UNSNA, Marilyn Waring, National Accounts, national income and product accounts, accountants, labour force, accounts, national accounting, National Bureau of Economic Research, non-market, developing nations, Economic Policy, Measurement Problems, Journal of Economic Literature, The State of U.S. Economic Statistics: Current, household services, Economic Council of Canada, Economic Measurement, Income and Wealth, Information Economy, Statistics Canada, labour services, Journal of Business & Economic Statistics, Feminist Economics, national income accounts, national income, United Nations System of National Accounts, Harper Collins Publishers, Anniversary Conference, U.S. Gross National Product, Economic Systems Research, household production, statistical surveys, feminist viewpoint, National Income and Product
Review of Income and Wealth Series 38, Number 2, June 1992 REVIEW O F IF WOMEN COUNTED: A NEW FEMINIST ECONOMICS Marilyn Waring, If Women Counted: A New Feminist Economics, Harper Collins
Publishers, San Francisco
, Paperback Edition, 1990. National income accountants are in deep trouble, or so claims the book by Marilyn Waring If Women Counted: A New Feminist Economics. The reason for this is very simple: sex discrimination in national income accounting. Here is a unique book that develops this main theme with much ingenuity and scholarship. The author writes from a strictly feminist viewpoint. Her arguments embody a considerable amount of insight and substance that should be of concern to readers interested in developing national accounts to better track the changing economy (and society) in which we live. Most of the book appears to have been written about 6 years ago. This means that c k a i n aspects of the book, briefly mentioned later, do not take account of the latest planned revisions of the United Nations
System of National Accounts (UNSNA).This should be kept in mind because a good deal of the book represents a severe criticism of the standards imposed by the United Nations SNA. Nevertheless, there is much in If Women Counted that is certainly not out-of-date and this review article will emphasize these important and timely aspects of the book. In Section 2 of this review I present a brief and rather selective summary of the book's highlights-a kind of "guided tour." In Section 3 I concentrate on what seems to be the main positive contributions of Marilyn Waring's approach to recommendations for future developments of national income accounts and related statistical surveys. There is also mention of some "biased aspects" of the book where the viewpoint lacks balance. In section 4 some technical and often overlooked issues are raised. In conclusion, it seems that the "honeymoon" is over for national income accountants! The book is essentially a popular introduction to economics and particularly to the assumptions and practices underlying national economic accounting. There are detailed discussions of such topics as: Labour Economics
, statistical surveys, the history of national accounting, special problems of poverty and developing nations
, and Environmental issues
(among others). All of these topics are treated from the viewpoint of feminist economics. What then, is the "viewpoint" of feminist economics? Crudely put, we might summarize the author's perceived motivation as follows: The field of economics,
and particularly the standards of national economic accounting, is a scheme by men to dominate women or, at least, keep women in "their place." For example, by not recognizing the economic value of women's household production services, the value of nonmarket (unpaid) household labour is excluded from both sides of national income and product accounts. This standard holds for both developed and developing nations and with important consequences. As well, Waring emphasizes the absence of an official economic valuation of women's volunteer services and women's "investment" in child-rearing and reproduction activities. The spirit of the book is clearly revealed in the following quotation (p. 7) from the Prologue: The rhetoric of these accounts (UNSNA) is so removed from us (women), and requires so much in the translation, that it is difficult to breathe any life onto the page when trying to demystify them. Yet demystify them we must, because the UNSNA is an essential tool of the male economic system. So I will have to trust my sisters to stay with the text and trust. . .I show how this system (mis)treats our lives ..." The author attempts to "go behind" the official national accounting standards by interviewing some key economic statisticians, but unfortunately, all the key economic statisticians turn out to be men interested in keeping women in "their place." One cannot help asking: Why did Marilyn Waring not interview Nancy Ruggles who was very active in national accounting circles at the time the book was