Feminist Explanations for the Feminization of Poverty
Pressman, Steven, Journal of Economic Issues
It is well known that women are much more likely to be poor than men. This is true in the United States (Pearce 1978; Pressman 1988) and in most developed nations (Casper, McLanahan, and Garfinkel 1994; Pressman 2002). But the causes of this phenomenon remain a matter of dispute. In a previous paper (Pressman 2002), I examined demographic and human capital explanations for the feminization of poverty and found them both lacking in empirical support. Instead, the impact of fiscal policy on the distribution of income was found to be the main reason that women in the United States are more likely to be poor than women in other countries.
This paper looks at two feminist explanations for the feminization of poverty. First, there is the issue of household structure. Parenthood, it is well known, leads to lower earnings for women (Budig and England 2001; Folbre 1987; Waldfogel 1997). There are many reasons for this. Female parents will have care-giving responsibilities for their children. This takes away from the time that they have available to earn incomes. It may also prevent women from taking jobs that require longer hours and substantial travel. These jobs, of course, are likely to come with higher pay. Furthermore, families headed by a single mother are likely to have just one adult earner. This not only reduces household income but also makes household income susceptible to large fluctuations as a result of either a bad labor market or bad luck. When there is only one earner and that earner gets laid off, gets sick, or gets reduced hours due to an economic slowdown, the household is more likely to wind up in poverty because there is no on e else in the household who can make up for the lost income.
Second, there is the issue of occupational sex segregation. If women are systematically excluded from higher-paying occupations, their wages and incomes will be lower than the wages of men (Bergmann 1986; Hudson and England 1986; Zellner 1972). In a series of controlled experiments, Rich and Riach (1995) found that women were systematically excluded from higher-paying jobs at the same time that men were excluded from lower-paying jobs. Because women are relegated to poorly paying jobs, households headed by women should stand a greater chance of being poor.
This paper seeks to examine whether either household structure or occupational sex segregation can help explain the relatively high poverty rates experienced by female-headed families. It should be noted that these are not two separate and distinct theses. Single motherhood may relegate women to certain low-paying occupations; likewise, occupational sex segregation may reduce the opportunities women have to meet men who might become their partner.
The empirical work below relies on the Luxembourg Income Study (LIS), an international database containing comparable socioeconomic data for more than two dozen nations. (1) Poverty is defined as having an adjusted household disposable income that is less than 50 percent of the median adjusted disposable income of one's country. (2) Income adjustments are necessary when measuring poverty to account for different income needs of households of different sizes. In the empirical work that follows we use the household adjustments suggested by Ruggles 1990, where household disposable income is divided by the square root of the number of people in the household. (3)
Gender Poverty Gaps
Table 1 presents data on poverty rates for female-headed households (FHHs) and all other households whose head is under 60 years old. I focus exclusively on non-elderly households in order to net out the impact of national retirement systems on my results. If the problem facing FHHs is occupational sex segregation, we need to restrict our attention to women who are employed rather than collecting retirement income. In many developed countries, eligibility for retirement programs begins at about the age of 60 (Blondal and Pearson 1995, table 6). Even in the United States, it is possible to begin collecting Social Security benefits at age 62.
Table 1 is divided into two parts. Looking at disposable income, 19.8 percent of non-elderly FHHs are poor, while 8.7 percent of other non-elderly households are poor. The difference between these two figures, "the gender poverty gap," shows that FHHs are about 11 percent more likely to be poor than other households on average.
But gender poverty gaps vary considerable from country to country. First, there are three countries where non-elderly FHHs are about as likely to be poor as other families (Poland) or are slightly more likely to be poor (Hungary and Russia). Most countries fit into the second grouping, where FHHs are about 10 percent more likely to be poor than other households. Third, in a few countries, non-elderly FHHs are more than 15 percent more likely to be poor than other families. These countries are Australia (17.2 percent), Canada (21.5 percent), Germany (17.5 percent), and the United States (21.9 percent).
