The federal statistical system has recently come under close scrutiny by government, business, the academia and the general user community. Questions have been raised about the usefulness, accuracy, accessibility and integration of federal statistics. There is more and more pressure coming from inside and outside the government for change. Indeed, I can't remember the last time there have been so many proposals to change the federal statistical system. The suggested changes range from improving data sharing capabilities among agencies to the establishment of a centralized statistical system. For example the House of Representatives Fiscal Year 1996 Budget Resolution states: Consolidate the Bureau of the Census and the Bureau of Economic Analysis Into an Independent U.S. Statistical Administration. U.S. Government statistics are collected and analyzed by at least 25 Federal offices, departments, and agencies; each constructs indices differently, uses different time periods and different base years. There is no central organization setting standards for the quality or consistency. Consequently, many statistics compiled by the U.S. Government are suspect. This proposal calls for consolidating many of the statistical organizations in the U.S. Government with the Census Bureau to achieve qualitative improvement and efficiencies. Because of the difficulty in scoring the sweeping consolidation this proposal would require, no savings are assumed.
The idea of a centralized system has been around for a quite a while. Back in 1990, the Office of Management and Budget took a close look at the idea, but came to the conclusion that there might be better alternatives. In an excellent "must read" book by Janet Norwood, former Commissioner of the Bureau of Labor Statistics (Organizing to Count, Urban Institute Press, ISBN 0-87766-634-2), Dr. Norwood proposes a number of approaches including "radical surgery to the organizational structure." One of these approaches, which she characterizes as revolutionary and swift, would be the carving out of a new "Statistics America" by combining several statistical agencies into a Central Statistical Board. I don't believe that the House's Budget language was based upon Norwood's book, but rather advice from the Heritage Foundation. As you can see, the idea of consolidation is coming from a number of quarters.
Perhaps the only major opposition to the idea of some form of consolidation would come from the departments of the agencies, and the agencies themselves. For example, it's hard to believe that the Department of Labor would readily say good-bye to the Bureau of Labor Statistics, which quite rightly is perceived as a jewel and an important resource of the department. Nor can we expect the statistical agencies to be sanguine about losing their identities and independence. Further, they have legitimate concerns about the ability to have a uniform confidentiality policy, and their desire to stay close to the policy issues of their specific departments.
Note that the House recognizes that at least for the first year, any consolidation would not save money. Would it be a money saver in the long run? Probably, since the elimination of duplication, economies of scale and better coordination should affect the bottom line. Most important to you, consolidation would be a boon to the user community. Just think: the ability to get an answer to a question on education or employment or health related data from one source instead of having to go to two or even three agencies.
Time to get back to reality. The idea of consolidation is so good, it probably won't happen - this is Washington, remember? Then again, with all that's happening - especially the move to abolish agencies such as Commerce, Education, etc. - such an option could slip through the cracks. A much more modest approach, yet one still requiring a lot of cooperation from the agencies, is what has been characterized as a "virtual" federal statistical agency. This organizational approach would keep the agencies independent but would improve coordination of planning, data collection, sharing of files, dissemination, research and personnel. Coordination would require new legislation. Also, any major changes in the sharing of files across agencies would require changing existing legislation, for example Title 13 of the Census Bureau. Such a change would allow for the sharing of confidential information for statistical purposes only. Changing legislation wouldn't be easy, but it's probably easier than trying to consolidate the agencies. With this legislation in place, the agencies could share information in ways that would yield the production of more useful data, better coordinate developmental research, share personnel, reduce redundancies and probably save the taxpayers money. A one-stop-shop should, and could be created for users. The coordination of this virtual agency might be carried out by the newly created Interagency Council on Statistical Policy, headed by the Chief Statistician. This Council is called for in the Paperwork Reduction Act of 1995, which has passed the Congress and signed by the President.
The time is now! Not because the issues are more critical. They've been around for a long time. It's the climate. If nothing else, there is a feeling around this town that changes are not only possible, but might actually happen. No doubt we won't like some of them, but then again, a better coordinated federal statistical system just might be one of the good ones that manages to come about.
In April, the House of Representatives passed the Family Privacy Protection Act, which states that no "program or activity funded in whole or in part by the Federal government" may, without the written consent of one parent or guardian, require or otherwise seek the response" of a minor "to a survey or questionnaire which is intended to elicit, or has the effect of eliciting" information about: (1) parental political affiliations or beliefs, (2) mental or psychological problems, (3) sexual behavior or attitudes, (4) illegal, antisocial, or self- incriminating behavior, (5) appraisals of other individuals with whom the minor has familial relationship, (6) relationships that are legally recognized as privileged, including those with lawyers, physicians, and members of the clergy, and (7) religious affiliations or beliefs. Exceptions are made for criminal investigations, good faith concern for the health, safety, or welfare of an individual minor, administration of the immigration, internal revenue, or customs law, and determining program eligibility. According to the Congressional Budget Office, the effect of the bill would be to increase costs some fifteen fold for any survey which includes minors (generally under 18) in the sample.
Economist William P. Butz has assumed the top position in the Division of Social, Behavioral, and Economic Research (SBER) of the National Science Foundation, effective August 21, 1995. Butz arrives at NSF with a diverse professional background as an economics and demographics expert who has served both the public and private sectors. Before his NSF appointment, he served as Associate Director for Demographic Projects at the US Bureau of the Census, a position he held since 1983. In that capacity, he led a staff of 700 and oversaw such high interest programs as population estimates and projections, major surveys, and international statistics.
Butz has also served as senior economist and Deputy Director of Labor and Population Studies at the Rand Corporation, and as a senior consultant to such organizations as USAID, NICHD, the World Bank, and the National Academy of Sciences. While at Rand, he also served as a visiting lecturer in economics at both UCLA and UC Santa Barbara.
Butz is Vice-President-Elect of the Population Association of America and a member of the American Statistical Association, the American Economic Association, and the American Association for the Advancement of Science.
As Director of the Social and Behavioral and Economic Research Division, Butz oversees a range of a dozen disciplines --including economics, anthropology, sociology, psychology, linguistics and anthropology ---in the Directorate for Social, Behavioral and Economic Studies.