written? Marilyn Waring's book is really unique because it is not only an introduction to the subject matter. Although a better balanced introduction can be found in Eisner (1988) or even Peskin (1982), economic experts in various fields could benefit from an examination of such provocative Chapters as: A Calling to Account: Who Developed the UNSNA and Why The Boundary of Conception: How the UNSNA Makes Women Invisible The Statistical Conspiracy: Sources for the National Accounts Villainy and Incompetence: Problems Economists See in Their Own System The Eye of the Beholder: The UNSNA as Applied Patriarchy By the term "national accounts" we mean the national income and product accounts (NIPA) and their sectoral counterparts, since these are the accounts with which Marilyn Waring's book is primarily concerned. The limitation to NIPA is unfortunate, because if national wealth accounts were introduced into the discussion some controversial issues could be rectified. (For example Chapter 7 "The Value of Death: How War, Poverty, and Poisons Help the Economy.") The following discussion of our review is mainly within the context of developed nations
. A prime concern throughout the book is the problem of lack of official measurement of household production services; their production, consumption and imputed labour income
. Although many national accountants, working on
their own, have performed such measurements (Eisner, 1988) there is still lack of complete recognition in, e.g. UNSNA. The present reviewer believes that the author's argument is entirely on the right track. We need measurements of the value of non-market household labour in order to correctly interpret trends in officially measured (market) GDP! A major change in the economy of many developed nations during the past 25 years has been the dramatic increase in the market labour force
participation rate of women (see, e.g. Clift and Wells, 1990 for the Canadian context). When women leave household non-market activity and enter the market economy including goverment, officially measured G D P becomes (in a sense) biased upwards in cases where the newly created market activity substitutes for or replaces household non-market labour. The bias, in effect, depends on whether household services
activities continue to be performed by the new labour force participants or are taken up by other members of the household. The best way, perhaps, to judge the impact of such radical changes in labour force market Participation rates
on measured GDP, is precisely to also have official measures of the value of non-market household labour and its changes over time. The two measures together furnish an indication of total economic well-being. The standard arguments often used to discourage such measures, e.g. ownaccount household services are difficult to quantify and standardize and are consumed as they are produced, do not stand up to in-depth inspection. There is, in fact, little reason to treat own-account produced services differently than own-account produced goods-a theme pursued at length in Postner (1991). So the author's arguments arejustified in principle. However, the feminist orientation is partly misplaced because time budget trends show a significant and increasing proportion of household labour services
performed by men, both employed and non-employed. Eisner (1988, p. 1676)notes that between 1965and 1981employed males in the U.S. increased their housework from 11.7 to 13.4 hours per employed week, on average, while housework by employed females fell from 32.3 to 27.7 hours- see also Juster and Stafford (1991, p. 477). While men are not actively involved, relative to women, in child-rearing and meal-preparation activities, they are very much involved in home-repair and home-budgeting activities (as in Peskin, 1982). The book under review does not present a balanced picture of this topic. Another theme that permeates the book is the problem of measuring the value of (unpaid) volunteer labour services. Once more the author's argument is a little distorted since she assumes that volunteer work is generally done by women (p. 27). The Canadian data actually show that women compose about 55 percent of volunteer labour-i.e. men do about 45 percent of volunteer work (as in Secretary of State, 1990, Table 2, p. 12). The issue here is that the lack of official measurement of the volunteer labour force, and its imputed service value, produces a gross underrepresentation of the importance of an economy's private non-profit sector, where the volunteer labour force is naturally concentrated. In fact, (private) accounting standards boards regulating the reporting activities of non-profit entities now recommend that the imputed value of labour services donated by volunteers should be recognized in the entities' accounting statements (the standards are explained in Skinner, 1987, p. 501).
