No abstract

We study an income tax enforcement problem using a principal-agent model where the government sets the tax and inspection functions. These are announced to the agents and there is no commitment problem. The penalty function for dishonest taxpayers is given exogenously and satisfies certain social norms. We prove that, for a large family of penalty functions, this policy is such that honesty implies regressiveness. This result does not depend on the fact that agents know the true probability of inspection.

This paper develops a political model to analyze the stability of income tax schedules. It is assumed that agents perceive any proposed alternative tax policy as more uncertain than the status quo. A tax policy is stable if it is a Condorcet winner. It is well known that in a model without uncertainty the existence of such a policy is very rare. We show, however, that in real cases this might not be a serious problem since small amounts of uncertainty can bring stability to the status quo. It is also shown that linear tax functions can only be stable in economies with very egalitarian income distributions and high taxation levels. © 1998 Elsevier Science S.A.

We provide an example of an outcome game form with two players for which there is an open set of utilities for both players such that, in each of the associated games, the set of Nash equilibria induces a continuum of outcome distributions.

Abstract. We study the microlocal structure of the examples of isospectral deformations of Riemannian manifolds given by D. DeTurck and C. Gordon in [DeT-Gl]. The Schwartz kernel of the intertwining operators considered by them may be written as an oscillatory integral with a singular phase function and product type amplitude. In certain instances, we identify them as belonging to the space of Fourier integral operators associated with various pairwise intersecting Lagrangians. After formulating a class of operators incorporating the most relevant features of the operators above, we establish a composition calculus for this class and show that is not necessary to introduce new Lagrangians in the composition.

a b s t r a c tWe show the generic finiteness of probability distributions induced on outcomes by the Nash equilibria in two player zero sum and common interest outcome games.

In economies with pub1ic goods, and agents with quasi-linear preferences, we give a characterization of the welfare egalitarian correspondence in terms of three axioms: Pareto optimality, symmetry, and solidarity. This last property requires that an increase in the willingness to pay for the public goods of some of the agents should not decrease the welfare 01' any of them. Journal ol Economic Literature Classitication Numbers: 063; H41.

We show the generic finiteness of the number of probability distributions on outcomes induced by Nash equilibria for two-person game forms such that either (i) one of the players has no more than two strategies or (ii) both of the players have three strategies, and (iii) for outcome game forms with three players, each with at most two strategies. Finally, we exhibit an example of a game form with three outcomes and three players for which the Nash equilibria of the associated game induce a continuum of payoffs for an open non-empty set of utility profiles.