The Evolution of Regional Labour Productivities in Japanese Manufacturing, 1968–2004

Abstract

Essletzbichler J. and Kadokawa K. The evolution of regional labour productivities in Japanese manufacturing, 1968–2004. Regional Studies. The success of the Japanese economy in the post-war period was predicated on the rapid productivity catch-up of its manufacturing sector to Western economies, although productivity varied widely among Japanese regions. This paper examines the evolution of labour productivity in Japanese prefectures over the period 1968–2004. Employing Markov chain analysis, variance shift–share analysis, and analysis of variance, the paper demonstrates that during the period of fast productivity growth, the Japanese economy was characterized by an increasing polarization of productivity levels, while the period of slow growth since the early 1990s produced convergence in regional productivity levels.

Essletzbichler J. et Kadokawa K. L’évolution de la productivité régionale du travail dans l'industrie japonaise de 1968 à 2004. Regional Studies. La réussite de l’économie japonaise pendant la période d'après-guerre repose sur le regain rapide de la productivité de son secteur industriel par rapport à celle des économies occidentales, bien que la productivité ait varié sensiblement au Japon à l’échelle interrégionale. Cet article cherche à examiner l’évolution de la productivité du travail dans les préfectures japonaises entre 1968 et 2004. A partir d'une analyse par chaîne de Markov, d'une analyse répartition-redistribution de variance et d'une analyse de la variance, cet article montre que l’économie japonaise se caractérise par une divergence accrue des niveaux de productivité pendant la période de croissance forte de la productivité, tandis que la période de croissance faible depuis le début des années 1990 fait preuve d'une convergence des niveaux de productivité régionaux.

Japon Industrie Productivité Convergence régionale Analyse par chaîne de Markov Analyse répartition-redistribution de variance

Essletzbichler J. und Kadokawa K. Die Entwicklung der regionalen Produktivität von Arbeitskräften im japanischen Produktionswesen, 1968–2004. Regional Studies. Der Erfolg der japanischen Wirtschaft in der Nachkriegszeit gründete darauf, dass das Land hinsichtlich der Produktivität seines produzierenden Sektors gegenüber den westlichen Wirtschaften rasch aufholte. Allerdings fiel die Produktivität in den verschiedenen japanischen Regionen stark unterschiedlich aus. In diesem Beitrag untersuchen wir die Entwicklung der Produktivität von Arbeitskräften in den japanischen Präfekturen im Zeitraum von 1968 bis 2004. Anhand einer Markow-Ketten-Analyse, Varianz-Shift–Share-Analyse und Varianz-Analyse weisen wir nach, dass sich die japanische Wirtschaft in der Periode des raschen Produktivitätswachstums durch eine zunehmende Polarisierung der Produktivitätsniveaus auszeichnete, während die Periode des langsamen Wachstums seit Beginn der neunziger Jahre zu einer Konvergenz der regionalen Produktivitätsniveaus führte.

Japan Produktion Produktivität Regionale Konvergenz Markow-Ketten-Analyse Varianz-Shift–Share-Analyse

Essletzbichler J. y Kadokawa K. La evolución de las productividades laborales a nivel regional en el sector manufacturero de Japón, 1968–2004. Regional Studies. El éxito de la economía japonesa en el periodo de posguerra se basó en el rápido aumento de productividad de su sector de manufactura hacia los niveles de las economías occidentales, aunque la productividad variaba mucho entre las diferentes regiones japonesas. En este artículo examinamos la evolución de la productividad laboral en las prefecturas japonesas durante el periodo 1968–2004. Con ayuda de una análisis de cadenas de Markov, análisis de varianza shift–share y análisis de varianza, en este ensayo demostramos que durante el periodo de rápido crecimiento de la productividad, la economía japonesa se caracterizó por un aumento de la polarización de los niveles de productividad, mientras que el periodo de crecimiento lento desde principios de los noventa produjo una convergencia en los niveles de productividad regionales.

Japón Industria manufacturera Productividad Convergencia regional Análisis de cadenas de Markov Análisis de varianza shift–share

Keywords:

• Japan
• Manufacturing
• Productivity
• Regional convergence
• Markov chain analysis
• Variance shift–share analysis

JEL classifications:

• R1
• R11
• R12

Acknowledgements

The authors are grateful to two anonymous referees and the Editors for valuable comments on earlier drafts of this paper.

Notes

It would have been preferable to use hours worked as the denominator, but hours are only available at the national level and since 1979.

All reported values are in constant 2000 yen.

There are forty-seven Japanese prefectures, but Okinawa has been omitted from the analysis because data were not available for this prefecture before 1979 and the share of manufacturing employment and output of Okinawa are negligible.

In order to check the robustness of the results, a reviewer recommended a finer grid with lower endpoints of 0.0, 0.5, 0.6, …, and 1.3. The results proved robust with respect to grid choice and the bipolar distribution was confirmed for this analysis.

LR statistics were also calculated for individual five-year transition periods, but the authors were unable to reject the null hypothesis of no difference for any of the sub-periods.

Significant reclassifications of Japanese manufacturing industries occurred in 1985 and 2002. The beverage industry split from food processing and the plastic industry split from miscellaneous in 1985. In order to assure consistency over time, beverages were merged with food processing and plastics with miscellaneous after 1985. In 2002, newspapers were removed from publication and printing industries and electrical equipment and machinery was split into electrical machinery, equipment and supplies, electric parts and devices, and information and communication equipment. Because newspapers moved into the service sector, the authors were unable to merge it back into publication and printing. The authors are grateful to one of the reviewers for pointing this out. Electrical machinery, equipment and parts were merged back into electrical equipment and machinery. In order to check whether the increase in the coefficients of variation after 2001 resulted from these reclassifications, the coefficients of variation were calculated for total manufacturing minus the printing and publishing sector. The trend in coefficients of variation is identical and the increase in the coefficient of variation after 2001 remains.