|
Article Excerpt
Abstract
This paper analyzes the redistributive and labor supply effects of extending at national level the Minimum Insertion Income, introduced experimentally in 1999 in some Italian municipalities. We develop a behavioral microsimulation tax-benefit model that allows for simultaneous labor supply decisions by household members, endogenous choice between dependent employment and self-employment, complete representation of the current Italian tax-benefit system. The results show a positive impact on both inequality and poverty, while overall labor supply reduces. On average magnitude of labor supply disincentive is small for married individuals, relatively larger for single persons and, within this category, for women than men. (JEL H23, I38)
Introduction
One feature of the Italian social protection scheme with respect to the ones prevailing in most EU states is the absence of a last resort benefit for individuals in their working age. (1) The Minimum Insertion Income (Reddito Minimo di Inserimento, hereinafter RMI) was introduced in 1999 on an experimental basis in 39 Italian municipalities with the purpose of filling this gap. (2)
As documented by recent reports from the European Commission [2002], most EU members have introduced reforms of their social protection systems to reduce costs and limit dependency by recipients of these systems through more active policies and better-targeted benefits. Greater emphasis on active policies is also found in the European Strategy for Employment and its recent developments.
As far as Italy is concerned, a government Commission, known as Onofri Commission [1997], pointed out that from a comparison between the Italian and other systems present within the EU two distortions stand out: The first regards the risks covered by the system, while the second focuses on the categories of beneficiaries. In Italy resources addressed to old age and survivors are higher than the EU average, while expenditure for the risks of unemployment/education, family/maternity, housing and social exclusion is lower than in the rest of Europe. Furthermore, the system is more generous towards dependent workers and pensioners, especially civil servants and workers of large enterprises, while it appears insufficient towards the other categories of workers and especially for the non-employed. This imbalance in covering risks and selecting beneficiaries makes the Italian social protection system poorly inclusive [Baldini et al., 2002].
On the basis of these considerations, the Onofri Commission [1997] proposed introducing a minimum income support scheme and suggested some specific guidelines. The main recommendation was to define eligibility in order to avoid moral hazard behaviors and to stimulate recipients not to depend on public support balancing adequately individual rights and responsibilities. The Commission proposed the Minimum Income:
1. Be addressed to people in poverty for reasons not depending upon their own choice.
2. Take into account all sources of incomes accruing to the family, as far as means testing is concerned.
3. Partially integrate the difference between family resources and the guaranteed income level, thus reducing the poverty trap effects of the scheme.
4. Stimulate recipients to actively seek work; individuals should accept a job when offered or they should be involved in social work or in vocational training courses; however, the scheme should allow the beneficiaries to take care of their family responsibilities.
5. Be administered by municipalities, thus allowing for a better integration between the program, local social assistance measures and active labor market policies.
These guidelines were followed in designing the RMI. The benefit addresses individuals who cannot support themselves and their families for psychological, physical or social reasons. It is granted (3) on a differential basis between family resources and a given threshold, both adjusted through a specific equivalence scale. In computing family resources, only 75 percent of after-tax employment income is taken into account in order to reduce unemployment/poverty trap effects. Families with income either from financial sources or real estate are not entitled to the benefit. (4) Eligibility for RMI is conditional to an activation program. Recipients in their working age who are not employed and able to work are required to be available for vocational training and work, with the exemption of individuals who are receiving therapeutic treatment, take care of children under three years of age, or of dependent disables.
In this paper, we discuss the effects that extending RMI at national level (5) would have on income distribution and on labor supply. Introducing a cash benefit can discourage labor supply or participation in the labor market depending upon the effective marginal tax rates underlying the tax-benefit system as a whole, although in the case of RMI the disincentive effect should be prevented by the activation program. The problem then becomes more complicated given that access to the benefit is means-tested against family resources and the activation program extends to all family members.
The global effect of the benefit on labor supply is largely an empirical issue. The analysis requires a model where the decisional unit is the family rather than the individuals and where non-paid work carried out within the family is taken into account. Then labor supply decisions can be analyzed as simultaneous choices of the partners. Both this aspect and the complexity of the tax-benefit system as a whole, make the application of the standard approach proposed by Hausman [1985] problematic. His proposal extends the traditional marginal analysis to the case of non-linear and non-convex budget constraints and consists in determining the optimal individual choice on each segment of the budget constraint and then in selecting, from all segments, the choice that maximizes utility. In principle, this approach can be used whenever the tax system can be approximated by piece-wise linear tax rules. Nevertheless, given the complex system of connections between the tax schedule, tax credits, income deductions, family allowances and other benefits the approach developed by Hausman [1985] cannot be applied whenever the objects of the analysis are the connections themselves. Moreover, the approach relies on the assumption that each individual is free to choose his or her optimal position on the budget set and, therefore, the number of hours he or she wishes to work given his or her hourly wage. In most cases, both wages and hours of work are defined by collective bargaining. It follows that individual choices regarding hours of work are strongly reduced. Dickens and Lundberg [1993] pointed out that non-consideration of constraints in the distribution of hours of work could lead to highly biased estimations of labor supply elasticity.
The labor supply model used in this paper builds and extends the one developed by Aaberge et al. [1998] who adopt an alternative approach, which has the advantage of being applicable independently of how complex the tax rules are. However, while the model of Aaberge et al. [1998] features labor supply decisions of married individuals and dependent workers, to offer a thorough analysis of the...
|
