Skip to main content


The Results of Quality Adjustment of the Corporate Service Price Index in 2000 and Future Implications for Handling Service Prices

November 2001
Hiroshi Ugai

Views expressed in Working Paper Series are those of authors and do not necessarily reflect those of the Bank of Japan or Research and Statistics Department.

Questions and opinions on the working paper should be e-mailed to each author whose address is indicated in the document.

Click on cwp01e22.pdf (95KB) to download the full text.


The Bank of Japan has published the Corporate Service Price Index (CSPI) since January 1991, and data used in the index has been collected since January 1985. Based on the Bank's ample experiences of compiling the CSPI, this paper has introduced the actual methods of quality adjustment to compile the CSPI and the results of quality adjustment in 2000. The Bank has been endeavoring to adjust the quality as much as possible by using various methods as well as to introduce many newly expanding service items at the time of revising the base year and collect as many samples as possible in compiling the CSPI. The small degree of declines in prices estimated from quality adjustment successfully indicates the features of the services industries--that is, their low productivity.

Still, such quality adjustment has faced many obstacles that are connected to how the output of services should be defined. Technological innovation induced by the progress of information technology and heterogeneity of services accelerated by deregulation and supplemented by information technology have become the two biggest recent changes and these changes have complicated quality adjustment in service prices. These obstacles are not confined to Japan only; rather, they should be considered as applicable to service prices worldwide.

This paper*1 has drawn out the many limitations involved in capturing the quality of services concretely, and has suggested directions toward expanding varieties of quality adjustment methods. As for services of which output can be determined by the amount of labor inputs and capital inputs, the existing production cost method is still valid, but wider application of the hedonic regression method should be developed. On the other hand, as for services for which it is difficult to use production functions and of which the output is determined by the behavior of demand, inventing econometric models that comprise the concept of saving or consuming time by using services, and utilizing survey data on customer satisfaction should be developed. Furthermore, when the assumption of perfect competition does not hold, inventing other econometric models that take account of the monopolistic behavior of service producers is necessary. Based on the above features of services, all of these new approaches focus on the demand side of services, not the producer side. It is important to continue to develop and utilize quality adjustment methods more positively and to further improve the accuracy of quality adjustment.

  • *1   This paper was originally presented to the 16th Meeting of the Voorburg Group on Service Statistics held at Örebro, Sweden, in September 2001 and has been revised thereafter.

(JEL classification: C43, C81, L15; keywords:
service price, quality adjustment, hedonic approach)