Printer Friendly

NABE presidential address: building improved official statistics for decisionmaking.

IT HAS BEEN a great honor to serve as your president during the past year. Throughout more than thirty years of membership in NABE, I have been impressed with the caliber of the membership, the professional contribution of the association and with the influence of the business economist in making key contributions to their individual organizations. In 1985, Dave Williams, NABE Executive Secretary-Treasurer, asked me to start a new committee -- the Statistics Committee -- in response to a recognition that the basic data of the business economist, the national economic statistics of the federal government, were in need of improvement and were being threatened by critical budget cutbacks occurring in governmental budgets. Today I want to use this occasion to draw more specific attention to our need for and responsibility to build improved economic statistics.


For the past seven years we have published a regular column in Business Economics entitled "The Statistics Corner."(1) This column was created at the suggestion of Ed Mennis at one of the early meetings of the Statistics Committee. The column has highlighted developments in official statistics that are of importance to business economists. Also, over the years many special reports, studies, task forces and technical papers have been made about weaknesses in existing statistics as well as recommendations for improvement and suggestions for new statistical measures. As a result, a regular flow of improvements in official statistics has occurred, but the pace of change has fallen behind the structural changes occurring in the global economy. New markets and industries are flowing from dramatic changes in technology and the new economic interactions among nation states that have been made possible by low cost transportation and increasingly efficient communications systems.

In my view, we need to take a comprehensive view of these changes and the needs for statistics in the future.(2) During the past eighteen months I have been engaged in an effort designed to illustrate the nature of the challenges that face us if we are to develop the type of information that will be needed in the future to aid governmental policymakers and business decisionmakers working in this new environment. My coauthor in this effort is Dr. Andrew C. Gross, Professor of Marketing at Cleveland State University. He has both national and international experience in marketing and in the information industry, has written texts on marketing demonstrating needs for market analysis, and he has been a student of international statistical systems.

Our report is entitled Statistics for the 21st Century. It was developed with the help of a number of experts and with excellent cooperation from many individuals who toil in the vineyards of official statistics. We assured our helpers that we would take full responsibility for the structure and content because we did not want to write another "committee" report with all of the necessary compromises and tempering of tone that is normally required to protect the collaborators in their individual institutional settings.

This book is, of course, still limited by the experience, perspective, insights, and imagination of the individual authors and those consulted. It is also limited by an inability to forecast the future accurately, and by the daunting task of relating all aspects of information needs from economic and social topics to specialized concerns in the areas of energy, environment, health care, education, welfare, crime, social behavior and all of the other dimensions of understanding the global economic system.

Nevertheless we thought it would be instructive to prepare a study of the changing conditions of the statistical system that generate the basic information used by business economists in their work as they strive to understand the environment for business in the years ahead.

Statistics for the 21st Century will be made available to all members of NABE who would like a copy.(3) It will also be distributed to key policymakers in government and business with an interest or role in improving the information base for economic policy. Our hope is that the document will serve as a starting point for debate, discussion and action on improving the full statistical system. It is not a final answer: rather it is an attempt to show the importance of improving official statistics and to build the case for comprehensive action. Many issues are not addressed, but we hope the ensuing discussion will be broader and that the overlooked areas will also receive attention. In developing this effort, our literature search identified over 2,000 articles and books that are relevant to the overall subject. Even though that literature review was limited(4) it illustrates the rich resource base that is available for any serious effort to improve basic statistical information.


In Statistics for the 21st Century we outline some of the major changes that are occurring in the collection, processing and dissemination of statistics, along with a few illustrations of dramatic change in basic economic relationships. These changes call for new sources and concepts of economic indicators. Our discussion of the major stakeholders in economic information follows the format of the now familiar SWOT analysis of "strengths, weaknesses, opportunities and threats." A full SWOT analysis of all of the stakeholders would take several volumes, but our brief analysis shows that a growing urgency exists for closer cooperation between business and government in statistical information development that is needed by both.

