"The New Strategies for Government Statistical Services for the Coming Decade in1995" - Toward development and increased efficiency of statistical work in response to changes: A progress report (Reproduction from "Statistical Notes of Japan" issued in October 1999)
Introduction
More than four years have passed since the Statistics Council(*) submitted its report entitled "The New Strategies for Government Statistical Services for the Coming Decade" (hereafter referred to as the "New Strategies") in March 1995.
The New Strategies was presented as the guideline for government statistical services as we move toward the 21st century, and its contents covers the entire range of statistical services.
With the understanding that it is of the highest priority to implement all of the contents of the New Strategies steadily, the government has sought to maintain close linkage with the work of the Statistics Council, and it has established the "Steering Committee for the Promotion of New Strategies" (hereafter referred to as the "Steering Committee"), and has been working to implement this work. (For the organizational framework for implementation of the New Strategies, see the appended diagram below.)
This report contains outline on the government's tackling of implementation of the New Strategies, and on overall situation of progress in concrete implementation of specific tasks.
(*) The Statistics Council is an advisory body attached to the MCA. It is in charge of investigation and deliberation on important matters concerning statistical surbeys, at the request of the Director-General of the MCA.
1. Establishment of the Steering Committee for the Promotion of New Strategies
In order for the Japanese government to be able to implement contents in the New Strategies in an unified manner, the Steering Committee consisting of the director-generals of the statistics bureaus in the concerned Ministries and Agencies was formed in April 1995 (the month following the Report of the Council), as well as the Coordinating Committee of the Steering Committee composed of directors of statistical coordination divisions of the Ministries and Agencies concerned, were established.
The Steering Committee does follow-up work on the implementation status of the New Strategies, as well as liaison and coordination work within the government.
Also, in order to examine more deeply the items that are common to the Ministries and Agencies, and items that have a higher technical level and higher level of specialization, that need to be examined more concretely, study committees consisting of scholarly experts, staffs selected Ministries and Agencies concerned and other concerned organizations have been formed under the Steering Committee.
At the present time, three ad-hoc study committees have been established by the Steering Committe, and six working groups and one research committee have been formed under the study committees with the aim of carrying out even more detailed examination of specific areas.
2. Concrete State of Progress of Various Study Issues
The New Strategies makes a number of specific proposals for the direction to be pursued in improvement of individual statistical surveys. With regard to these specific tasks, study is being carried out by the Ministries and/or Agencies that are in charge of the concerned statistical surveys, and considerable results have been achieved already.
In the work of improving such individual statistical surveys, major items are placed again for deliberation by the Statistics Council, and efforts are made to ensure that adequate linkage is achieved through such deliberations by the Council.
Below, an overview of the state of progress of the major items proposed along the chapter composition of the New Strategies is given in the report.
Chapter 1. Review of statistical surveys to adjust to social and economic changes
In order to respond accurately to changes in social and economic situations that affect the government statistical services such as internationalization, aging of society, diversification of values, etc., and in order to develop statistical surveys in a systematic way, to achieve the necessary simplification and increase in efficiency, and, finally, to secure a comprehensive and integrated character of statistics administration for the government overall, the following review and modifications have been made.
1) Improving statistics on business enterprises
[1] In relation to the Establishment Census carried out by the Management and Coordination Agency (which is a designated statistical survey), items that make the government statistical offices possible collation of names of business enterprises, and items that are useful for them to grasp the activity situation of business enterprises have been added, with the aim of contributing further to obtaining grasp and analysis of business enterprise activity. And the name of the census has been revised to "Established and Enterprise Census" and it was implemented in fiscal 1996.
Also, based upon the results of this, the "Establishment and Enterprise Information Database for the Business Register" has been built, and a part of it has begun to be put to service starting in January 1998.
[2] The Ministry of International Trade and Industry (MITI) abolished the Mode of Trade Statistics Survey (a designated statistical survey based on the Statistics Law), and incorporated its major investigated items into the Basic Survey of Business Structure and Activity (a new designated statistical survey) with the aim of arranging business statistics systematically, and grasping accurate trends of enterprise activities in a time series. MITI conducts this new survey annually since fiscal year 1995.
