Since you are a consumer of information as well as commodities, you may know that economists typically measure “consumption” as money spent purchasing goods and services. Guess what. This is misleading.  We all produce and consume services that we provide for ourselves, other household members and especially for children and adults in need of care. We also produce and consume many services provided by government that don’t have an explicit price tag. In both cases, we can ask what these “nonmarket” services would cost if we had to purchase substitutes for them. An estimate of total consumption would be a better measure of living standards than money expenditures alone, and it is based on an estimate of the value of unpaid work.

It’s not easy to provide such an estimate since the entire infrastructure of survey data in the U.S. (as in other countries) was constructed on the assumption that unpaid work has no economic value. This infrastructure also reflects disinterest in the economic value to households of  government. The taxes that people pay are far more visible than what they would have to pay out of pocket for public services such as health care, education, and social services as well as spending on roads, utilities, the military and sundry other items).

I recently coauthored a report from the Levy Economics Institute commissioned by the U.S. Bureau of Labor Statistics aimed at overcoming these limitations by combining data from several different surveys. The report itself is highly technical, largely focused on description of the statistical methods used to combine data from similar households in the American Time Use Survey,  the Consumer Expenditure Survey, and the Survey of Income and Program Participation. This new data set will hopefully become available to the public sometime later this year and it includes some illustrative results. A plausible lower-bound valuation of the unpaid household work that is predominantly performed by women, for example, increases the estimate of average household consumption by about two-thirds.

This news sparked controversy on the listserv of the International Association for Feminist Economics. Several posts expressed objections to any market valuation of non-market work, on the grounds that it is implausible, politically  dangerous, or irrelevant to feminist public policy concerns. These objections deserve serious discussion and debate, because they reflect some very real risks of imputing value based on replacement costs.

Such an imputation can only offer a lower-bound estimate of unpaid work, because not all unpaid work has market substitutes, and much of it is qualitatively different in both motivation and quality. To equate market and non-market work is to ignore differences in the social and cultural organization work across these realms. However, these two forms of work are related. Consider the following counterfactual question: If you did not perform a specific non-market service (such as home cooking or child care) how much would you have to pay someone to provide the closest available substitute?

Many of us buy just about as many meals as we prepare at home, and we purchase child care services as well as services for the frail elderly and people experiencing disability.  Any call for greater public provision of care services presumes that paid care workers  can lighten the load of unpaid family care. It doesn’t presume that it can or should completely eliminate unpaid family care. 

It seems contradictory to insist that unpaid work has value but then decline to offer even an approximation of that value. How else can we make the case for greater public support for it—whether this takes the form of family allowances, a larger child tax credit, universal basic income or greater provision of public services?

Still, there’s a horrible circularity in play. Partly because many women are expected to commit to family care, they face discrimination and lower earnings in the labor market. So, when an economist like me turns around and uses their hourly wage to estimate the value of family care, I am, in a way, ignoring an underlying problem. Except that, if no lower-bound estimate is offered, the effective valuation is a big fat zero.

One consequence is that women’s earnings are counted as a contribution to Gross Domestic Product, but their unpaid work is not. When women or men decrease their hours of wage employment to tend to family and community needs, Gross Domestic Product goes down, implying that we’re all worse off. We’re not.

Which is why I think we should try to improve methods of valuing unpaid work rather than abjuring valuation altogether. The Levy Institute report moves in this direction by including supervisory as well as active child care and using relatively high wage rates (those of preschool and kindergarten teachers rather than general housekeepers) as a replacement value.

Another possible risk is that assigning a value to unpaid work could make poor people appear to be much better off relative to the rich than they are in money terms, and therefore sap anti-poverty efforts. This is not a far-fetched concern, giving the shocking level of misinformation circulating in the world today. Surprisingly, however, in the U.S., low-income households don’t devote significantly more time to unpaid work than their higher-income counterparts.

Also, imputation of the value of non-market work has a somewhat equalizing effect on the distribution of household income in the cross-section but probably accentuates trends in inequality over time. Whenever households have the opportunity to produce more for their own use, they gain a buffer from the ups and downs of the business cycle, ability to improve their own living standards independently of earnings. This opportunity declines in the course of economic development.

Back in the 1950s, many U.S. households included a full-time housewife/ mom who performed similar quantities of unpaid work of similar replacement cost value. Their contributions had an equalizing effect on the household income distribution. As more women entered wage employment in succeeding years, this equalizing effect diminished, because there is far more variation in women’s market earnings than in the lower-bound valuation of their non-market work.

Also, it’s important to note that if the definitions of household consumption and income changed, the definition of the poverty line itself should be modified. 

Feminist policy concerns shouldn’t be dominated by dollar values. The length of women’s total work days and the quality of their family and social environments clearly matter. Does attention to dollar values crowd out consideration of such larger concerns? Maybe. However, avoiding any dollar valuation reinforces the dominant view that unpaid care is not really work and therefore deserves no public support. This, in my view, is the more dangerous risk.

The cover picture is designed by Nancy Folbre
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.