One important mission of the Research Institute of Economy, Trade and Industry (RIETI) is to realize Evidence-Based Policy Making, or EBPM. To devote increased efforts to the EBPM approach, RIETI has established the EBPM Center. In 2023, RIETI will further develop the EBPM approach.
Government statistics play a significant role in implementing effective policy measures. At the beginning of the “Basic Plan for Development of Official Statistics” (2020), it is written that official statistics form the foundation of EBPM. While the use of “alternative data,” or data other than government statistics, including POS (point of sale) data on retail sales, credit card transaction data, and mobile phone location information, is spreading, the importance of government statistics, and particularly panel data, for policy evaluation, remains unchanged. However, it is difficult for ordinary people to understand discussions about statistics, which tend to be quite technical. Therefore, let us consider the significance of government statistics for EBPM by looking at the case of “investment in people,” one of the current government’s key economic policy initiatives, as an example.
Investment in Employer-Provided Education and Training and Productivity
Few economists would dispute the point that innovation and human capital investment are the twin engines that drive long-term productivity improvement and economic growth. Therefore, the government is right on the mark in placing policy emphasis on “investment in people.” Human capital investment includes various sorts of investment, such as investment in home education, preschool education, school education and vocational training, but under the government’s current policy, the emphasis is on improving vocational skills, for example by increasing employers’ expenditure on education and training.
The effects of employers’ human capital investment for employees can be measured in terms of the productivity improvement achieved by the employers. According to the results of the estimation made by this author based on data available from the Basic Survey of Japanese Business Structure and Activities (Ministry of Economy, Trade and Industry), companies’ investment in employee education and training has a significant impact on their productivity. In terms of the return on investment, investment in employee education and training has a much stronger impact than capital investment (Morikawa, 2021). Given that employee education and training has also been found to be effective in raising employee wages, it makes significant contributions to increasing the level of real wages.
Skills acquired by workers through education and training are not lost over a short amount of time, but are maintained over several years while their value gradually depreciates. Therefore, in order to accurately measure the effects of investment, it is desirable to look at the results of education and training on a stock basis—that is, the cumulative figures of the education and training expenses. The Basic Survey of Japanese Business Structure and Activities accumulates data based on a continuous tracking survey of human capital investment over time by individual companies, including unlisted ones (panel data) and as a result, the cumulative value of human capital investment can be calculated. Moreover, because this is the Japanese survey that is best suited to measuring the level of companies’ productivity and time-series changes, it is possible to quantitatively identify the contribution to productivity made by human capital investment.
Even so, analyzing the relationship between those two factors is not sufficient in analyzing policy effects. To clarify the effects of the government’s “investment in people” initiative, it is necessary to evaluate the extent to which specific policy measures have increased companies’ human capital investment. To that end, company-level panel data, including information on human capital investment, is indispensable. It is desirable to increase resource allocation for effective policy measures and to reduce allocation for ineffective ones in light of the evaluation results. What kind of data does the government use and how is the data used? Has the government developed a concrete action plan before policy implementation? These points would be the touchstone of the strength (or lack) of the government’s commitment to the policy.
Return on Investment in Postgraduate Education
In many cases, the effects of vocational training for workers are realized relatively early because the training is customized to suit the needs of actual jobs. However, from the perspective of long-term human capital development, school education plays a significant role. In particular, the importance of postgraduate education for nurturing innovators is growing around the world. According to a rough estimate based on public data available from the School Basic Survey (Ministry of Education, Culture, Sports, Science and Technology), the postgraduate enrollment rate in Japan is higher than 10%, and around 13% of people who start working after graduating from university or higher are those who have completed postgraduate courses (master’s or doctor’s courses).
Is postgraduate education an effective means of human capital investment? To measure the effects of postgraduate education, it is necessary in the first place to gather data that can identify whether workers have completed postgraduate courses. On that premise, if the working conditions and wages of people who have completed postgraduate courses can be understood, it will be possible to calculate the return on postgraduate education investment. At a relatively early stage, the Employment Status Survey introduced a questionnaire that distinguishes between workers with only an undergraduate degree and workers with a postgraduate degree. According to the results of the estimation made by this author based on micro data available from this survey, the return on postgraduate education was substantially higher than 10% (Morikawa, 2015). Several subsequent studies also found that postgraduate education has significant effects on wages.
The Basic Survey on Wage Structure (Ministry of Health, Labour and Welfare), which is used most frequently by researchers for wage analysis, also introduced, in 2020, a questionnaire that distinguishes between workers with an undergraduate degree and workers with a postgraduate degree and, at the same time, started asking respondents who identified themselves as part-time workers about their academic achievement. These refinements of surveys are expected to contribute to the development of analysis of the effects of the “investment in people” policy initiative.
Human Capital Investment for Construction of Official Statistics
In recent years, the government bureaucracy has been disgraced by a succession of misconduct cases related to government statistics, such as inappropriate data collection and processing. As a result, some ministries and agencies are moving to stop preparing statistical surveys as a risk avoidance measure. Avoiding the risk of suffering disgrace from possible cases of wrongdoing is a poor excuse for abandoning the task of preparing statistics, but the underlying factor is the downsizing of the manpower allocated for the preparation of statistics at ministries and agencies. On the other hand, in society at large, there are growing moves to train highly-skilled workers who are adept at preparing and using various data, including government statistics—known as data scientists—as evidenced by the fact that many universities have established data science departments. Under these circumstances, securing bureaucratic manpower capable of preparing statistics and improving the data literacy of government staff at policy planning and implementation divisions is a sort of human capital investment that is essential for ensuring that the EBPM approach takes hold.
Meanwhile, increasing the availability of micro data that comprise statistics is another challenge. Among academic and research communities, there is widespread discontent with the cumbersome, time-consuming procedures that are necessary for using micro data related to government statistics, and this problem has undermined the international competitiveness of the Japanese research community. While shortages of manpower related to providing these data appear to be creating a bottleneck in this respect, there is ample room for invigorating policy-oriented research activity by simplifying the procedures and paperwork required of prospective data users.
At present, a study is ongoing with a view to developing the next basic plan for government statistics. Government statistics have a significant role to play not only in promoting the “investment in people” initiative discussed in this article but also in developing policies that address such critical challenges as innovation and globalization and evaluating policy effects. In particular, long-term panel datasets, which are created through continuous additions of new data to already accumulated data, represent an asset that is very important for policy evaluation, and it is thus highly important to keep collecting data. The burden on respondents is often highlighted as a problem related to government statistics. If the EBPM approach is promoted with no regard for data continuity or data quality improvement, respondent companies may have to put up with an even more onerous burden as they are more frequently asked to respond to ad hoc surveys or are subjected to increased reporting requirements.
Statistics is a technicality-riddled area, so it only rarely comes to the public’s attention, except when negative publicity is created due to the revelation of some wrongdoing or other on the part of those who collect and prepare data. Since statistics represents a policy infrastructure that is valuable for ensuring that the EBPM approach produces actual benefits, this author looks forward to more and more people understanding its significance.
