Is Productivity in the Service Industries Low? An Analysis Using Firm-level Data on the Dispersion and the Dynamics of Productivity (Japanese)

In this paper, by using firm-level panel data from the Basic Survey of Japanese Business Structure and Activities over the four years from 2001-2004, I first analyze the dispersion of productivity among Japanese companies. Then I decompose the productivity growth by industry to analyze the contributions of company turnover and reallocation to industry-level productivity. The focus of this paper is to compare the structure of productivity between service industries and the manufacturing industry. According to the results of the analysis, the total factor productivity (TFP) level in service industries cannot be said to be lower than that of the manufacturing industry. In the service sector, there are numerous companies with high levels of productivity. In addition, the TFP growth rate in companies belonging to the service sector is by no means inferior to that of manufacturing companies. The majority of service companies shows a higher TFP growth rate than the median manufacturing company. Nevertheless, because productivity growth rates of large-scale companies in the service sector are low, the aggregate TFP growth rate is lowered significantly when companies are weighted by their size. Productivity dispersion between firms in service industries is larger than in the manufacturing industry. Most of the dispersion (log variance) of productivity level and growth rate among companies is intra-industry rather than inter-industry. The large dispersion of productivity between companies in the service sector is not attributable to the variety of types of businesses in the sector. Both service industries and the retail industry have a remarkably smaller "within effect" (productivity growth that keeps each continuing company's market share constant) than the manufacturing industry and the wholesale industry. Also, in the service industry "entry effect" and "redistribution effect" between companies make negative contributions to increases in productivity, and in this respect it differs from manufacturing and other industries. That is to say, in the service sector the market shares of companies with relatively low productivity are increasing. The above findings mean that there is substantial potential to raise the productivity of service industries as a whole. At the same time, the fact that the mechanism, whereby more productive companies expand their market shares and enhance the productivity of the sector as a whole is not necessarily functioning, suggests that the market selection mechanism in the service industry should be improved.