For example, while business resists governmental regulation and is resistant to more governmental "paperwork," it is clear that new electronic-based information systems, including universal product codes on goods moving through the transportation and distribution system to the ultimate consumer, make it possible unobtrusively to collect better information about transactions that form the economic lifeblood of the business system. Newly adopted standard classification systems like the Harmonized Commodity Classification System make it possible to redefine data collection on a consistent basis around the world, thus improving the accuracy and consistency of information on trade flows. While these are technical developments, they need to be merged with the definition of needs for information to build "decision-oriented" data systems that give immediate benefits to businesses and that make possible better description of the macro- and microeconomic situation. Then it becomes clear that the issue of "paperwork burden" diminishes in the face of information benefits and the capabilities to undertake unobtrusive information collection.


It is clear that economic theory is lagging the revolution in economic relationships that is occurring. For example, development theory is inadequate to the task of mapping a strategy for countries that wish to shift from "planned" or "command" economies to market-based economic systems. Likewise, the concept of "stages of economic growth" is inadequate today as an explanation of the process of transformation from developing to developed countries. The emergence of the newly industrialized countries has demonstrated how low-cost labor and sophisticated technology can be combined with inexpensive transportation and rapid communications to bypass the reliance on natural-resource based export industries in moving to higher stages of economic development. As this new practice spreads, a need exists for new theory and understanding to design statistical systems that will better describe the state of affairs in these "transforming" and "newly industrialized countries." While new work on theory and understanding is essential, it is true that a decade of work is now coming to fruition with the recent agreement on a revised and improved System of National Accounts (the SNA)(5). The SNA is an accounting framework for defining economic relationships within nations and among nations so that a consistent set of statistics can be developed about economic (and some social) developments.(6) It is a cooperative effort of the United Nations, Organization for Economic Cooperation and Development, the World Bank, the International Monetary Fund, and the European Economic Communities. The SNA has several key elements that deserve this focus. They are:

1. This revision has built on twenty-five years of experience with the earlier SNA that was part of the statistical program of the United Nations. The 1993 recommendations have been endorsed by all of the cooperating agencies, greatly expanding its use and role in statistical comparability.

2. The U.S., which has long used a different framework (the National Income and Product Accounts-NIPA), has already announced plans to adopt the concepts of the SNA.

3. The SNA fosters standard definitions and concepts along with a flexible system that can be modified as required by future conditions.

4. Expanded use of the SNA by other nations will contribute to improved international comparability.

Consequently, the recent decision by the Bureau of Economic Analysis to implement the 1993 System of National Accounts should be accelerated as part of building a statistical system for the future.

A number of recommendations for improvements in data collection are required before developing better national income estimates. We identify nineteen specific recommendations that are long overdue and that will contribute to the implementation of the System of National Accounts. Specific recommendations are listed in Appendix 1 of this article.


The interdependence of the world economy has increased dramatically and will intensify in the future. Major business organizations now make decisions on a global basis both for inputs to their products and services and for markets. In a statistical sense, linkages among countries are captured in the balance of payments accounts of individual countries as a reflection of flows of trade, services and capital. With the differing systems among countries for banking regulation, for taxing of individuals and industry, for stimulating and controlling trade and for control of currency and exchange rates, many difficulties occur in building the type of information system that will adequately meet the needs of public and private decisionmakers involved in international commerce.

The need for substantial improvement in international statistics is clear from the failure of reports from all countries to balance. The individual categories in the world tabulation of the balance of payments should, in principle, sum to zero. The world discrepancy in the current account topped $100 billion in 1990 and decreased somewhat in 1991. Errors of this magnitude indicate that many of the world's balance of payments tabulations are subject to considerable errors.(7)

Statistics for the 21st Century discusses selected problems with data collection for trade statistics, especially in the services area, for international capital flows including foreign direct investment and data on international capital markets. Two recent studies by the National Academy of Sciences have drawn attention to needs for improving trade statistics(8) and for dealing with international capital flows.(9) We make nine additional specific recommendations we believe can be implemented as a first step in addressing these issues (see Appendix 2).


The normal lags in creating new statistical series are well documented. As a minimum it takes ten years to bring a new statistical series into use. As noted earlier, agreement on the SNA revision took over a decade and implementation will take a long time. Yet there are some things that can be done now. Many of the recommendations made in Statistics for the 21st Century are the result of many years of study and recommendations from many groups. What is needed is an aggressive program to push for a faster pace of change and to consider the implications for the longer term.

Several things can be done now. We recommend the following:

1. Create a permanent Statistical Advisory Committee to consult on a continuous basis with the major statistical agencies.