The results for Australia, Canada, and the United States are not surprising, given previous work on the issue of women and poverty in an international context. But the results for Germany stand in sharp contrast to estimates of FHH poverty using other LIS waves. These results may be due to the more rapid changes in the eastern half of Germany following the end of socialism.
My previous work (Pressman 2002) found fiscal policy to be a key cause of gender poverty gaps across nations. The right-side columns of table 1 support this result. These figures were derived using factor incomes alone, ignoring any impact of government redistributive effects through taxes and transfers. They also ignore any inter-household transfers, which turn out to be minimal for most households. The figures show what fraction of households do not have poverty-level disposable income based on their factor income alone, and support several conclusions. First, without any government redistributive efforts poverty rates are much higher for both FHHs and other households. Second, poverty gaps are much larger without fiscal policy; fiscal policy cuts the gender poverty gap in half. Third, poverty rates and poverty gaps are relatively uniform across countries when measured in terms of factor incomes. Therefore, differences in government tax and spending policy account for the large cross-national differences i n the gender poverty gap that we saw in table 1.
Some Feminist Explanations of the Gender Poverty Gap
We now examine the two feminist explanations of the gender poverty gap discussed in the first section.
Table 2 employs shift-share analysis to examine the impact of the number of earners on poverty rates and the gender poverty gap. FHHs generally have just one adult to work and earn income while a large fraction of other households have two adults. For this reason FHHs should be greatly disadvantaged when it comes to earnings and factor incomes. Moreover, as noted above, females heading a household have child-rearing responsibilities which limit the number of hours they can work and the sort of incomes they can earn.
The results of table 2 appear in two parts. First we look at disposable incomes, and then we look at factor incomes. Column 2 repeats the results of table 1. Column 3 examines the extent to which poverty for FHHs is due to their lack of earners. It recalculates poverty rates for FHHs as the weighted average of poverty rates for households with different numbers of earners but assumes that the distribution of the number of earners for FHHs is the same as that for other households. The poverty rate for FHHs of each type is assumed to remain constant in this shift-share exercise; only the distribution of the number of earners changes. The results in column 4 show the change in the poverty gap had FHHs had the same number of earners as other households.
The results of this exercise are quite striking. The poverty rate for FHHs in most all countries approaches the poverty rate for other households, and the gender poverty gap virtually disappears in most countries. The main exceptions are the United States and Canada, the two countries that began with the largest gender poverty gaps.
The last columns of table 2 repeat this analysis using factor income instead of disposable income. Again, the results are quite striking. Had FHHs been able to work as much as other families, the gender poverty gap would have been about two-thirds lower. Since the poverty rates for other families remain the same in this analysis, the entire decline is the result of lower poverty rates for FHHs as a result of these families having more earners.
Table 3 employs shift-share analysis to examine the impact of occupational sex segregation on poverty rates across our LIS countries. One key question here concerns the level at which occupational sex segregation applies and thus the level at which shift-share analysis should be employed. At one extreme, it is possible to view every job as a separate occupation. This makes the theory trivially true. Occupations with women, by definition, have only women in those jobs, and occupations with men have only men in them. In this case, if women were shifted to men's jobs and received the same income that men receive at these jobs, they would have higher incomes and poverty rates approaching those of non-FHHs. At the other extreme, however, it is possible to view everything as just a job or occupation. This makes the theory of occupational sex segregation trivially false. Women do not have their own occupations, since all occupations are one. This being the case, shift-share analysis is impossible to do, and no occup ational shifting would affect women's earnings or their poverty rates. As we move from one extreme to the other, as we define occupations more and more broadly, the occupational sex segregation hypothesis will turn out to be false to a greater and greater extent. Further complicating matters, there is no agreed upon convention on where to draw occupational boundary lines, and theory cannot help whatsoever in resolving this question.
In what follows I attempt to steer a middle course between these two extreme positions. Ten or so broad occupational categories are distinguished for each country. The categories used for Australia are fairly standard across countries-(1) managers and administrators, (2) professionals, (3) para-professionals, (4) trade persons, (5) clerks, (6) salespersons and personal service workers, (7) plant and machine operators, (8) laborers, (9) other, and (10) not applicable. We now ask what would happen if FHHs were distributed among these occupations as male household heads are distributed among them.
