Working papers

"Occupational polarisation and endogenous Task Biased Technical Change"   latest draft here (was my Job Market Paper in 2021)

Abstract

Since the 90s many developed countries have experienced job polarisation, which is defined as employment growth in both high-paid and low-paid occupations and a relative decline in middle-paying occupations. The most popular explanation is that recent technological change has been biased against routine tasks, which are more important in middle-paying occupations. This paper offers a new and complementary explanation that emphasises increasing skill supply and endogenous adoption of technology. I exploit the large policy-driven increase in education in the UK and argue that this supply shift has caused the adoption of routine-biased technology and thereby employment polarisation. This framework is consistent with two additional facts in the UK labor market. First, there were relatively little movements in occupational wages and the pattern is certainly not polarising. Second, over a period of rapidly increasing supply of graduates, occupational outcomes among graduates have not deteriorated much. I build an equilibrium multi-sector model of occupational labor and fit it to UK data over 1997-2015. I find that in most industries, technical change over the period was biased against routine tasks and favoured managerial and professional tasks. Allowing endogenous technological change, the shift in skills supply alone can account for over half of the actual decline in routine manual occupations, and 60% of the increase in abstract occupations.

"Female labour supply in urban China" 

Abstract

Women in urban China have much lower employment rates at older ages, compared to younger women and to other countries. Based on a wide range of descriptive correlations, I argue that there are two main reasons for this low employment rate. One is the early age at which urban women become eligible for pensions. One is financial transfers from their children. I build and calibrate a life-cycle model of female labour supply, incorporating income uncertainties and income-contingent transfers from children. The model can generate realistic patterns of employment, including a lot of bunching in the timing of labour market exit at the point of pension receipt. I use the model to simulate the effects of increasing pension age. The latest version is Chapter 4 in my PhD thesis.