ordinal approach to welfare comparisons. Given a poverty line, a priori, this index has several appealing properties: (i) it can be computed using readily available information, (ii) it is sensitive to the percentage of the population that is below the line (the head-count ratio), (iii) it depends on the income of the average poor person, and (iv) it depends on the amount of inequality among the poor themselves. In this note, we offer an alternative interpretation and a generalization of Sen's index as an index. These are indices, usually of inequality, that are exact for social evaluation functions. Each index is thus implied by and implies at least one social evaluation function. Essential to the construction of these ethical indices is the notion of the

[This paper investigates second-best (transfers in kind) and third-best (subsidies and taxes) Pareto optima in a simple model where government lacks full information about consumer types (who is able, who is infirm). These Pareto optima rely on self-selection. We show that every third-best optimum other than the equal-income Walrasian equilibrium is Pareto-dominated by a second-best optimum and that standard "willingness-to-pay" criteria are inappropriate in this environment.]

This paper investigates second-best (transfers in kind) and third-best (subsidies and taxes) Pare to optima in a simple model were government lacks full information ab out consumer types (who is able, who is infirm). These Pareto optima rely on self-selection. The authors show that those second-best Paret o optima which are not also first-best (some do exist) can only be su pported by rationing. They also show that every third-best optimum, o ther than the equal-income Walrasian equilibrium, is Pareto-dominated by some second-best optimum. In addition, standard "willingness-to- pay" cost-benefit tests are inappropriate in this environment. Copyright 1988 by American Economic Association.

This paper shows that the discounted sum of instantaneous equivalent or compensating variations (generalized to allow for expenditure changes) is never an exact welfare indicator for a consumer whose preferences are represented by a continuous, increasing inter-temporal utility function.

"This paper considers the problem of social evaluation in a model where population size, individual lifetime utilities, lengths of life, and birth dates vary across states. In an intertemporal framework, we investigate principles for social evaluation that allow history to matter to some extent. Using an axiom called independence of the utilities of the dead, we provide a characterization of critical-level generalized utilitarian rules. As a by-product of our analysis, we show that social discounting is ruled out in an intertemporal welfarist environment. A simple population-planning example is also discussed."

This paper introduces and characterizes a new class of solutions to cooperative bargaining problems that can be rationalized by generalized Gini orderings defined on the agents' utility gains. Generalized Ginis are orderings that can be represented by quasi-concave, nondecreasing functions that are linear in rank-ordered subspaces of Euclidean space. In the case of three or more agents, the authors' characterization of (multivalued) generalized Gini bargaining solutions uses a linear invariance requirement in addition to some standard conditions. In the two-person case, the generalized Gini bargaining solutions can be characterized with a weakening of linear invariance. Copyright 1994 by The Econometric Society.