2. Simplify statistical reporting to the public, especially by eliminating the second and third preliminary estimates.

3. Develop programs to get broader public input into the definition of statistical needs for the future.

A Permanent Statistical Advisory Committee.

The current system of statistics has fallen woefully behind the pace of change of the realities those statistics purport to represent. And while a conservative and incremental approach must be the constant watchword, truly effective change can only come about if it is formed by a larger vision: a vision that is clearsighted in its understanding of the current statistical system, bold in its projections of what a more adequate system will look like and aggressive in its determination to move all affected and interested parties toward making that vision a reality.

Such a layered and comprehensive vision of the required changes to the statistical system will not--to be effective it cannot--spring full grown from the mind of any single individual, nor from any single agency, bureau or institution. It must be housed in one place that can serve as a clearing house to accumulate, sort and disseminate the best ideas among all interested parties. The role of that kind of clearing house will not be to preempt a wide-ranging conversation about needed reforms but facilitate it.

It's a job for the Office of Management and Budget (OMB). Such a task is not new for OMB. While resources for statistical coordination have been diverted to other roles in the past couple of decades, we are suggesting that OMB should reprise its central role of statistical and information policy leadership. But the leadership role is not the sole responsibility of the Executive Office of the President. We have pointed out in our review that there are a large and ever-increasing number of producers and users of statistical information. Each has a crucial stake, a unique perspective and a valuable contribution to make. We propose that OMB appoint a permanent Statistical Advisory Committee to serve as a focal point for a continuous dialogue with these stakeholders which include:

1. Academicians

2. State and local governments

3. Trade associations

4. Unions

5. Environmentalists

6. Business, including both large and small private companies(10)

Improving Statistical Reporting to the Public

In recent years statistical agencies have made major improvements in describing statistical error, in explaining the difference among various series and in publishing statements and reports on important methodology issues. Yet the fact remains that much of the public does not follow these topics. Consequently, a need exists to make certain that the interpreters of the statistics have clear concepts of what is being measured and how it is being measured. This problem can be partially attacked by members of the National Association of Business Economists. Members can take special care to be certain that they are well informed on such issues and that they explain these issues to the press and other representatives of the public with whom they come in contact.

The press also has a responsibility. They should seek out the experts in statistical agencies and they should challenge the knowledge of the commentators they are interviewing. In investment circles there is a tendency to seek a consensus before official statistics are released. Yet many of the sources of consensus have no special independent information; rather they simply offer "guesstimates."

Finally, the statistical agencies should take care to help present clearer indications of what the numbers mean. I will illustrate this with the many revisions that are published concerning the Gross Domestic Product (GDP). All members of NABE are familiar with the many revisions published as part of the GDP estimates. The key estimates are the preliminary (25-day estimate, i.e., 25 days after the end of the quarter), the revised (the 55-day estimate), the final current (85 days), the 1st July revision (the July following the year end), the 2nd July revision (one year later) and the 3rd July revision (an additional year later). There are also revisions as a result of comprehensive benchmarking after data from the quinquennial results from the Census of Business are available. These revisions have long been a subject of controversy. Many observers who do not understand statistical programs think that someone is "manipulating" the numbers, yet every analysis of the process reveals that these revisions are the final information. The classical study of the revisions was conducted by Roseanne Cole.(11) A more recent study was conducted by Stephen K. McNees of the Boston Federal Reserve.(12) Currently, Martin Fleming, Chairman of the Statistics Committee of the National Association of Business Economists, has completed a more recent review.(13)

As part of the study for the present report, data were developed by the Bureau of Economic Analysis for Haver Analytics of New York. That data base was used by Martin Fleming for his study. The data were also analyzed by George Feeney of Haver Analytics as an input to our book. The Feeney study used the third year revision as an estimate of truth for the individual GDP components, while Fleming used the second year revision as the benchmark for the "true" estimate. In the Feeney study the early quarterly estimates (estimates 1, 2, and 3) and the 1st and 2nd July revisions were compared with the 3rd July revision (assumed truth) to see the pattern of variance. The analysis showed little variance reduction in the 45- and 75-day estimates.

