National Study of the Developed World. Polity 42, 511–541 (2010). https://doi.org/10.1057/pol.2010.14
Published:
Issue Date:
DOI: https://doi.org/10.1057/pol.2010.14