Also, in order to be able to grasp various activities of small and medium enterprises in a cross-cutting way, the Ministry reorganized and consolidated the Basic Survey of Manufacturing Activity and Industrial Structure and the Basic Survey of Commercial Structure and Activities (both of which are signated statistical surveys), and started the Basic Survey of Commercial and Manufacturing Structure and Activity (a new designated statistical survey) every 5 years since fiscal year 1998.
[3] Of the statistics on stock, the work on household assets and land assets has made progress, but measuring the tangible fixed assets held by business enterprises is left as an unfilled task. For this reason, the National Land Agency conducted the Survey on Corporation’s Buildings (an approved statistical survey based on the Statistical Reports Coordination Law), as an auxiliary survey to the Survey on Corporation’s Land (which is conducted as a designated survey in fiscal year 1998), which is expected to contribute to the grasp of the tangible fixed assets.
2) Improving statistics on household and family economy
[1] At the Management and Coordination Agency (Statistics Bureau), the National Survey of Prices (a designated statistical survey) was examined with a point of view corresponding to the changes in the role of the system for price statistics. As a result, it was decided to abolish survey on wholesale price, and to increase number of sample stores and items priced for the retail price survey. New survey with such improvements is conducted every five years since fiscal year 1997.
[2] With the aim of getting an overall grasp and analysis of the households' residential situation and residence-related assets, the Management and Coordination Agency added items related to housing and land other than those for their present residence use in the Housing Survey (a designated statistical survey), and changed its name to the Housing and Land Survey, and decided to conduct it every five years since fiscal year 1998.
In this new survey, it was also decided to adopt the so called long-form/short-form questionnaire system (to be described later in Chapter 5-2-(2)).
The National Land Agency will produce a basic statistics on land and buildings by using the data obtained from the long-form questionnaire of this survey and above-mentioned surveys on Corporation’s Land and Buildings.
Chapter 2. Timing and frequency of major statistical censuses and surveys
The fact that implementation of many large-scale censuses and statistical surveys are concentrated in the same timing means that securing of cooperation for reports is made more difficult, and that a great deal of excess burden is placed upon office work of local government and other bodies involved, as most of above surveys are conducted through them. There have been concerns that such problems can create grave problems in ensuring accuracy of the statistics.
Because of this situation, in the New Strategies, a more evened-out planed schedule is set for implementation timing of major censuses and statistical surveys, such as providing for an uniform 5-year cycle for major statistical surveys.
Today, major censuses and statistical surveys are implemented more or less according to a schedule in line with the Plan.
In fiscal year 1999, there is, however, an overlapping of a number of large-scale censuses and statistical surveys conducted through local government including the Establishment and Enterprise Census, the Census of Commerce, and the Survey on Service Industries. If all of these major surveys concerning business enterprises were to be implemented without careful coordination and adjustment, it could have been anticipated that the burdens placed upon the respondents, local governments, etc., would have become excessive.
Therefore, the way of solution on the issue was discussed again at the permanent working group of the Statistics Council, and it was decided that the questionnaires for the Established and Enterprise Census, and the Census of Commerce should be consolidated into one questionnaire and the two censuses should be carried out simultaneously in July, and the surveying day of the Survey on Service Industries should be postponed about 2 weeks (19th November).
Chapter 3. Reducing reporting burdens
Statistics is the numerical information which tabulate the data submitted by the respondents and so, in order to carry out statistical survey efficiently, and to ensure the accuracy of the statistics obtained, it is one of the important factors to reduce the reporting burdens.
The reporting burdens can be classified approximately into the psychological burdens especially in statistical surveys directed toward individuals and households, on one hand, and the physical burdens created mainly in statistical surveys directed toward business enterprises and establishments, on the other.
[1] Review of statistical surveys for reducing the respondent’s burden
In order to reduce the reporting burdens an agreement reached at a meeting of directors in charge of statistics of ministries or agencies that all the concerned government Ministries and Agencies review all statistical surveys under their jurisdiction which is based upon the Planned Review of Statistical Surveys.
359 surveys are scheduled to be reviewed for possible change in a 3-year period starting in fiscal year 1995 in the agreement, and of this total, such review has been completed already for 229 surveys.
When we look at the steps that were taken as a result of this review process (which comes to a cumulative total of 382 steps), the largest number related to "reduction of survey items" (about 42%), followed by - improvement of the survey method" such as introduction of online surveys and mail surveys (about 20%), and "reduction of respondents" (about 15%).