In a forthcoming study(14) Allan E. Young notes, "...the incorporation of additional or more accurate source data in the preliminary and final current estimates of GDP does not result in substantial improvement." He notes that most observers are puzzled by this lack of improvement, and he identifies two factors that are overlooked. They are:

1. "the small role played by the second and third months of the quarter in determining the change from the previous quarter,(15) and

2. "certain sources of errors in the preliminary and final current estimates to which the advance estimates are immune."(16,17)

Because the estimates of growth in GDP show little change between the first, second and third preliminary estimates (when compared with the more complete estimates provided by the July revisions), we strongly endorse the recommendation of earlier studies, including Cole of the Federal Reserve Bank of Boston, and Fleming, who have recommended that the early revisions be simplified to one. Specifically, we recommend publishing only the first (the 25-day) estimate and dropping the 55- and 85-day estimates. Instead, the resources of the Bureau of Economic Analysis could be utilized better in refinements that contribute to the July revisions.

This illustrates the need for statistical agencies to review their releases and to seek means of reducing confusion when the press for timeliness conflicts with accuracy, and when users need to understand better the sources of errors. While many improvements have been made in this regard in recent years, it requires a continuing effort to explain and educate what the statistics really mean. Defining Statistical Needs

While a Statistical Advisory Committee will be a good focal point for learning about statistical needs, it is not sufficient. A need exists for all agencies (and, in fact, for all major individual statistical programs) to reach out to users, potential users, and nonusers to learn about what is really needed. In the terms of Vice President Gore's Reinventing Government program, the agencies need to practice "customer focus" to learn what is really valuable and what is needed that is not now available. In the past decade, U.S. business has developed a policy of being sensitive to the "customer." Official statistical agencies and other producers of statistics need to follow the same practice. Otherwise, the relationship between cost and benefits and the incentive for cooperation are reduced.


All of us have a potential role to play. Statistics for the 21st Century is an initial view to stimulate debate, discussion and new ideas. Members of the National Association of Business are key players. First, they are intensive and informed users of economic statistics. Second, they are key players in the corporations that provide so much of the basic information.

We invite you to review your company's response to governmental statistical inquiries, and we encourage you to send us your ideas about problems with existing series and needs for new series that will help you make better decisions in the future.

Appendix 1

GDP Recommendations

Recommendation 1: Eliminate the 85-day estimate. If a decision is made to continue publication of the 85-day estimates, the Census Bureau should collect quarterly data on sales of service industries for use in the final 85-day GDP estimates. In planning this survey of service industry establishments, Census should examine the ongoing Consumer Expenditure Survey of households that it conducts for the Bureau of Labor Statistics to determine if some outlays reported by households can be used to supplement the industry estimates.

Recommendation 2: If a decision is made to continue publication of the 85-day estimates, the Census Bureau should conduct a quarterly survey of expenditures of state and local governments for use in both the final 85-day GDP estimates and in the 85-day state and local budget estimates. The survey should cover compensation other than wage and salary compensation, which is the only component currently reported.

Recommendation 3: The Internal Revenue Service should tabulate data from a probability sample of the Employer's Quarterly Federal Tax Return (Form 941) on wages and social security payments for use in the final 85-day GDP estimate. Recommendation 4: The Census Bureau should conduct a comprehensive review of the scope of the Quarterly Financial Report (QFR) in consultation with the Bureau of Economic Analysis to determine where it can feasibly remedy major data gaps in the quarterly estimates of corporate profits. Based on an assessment of the quality of the existing data sources used in estimating corporate profits, the review should include the benefits and costs of: (1) expanding coverage to the construction, transportation, utilities, finance, and services industries; and (2) expanding the coverage to unincorporated businesses in those industries where sole proprietorships and partnerships are prominent. If some expansion of the QFR is warranted, research and pilot surveys will be necessary prior to operational surveys. The inclusion of banks in the survey would have to be coordinated with the banking regulatory agencies. The FY 1994 Census Bureau's budget includes a request for funds to add the business services industries (SIC 73) to the QFR. We support the request. Recommendation 5: The Bureau of Economic Analysis should examine possible sources of the large increases in the statistical discrepancy in 1991 and 1992 for: (1) clues of problems that may be developing in the GDP data base or in the BEA estimating procedures; and (2) the effect of incorporating new data improvements in the national accounts. If the increases in the statistical discrepancy are considered to reflect particular data problems, remedial action is necessary. To the extent the increases are considered to reflect the incorporation of new data improvements, no remedial action is warranted.