Table 3 shows the results of this analysis for the ten LIS countries where sufficient data exist for such an analysis. Table 3 shows that occupational sex segregation does make some difference, but it does not make as much of a difference as the number of earners per household. Looking at disposable income, the gender poverty gap would have been nearly 3 percentage points (or about 20 percent) lower had women household heads been employed in the same occupations as male household heads. The decline is especially pronounced in two countries with very large gender poverty gaps. In the United States, the gender poverty gap falls by around one-third, while in Australia the gender poverty gap falls by nearly two-thirds.
The last three columns of table 3 focus on factor incomes rather than disposable incomes. It shows that the gender poverty gap would have been 5.7 percentage points (or about 25 percent) lower had women household heads been employed to the same extent as male household heads in high-paying occupations. Again, the decline is most pronounced in the United States and Australia. As with table 2, the declines in table 3 result from lower poverty rates due to better jobs held by FHHs.
Conclusion and Policy Implications
We have seen that three inter-related factors can explain the gender poverty gap for non-elderly households. First, for a number of reasons, the labor force participation of FHHs is not likely to be the same as that of other households. This reduces the household income of FHHS and increases the chance the FHHs will be poor. This was found to have a major impact on female poverty and to be a major cause of the gender poverty gap. Second, women household heads work in different sorts of jobs from male household heads. Had this not been the case, female poverty rates and the gender poverty gap would each have been around 20-25 percent lower on average across several countries.
Finally, due to these labor market facts, FHHs must rely on government support and assistance to stay out of poverty. This requires that the government assure inter-household transfers to single parents or sufficient transfers themselves through government tax and spending policy. In many countries throughout the world, this has not occurred. The result is high poverty rates for FHHs and large gender poverty gaps.
Table 1 Poverty Rates of Non-elderly Households Country Based on Disposable Income Poverty Poverty Rate Gender Poverty Gap Rate of of Other (Female Poverty Female-Headed Households Rate Minus Other Households Poverty Rates) Australia (1994) 27.6 10.4 17.2 Austria (1995) 22.0 8.5 13.5 Belgium (1997) 16.5 6.0 10.5 Canada (1994) 32.1 10.6 21.5 Czech Republic (1996) 15.9 2.5 13.4 Denmark (1995) 22.2 9.3 12.9 Finland (1995) 11.4 4.9 6.5 France (1994) 18.9 6.6 12.3 Germany (1994) 22.8 5.3 17.5 Hungary (1994) 11.7 8.8 2.9 Israel (1997) 23.5 10.5 13.0 Italy (1995) 19.4 13.6 5.8 Luxembourg (1994) 11.9 2.5 9.4 Netherlands (1994) 20.5 7.5 13.0 Norway (1995) 17.3 6.6 10.7 Poland (1995) 11.2 11.4 -0.2 ROC Taiwan (1995) 12.4 3.8 8.6 Russia (1995) 23.2 21.0 2.2 Sweden (1995) 18.7 9.8 8.9 United Kingdom (1995) 21.7 9.6 12.1 United States (1994) 34.8 12.9 21.9 Averages (unweighted) 19.8 8.7 11.1 Country Based on FactorIncome Poverty Rate Gender Poverty Gap of Female- Poverty Rate (Female Poverty Headed of Other Rate Minus Other Households Households Poverty Rates) Australia (1994) 42.0 17.1 24.9 Austria (1995) N.A. N.A. N.A. Belgium (1997) 