IN MY OWN CLOUD of economics anxiety, I never would have thought I could get excited about two such dry and numerical entities as a census and an accounting system. Certainly, I didn’t imagine there were two semiconsistent sets of values being imposed on all the diverse cultures and corners of the globe. But that’s exactly what has been happening with the census that each country conducts every five or ten years, and with the United Nations’ System of National Accounts (UNSNA), a Western-invented accounting system that has been imposed worldwide. They sit at the heart of obscurantist economics like twin minotaurs at the heart of a maze. The path to them may be tortuous, academic, and mysterious, but once you get there, you’re confronted with two forces that are all-powerful but also simple. The census decides what is visible. The national system of accounts decides what is valuable. Anyone who is concerned about where the world is being led by current values, whether it’s rendering unimportant some groups of people or rendering invisible the entire environment, will have to convert these guardians.

In the United States, the once-a-decade national census was a way of apportioning electoral representation. That was the purpose for which it was originally mandated by the Constitution. By the mid-nineteenth century, however, its questions had grown from six to seventy, and the answers had become the main source of information for legislation, economic planning, educational systems, and social services. Because it purposefully started out to include only a percentage of slaves as property and to exclude Native Americans, its greatest controversies have centered not around what kinds of questions it asked about work, but whether it even included everyone in the country. Most recently, the question has been whether its methods undercount racial minorities, migrants, immigrants, the homeless, and other groups, whose census invisibility deprives them of economic, social, and political power.

In the 1970s, the modern women’s movement waged its first battle against sexist census categories and information-gathering methods, mainly the “head of household” category, which imposed a hierarchy on families and meant that men were more likely to answer the questions on everyone’s behalf. Now, one “reference person” may answer questions for the household in the United States, as in Canada and other countries; a great improvement, but still a problem. Until the census is focused on individuals, not households, the situation of women and children may continue to be distorted—just as it might be if there were only one vote per household. There is such a wide range of constituencies with an interest in Census Bureau policies that journalists have coined the phrase “census politics.” But social justice movements haven’t yet focused on the fact that census categories also determine what is counted as work, who is defined as a worker, how we conceptualize the health and progress of the country, how class is measured, what is counted as social mobility—and much more. In fact, the census is the one populist instrument that feeds almost every other defining and decisionmaking process in the country. What doesn’t get included may continue to exist, but it exists in the dark.

In the 1980s, when the United Nations’ Decade for Women exerted its educational force on both governments and women’s groups, women in many countries began to stage a not-so-quiet revolt against census definitions of what was and wasn’t productive work. Sometimes they worked within the government, changing definitions to include what they themselves knew to be productive. In Burkina Faso, the West African nation known as Upper Volta when it was a French colony, the economic activity of women hadn’t been counted at all unless they could be categorized as functionnaires—rare in a 92-percent agricultural country. But the end of colonialism, a progressive government, and more women in government jobs (partly because many men had been forced to find better-paid work in other countries) combined to produce a new determination to count all productive work, whether it was “paid” in a Western sense or not. Everyone over ten was asked how most of their time (principal occupation) as well as the next largest time share (secondary occupation) was spent. The result was stunning. Only in cities where jobs were salaried were women more likely to be more economically inactive than men. One result was a 1991 resolution to restore the women’s land and irrigation system rights that had been taken away by Western, patriarchal land registry systems.* In other African and Asian countries, sophisticated women’s groups campaigned outside the government to change the census takers as well as their questions, and used the work of specialized United Nations agencies such as the U.N. Statistical Commission and the International Research and Training Institute for the Advancement of Women (INSTRAW) to buttress their arguments. In India, women’s organizations successfully agitated for more women to be hired as census takers and for the training of all census takers, male or female, in “special efforts for capturing women’s work by asking probing questions,” as its census instructions now say. The parameters of productive work still don’t include “producing or making something only for the domestic consumption of the household,” which is more likely to be done by women, and do include “persons who cultivate land to produce for domestic consumption only,” who are more likely to be men. Nonetheless, the inclusion of a wide variety of unpaid productive work has made women more economically visible in India than ever before, and thus increased the possibility of accurate economic planning. Just by switching from a more Western-influenced, monetary definition of productive work to the 1982 International Labor Organization’s criteria,† for example, the estimated proportion of economically active women in India went from 13 percent to 88 percent.

In more industrial and technological countries where the exclusion of homemaking and other unpaid work carries less penalty for government planning than the exclusion of female food producers in agricultural countries, census methods have changed much less. Protests have to be more forceful. From postcard campaigns to demonstrations and even refusing to answer census questions as a form of civil disobedience, protests have been getting more informed and spirited in the last few years. In Canada, for instance, Carol Lees, a homemaker in Saskatoon, looked at a 1991 census question about “number of hours worked in the past week,” realized she would have to answer “zero” by census definitions, and decided to celebrate March 8, International Women’s Day, by writing the following letter to the minister in charge of the census that Canada takes every five years:

I am attempting to initiate a national campaign that would be very troublesome for you. It has come to my attention that in the upcoming census there is no classification recognizing labour performed in the home as work. Since I have worked full-time within the home for the past 13 years raising three children, I take exception to the fact that my labour is not defined as productive. As a result of the exclusion of women’s labour from information gathering and dissemination, we are denied proper access to programs and policy at every level of government in every country.

I am aware of the penalties I may face for my actions, and will have no trouble in dealing with them. It is my understanding that I face either a $500 fine or three months in jail for refusing to provide information for the census. I could not pay such a fine as my income as a home manager is so limited. The government will not show well if it levies a fine on a mother of three with no income because she is refused recognition for her labours in raising her children. Removing the mother from the home to send her to jail will not go down well either.

