Reports of Study Groups, etc. No.06

Report of a Study Group on Economic Analysis and Statistics
in connection with Information and Communication Technology


August, 2002
[Committee]
Takeshi Hiromatsu
(Professor, Graduate School of Arts and Sciences, College of Arts and Sciences, The University of Tokyo)
Fumio Funaoka
(Professor, Faculty of Economics, Shinshu University)
Tsutomu Miyagawa
(Professor, Faculty of Economics, Gakushuin University)
Yasuhiro Murota
(President, Shonan Econometrics Inc.)
Naoki Tsubone
(Deputy General Manager, Information Science Laboratory, Mathematical & Computing Science Laboratory,
Information Technology Center, Daiwa Institute of Research Ltd.)
Kazunori Minetaki
(Research Fellow, Economic Research Center, Fujitsu Research Institute)
Yasuhiro Ishimaru
(Manager, Research Office, The Bank of Tokyo-Mitsubishi, Ltd.)

[Secretariat]
Masakatsu Tamaru
(Economic and Social Research Institute, Cabinet Office)
Naoki Okumura
(Economic and Social Research Institute, Cabinet Office)
The full text is written in Japanese.     
(Abstract)

    Chapter 1 surveys the spread of Information and Communication Technology (IT) in Japan, some trends in economic analysis, and improvements in economic statistics in connection with IT. Chapter 2 examines what kind of improvements should be made in the nearest future regarding IT-related economic statistics from the following three aspects:

(a) to assess the actual situation
(b) to upgrade economic analysis in connection with the spread of IT and productivity
(c) to meet the new needs of economic analysis

    Chapter 3 considers the issues to be kept in mind while working on the development of new statistics in compliance with the spread of IT and sums up the recommendations on the key principles of that development. Chapter 4 works out a plan for an experimental survey. Moreover, the report includes, for reference, detailed discussions and basic materials for examination in the Study Group meeting as well as a list of IT-related statistics in Japan and other documents. The main issues and recommendations are summarized at the beginning of the report.
 
Chapter 1 The Main Trends in the Spread of IT in Japan and Improvement of IT-Related Statistics
 
    In the last half of the 1990s, commencement of commercial uses of the internet triggered a mass connection of computers to the internet. IT became an essential infrastructure, exerting great influence on the economy and society. It can be thought of as the first stage of the IT revolution. Today, the IT revolution has entered the second stage, one that centers on broadband and continuous connection to the internet. It has been developing along the lines of P2P (peer to peer) as a system, multimedia as content, and ubiquitous computing as a usage method. The internet is expected to apply to a wide range of people's lives and business activities. On the other hand, the security problem has become obvious, and the problem of "digital divide" also attracts attention.

    Concerning the impact exercised by IT on the economy, the analysis in connection with productivity began to be implemented from the last half of the 1990s, inspired by the U.S. "new economy" theory. While focusing on IT-related investments, this analysis aims at verifying whether the increase in the IT capital stock led to a significant increase in labor productivity and/or total factor productivity. The common point in the results of the series of researches is that IT-related investment alone does not trigger an increase in productivity in the IT- using industries; the quality of human capital, business structure, and other factors seem to play important roles as well. Nevertheless, the statistical data is far from complete, and it is difficult to draw definite conclusions.

    The current status of IT-related statistics in Japan is shown in diagrams 1-3-1 to 1-3-11 in the body text; the list of IT-related statistics attached is to be used for reference.
 
Chapter 2 IT-Related Statistics Must Be Developed and Improved
 
    The Statistical data necessary to assess the actual situation include the size of production activity and capital stock, value added, expenditure, employment, and other basic data of IT-related industries on the base of the SNA (System of National Accounts), production and sales turnover by customer of software, including package software and self-developed software, detailed price index in connection with IT, inventory index of computers and mobile phones and others.

    It is necessary to examine how to improve the precision of the Survey of Selected Service Industries to assess the real state. This would include revenues and other data points of game software, which characterize Japan's software industry. Furthermore, it is indispensable to assess the export and import of software and the amount of copyright royalties in the Foreign Trade Statistics and Balance of Payments statistics. Surveys on leading-edge trends in revenues and service charges in connection with Internet service, content services, accounting and copyright management services, and others should be implemented. The survey on the spread ratio of IP telephones, wireless LAN, IP-VPN, IPv6, and home information appliances should not fall behind.

    As for the statistical data necessary for economic analysis in connection with spread of IT and productivity, data on capital stock by industry and by asset need to be produced urgently. For that purpose, data on the investment amount and economic depreciation ratio are crucial. It is necessary to investigate this problem in a comprehensive way, from data collection to processing and estimation. There is also the problem that self-developed software and package software are not included in SNA capital investment and capital stock. In connection with these problems, further investigation on quality adjustment of deflators, depreciation ratio of software, and other factors is deemed essential. In regard to the productivity analysis, since it is desirable to use user's data for the production factors, it is fundamental to compile data on leases, dispatched workers, outsourcing and others, and necessary to adjust the owner's data to the user's data, in response to the classification of IT. It is also indispensable to examine the compilation of data on retirement and the capacity operating ratio necessary to measure the actual operation of IT capital stock and intellectual capital stock. Moreover, to analyze the conditions under which IT-related investment leads to increases in productivity, it is necessary to develop data concerning human capital, to develop the authority and competence of information division, and to collect data on the adoption ratio of organizational structure and work patterns complementary to the positive use of IT.