42.1 14.2 27.9 Canada (1994) 45.8 18.1 27.7 Czech Republic (1996) 30.2 7.7 22.5 Denmark (1995) 44.2 18.2 26.0 Finland (1995) 42.3 21.3 21.0 France (1994) 43.0 19.4 23.6 Germany (1994) 37.1 11.2 25.9 Hungary (1994) 44.4 30.4 14.0 Israel (1997) 48.3 17.8 30.5 Italy (1995) 41.0 21.6 19.4 Luxembourg (1994) 31.9 13.1 18.8 Netherlands (1994) 48.2 16.1 32.1 Norway (1995) 41.7 13.7 28.0 Poland (1995) 50.9 36.9 14.0 ROC Taiwan (1995) 26.5 7.7 18.8 Russia (1995) 50.1 33.8 16.3 Sweden (1995) 47.0 24.4 22.6 United Kingdom (1995) 55.9 20.6 35.3 United States (1994) 40.3 15.9 24.4 Averages (unweighted) 42.6 19.0 23.7 Source: Luxembourg Income Study, Wave IV. Table 2 The Impact of Household Earners on the Gender Poverty Gap Country Gender Poverty Gap Gender Poverty Gap Change (Using Disposable) with Other # Earners Australia (1994) 17.2 4.5 -12.7 Austria (1995) 13.5 1.5 -12.0 Belgium (1997) 10.5 1.3 -9.2 Canada (1994) 21.5 7.6 -13.9 Czech Republic (1996) 13.4 3.2 -10.2 Denmark (1995) 12.9 2.9 -10.0 Finland (1994) 6.5 -0.1 -6.6 France (1994) 12.3 5.7 -6.6 Germany (1994) 17.5 5.5 -12.0 Hungary (1994) 2.9 2.9 0.0 Israel (1997) 13.0 0.4 -12.6 Italy (1995) 5.8 0.8 -5.0 Luxembourg (1994) 9.4 4.3 -5.1 Netherlands (1994) 13.0 4.6 -8.4 Norway (1995) 10.7 1.0 -9.7 Poland (1995) -0.2 -3.8 -3.8 ROC Taiwan (1995) 8.6 4.1 -4.5 Russia (1995) 2.2 -5.5 -7.7 Sweden (1995) 8.9 0.5 -8.4 United Kingdom (1995) 12.1 -1.2 -13.3 United States (1994) 21.9 10.5 -10.4 Averages (unweighted) 11.1 2.4 -8.7 Country Gender Poverty Gap Gender Poverty Gap Change (Using Factor with Other Income) # Earners Australia (1994) 24.9 8.4 -16.5 Austria (1995) N.A. N.A. N.A. Belgium (1997) 27.9 0.1 -27.8 Canada (1994) 27.7 10.9 -16.8 Czech Republic (1996) 22.5 2.0 -20.5 Denmark (1995) 26.0 10.2 -15.8 Finland (1994) 21.0 3.5 -17.5 France (1994) 23.6 14.8 -8.8 Germany (1994) 25.9 9.0 -16.9 Hungary (1994) 14.0 3.7 -17.7 Israel (1997) 30.5 9.0 -21.5 Italy (1995) 19.4 8.4 -11.0 Luxembourg (1994) 18.8 5.8 -13.0 Netherlands (1994) 32.1 17.5 -14.6 Norway (1995) 28.0 9.6 -18.4 Poland (1995) 14.0 -0.7 -14.7 ROC Taiwan (1995) 18.8 11.1 -7.7 Russia (1995) 16.3 0.7 -15.6 Sweden (1995) 22.6 11.6 -11.0 United Kingdom (1995) 35.3 5.6 -29.7 United States (1994) 24.4 10.8 -13.6 Averages (unweighted) 23.7 7.2 -16.5 Source: Luxembourg Income Study, Wave IV Table 3 The Impact of Occuaptional Sex Segregation on the Gender Poverty Gap Country Gender Poverty Gap Gender Poverty Gap Change (Using Disposable) with Male Occupa- tional Distributions Australia (1994) 17.2 6.3 -10.9 Canada (1994) 21.5 18.0 -3.5 Czech Republic (1996) 13.4 9.7 -3.7 Finland (1994) 6.5 4.3 -2.5 France (1994) 12.3 9.2 -3.1 Germany (1994) 17.5 16.8 -0.7 Hungary (1994) 2.9 1.2 -1.7 Poland (1995) -0.2 2.3 +2.5 ROC Taiwan (1995) 8.6 9.7 +1.1 United States (1994) 21.9 14.8 -7.1 Average (unweighted) 12.2 9.2 -2.9 Country Gender Poverty Gap Gender Poverty Gap Change (Using Factor In- with Male Occupa- come) tional Distributions Australia (1994) 24.9 9.7 -15.2 Canada (1994) 27.7 24.0 -3.7 Czech Republic (1996) 22.5 14.0 -8.5 Finland (1994) 21.0 11.6 -9.4 France (1994) 23.6 20.8 -2.8 Germany (1994) 25.9 18.8 -7.1 Hungary (1994) 14.0 12.8 -1.8 Poland (1995) 14.0 11.5 -2.5 ROC Taiwan (1995) 18.8 19.5 +0.7 United States (1994) 24.4 17.3 -7.1 Average (unweighted) 21.7 16.0 -5.7 Source: Luxembourg Income Study, Wave IV
(1.) For more information about the Luxembourg Income Study see Smeeding 2002 and the LIS website at http://www.lisproject.org.