The private non-profit sector is potentially important, probably of growing importance, and deserves a distinct analysis over and above the traditional dichotomy between the private for-profit sector and the public sector
. Marilyn Waring, the author of If Women Counted, deserves credit for focusing attention on this matter with key implications for public policy
(as seen in Weisbrod, 1988). However a balanced analysis would show that there are no special implications for economic feminism. Marilyn Waring also deserves credit for her critical evaluation of the hidden assumptions
underlying statistical surveys. Through this aspect of the book is not well developed, there are a number of (implicit) suggestions that should be spelled out. With current trends towards self-employment and operating a business from the home, trends that apply to men as well as women, it is not clear that standard census and survey procedures are sufficiently sensitive. The selfemployed and small domestic producers may operate at a loss (or even negative Value Added
) after taking account of the opportunity cost
of their labour time and contributed capital as noted originally in Ruggles and Ruggles (1970). There is no need to be "romantic" about the self-employed entrepreneur who is, in effect, living off her or his capital savings! As well, the "hidden (unreported) economy" of (1) legal and voluntary transactions (such as paid household domestic services) and (2) illegal and possibiy involuntary transactions (namely criminal actions) are meant to be included in the official UNSNA, but it is not clear whether our statistical reporting systems are capable of doing the job (a point a k o recently made by Brenner, 1991). Marilyn Waring deserves credit for noting that household surveys permit sexual discrimination when the reporting unit is chosen to be the "head of the household" rather than the individuals within the household, There may be a need to deconsolidate and disaggregate the structure of household statistical reporting to be more sensitive to individual responses within the modern household since individuals may have little or no economic information about other members of the same household. (A similar position was also recently taken in the 1988 Juster Report.) A major and persistent theme throughout the book is sexual discrimination in national accounting with respect to the special conditions of developing nations. This is essentially due to the attempts of economic statisticians to draw the production boundary along similar lines for both developed and developing nations. However, if the household of developing nations is not recognized as a production unit or if a good deal of basic production activity in developing nations is not marketed, then the official statistics
would be misleading. Moreover, the evident dominating role of women in such household and non-marketed production of basic goods in developing nations would be neglected in a discriminatory manner. The argument of the book, which is here very briefly summarized, seems essentially valid. It appears, however, for the latest planned revisions of UNSNA, that serious attempts are being made to unconditionally meet Marilyn Waring's objections,
at least with respect to goods production within developing nations (see United Nations, 1990, pp. 8-9). There is still the problem of recognizing, within the production boundary, the services consumed and produced within the household-an issue, which is shared (but not equally shared) by both developed and developing nations. One should not look for technical solutions to national accounting and statistical problems in If Women Counted. There is, rather, an emphasis on the fact that national economic statisticians appear to be "dragging their feet" when it comes to establishing appropriate standards on important issues of interest to feminists (but not exclusively to feminists). There is, in effect, a tendency towards "passing the buck," as illustrated in the following quotation from Waring (1990, p. 92): "The policymakers said that they could only work with and from, the economists' models and the statisticians' figures. The economists said that they worked within policy guidelines that required particular statistics. And the statisticians said that they just produced the figures the policymakers and the economists' models required of them. No one, it seems, is responsible for the dismal state of economics." Moreover, she also notes (p. 57) that the traditionally accepted views of one of the founders of national economic accounting, Milton Gilbert: "leaves no room for alternative perceptions, for change, for flexibility, for the human response to life. It admits to a fascistic forced (statistical) consistency." One of the key assumptions of national income and product accounts is that the two sides of every economic transaction must be viewed identically by the parties involved-to make national accounting "work" (via inter-articulation). The technical implications of this universally accepted assumption have only recently been questioned and the need for