This is a very interesting issue to be resolved and a challenge to both of us. I wish us well in finding an amicable solution.40

Though Lees and thousands of other grassroots census resisters did not succeed in changing work definitions, the government did admit defeat by levying no charges against them, though their postcard campaign made their civil disobedience very clear. When Statistics Canada proposed sampling the population on homemaking work instead (Canada is already ahead of the United States in its ability to do this, as well as in its understanding of comparability standards), Lees and others made clear that wasn’t good enough. “No other sector of the labor pool would consent to be ‘sampled’ rather than fully included,” as she explained to me, “so why should homemakers? I’m good and damn mad.” Her activism had come about as a result of the distance between what she experienced as the reality of productive work and what the census and the economic structure admitted as work. “I’d just been meeting with a group of other homemakers,” she explained, “and we talked about how to answer when someone says, ‘What do you do?’ Then I got the census question telling me I did nothing—and it was just too much. I’m an ordinary person who wants to be counted, as everyone deserves to be. There’s a lot of sentiment and support on our side.”

In anticipation of the 1996 census, she and other Canadian women have organized a group called Work Is Work Is Work and are circulating test questions for inclusion in the census. As she says, “A lot of us are never going to give up until we’re counted.”41

The Beti women of southern Cameroon have more in common with Carol Lees and her colleagues than economic institutions would like to admit. They, too, have been rendered invisible. Their plight has been recognized in this United Nations description: “Beti women labour for 11 hours a day. Five hours are spent on food production … they devote three or four hours a day to food processing and cooking and two or more hours to water and firewood collection, washing, child care and tending the sick. In addition to their family specific duties, women are often involved in community projects, such as the installation of pumps, wells, schools, and health care centres.” Beti men work about seven and a half hours on their cocoa plots, palm wine production, and house building or repair, yet as the U.N. report points out: “most conventional studies would count the male as the active labourer and the wife as simply a helper to her husband. Moreover, as tangible evidence of their status, the men would be credited with decision-making power.”42

This international revolt against patriarchal definitions of work is growing. Since the late 1980s, women’s international pressures on their national census and the UNSNA (United Nations System of National Accounts) categories have taken the form of petitioning their governments, boycotting the census, or writing in what women really do, whether there were categories or not. Imputing value to homemakers’ productive work has focused on everything from social security and disability benefits to the still losing battle to define women on welfare who care for young children as working. There has yet to be much focus on “the second shift,” that is, valuing the second job of homemaking done by women (and a few men) who are also in the paid labor force. Except for a few U.N. studies on the economic importance of breast-feeding, there has been no emphasis on valuing reproductive work. But now, there is also a growing cadre of women who crisscross the globe, carrying economic organizing advice from one country to the next.*

In New Zealand, Marilyn Waring wrote about the objections of “conceptual difficulties” and “problems of data collection” that were raised against including unpaid work in the 1986 census; difficulties that arose because “too many women did too much work.” But she and others were successful in adding the question: “What is your main work or activity?” Possible answers included “home duties” and “looking after children,” as well as such alternatives as “unpaid work in a family business.” As Waring wrote: “The women of New Zealand have shown how to challenge the system, at least in regard to unpaid productive work. But a woman still has no way of informing the census about her reproductive work.”43

“The combination that works,” said Waring in 1994 after a decade of such efforts, “is grassroots women organizing in a forceful way, plus feminist experts inside the system—someone like Joann Vanek, in the U.N. Statistical Office in New York, and a few other hardy souls—who can have enormous impact if they have international women’s efforts to back them up and keep them informed. Changing a phrase or adding a concept to a social instrument like the census has leverage like nothing else. Individual women just have to pressure their governments, refuse to cooperate in their own invisibility, and insist on reporting the work they really do, whether it fits the census categories or not.”44

What would happen if the women of the world answered questions like those in the Burkina Faso census about the work really done and the time really spent? I always knew Muriel Rukeyser was right when she said:

What would happen if one woman told the truth about her life?

The world would split open.45

I just didn’t know her poetry applied to the census.

* The imposition of patriarchal and individualized ownership has been a major cause of land loss in parts of the world where land was passed through the woman’s family or owned communally. In the U.S. even jointly worked family farms were considered to be the property of the husband. If he died first, the wife was required to pay inheritance taxes, and her inability to do so often caused her to lose her land. If the wife died first, there were no such taxes, since the farm was considered his. It was for this reason that rural women’s groups supported the Equal Rights Amendment, which would have eliminated the basis for these laws and the need for fighting them one by one. These and many other stories lie behind the United Nations statistic that women own only one percent of the world’s property.

† In the ILO definition, “the economically active population includes all persons of either sex who provide labour for the production of economic goods and services. All work for pay or in anticipation of profit is included. In addition, the standard specifies that the production of economic goods and services includes all production and processing of primary products, whether for the market, for barter or for home consumption.” The ILO acknowledges that this still does not capture the domestic service work that occupies 63 hours a week for rural women in Pakistan and 56 hours a week for women in advanced technological societies, but compared with most current practices, enacting this definition in all countries would be a great leap forward.

* And who are part of a fund of expertise that has grown over the last twenty-five years, from a few pioneers trying to make the overdeveloped world listen—for instance, Swedish economist Ester Boserup in her 1970 book, Women’s Role in Economic Development, which added a new gender analysis to neoclassical economic assumptions—to networks of feminists from developing countries who are changing those assumptions in research, analysis, policy making, and activism—for instance, DAWN (Developing Alternatives with Women for a New Era) with leaders like Devaki Jain from India and Peggy Antrobus from the Caribbean. For information, write DAWN, School of Continuing Education, University of the West Indies, Barbados, West Indies. Fax 809-426-3006. For suggested reading, see endnotes.