    The new needs of economic analysis and the development of IT-related economic statistics capable of meeting these needs are rigorously considered. First of all, we focused attention on the possibility of social and institutional environments in exerting an influence on the spread of IT and its results. To this end, data on the improvement of education, mobility of labor market, establishment of e-government, environmental conditions facilitating business promotion, new entry and development of new goods, and others are fundamental. Second, the analysis based on network externalities representing the distinction of IT such as analysis in connection with production, development, competition, and market of software, analysis concerning the agglomeration of IT-related industries, analysis of the agglomeration paradox, and analysis of protection and exploitation of intellectual property are all extremely important. It is also necessary to compile the relevant data. The change in behavior of both enterprises and households also requires analysis. Consequently, the necessary data are data concerning SCM and e-marketplace, data on households' continuous and dial-up internet connections, buying behavior and price, data on diversification of working pattern and place, data on characteristics of the spread of IT in Japan such as mobile use of the internet, data necessary for the analysis on the impact of the leading-edge trends on economy and society, and other data.
 
Chapter 3 The Key Principles of New Statistics Development in Compliance with the Spread of IT
 
    The present system of statistics attaches much importance to objectivity and consistency; especially, it is characterized by fixing classification for a long period, by securing precision, and by eliminating bias through rigorous sampling. However, the rapidly changing IT sector poses serious limitations to the usage of this approach. Depending on the survey target it is necessary to emphasize prompt publication and on sensor functions, allowing a good grip on the changing situations. To this end, a more flexible response in connection with classification and sampling and complementary usage of case study and significance sampling survey are possible.

    In selecting survey items it is usually necessary for the weight of expenditure and other data points to have already reached a certain standard level. Nevertheless, it is necessary to study other selection standards to assess new IT-related developments and trends important from the standpoint of policy.

    As it has become more difficult to make an analysis based on aggregate data of long-time series while maintaining constant data definitions, it becomes necessary to make a microeconomic analysis on the basis of recent cross-sectional data. It is desirable to develop a research design under which these data can be used as aggregate time series data as well when the data have been accumulated. Usage of panel surveys should also be considered. Moreover, links of the existing statistical data are important. To support this microeconomic analysis, we should actively proceed in work on promotion of use of individual survey data, establishment of re-aggregation in compliance with the user's demand, and the development of micro data sets.

    Moreover, it is necessary to undertake research making the most use of IT (use of POS data, survey by means of Internet such as Web surveys) and to take users' convenience into consideration (prompt report on statistical data, use of data through the internet, disclosure of information such as standard error, and supply of bridge tables for different classifications).

    The following types of data are given higher priorities in improvement and development of IT-related statistics. The first type is the data necessary for assessment of the actual situation or economic analysis important for policy planning. The second type is the basic data that are used very frequently. The last type is the data that are not yet compiled in Japan but compiled in most other countries. Thus, after fixing the order of priorities, it is necessary to promote the immediate development of those statistics while keeping in mind the necessity of theoretical research mentioned in Chapter 2-2 and the problems arising during implementation of statistical surveys such as respondent's burden. The practical approach is that especially important data are developed by priority, while other data mentioned in this report are developed in turn.

    Based on the above discussion, the Japanese government should give priority to improving and developing the following data:
      Concerning the other data mentioned in the Chapter 2, each ministry should examine on its own and begin to implement possible data development as soon as possible. Furthermore, with regard to improvement and development of statistical data related to multiple ministries, it is expected that Cabinet Office, Statistical Bureau of Ministry of Public Management, Home Affairs, Posts and Telecommunications and each ministry and agency collaborate. The Cabinet Office should take the lead in research from an economic analysis perspective.
 
Chapter 4 About a Project of Experimental Survey
 
    Based on the issues discussed in this report, we propose to choose several data from the set of the most important data and to implement an experimental survey. The progress and limit of the economic analysis implemented on the basis of this experimental survey are expected to provide valuable information to the government on full-scale statistical development.

    According to the discussion in Chapter 1 on recent trends in IT-related studies, it is necessary in Japan to analyze the relations between investment and the positive use of IT and productivity in the context of characteristic features of business establishments or enterprises, the quality of human capital, and capability of practical use of IT. To this end, it is necessary to collect data that can be good indices of these characteristics, to make basic aggregation by industry and by scale and various aggregations by cross-section, and to clarify correlation on the basis of econometric analysis using the indices. At the same time, it is crucial to collect data in connection with the development of software necessary for improvement of SNA estimation and data necessary for estimation of value added of IT-related industries constituting the base to assess the actual status of and to make analysis on the spread of IT. Moreover, it will be useful if we can gain some information helpful for future survey design through the analysis of the results of this survey.