Esben Scriver Andersen defends his PhD thesis

Candidate

Esben Scriver Andersen

Title: "Essays on Structural Microeconometrics. Perturbed utility and equilibrium models".

Time and place: 11 May 2022 at 14:00 in CSS 26.2 21.

An electronic copy of the thesis can be obtained here: charlotte.jespersen@econ.ku.dk
ZOOM-link: https://ucph-ku.zoom.us/j/2757553329, code: 1234.

Assessment Committee

Professor Mette Ejrnæs, Department of Economics, University of Copenhagen, Denmark (chairman)
Professor Arnaud Dupuy, University of Luxemburg
Associate Professor Ismir Mulalic, CBS, DK

Abstract

This thesis consists of three independent chapters. Each chapter analyzes the equilibrium outcomes of the labor and housing market by describing the behavior of the participants in the markets by the perturbed utility model. This class of models allows for a more general substitution patterns as opposed to the mostly used model class in the literature, the additive random utility model. Substitution patterns is important, as wages and housing prices not only reflect the productivity of labor and the quality of the properties, but also reflect the availability of close substitutes.

The first chapter Multidimensional matching and labor market complementarity is written in cooperation with Young Jun Lee. We propose an empirical framework for matching in a frictionless labor market, where workers and firms negotiate over wages and other contractual terms (e.g. whether to work part time or full time). The behavior of workers and firms is described by a class of static perturbed utility models in which alternatives may be complements or substitutes. We show that a unique equilibrium exists and how the equilibrium of the model is obtained. As a proof of concept, we estimate the model based on aggregated data for the Danish labor market and find that workers with different educational levels can be complements in the production, which the additive random utility model rules out.

The second chapter A perturbed spatial equilibrium model proposes an empirical spatial equilibrium framework for the housing and labor markets, where households can only move and commute between locations at a cost. Household behavior is described by a static perturbed utility model, which implies that any forward-looking motive is ignored. The proposed static equilibrium model allow spatial substitution patterns of housing prices and wages to be analyzed, and due to the simplicity of the model it is possible to solve it for a large choice set of locations. As a proof of concept, the model is estimated on Danish data, where the households are assumed to choose their residential and work location among 92 municipalities. Based on the estimated model, a counterfactual analysis of an increase of 1 percent in the supply of housing in the capital municipality, Copenhagen, is conducted. As expected this shows analysis that the square meter prices and wages are most affected in Copenhagen and the nearby municipalities. Due to the higher supply of housing and labor the average square meter prices and wages decrease by 10.3 and 0.5 percent in Copenhagen.

Moving is associated with large financial and none-financial costs. Hence, the location decision is inherently a dynamic decision, as the households take their future welfare into account when they making a decision. The third chapter A dynamic spatial equilibrium model for the housing and labor market shows that perturbed utility can be used to describe dynamic decision. Based on a dynamic perturbed utility model, the chapter then proposes a spatial equilibrium model similar to the model in chapter 2. As dynamic models are computationally costly to estimate, the proposed model focuses on the capital municipalities that constitutes of 23 municipalities. The counterfactual analysis shows that the average square meter prices and wages in Copenhagen decrease by 10.8 and 0.2 percent, when the supply of housing increases by 1 pct. The similar sizes of the counterfactual changes in the two chapter are partly due to the maintained assumption of static expectations with respect to square meter prices and wages in chapter 3. Initially this leads prices and wages to respond too strongly. Incorporating forward-looking expectations into the model is left for future research.