The distribution of income,incomplete information and the rank and Pareto criteria

One income distribution dominates another according to the rank criterion if the income in each position, ordered from lowest to highest, is at least as great in the former distribution as the corresponding income in the latter, with the strict inequality holding at least once. Pareto dominance implies rank dominance, but not conversely. But rank dominance does imply Pareto dominance where agents are expected utility maximizers relative to subjective probability distributions that characterize incomplete information regarding agents' positions in income distributions. This suggests the rank criterion as a way of evaluating income distributions without resorting to interpersonal comparisons of utility.I have benefitted from presenting an earlier version of this paper in the seminar series of the Economics Department of Emory University and in the lecture series sponsored by the Economics Institute at the University of Colorado, Boulder. I am grateful to Peter Aranson and three referees for many helpful comments and suggestions.     

Budget Control in Virginia Public School Districts

This article examines the selection and use of a budget approval technique. Specifically, two research questions are addressed: (1) Does the use of a particular budget approval method by a governmental body impact the execution of the budget? (2) What characteristics of a local government are associated with its decision to use a particular budget approval method? Both research questions are addressed by analyzing data from public school systems in the Commonwealth of Virginia, specifically, whether using categorical budget approval differs in the accuracy of budgeting versus using a lump-sum approval method. Characteristics of Virginia school districts are also analyzed according to whether any of them are likely to be associated with school districts using a particular approval method. The remainder of this article is organized as follows: the next section describes the role of budgeting control in schools; subsequent sections review Virginia school district budgeting practices, develop the empirical model used to test the hypotheses, describe the sample, analyze the results of empirical tests and discuss implications of the findings.     

The electoral college and the rational vote

We have shown first, that if the electoral college was abolished the theoretically measured power of voters would increase and second, that in presidential elections the measure of voting power used does in fact have a highly significant impact on the decision as to whether or not to vote. Thus, the analysis predicts that the abolition of the electoral college would have a significant impact on voter participation. From a policy viewpoint, if we view participation in elections as desirable, this could be used as an argument in favor of direct election of the president. From a scientific viewpoint, we are able to make a strong and unambiguous prediction about the results of a (possible) future event from theoretical considerations. If the electoral college should be abolished, it will be possible to test our predictions. In addition, we have provided a further test of the rational behavior view of electoral participation and have shown that this model applies to presidential elections. Finally, we have shown that the theoretical measure of voting power does predict actual behavior.