Also, some surveys including the Mode of Trade Statistics Survey, the Statistical Survey on Registered Hotels, Ins, and Lodging (an approved statistical survey, the Ministry of Transport) have been abolished, while others have been integrated, such as the integration of the Basic Survey of Manufacturing Activity and Industrial Structure and the Basic Survey of Commercial Structure and Activities into the Basic Survey of Commercial and Manufacturing Structure and Activities. In all of these ways, efforts are continuing to be made for reduction of reporting burdens.
[2] Utilizing use of administrative records for statistics
All the concerned government Ministries and Agencies collect and possess data (in the form of administrative records) from work for approval and permits based on laws, cabinet order and so on, in addition to statistical surveys as such. By tabulating and processing such data, and making good use of them as statistics, the burdens imposed upon respondents for statistical surveys can be reduced.
The Ministry of Health and Welfare has reduced the number of survey items for the Basic Survey on the Pension System (an approved statistical survey) by making use of data from various administrative records gained by the management of the system.
Chapter 4. Expansion of the use of survey results
Results of statistical surveys are precious information obtained with great cost and public cooperation. Therefore, these results should be disseminated widely as a public asset, protecting fully the privacy of the individual.
In this connection, it is important to improve the system for users considering the progress of information technology, and to provide statistical information quickly in a form that is easy for users to access.
[1] Quick public release of statistical survey results
All the concerned government Ministries and Agencies are making efforts to speed up the public release of survey results, by shortening the time taken up in processing of the survey data. In the New Strategies, it is requested as a working target that the preliminary report on the results of a designated statistical survey is to be released the public made within 60 days for monthly surveys, and within one year for annual surveys and surveys conducted at interval over one year. As of 31st March 1998, 50 out of the total of 60 designated statistical surveys implemented periodically have already met this target.
[2] Use of new information and communication technology
All the concerned government Ministries and Agencies are making efforts for convenience to providing the results of the surveys by electronic means such as online communication and magnetic media, in. As of 31st March 1998, out of 60 designated statistical surveys mentioned above, the results have been provided electronically for 49 statistical surveys. And the efforts to expand and improve such services will be continued in the future.
[3] Improvement of guidance on sources of information
All the concerned government Ministries and Agencies are striving to improve guidance on sources of statistical information under their jurisdiction by providing overall listing on the service offices and reference materials, etc. As of 31st March 1998, 6 Ministries and Agencies (the Management and Coordination Agency, the Ministry of Education, Ministry of Health and Welfare, Ministry of Agriculture, Forestry, and Fisheries, Ministry of International Trade and Industry, and the Ministry of Construction) are providing such guidance via Internet and guidebooks.
Also, the Statistical Standards Department of Management and Coordination Agency grasping the information on sources of designated statistical surveys by the government began to provide their information to the public via Internet since May 1997.
Chapter 5. Efficient implementation and quality improvement of statistical surveys
While quick and timely public release of the results of statistical surveys meet the diversifying statistical needs, there is also a heightened claim for accuracy of such statistical information. This problem has become especially acute in the recent years, due to the increase in the number of households where no one is at home during the daytime, and other complex changes in the environment for making statistical surveys, as well as to financial circumstances. These factors have increased the claim for efficiency and accuracy in survey-taking, along with the ever-increasing claim for lightening the burdens of the respondents.
In order to respond to such problems in an appropriate way, it is key to make good use of the fruits of innovation in information and telecommunications technology, survey technology, and information management technology, in collecting and tabulating statistical data.
1) Capturing statistical data through magnetic media or/and telecommunications
[1] At the Ministry of International Trade and Industry, five years research program has been underway since 1995 to develop "a new generation statistical system" which can collect survey data online from the business enterprises covered by a given survey, and can return the results of tabulation back to the respondents online as well.
[2] The method of collecting statistical data from respondents mainly business enterprises and business establishments in the form of floppy disks and online communication is being introduced at some Ministries and Agencies.
In fiscal year 1997, the method of collection of statistical data with the use of magnetic media and telecommunications line was introduced for seven surveys including the following: the survey of investment by corporations and business enterprises in Hokkaido (an approved statistical survey, the Hokkaido Development Agency), the statistical survey on orders received for machinery (an approved statistical survey, the Economic Planning Agency), the survey on the movement of international air cargo (an approved statistical survey, the Ministry of Transport), etc.
