The recent process of job polarization in Japan: evidence from 1980 to 2010

ABSTRACT

This study investigates the recent process of job polarization in Japan. We focus on three particular aspects: the relationship with business cycles, total hours rather than employment and age cohorts. We find that, regardless of whether the focus is employment or total hours, job polarization is concentrated in recessions and that job polarization has occurred mainly in younger-age cohorts.

KEYWORDS:

• Job polarization
• business cycles
• age cohorts
• the labour market in Japan

JEL CLASSIFICATION:

• E24
• E32
• J23
• J24

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Acemoglu and Autor (2011) present a survey of the job polarization literature.

² In this article, ‘total hours’ indicates total working hours per capita in each period. This is important in Japan’s labour market because hours per worker are considered more important factors in labour adjustment in Japan (e.g. Rogerson and Shimer 2011).

³ Data for the period from January 2000 to December 2010 were downloaded from http://www.stat.go.jp/english/data/roudou/index.htm. Data for the period from April 1984 to December 1999 were obtained from the Statistical Library in Tokyo.

⁴ As the behaviours of routine cognitive and routine manual occupations are similar, we group them into one class.

⁵ Owing to limitations of the data, we focus on trends in job polarization and employment, not total hours. As the Population Census is conducted every 5 years, it is difficult to use for examining business cycle properties.

⁶ The high growth rate of employment in NRC occupations until 1991 may be related to major revisions of the Labour Standards Law, which added one day to paid vacation entitlements. The statutory workweek was reduced from 48 to 40 h. See Hayashi and Prescott (2002).