Recommendation 6: The Bureau of Economic Analysis should publish the GDP as estimated from the income side of the national accounts and the income-side growth rates. This will provide an immediate comparison with economic growth rates based on the product-side GDP. The income-side measure can be calculated using the data published on the statistical discrepancy. However, only sophisticated users are likely to perform this calculation and then also calculate the income-side growth rates. The income-side GDP has the same statistical validity as the product-side GDP.

Recommendation 7: The Bureau of Labor Statistics should improve the measures of price change for various high-tech products in the producer, consumer, and import/export price indexes. First, it should identify electronic, communications equipment, medical equipment, pharmaceuticals, and other high-tech products in which quality improvements have been accompanied by lower prices. Second, for these high-tech products it should develop new methods for measuring quality change similar to the hedonic technique it uses for computers in the producer price index. Third, when the same high-tech products are included in two or more of the producer, consumer, and import-export price indexes, the new procedures should be included in all of the affected indexes. Recommendation 8: The Census Bureau should develop price indexes for new nonresidential building and nonbuilding structures. These should refine the presently used input-cost indexes if feasible. However, it may be necessary to develop completely new output-price indexes for some types of structures. The FY 1994 Census Bureau's budget includes a request for funds to develop a price index for nonresidential buildings. We support the request.

Recommendation 9: The Bureau of Economic Analysis should focus on developing output measures of the productivity of government workers. The Bureau of Labor Statistics' studies of the productivity in twenty-eight federal government functions and services are a major data source for preparing federal government estimates. These studies may also be appropriate for estimating productivity in related state and local government programs.

Recommendation 10: The Bureau of Economic Analysis should revitalize the input/output program to reduce the completion time of the benchmark tables to five years after the reference year. This includes motivating experienced employees to stay with the program and developing automated procedures for estimating the intermediate production quadrant cells of the table.

Recommendation 11: The Census Bureau should tabulate certain data from the microdata files of the economic censuses for use in preparing the benchmark input/output table. Examples of such tabulations are those for small companies and central administrative offices. The FY 1994 Bureau of Economic Analysis budget includes a request for funds to reimburse the Census Bureau for such tabulations. We support the request.

Recommendation 12: When it can be foreseen that existing data sources are inadequate to develop the satellite accounts further, BEA should prepare an inventory of high-priority data needs for these accounts. Because the satellites involve the program and statistical activities of their agencies, BEA should consult with these agencies in drawing up the data-needs inventory through an interagency committee. Such an inventory will facilitate the planning of future statistical programs by putting them in an overall context. Recommendation 13: The Bureau of Economic Analysis should convene a standing group of users of the national accounts with whom it consults on a regular basis. The group should include representatives from the Council of Economic Advisers, Federal Reserve Board, Congressional Budget Office, universities, private industry, labor unions, and research organizations. The subjects would include conceptual innovations, technical issues, data presentation, or other aspects of the accounts.

Recommendation 14: The Economic Classification Policy Committee composed of the Office of Management and Budget (ex officio), Bureau of Economic Analysis, Bureau of the Census, and Bureau of Labor Statistics should make a concerted effort to complete the ongoing revision of the Standard Industrial Classification system to be incorporated in the 1997 economic census. If there is a slippage of more than a couple of months in this complete data, the revised SIC would first be incorporated in the economic statistics of the 2002 economic census.

Recommendation 15: When submitting their budgets to the executive branch and to Congress, statistical agencies should give more attention to relating the need for the data to existing or potential problems and issues of the nation. Requests should provide an example of how the data will address a concrete problem, in contrast to explaining the consequences of not having the information. This is particularly important in requests for a new program or to expand an existing one.

Recommendation 16: Federal statistical agencies should make broad-based efforts to raise the response rate on voluntary business surveys. One approach is for the agencies to convene regional or industry conferences of high-level company officials, stressing the importance of the data for public policy and the direct or indirect benefits to the company. The officials should be encouraged to institute policies within their companies to ensure that survey forms are completed and that care is given in supplying accurate information. Such private industry groups as the National Association of Business Economists, U.S. Chamber of Commerce and local Chambers of Commerce, National Association of Manufacturers, and National Federation of Independent Business should work with the federal agencies in this effort.

