Treble correct when he states that the integral representing consumer surplus is increasing in D(, where Do stands for market radius for the firm. But he incorrect in his speculation that the CV solution under L6schian competition according/v yields an increasingly smaller consumer surplus in the aggregate. Note that the decreasing surplus with decreasing Do created by each one of the individual firms which are increasing in number under free entrv. The aggregate consumer surplus therefore not necessarily greater under spatial monopoly than it under conditions of spatial competition. In fact, it can readily be shown to be increased, not decreased, with an increasing entry under the CV model. To prove our contention consider the average consumer surplus a la W. Holahan [3] which given below by evaluating Treble's integral S in [4, p. 1328] and in turn dividing both sides of the resulting equation by Do:

This paper establishes a natural and satisfying characterization of the class of collective choice rules which are acyclic and satisfy the Arrow axioms (unrestricted domain, independence of irrelevant alternatives, and the weak Pareto principle). We show that, when the number of alternatives is larger than the number of individuals, there must exist an individual who can at least some critical number of pairwise decisions. This critical number of veto pairs depends on the number of alternatives and individuals, and, as the number of alternatives increases without limit, the fraction of all pairs which some individual can veto approaches unity. We also present a global veto theorem and an axiomatic characterization of the Pareto extension rule which utilizes acyclicity rather than quasi-transitivity. ARROW [1] SHOWED that the only collective choice rules that yield weak order social preference relations and satisfy unrestricted domain, independence of irrelevant alternatives, and the weak Pareto principle are dictatorial. Gibbard [9] demonstrated that by relaxing the rationality requirement from transitivity to quasi-transitivity (i.e., transitivity of the strict preference relation) we can evade the letter though not the spirit of the Arrow dictatorship result: oligarchy, a weaker form of dictatorship, still obtains when the other three axioms are imposed. In this paper we prove a theorem parallel to those of Arrow and Gibbard for the weaker rationality requirement of acyclicity (i.e., the absence of cycles of strict preference). Since acyclicity is a necessary and sufficient condition for the existence of a nonempty set of maximal elements in every finite feasible set, there are powerful reasons for imposing it. Moreover, as we argue in Blair and Pollak [2], it is difficult to justify any stronger rationality property such as quasi-transitivity without at the same time justifying some even stronger rationality condition which implies dictatorship. Our principal result shows that, when the number of alternatives is larger than the number of individuals, there must exist an individual who can at least some critical number of pairwise decisions. (We say that individual i has a veto over the ordered pair (y, x) if he is weakly decisive for x against y-that is, if his strict preference for x over y implies weak social preference for x over y, regardless of the preferences of other individuals.) This critical number of veto pairs depends on the number of alternatives and the number of individuals. As the number of alternatives increases without limit, the fraction of all pairs that some individual can veto approaches unity. There may be more than one individual who can veto at least the critical number of pairs; indeed, it is possible for every individual to have a veto over every ordered pair of alternatives.

We give necessary conditions for a neutral social choice function to be partially implementable by means of a strong equilibrium (i.e., implementable by cooperative agents): the veto power of the various coalitions should be maximally distributed. If moreover the social choice function is veto-anonymous, then the veto power of a coalition must be (roughly) proportional to its size: x per cent of the agents have the power to veto x per cent of the candidates. The procedure of "voting by successive veto" is an example of a neutral and (nearly) veto-anonymous social choice function which is implementable.

The paper derives and compares the local power of three different methods of testing non-nested regression models that are available in the literature. It shows that the asymptotic power of the orthodox F test against local alternatives is strictly less than that of Cox's non-nested test or the J test recently proposed by Davidson and MacKinnon [5], unless the number of non-overlapping variables of the alternative hypothesis over the null hypothesis is unity, in which case all three tests are shown to be asymptotically equivalent. The final section of the paper gives results of Monte Carlo experiments designed to check the validity of the theoretical findings of the paper and also to shed light on the small sample properties of the Cox test and the orthodox test.

[In a portfolio problem with given asset returns, the portfolio efficient set is the set of portfolios chosen by any risk averse agent. Using an approach of Peleg and Yaari [13], we characterize the portfolio efficient set and derive some of its properties. In particular, we show that it may not be convex, proving that a central result of mean variance theory, the efficiency of the market portfolio, does not generalize. Finally, a characterization of the efficiency of several observations gives a version of revealed preference theory for incomplete markets.]