Essays in applied economics : new techniques in aggregate data analysis

Citation

Udell, Michael Alan (1995) Essays in applied economics : new techniques in aggregate data analysis. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/4d5q-ej78. https://resolver.caltech.edu/CaltechETD:etd-10262007-105829

Abstract

This dissertation consists of three essays in applied economics. Common to each essay is the use of aggregate data. The first two essays address the demand for tax return preparation services and the effect of that demand on tax evasion. The third essay is an analysis of the effect of general economic conditions on congressional House elections.

In the first chapter, I analyze taxpayer choices of return preparation services. In particular, I distinguish between two types of nonpaid preparation, six types of paid third parties, and self-preparation. Among other things, I find significant differences in the factors which explain the demand for paid third parties who are and are not able to represent clients before the IRS. Among these factors are increases in IRS audit rates, the frequency of IRS penalties, and the complexity of the tax return.

The second chapter builds upon the results of the first chapter and analyzes the effect that different modes of tax return preparation have on tax evasion. Specifically, I allow for three assisted modes of return preparation and self-preparation to effect the level of tax evasion detected on returns they prepared. Chief among my findings is that while returns prepared by third parties who are able to represent their clients before the IRS are characterized by the greatest amounts of income and the most complex tax situations, after controlling for these factors, these returns exhibit less non-compliance than those prepared by the other modes of return preparation. The third chapter addresses a long standing controversy as to whether general economic conditions effect the outcome of congressional House elections. I test this general hypothesis by applying new estimates of historical gross national product and unemployment, as well as robust regression techniques useful for the identification of influential observations, to a variety of models in the political science literature that address the outcomes of congressional House elections. The use of new data strongly supports the proposition that changes in gross national product and unemployment have a significant impact on the congressional House election outcomes.

Thesis Files