Productivity Analysis Using Employer-Employee Matched Data

2023 Fiscal Year Final Research Report

• Project/Area Number: 19K01584
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 07030:Economic statistics-related
• Research Institution: Tohoku University
• Principal Investigator: SUZUKI Michio 東北大学, 経済学研究科, 准教授 (40580717)
• Project Period (FY): 2019-04-01 – 2024-03-31
• Keywords: 雇用者被用者マッチデータ / 生産性

Outline of Final Research Achievements

This study combines microdata from the Census of Manufactures, which contains information such as shipment value and the number of employees for manufacturing establishments, with microdata from the Basic Survey on Wage Structure, which contains detailed employee attributes, to construct employer-employee matched data for more detailed productivity analysis. The matching rate was relatively high between 2006 and 2014. For the production function estimation, we confirmed the non-parametric identification conditions for a finite mixture model that considers unobserved heterogeneity in production technologies across establishments and proposed a maximum likelihood estimation method. The estimation using data from the Census of Manufactures revealed significant heterogeneity in production functions among establishments within narrowly defined industries. It was noted that ignoring this heterogeneity could lead to systematic bias in the estimates of productivity growth rates.

Free Research Field

マクロ経済学

Academic Significance and Societal Importance of the Research Achievements

本研究は、工業統計調査と賃金構造基本統計調査の個票データを接合し、詳細な生産性分析を可能とする雇用者被用者マッチデータを構築した。生産関数推定では、事業所間の生産技術の異質性を考慮した有限混合モデルを用い、より正確な生産性推定を実現した。科学的意義としては、生産性分析の精度向上が挙げられ、社会的意義としては、精確な生産性分析が様々な政策設計に寄与することが期待される。