Dissertation Defense: “Essays on the Labor Market”, Yeqing (Ariel) Gao
Speaker
Yeqing (Ariel) Gao, PhD Candidate, University of California, Santa Barbara
Biography
Ariel is a PhD candidate studying Economics at UCSB. Her research interest lies in labor economics. Her current research focuses on studying the effect of work-from-home on workers’ relocation, reservation strategy, and wage inequality. Before pursuing her PhD at UCSB, Ariel completed her Bachelor of Arts in Economics degree from UC Santa Barbara.
Abstract
The first chapter proposes a search model augmented with core-periphery location to study the impact of work-from- home (WFH) on workers’ reservation strategy and their relocation within metropolitan areas. The model reveals that only high-wage workers are willing to accept wage cuts for WFH, and the magnitude of the wage cut they are willing to accept increases with their earnings. On average, workers are willing to accept a 3.78% wage cut for working from home for half of the full workdays. Furthermore, the analysis of the steady state wage distribution suggests that high-wage workers lowering their reservation wages to work remotely and low-wage workers climbing the job ladders through job-to-job transitions jointly contribute to a compression of the wage distribution. Additionally, the model predicts that the rise of WFH would result in relocations from city centers to suburbs.
The second chapter investigates the contribution of WFH to the post-pandemic decline in wage inequality by constructing a search-and-match model with heterogeneous workers and firms. Workers' wages and their allocation of hours between working from home and in the office are determined by Nash bargaining. WFH affects workers' wages through two channels: preference and productivity, which are estimated based on workers' allocation of work time between home and office. The findings indicate that, on average, workers are 55.3% as productive when working remotely as they are in the office and experience 42.8% of the disutility associated with in-office work. Consequently, the rise of WFH accounts for 15.9% of the post-pandemic decline in wage inequality.
In the third chapter, I develop a package in Python that integrates the Augmented Inverse Probability Weighting (AIPW) method with the T-learner algorithm. Most existing packages that combine machine learning models with the AIPW method rely on the S-learner. However, the S-learner has a limitation: if the treatment effect is relatively small in predicting the outcome compared to other features, the treatment variable may be entirely discarded when using machine learning models that employ regularization for feature selection. To address this issue, I develop an AIPW package that estimates the outcome models using the T-learner approach. I then apply this package to examine the relationship between unemployment insurance and unemployment duration and explore the heterogeneity of this effect.
Event Details
Join us to hear Ariel dissertation defense. She will be presenting her dissertation titled, “Essays on the Labor Market”.
To access a copy of the dissertation, you must have an active UCSB NetID and password.
ID: 812 9535 0777