(2.) This relative definition of income is preferable in cross-national studies for a number of reasons (see Pressman 2002, 19-22).
(3.) Sensitivity analyses, using different equivalence scales such as that proposed by the OECD, find no change in the over (relative) results; however, individual numbers will differ based upon the equivalence scale chosen.
Bergmann, Barbara R. The Economic Emergence of Women. New York Basic Books, 1986.
Blondal, S., and M. Pearson. "Unemployment and Other Non-Employment Benefits." Oxford Review of Economic Policy 11(1995): 136-169.
Budig, M., and Paula England. 'The Wage Penalty for Motherhood." American Sociological Review 66, no. 2 (2001): 204-25.
Casper, Lynn, Sara McLanahan, and Irwin Garfinkel. "The Gender Poverty Gap: What Can We Learn from Other Countries?" American Sociological Review 59 (August 1994): 594-605.
Folbre, Nancy. "The Pauperization of Motherhood: Patriarchy and Public Policy in the United States." In Families and Work, edited by N. Gerstel and H. Gross, 491-511. Philadelphia: Temple University Press, 1987.
Hudson, R., and Paula England. "Industrial Structure and Sex Differences in Earnings." Industrial Relations 25, no. 1(1986): 16-32.
Pearce, Diana M. "The Feminization of Poverty: Women, Work, and Welfare." Urban and Social Change Review 11 (1978): 28-36.
Pressman, Steven. "The Feminization of Poverty: Causes and Remedies." Challenge 31(1988): 57-61
-----. "Explaining the Gender Poverty Gap in Developed and Transitional Economies." Journal of Economic Issues 36, no. 1 (March 2002): 17-40.
Rich, Judith, and Peter Riach. "An Investigation of Gender Discrimination in Labor Hiring." Eastern Economic Journal 21(1995): 343-356.
Ruggles, Nancy. Drawing the Line. Washington, D.C.: Urban Institute Press, 1990.
Smeeding, Timothy. The LIS/LES Project: Overview and Recent Developments, LIS Working Paper #294, January 2002.
Waldfogel, J. "The Effects of Children on Women's Wages." American Sociological Review 62(1997): 209-217.
Zellner, H. "Discrimination against Women, Occupational Segregation, and the Relative Wage." American Economic Review 62 (1972): 157-60.
The author is Professor of Economics and Finance at Monmouth University in West Long Branch, New Jersey, USA. This paper was prepared for the annual meeting of the Association for Evolutionary Economies at the Allied Social Science Association meetings in Washington, D.C., January 3-5, 2003.…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Feminist Explanations for the Feminization of Poverty. Contributors: Pressman, Steven - Author. Journal title: Journal of Economic Issues. Volume: 37. Issue: 2 Publication date: June 2003. Page number: 353+. © 1999 Association for Evolutionary Economics. COPYRIGHT 2003 Gale Group.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.