alternatives suggested (see Postner, 1988). Marilyn Waring too appears to favour a relaxation of this assumption, but the technical consequences are not spelled out. (See also Waring, 1990, pp. 301-302). Nevertheless, this is an important insight. There are also suggestions throughout the book that national economic accountants have not been entirely open in declaring the reliability of their "balanced" accounts (the income and product sides are supposed to balance) even with an official statistical discrepancy. See, for example, Waring (1990, pp. 144-147) which is too extensive to quote here. However, a more concise reference comes from the recent article by de Leeuw (1990, p. 200): "Two factors tend to invalidate this statistical discrepancy as a measure of (balancing) error.. .The other is that the BEA tends to modify either the product or the income estimates when there is a troublesome swing in the discrepancy.. .the small size of the discrepancy is due to errors on both sides that offset.. ." How much do the users of national income and product accounts really know about the "modifications" of the accounts introduced by their producers in order to show "reasonable looking" statistical discrepancies? Do national accounts adjusted and modified in this "unknown" manner, reflect the real world of economic behaviour? Although the above quotation is with respect to the work
of the U.S. Bureau of Economic Analysis
(BEA), the situation in other national statistical agencies is probably no better. So Marilyn Waring is not alone. In fact we are just beginning to realize that the traditional emphasis on building a fully-integrated system
of national accounts for analytical purposes is somewhat misplaced. The old assumptions underlying, e.g. input-output analysis are increasingly difficult to apply in today's information economy (as illustrated so well in Carter, 1989). The book under review is on the right track-in a very insightful way! It is hoped that the author's uninhibited insights will stimulate technical research in areas that are currently being kept "underground." The book by Marilyn Waring shows that conventional national income and product accounts have severe limitations and must be carefully interpreted. Indeed it is becoming evident that conventional measures of GDP have been significantly "oversold7' and "overrated." The standard national income and product accounts, that are presently available, omit some vitally important aspects of the national economy that need to be revealed both for economic policy purposes and for merely tracking economic trends. The survey methodologies
and reliability indicators underlying our national accounting data, are increasingly out-of-date and still seem to be produced with an "air of secrecy." The "forced-consistency" elements that permeate the publicly available national accounts are at "odds" with economic behaviour. The days when the performance of a national economy can be measured by a unique GDP are over! HARRYH. POSTNER Economic Council of Canada
Brenner, R., Measurement Problems and Economic Policy, C.R.D.E., Cahier 0991, UniversitC de
Montrbal, March 1991.
Carter, A. P., Input-Output Recipes in an Information Economy, Economic Systems Research
, 1 (I),
Clift, B. and Wells, S., The Reliability of the Canadian National Accounts Estimates, in Canadian
Economic Observer, Statistics Canada
, Catalogue 11-010, February 1990.
de Leeuw. F.. Reliabilitv of the U.S. Gross national product
. Journal of Business & Economic Statistics,
~ ~ r19i90l.
o f Economic
Juster, F. T., The State of U.S. Economic Statistics: Current and Prospective Quality Policy Needs,
and Resources, Conference on Research in Income and Wealth, 50th Anniversary Conference,
Juster, F. T. and F. P. Stafford, The Allocation of Time: Empirical Findings, Behavioral Models, and
Problems of Measurement, Journal of Economic Literature
, June 1991.
Peskin, J., Measuring Household Production for the GNP, Family Economics Review, 3, 1982.
Postner, H. H., Linkages Between Macro and Micro Business Accounts: Implications for Economic
Measurement, The Review of Income and Wealth, 34 (3), 1988.
Postner, H. H., The Goods/Services Convergence Hypothesis: An Analysis, Working Paper
Economic Council of Canada, 1991.
Ruggles, N. and Ruggles, R., Design of Economic Accounts, National Bureau of Economic Research
General Series 89, Columbia University
Press, New York, 1970.
Secretary of State, Economic Dimensions of Volunteer Work in Canada, paper prepared under contract, Department of Secretary of State, Government of Canada, January 1990. Skinner, R. M., Accounting Standards in Evolution, Holt, Rinehart and Winston of Canada, Ltd., Toronto, 1987. United Nations Statistical Office, System of National Accounts (SNA) Review Issues: Discussion Paper for 1990 Regional Commissions Meetings on SNA, prepared by Inter-Secretariat Working Group on National Accounts, February 1990. Waring, M., If Women Counted: A New Feminist Economics, Harper Collins Paperback Edition, 1990. Weisbrod, B. A., Nonprojit Economy, Haward UNIVERSITY PRESS
, Cambridge, Massachusetts, 1988.