An examination of the sectoral shift hypothesis

Deanne M Short, Purdue University

Abstract

In this thesis we set out to analyze the sectoral shift hypothesis originally postulated by Lilien (1982). Lilien proposed that sectoral shocks, be they changes in the relative demand for goods or changes in relative productivity, can lead to higher aggregate unemployment if labor markets are decentralized, making the reallocation of labor a costly and time consuming process. The most significant critical response to his hypothesis and findings was from Abraham and Katz (1986). They claimed that the positive correlation Lilien found could have been induced by general aggregate demand shocks instead. The first of three issues addressed in this thesis is the lack of a model which details the process by which matches between workers and firms form. In Chapter 2 we propose such a model. Extending this model to a multisector model in Chapter 3, we are able to examine the effects of both a sectoral shock and an aggregate demand shock on employment, unemployment and vacancies. We therefore are able to examine the competing hypotheses of Lilien and Abraham and Katz in terms of a single, coherent theoretical framework. This is the second issue addressed. Prior analyses have been based on various theoretical models which are noncomparable in structure and definition of variables. This leads us to the concern that a shock, regardless of its source, causes an increase in the necessary reallocation of labor across sectors. The third issue addressed is therefore an inquiry into how a sectoral shift works through the market to cause fluctuations in the unemployment level. In Chapter 4 we discuss some of the existing explanations and contribute our own. We propose an extention of our model which incorporates the effects of workers switching sectors. (Abstract shortened with permission of author.)

Degree

Ph.D.

Advisors

Barron, Purdue University.

Subject Area

Labor economics

COinS