Regional patterns of employment changes of less-educated men in Japan: 1990–2007

Abstract

We investigate regional patterns in employment of less-educated men in Japan from 1990 to 2007. The employment-population ratio of junior high school graduate men (9 years of compulsory schooling) decreased from 1990 to 2007. Wage growth across regions had a unique pattern during this period: it was high in the low-wage regions in the 1990s but high in the high-wage regions in the 2000s. We use these regional variations in wage growth to identify the labor supply elasticity of less-educated men. The estimated elasticity of the employment-to-population ratio of junior high school graduate men is around 0.15. JEL classifications: J21, R11

The educational upgrading of Japanese youth, 1982–2007: Are all Japanese youth ready for structural reforms?

“…Moreover, Abe and Tamada (2010) report that the extensive margin of wage elasticity among junior-high-school-graduate males is about 0.15, which is reasonably high. To summarize the findings for male youth, the fraction of high-school graduates dropped significantly and that of junior/tech college or vocational school graduates soared almost by the same amount as found in Figure 11.…”

“…These explanations include labor-market segmentation (Abe and Ohta, 2001), worker flows (Kuroda, 2003, Ling andMiyamoto, 2012), cyclical behavior (Miyamoto, 2010), and regional patterns of low-skilled labor (Abe and Tamada, 2010). Among the various plausible causes, Genda (2007)'s influential book points out that lack of demand after the burst of the economic bubble in 1991 and the financial crisis in 1997 significantly shrank the port of entry to permanent jobs.…”

A cohort analysis of male labor supply in Japan

“…Distinguishing between regular full-time work and non-regular work is quite important, especially in light of recent concerns regarding the part-time/full-time wage gap and increasing earnings inequality in Japan (Ohtake, 2005;Fukawa and Oshio, 2007). The data of the ESS are particularly suitable for this purpose because the survey questionnaire asks whether each 5 The models used by Kuroda and Yamamoto (2008) and Abe and Tamada (2010) are based on cohorts, but they do not distinguish explicitly between regular full-time employment and non-regular employment. Genda et al (2010) conduct a cohort-based analysis of young men, although their main focus is the relationship between the cohort effects and the unemployment rate at the time of labor market entry by each cohort.…”

“…To understand the overall employment, the appropriate measure would be the E-P ratio, in which the self-employed and those working in the family businesses are included in the set of the employed. A detailed analysis of the E-P ratio of men using the recent ESS data is found in Abe and Tamada (2010). 12 The Data Appendix of Abe (2011) explains the issues concerning this treatment.…”

“…Alternatively, one could use the number of employed as the denominator. I choose population because the number of employed is potentially endogenous; for example, Abe and Tamada (2010) provide such evidence for the same data. Using a more exogenous variable (population) to deflate the number of regular and non-regular employees would allow me to gauge their movement more consistently.…”