Recommendation 17: The federal statistical agencies should work with economics and statistics departments in universities to introduce meaningful courses in economic measurement. These may include interdisciplinary programs between the economics and statistics departments. It will involve convening conferences with college faculties, perhaps through professional organizations such as the American Economic Association and the American Statistical Association, to present the dimensions of the problem and to design appropriate course programs. Follow-up conferences should monitor the progress and modify or expand the teaching programs as experience is gained with them. The new Joint Program in Survey Methodology of the Universities of Maryland and Michigan, and Westat, Inc., would be an appropriate vehicle for organizing the conferences. Recommendation 18: Private industry companies should reverse their declining participation in government business surveys and ensure that the information they report is accurate. While industry makes only limited direct use of the national accounts, the economic well being of businesses is significantly influenced by use of the accounts to formulate and evaluate fiscal and monetary policies. Government business surveys are the lifeblood of the accounts, and the accounts underlie basic policies used to create a climate of economic growth in which businesses prosper.

Recommendation 19: The Office of Management and Budget or the Joint Economic Committee of Congress should sponsor a comprehensive assessment of the national accounts by a nongovernment group of experts from universities, research organizations, industry, and labor unions. The study would examine the concepts, methodologies, data bases, and data presentation of the accounts to reflect current and prospective needs of economic analysis and policy formulation. The beneficial results of the Creamer Report and the passage of sixteen years since the report was completed underscore the usefulness of having an independent review in the 1990s.

Appendix 2

International Recommendations

Recommendation 1: Use a supplemental balance of payments framework that captures transactions between foreign affiliates of U.S. firms abroad and U.S. parents and foreign affiliates in the U.S. and their foreign parents. This supplemental framework would dramatically change a country's international trade balance. For example, in 1987, the U.S. trade deficit was $148 billion, but computed under a supplemental framework the deficit was $68 billion. For policy purposes and trade negotiations, this dramatically alters the position of the United States. Recommendation 2: Incorporate the annual surveys of multinational companies into the balance of payments accounting. Also, incorporate data from the annual surveys of direct investment for more detailed data on affiliated trade. No distinction is made between raw, intermediate, or final goods. Components for various products are often imported from various countries and then assembled for export. With the present statistics, it is not possible to derive a value-added component.

Recommendation 3: Use universal bar coding to track the components of trade. Such an automated system would produce information about the origin of various factors of production and assist in improved classification.

Recommendation 4: Simplify the amount of data that is compiled. Although this might involve a loss of accuracy and completeness, the considerable cost savings that would be generated could be used to improve the collection of other international statistics, notable in the capital accounts.

Recommendation 5: Consistent estimation and sampling procedures need to be established. The framework developed by the U.S. should be coordinated with the major industrial countries to develop ultimately a consistent framework for use on a worldwide basis. Using estimates and simplifying the data collection process should be considered a reliable alternative to survey methods with considerable cost savings.

Recommendation 6: Integrate Immigration and Naturalization Service (INS) data - which are already being collected at the Customs Service - with a balance of payments statistics.

Recommendation 7: Data on imports need to track not only freight, but also the country of origin. A five-year benchmark survey would improve the statistical accuracy of these expenditures.

Recommendation 8: It is widely acknowledged that serious shortcomings exist in the present system of capital accounts. There are three basic approaches to revising the entire system of capital accounts:

1. Make revisions and changes to improve the accuracy of the present system. These types of changes are the easiest to implement, but the usefulness of such a piecemeal effort is questionable. A marginal improvement in the accuracy of the capital flows would not necessarily be more reliable than what presently exists;

2. Restructure the domestic system entirely and change the method of reporting. This may involve a change to annual statistics via benchmark surveys. The quarterly data are presently so volatile and subject to such tremendous revision that they are of limited use. A more accurate framework that is produced annually would represent an improvement for users, although the intrayear changes in financial conditions could not be well tracked. It may be true that a full tabulation of international capital flows is not possible. Use sampling techniques that are less costly but perhaps just as accurate in tracking capital flows. Finally, any new system must have the flexibility to include new financial instruments: and

3. Develop a worldwide system to monitor capital flows. In view of the tremendous linkages of global capital, it is becoming increasingly difficult for the U.S. or any single country to monitor adequately capital flows without international coordination and cooperation. As an alternative to the present methodology, a new system could be developed by international monetary and financial agencies with consistent concepts and definitions. In developing a worldwide system, serious consideration should be given to whether the present classifications are meaningful. Of course, there would obviously be a number of difficulties and objections to a global system, such as enforcing compliance among participating countries, ensuring confidentiality, and managing the huge number of transactions.

