Do more equal slices shrink the cake? An empirical investigation of tax-transfer reform proposals in Italy

Abstract

A crucial issue in efficiency-equality evaluations of tax reforms resides in the possibility that the level as well as the distribution of welfare may change, where the household-specific measures of welfare capture the value of income as well as the value of leisure. A better-designed redistribution and income support system may not only foster equality but also improve the configuration of incentives and by this route contribute in its turn to efficiency. This paper presents an empirical analysis of the welfare effects for married couples of replacing the Italian tax system by three alternative hypothetical reforms: a flat tax, a negative income tax, and a work fare scheme. We employ a microeconometric model of household labour supply that represents partners' simultaneous choices, allows for constraints in the choice of hours of work, and is sufficiently flexible to capture a large variety of supply responses. These features appear to be crucial in the evaluation of reform effects. The results suggest that there is scope for improving upon the current system under both the efficiency and the equality criterion. The benefits from the reforms, however, come from unexpected directions since the largest contribution to the increase in welfare come from poor and middle class households whereas rich households appear to be much less responsive to changes in the tax rates. The simulation results reveal that a crucial role in shaping the results is played by the relatively higher behavioural responsiveness of married women living in low and average income households.

Empirical welfare analysis: when preferences matter

“…To this end, we start by presenting a utility model close to the one developed by Aaberge et al [1,2] and Decoster and Haan [8]. We then extend these approaches and detail how individual preferences can be genuinely approximated.…”

“…We here rely on the models developed by Aaberge et al [1,2] and Decoster and Haan [8] where the labour time is discretized into a finite number of intervals. We assume that the individual's well-being is representable through a utility function that depends on consumption C, leisure L, socio-demographic variables X and a random error term ij that varies independently among individuals.…”

“…Regarding the deterministic part of the utility function, we use the same specification as Aaberge et al [1,2] and Decoster and Haan [8]:…”

“…In the end, with this estimation proposed by Aaberge et al [1,2] and Decoster and Haan [8], we estimated groups' preferences for leisure. However, Fleurbaey and Maniquet recommend to identify individuals' preferences, as the criteria they propose allow us to establish individuals' rankings.…”

Measuring long-term inequality of opportunity

“…been proposed by Deaton and Muellbauer (1980) and Hammond (1991). It also forms the basis for the definition and measurement of a moneymetric measure of utility in for example King (1983) and Aaberge et al (2004). 8 It is straightforward to extent the budget constraint to account for wealth, e.g.…”