In the business world, a variety of business intelligence (BI) tools have been developed and introduced in recent years with an array of mechanisms worked out to facilitate the use of big data for business decision making. Meanwhile, the government and administration have come to recognize the importance of evidence-based policymaking, and various efforts are being made both within individual ministries and across the entire government. For instance, the Ministry of Education, Culture, Sports, Science and Technology (MEXT) launched a program called "Science for Redesigning Science, Technology and Innovation (SciREX)" in FY2011. In this article, I would like to explore a new direction for developing an innovative environment for the use of public data.
Path to a New Future of Evidence-based Policymaking Paved by a Public Data Platform
Realizing an environment for analysis capitalizing on the strengths of Japanese public data
One of the advantages of policy research in Japan is the presence of statistical data accumulated over many years and well-established rules and procedures for their use. As a particularly notable development in recent years, the establishment of rules and procedures for the secondary use of microdata (primary data) collected in statistical surveys has prompted many researchers to conduct empirical studies on various policy issues (including many of those published as RIETI discussion papers) using newly available data on individual objects. On the other hand, a major challenge in conducing policy research in Japan is that all such microdata collected by statistical surveys are kept and managed separately by respective government ministries or sections in charge, making it difficult to analyze multiple sets of data linked at the micro level, i.e., at the level of firms, institutions, or individuals. Depending on the subject of research, it may be necessary to analyze not only public statistical data but also data held by administrative organizations and private-sector companies (e.g., patent data and a list of companies receiving public subsidies). However, it is difficult to link those different sets of data with one another. By not only using common classification systems for industries and regions but also managing microdata on firms, institutions, individuals, and so forth in an integrated way with common identification information across different sets of statistical data, it would facilitate the development of a common list of respondents that enables or accommodates linking among both public statistical data as well as with other data. When such big data that contain public statistical data and various other data held by public and private sector organizations are made available for use in research, more and higher quality policy research will be conducted using Japanese data.
However, ensuring the security of information is crucial to the unified management of public statistical data. It is necessary to promote technological advancement and proactively utilize the latest data security technologies such as those for encryption, identify verification, and the highly effective detection and prevention of unauthorized access. At the same time, in order to handle a large volume of data, we need to have computers capable of high-speed computation and communication. Japan has a good chance to realize an environment that enables the safe and effective use of data by mobilizing the best of information and communications technologies from industry, academia, and the government. It is hoped that the government will make a concerted effort—under the strong leadership of the prime minister and through extensive collaboration with relevant administrative agencies, policy think tanks, universities, and other academic research institutions—to build a single government-wide data management system toward the realization of an evidence-based policymaking process. For instance, in the area of science, technology, and innovation policy, discussions are underway for the development of a cross-ministerial data platform as a joint initiative of the Council on Economic and Fiscal Policy (CEFP) and the Council for Science, Technology and Innovation (CSTI) (see a draft interim report (in Japanese) submitted to the second meeting of the experts' committee on the economy, society, and the revitalization of science, technology, and innovation on October 6, 2016).
Case examples of policymaking systems utilizing big data
Next, I would like to introduce two recent attempts to utilize big data in the process of policymaking.
Regional Economy and Society Analyzing (RESAS)
RESAS is an online portal launched in April 2015 by the Council for Overcoming Population Decline and Vitalizing Local Economy in Japan. Speaking at a news conference after the Cabinet meeting on April 21, 2015, Shigeru Ishiba, then minister in charge, said that with the introduction of RESAS, the administrative authorities can no longer come up with policy measures just for show. Indeed, the portal is perceived to have been designed with the clear intention of promoting evidence-based policymaking. Various efforts are being made on an ongoing basis to improve user convenience and promote further use, with recent initiatives including the launch of e-learning services and the hosting of a contest for local revitalization policy ideas and for RESAS applications (see the website of the Council for Overcoming Population Decline and Vitalizing Local Economy (in Japanese) for further details).
SciREX Policymaking Intelligent Assistant System (SPIAS)
Under development since FY2016, SPIAS is intended to serve as a platform that provides policymakers and researchers with open access to data on science, technology, and innovation policies. The development of SPIAS is being promoted as a joint project of three national research institutes: the National Graduate Institute for Policy Studies (GRIPS)'s SciREX Center, the National Institute of Science and Technology Policy (NISTEP), and Japan Science and Technology Agency (JST)'s Center for Research and Development Strategy (CRDS). By linking data on competitive funding—i.e., kakenhi grants-in-aid scientific research and competitive research funds offered by the JST and the New Energy and Industrial Technology Development Organization (NEDO)—with those on patents, academic research papers, and press releases, SPIAS aims to enable the visual representation of how competitive research funds are allocated among universities and across different domains of research or how leading researchers have been awarded research funds, and thereby realize a mechanism allowing policymakers and policy researchers to see trends in science technology and utilize the information on a real time basis in accordance with their respective needs (see SciREX Center's website (in Japanese) for further details).
There remain many institutional and technical challenges that must be overcome before realizing evidence-based policymaking. Meanwhile, accelerated efforts have been made both in Japan and abroad to develop micro-level data relevant to policy research. As a researcher, I would also like to accelerate my efforts to contribute to the realization of evidence-based policymaking, sharing visions and information with other researchers and policymakers in Japan and other countries.
January 20, 2017
Article(s) by this author
June 30, 2017［VoxEU Column］
January 20, 2017［Column］