Revisions to the present system, or development of a new system, must begin with clarifying and redefining important concepts for consistency. There needs to be an improvement in the interaction between agencies assigned to collect, report, and analyze data. Innovations will require frequent adaptations of reporting systems. Portfolio capital needs to be expanded to include derivative financial instruments.

Recommendation 9: Under the present system, countries should adhere to the definitions in the International Monetary Fund's Balance of Payments Manual to achieve greater consistency, but there is no realistic way to enforce this. In the future, a total revamping of the international balance of payments is needed to improve timeliness and accuracy. The task of revamping the world's balance of payments system is highly complex and would require global coordination and cooperation. Changes in the U.S. could serve as a model for a new global system that reflects the linkages in the world economy.


1 The column first appeared in the July 1986 issue of Business Economics.

2 See Joseph W. Duncan, "The Statistics Corner: Planning for Statistical Developments in the Twenty-First Century," Business Economics, October 1992, pp. 63-65; and Joseph W. Duncan, "The Numbers are Starting to Lie; Restructuring Economic Statistics," Leaders Magazine, July, August, September 1992, pp. 44-45. 3 A form was enclosed in the latest issue of NABE News. If you don't have a form, see The Statistics Corner in this issue for further details about obtaining a copy.

4 For a discussion of limitations of the literature search see Appendix 1 of Statistics for the 21st Century.

5 See The Statistics Corner in this issue for a brief overview of the System of National Accounts.

6 Social conditions can be evaluated on the basis of the Social Accounting Matrix (SAM), which is a set of satellite accounts of the SNA as briefly described in Statistics for the 21st Century, pages 165-66. More details will be available in the forthcoming text to be published jointly by the United Nations, the Organization for Economic Cooperation and Development, the World Bank, the International Monetary Fund, and the European Economic Communities.

7 The International Monetary Fund has addressed many of these questions in two major studies: Report on the World Current Account Discrepancy and Report on the Measurement of International Capital Flows.

8 Anne Y. Kester, Ed. Behind the Numbers: U.S. Trade in the World Economy, Washington, D.C.: National Academy Press, 1992.

9 Anne Y. Kester, Ed. Following the Money: U.S. Finance in the World Economy, Washington, D.C.: National Academy Press, forthcoming.

10 Special attention should be given to smaller businesses whose importance and input are still underrepresented in statistics, especially as their importance in the economy as sources of jobs and innovations are growing.

11 Roseanne Cole, "Data Errors and Forecasting Accuracy," Chapter 2 in Economic Forecasts and Expectations: Analyses of Forecasting Behavior and Performance (Jacob Mincer, Ed.). New York: National Bureau of Economic Research, 1969. 12 Stephen K. McNees, "Estimating GNP; The Trade-off between Timeliness and Accuracy," New England Economic Review, January/February 1986, pp. 3-10.

13 His report to the 1993 NABE Annual Meeting in Chicago is entitled "Measurement Error in the National Income and Product Accounts--Its Nature and Impact on Forecasts, 1993."

14 Allan H. Young, "Accuracy and Reliability of the Quarterly Estimates of GDP," Conference of the Atlantic Economic Society, Philadelphia, PA, October 8, 1993 (preliminary draft).

15 "The third month of the source data in the quarter...receives a weight of only 1/9th in the determination of the quarterly change. The weight of the second and third months together is only 1/3rd." Ibid, page 19 of the preliminary version.

16 Ibid, page 18 of the preliminary version.

17 For example, seasonal adjustment of source data used for the final current quarterly estimate introduces an error that is not contained in the judgmental projections. Ibid, page 20 of the preliminary version.

Joseph W. Duncan is Vice President, Corporate Economist and Chief Statistician, The Dun & Bradstreet Corporation, NY.
COPYRIGHT 1994 The National Association for Business Economists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1994 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:National Association of Business Economists
Author:Duncan, Joseph W.
Publication:Business Economics
Article Type:Transcript
Date:Jan 1, 1994
Previous Article:The Adam Smith address: competing in the world economy: trade competition vs. growth competition.
Next Article:U.S. trade deficits and international competitiveness.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters