Multiple regimes and preferences for redistribution

Abstract

This paper investigates the presence of multiple regimes in the preferences of redistribution, which is a key feature of a welfare state. The size of the redistributive government depends on the demand for redistribution, that is, the willingness of individuals to tax the rich more heavily and transfer resources to the poor. Surprisingly, while several theories imply nonlinear models (e.g., Piketty (1995), Benabou (2000), Benabou and Ok (2001), Alesina and Angeletos (2005)), the existing empirical work (e.g., Alesina and La Ferrara (2005), Alesina and Giuliano (2011)) has paid little attention to nonlinearities in the preferences for redistribution. Instead, the existing empirical evidence establishes linear associations that cannot identify the above nonlinear mechanisms of the preferences for redistribution. For example, Alesina and Angeletos (2005) argue that when there is a demand for fairness the complementarity between the optimal level of taxation and the equilibrium signal-tonoise ratio in the income distribution generates multiple equilibria. In the one equilibrium, taxes are higher, individuals invest and work less, and inequality is lower and as result the society desires high redistribution. In the other equilibrium, taxes are lower, individuals invest and work more, and inequality is higher and as result the society desires low redistribution. In general, all these theories imply threshold-type mechanisms in the sense that the preferences for redistribution may follow a different process depending whether the country is above or below that threshold value. In this paper we do three things. First, we extend the existing work on preferences for redistribution by allowing for multiple regimes using a generalized threshold regression model. Second, we focus argue that social norms may have subtle impact on the preferences for redistribution when they are modeled as endogenous social interactions. By social interactions we mean that the model will allow for interdependences among individuals in the preferences, beliefs, and constraints. Third, we explore various groups based on various socioeconomic distances at the regional or country level along the lines of Conley and Topa (2002). Using individual data from two complementary surveys the General Social Survey (GSS) and the World Values Survey (WVS) we employ threshold regression model to uncover the presence of multiple regimes in the preferences of redistribution. We find strong evidence that countries are organized into two regimes according to the degree of fairness, that is, the belief that hard work is more important in order to succeed in life. Countries with dominant beliefs that luck and connection is the main determinant for succeeding in life have higher preferences for redistribution. The marginal effect of fairness for this group of countries is positive as in Alesina and La Ferrara (2005). However, for countries with dominant beliefs that hard work is the main determinant for succeeding in life we obtain different results. While this set of countries has lower preferences for redistribution, the marginal effect of fairness is negative, indicating that individuals who believe that hard work is more important in order to succeed in life demand more redistribution. This finding is consistent with a modern version of Weber's hypothesis of a Protestant work ethic, combined with a charitable attitude towards the poor (Bjrnskov, Dreher, Fischer, Schnellenbach, and Gehringe (2013)) rather than Alesina and La Ferrara (2005). ABSTRACT FROM AUTHOR]

Copyright of Proceedings: Ioannina Meeting on Applied Economics & Finance is the property of University of Ioannina, Department of Economics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)