Poster #37 - Evaluating the Impact of School Work Style Reform in Japan

• In Event: Friday Poster Reception

Abstract

This study investigates the impact of Japan’s national work style reform policy introduced in 2019, which aimed to improve working conditions in public schools by reducing excessive teacher workloads. Although international surveys have highlighted Japan’s long working hours and low teacher well-being, few studies have rigorously examined the causal effects of such policy interventions. To address this gap, I use a difference-in-differences (DiD) approach to estimate the effects of the reform on key teacher outcomes.   
　　　This study aims to answer the following research questions:

1. How does work style reform affect teacher outcomes?

2. To what extent do the effects of the work style reform vary by school/ teacher characteristics such as school location, years of experience, and gender? 

This study draws on three waves (2013, 2018, and the upcoming 2024) of the OECD Teaching and Learning International Survey (TALIS), which provides nationally representative, cross-sectional data on schools and teachers. The treatment group comprises public middle school teachers in Japan. Two control groups are used: (a) Japanese private middle school teachers and (b) public middle school teachers in Korea. I select Korea because of its similarities to Japan, such as its centralized education system and lack of education policy similar to that of the country during the period, according to previous research.

A DiD model is estimated with year-fixed effects, country-fixed effects, and teacher-level covariates, including gender, years of teaching experience, school location, full-time status, and others. The key outcome variables include (1) weekly working hours, (2) job satisfaction with the profession (a composite variable), (3) job satisfaction with the school environment (a composite variable), and (4) teacher-student relations (a composite variable). While the analysis relies on repeated cross-sectional data, I include year and country-fixed effects to mitigate concerns about unobserved heterogeneity. I also use sampling weights to ensure national representativeness.

The identification strategy relies on a DiD model, comparing teacher outcomes before (2013 and 2018) and after (2024, forthcoming) the reform. While post-reform data are not yet available, this proposal presents findings from descriptive analysis, and a placebo DiD analysis to assess the parallel trend in the pre-intervention period.

Preliminary results indicate that public and private school teachers in Japan exhibit statistically similar pre-treatment trends across the three outcomes except for working hours. Although the private school sample is smaller, the trend alignment supports their use as a credible counterfactual. By contrast, while Korean and Japanese public schools offer comparable sample sizes, pre-trend consistency is weaker—especially for job satisfaction with the teaching profession—suggesting that Korean teachers may be better suited for robustness checks rather than primary comparison.

　　　This study contributes to the limited but growing body of causal research on teacher policy in Japan. By leveraging internationally comparable data and a quasi-experimental design, the study provides timely evidence of how national reforms shape teachers’ work lives. In doing so, it informs domestic and international discussions on teacher policy, work-life balance, and the sustainability of public education systems.

Author

• Kenta Nagasawa, Florida State University
  Presenting Author