Publication
Rethinking the skills gap
November 2023, IZA World of Labor
Joint with Roland Rathelot and Thijs van Rens
Evidence suggests that productivity would be much higher and unemployment much lower if the supply of and demand for skills were better matched. As a result, the skills mismatch between workers (supply) and jobs (demand) commands the ongoing attention of policymakers in many countries. Policies intended to address the persistence of skills mismatch focus on the supply side of the issue by emphasizing worker education and training. However, the role of the demand side—specifically, employers’ rigid skill requirements—garners comparatively little policy attention.
Working Papers
How Search Technology Shape Wage Inequality and Job Allocations
May 2026
Previously circulated as “Recruitment, Wage Inequality, and Shrinking Task Differences“
\noindent Advancements in search technology—from print ads to online platforms—have reshaped labor market matching. Exploiting Craigslist’s staggered expansion in the 2000s, I show that the availability of new search technology improved matching efficiency and raised the local college wage premium by 3.6-5.5% but had no significant effect on local college employment share. I rationalize these empirical facts using a general equilibrium search model with a task-based framework. While improved matching efficiency for college workers raises the wage premium by increasing job-matching probability, it protects less‑educated workers against automation by reducing vacancy creation and raising firms’ job-filling probabilities with them. To understand how search technology tends to favor college workers, I endogenize matching efficiency and show that its level rises with the number of vacancies assigned to a worker type. This mechanism amplifies inequality when labor demand shifts toward college workers and mitigates downward wage pressure when college supply expands. Quantitatively, it helps reconcile the observed shift in college employment toward less abstract-intensive jobs despite the rise in wage premium. It also accounts for about 12% of the rise in wage premium since 1980 and informs the design of active labor market policies.
What is stopping you?
The falling employment-to-employment mobility in the UK
April 2024
Tendencies of British workers switching to new jobs dropped by one-third in the two decades since 2000. Mobility was held low after 2010 as firms were not making offers attractive enough to poach workers from other firms. Meanwhile, compositional changes, such as an aging population and structural transformation, did not contribute to the fall in worker mobility.
Why did we think wages are rigid for all those years?
September 2025
Joint with Stephan Hobler and Thijs van Rens
Although survey data readily reveal downward nominal wage rigidity, we show that reported wages are often distorted by recall and rounding errors. Our correction method brings survey-based estimates of nominal wage rigidity in line with those derived from administrative data.
Work-in-Progress
Engineering Economics
Joint with Abdullah Almansour and
Thiemo Fetzer
Saudi Arabia’s Nitaqat policy, which imposes soft minimum quotas on domestic employment in the private sector, was introduced in 2012 to promote Saudi national hiring. In subsequent years, the policy has evolved to include detailed requirements based on firm size and occupations. This approach, resembling “Engineering Economics,” reflects policymakers’ attempts to regulate the labor market mechanistically. Empirical analysis of administrative data reveals that firms often responded by downsizing or exiting the market, contrary to the policy’s intent. We develop a policy choice framework demonstrating that Nitaqat could only achieve its goal of increasing private-sector employment for locals if policymakers had superior knowledge of the production technology compared to producers.
Labor surplus area, the US Federal procurement and local labor markets.
Joint with Thiemo Fetzer
The Labor Surplus Area (LSA) program was introduced in the US to act as an automatic fiscal stabilizer for local labor markets, with the goal of directing federal procurement funding towards high-unemployment areas. This paper examines the intended effects of the LSA program and its subsequent impacts on local economies. Using Federal procurement data, our empirical results indicate the LSA program successfully directs procurement contracts to high-unemployment areas.
See-Yu Chan 