Academic journal article Public Finance and Management

Income Inequality & Redistributional Spending: An Empirical Investigation of Competing Theories

Academic journal article Public Finance and Management

Income Inequality & Redistributional Spending: An Empirical Investigation of Competing Theories

Article excerpt


The link between income inequality and governmental redistribution is still subject to intense research and debate. Starting with the median-voter-hypothesis, a plethora of theoretical models have been developed during the last three decades to identify and explain possible causal relationships. The empirical evidence so far, however, has been mixed. The aim of this paper is to review the existing literature on inequality and redistribution, to explicate the theoretical causal mechanisms identified so far, and to provide a comprehensive and rigorous empirical test that overcomes some of the shortcomings of previous empirical studies. Using panel data on 23 OECD countries over the time period of 1971-2005, we not only look at different spending categories but also at the share gains of different income deciles and the median-to-mean ratio. We find robust evidence that income inequality affects redistribution. However, the precise degree of this relationship differs for different parts of the income distribution.


Over the last three decades, a large amount of litera-ture has been devoted to explaining the relationship be-tween redistribution and (income) inequality. Diverse theo-retical mechanisms relating redistribution to inequality have been proposed. The theoretical results are often de-rived from median voter models in the vein of Meltzer and Richard (1981), Romer (1975) and Roberts (1977) or one of their extensions. The general conclusion is that redistribution increases with a rising income gap between the me-dian and the mean voter, and so does redistributive gov-ernment taxation. However, not all models predict that re-distribution runs from rich to poor as predicted by the me-dian voter approach. By contrast, focusing on the insurance motives of public transfer spending, Moene and Wallerstein (2001, 2003) predict a negative relation between income inequality and government transfers, implying that redistri-bution runs from poor to rich. Furthermore, according to the relations derived by other models, redistribution runs from the ends of the income distribution towards the mid-dle class (Stigler, 1970; Dixit and Londregan, 1998; Epple and Romano, 1996).

From an empirical point of view, there are no clear-cut results, either. While some authors find empirical evi-dence for a positive relationship between redistribution and inequality (Alesina and Rodrik, 1994; Persson and Tabelli-ni, 1994; Milanovic, 2000) others do not find evidence in support of the median voter model (Perotti, 1996; Kenwor-thy and McCall, 2008). This has led some to believe that a robust empirical relationship does not exist. Many of the existing studies, however, only look at different types of expenditures as dependent variables. Others do not make the effort to look beyond a coarse summary indicator such as the Gini coefficient, thereby failing to take note of the varying dynamics at the lower or higher end of the income ladder. Still others only use gross or even disposable in-come instead of factor income to derive inequality meas-ures. Finally, many results seem to stem from the poor data quality of the inequality measures.

Against this background, the aim of this paper is twofold: first, to survey the main causal mechanisms be-tween inequality and redistribution and to provide an over-view of the existing empirical evidence; and second, to provide a comprehensive empirical test of the main hypo-theses, employing more precise measures of redistribution using the high quality data of the Luxembourg Income Study (LIS).

Analyzing a panel dataset of 23 OECD countries over the time period 1971-2005, we not only look at differ-ent spending categories but also at the share gains of differ-ent income deciles and the median to mean ratio. We find evidence that income inequality is indeed associated with redistribution. Overall, it clearly emerges that the main driver behind redistribution is not only general inequality as measured by the Gini coefficient, but the ratio of (factor) income between the top income decile and the middle class. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.