2) Increased efficiency of statistical surveys achieved through various measures
[1] Introduction of the pre-printed questionnaire
Some concerned Ministries and Agencies has adopted the pre-printed questionnaire system, in which the data already obtained by them such as the names of business enterprises, addresses, etc. are printed on the questionnaire before their distribution to the respondents, as another means for reducing the respondent’s burden. This type of statistical survey has been increasing year by year.
[2] Use of the long-form/short-form method
The Management and Coordination Agency prepared two types of forms in conducting the 1998 Housing and Land Survey: the regular form, and the expanded form with some new items added to it, and distributed them differentially by the survey district based on some sampling rules. That is, the so-called long-form/short-form method. By using this method, the additional items are surveyed only for some of respondents and new statistical needs are satisfied without unnecessarily increasing the burden for all of the respondents.
Chapter 6. Promotion of International Cooperation
In order to strengthen international efforts to tackle global issues such as environment problems, as well as to set up policies in each country, internationally comparable, accurate, detailed and timely statistics are indispensable. Thus, it has become urgent to promote development and preparation of such statistics in the desired manner.
Also, it is equally important for developed countries to improve statistics that are necessary for their social and economic development to actively support developing countries to enhance their statistical capabilities.
1) Improvement of international comparability of Japan's statistics and enhancing the external use of these
[1] In various Ministries and Agencies, the results of statistical surveys under their respective jurisdiction and major statistical indicators are now being provided in English on their Web Site.
[2] Statistical Standards Department, Management and Coordination Agency, prepared the correspondence table in English version between the International Standard Industrial Classification of All Economic Activities (the third revision) and Standard Industrial Classification for Japan (the tenth revision) in 1995 and also carried out the fourth revision of the Standard Occupational Classification for Japan considering the increase of international comparability in 1997.
[3] The Ministry of Finance reviewed the balance of payments statistics in accordance with IMF's Balance of Payments Manual (fifth edition) in 1995.
2) Convening of Inter-ministerial Meeting on International Statistical Affairs
In order to comprehensively promote international statistical cooperation, an inter-ministerial meeting on international statistical affairs among concerned Ministries and Agencies was set up in December 1995, and report and liaison on various types of international statistical conferences and exchange of other relevant information have been implemented in an attentive manner.
3. Promotion of Work in the Statistics Council
The Statistics Council, which main role is to make recommendations in response to the requests of Minister of MCA, is underway to promote the New Strategies in its work. Actually the work as follows has been promoted by its permanent working groups’ members.
[1] Research on the estimation method for sampling errors
In response to the need to "set up standards for information related to the quality of statistics (standards related to sampling design and sample error, etc.)", which was presented in Chapters 4 and 5 of the New Strategies, the Statistics Council set up "the research group on the estimation method for sampling error" consisted in some permanent working groups’ members. With regard to statistical surveys for which the sampling error cannot be estimated theoretically, the group will verify the extent of validity of sampling error estimated by a re-sampling method such as the boost strap method, and examine the extent of practicality.
[2] Promotion of data linkage
Two permanent groups, on development of statistical survey technology and on information processing, have jointly studied and examined the promotion of data linkage.
The Management and Coordination Agency has also commissioned Tokei Kenkyukai (the Institute of Statistical Research) a research on "promotion of data linkage" which was shown in Chapter 5 of the New Strategies, so as to contribute to the study and examination of the joint permanent working group.
There are two types of cases for data linkage: one type of cases in which the data from different statistical surveys related to the same object of survey are connected (exact matching), and the other type, where different objects of a survey have such characteristics so that they can be regarded as the same in a statistical sense, and the data concerning these objects are connected (statistical matching). The Institute has set up "the committee for investigation and research on verification of factors contributing to error arising from statistical matching" and the usefulness of the statistical matching is considering in it
In Conclusion
The New Strategies looked at the period of next ten years or so after itself, and presented guidelines as concretely as possible. Already four years have past since the Strategies was submitted, and it has been implemented in a steady manner, but many problems also remain which need to be examined in the coming period.
From this point on, the state of progress thus far needs to be evaluated appropriately, and the each item in the Strategies must be pushed forward in a prioritized manner. MCA, in charge of the coordination for the progress of this Strategies in the government, would like to maintain adequate linkage and cooperation with the Statistics Council, concerned Ministries and Agencies, the prefectural governments, etc., and implement this Strategies in a cooperative manner.