Job Search and Labour Market Risk
How much Heterogeneity is there in Job Search? Does it Matter for Understanding the Labor Market Risk faced by Individuals in the Economy?
Searching for a better job is one important way to invest into one’s career. In any given period, many individuals move from job to job and between non-employment and employment. Understanding this constant churn process is crucial for understanding how careers evolve and for analyzing how the labour market at large responds to the shocks hitting it. This project aims to quantify how much heterogeneity there is in this process of searching and separating from work. The project aims to measure how labour market risk is distributed in the population and how significant it is: how does the risk of job loss vary across the population? How much do the economic costs associated with job loss vary? And, how does the likelihood of finding a new job and specifically of finding a new good job vary across the population?
Challenges
Heterogeneity in the labour market can be associated with socio-economic variables available in many data. At least as important is heterogeneity that are not easily observed but that nevertheless leave their traces in the career histories of individuals. To estimate such differences requires access to high-quality data covering long histories of individual careers. In this project, my collaborators and I will use test and estimate simple models of search in the labour market that capture different sources of heterogeneity in the labour market. To do so, we will use administrative data from Canada (the Canadian Employer and Employee Dynamics Database, CEEDD) and the US (the Longitudinal Employer Household Dynamics Data, LEHD).
The first model I will estimate is a dual labour market model. The primary market consists of skilled employees and firms with high value jobs. Search in this market is intense and non-employment spells are few and short. In the dual market, jobs are low value and unstable and non-employment spells are frequent and sometimes long-lasting. Using the CEEDD/LEHD data we will estimate how large the primary and the dual labour are and how labour market transition rates such as job loss rates differ across these groups. The second project, with co-applicants Rui Castro and Markus Poschke, tests different job ladder models and measures the difference in outcomes associated with the position on the job ladder. Job ladder models capture the risk that individual face not just from being unemployed for specific periods, but also from losing « good jobs » found in previous search including on-the-job search. The third project will exploit the measurement tools developed as part of the first two projects to estimate the incidence of the economic costs of the COVID downturn across the primary and dual labour market and across the job ladder.
Potential Contribution
This research will specify search models of the labour market with heterogeneity. We will then quantify the heterogeneity in these models to better understand how risk and uncertainty is distributed in the labour market. This will provide insights into what types of heterogeneity are important to allow for when modeling the performance of the labour market in aggregate. We also hope it will allow improve existing social insurance policies such as Employment Insurance.
Broader Potential Benefits
Risk in the labour market is of interest for all the social sciences. This research will directly contribute to the macro-economic literature on search markets, but also to micro-economic studies trying to understand consumption-savings behavior. The research on the dual labour market and how it performed in the COVID downturn will be helpful for designing social programs to insure against risk of job loss.
