The Diminishing Gender Wage Gap in the U.S.

Natalia Kolesnikova and Yang Liu of the St. Louis Fed have an interesting overview of the evidence: \”Gender Wage Gap May Be Much Smaller Than Most Think.\”

Start with a provocative figure, comparing median weekly earnings of full-time male and female workers from 1979 to 2011. Back when I was starting college in 1979, it was common to hear the claim that women earned only about 70% of what men earned. The data from the figure in 1979 showing a wage gap of about 35% at that time backs up that claim. But since then, the gap has fallen to 16.5%. 

Of course, this sort of graph is just the beginning of a serious discussion. An obvious next step is to adjust these median wage differentials for demographic characteristics like educational attainment, work experience, occupation, career interruptions, overtime worked, availability of fringe benefits, and the like. These sorts of adjustments typically push the remaining gender wage gap down into low single digits. Moreover, the higher levels of women now attending college certainly suggest that the wage gap will diminish further in the future.

The standard response is to point out that a number of these adjustments to the wage gap are not exogenous choices by women, but instead are part of societal pressures. For example, the ease with which women can leave or re-enter the labor force is related to social, legal, and government support that makes it easier to do so. Adjusting for occupation means adjusting away the fact that women are still more likely to be teachers, nurses, and office clerks than men, and less like to be lawyers, doctors and top executives. Indeed, using the median wage in the figure above, rather than an average, means that the wage ratios are not affected by the much higher growth of incomes in the top few percentage points of the wage distribution–wage growth that has disproportionately benefited men.

Decades ago, newspapers used to run separate help-wanted ads for men\’s jobs and women\’s jobs, and if a woman who was teaching school married, she often was required to quit her job. That sort of egregious gender discrimination is largely in the past. But a more delicate interplay of gender roles, legal rules, and labor market outcomes remains. 

Millions of Missing Women: WDR #2

The most recent World Development Report from the World Bank is centered on the theme: \”Gender Equality and  Development.\” The first chapter of the report focuses on gains that have been made, and I posted earlier today on surprising (to me) conclusion that around the world, gender equality has been largely attained in education and health. However, the second chapter focuses on dimensions of inequality that persist. The chapter focuses certain contexts where a high degree of gender inequality persists: for example, in lack of female participation in certain occupations and in political leadership, and in certain areas or economic groups where females are disadvantaged in many dimensions of life. But to me, the most appalling example of gender inequality is expressed by the problem of the \”missing women.\” Based on evidence from biology and experience in high-income countries, we know that on average, slightly more men are born than women, and that women tend to have longer life-expectancies than men. But when we then apply those known proportions to certain countries and regions, we find that girls are missing at birth, and women are dying too frequently. Some excerpts, with footnotes and most references to tables and figures omitted, as usual:

Skewed sex ratios at birth and 3.9 million women missing under the age of 60
\”First, the problem of skewed sex-ratios at birth in China and India (and in some countries in the Caucasus and the Western Balkans) remains unresolved (table 2.1). Population estimates suggest that an additional 1.4
million girls would have been born (mostly in China and India) if sex ratios at birth in these countries resembled those found worldwide. Second, compared with developed economies, the rate at which women die relative to men in low- and middle-income countries is higher in many regions of the world. Overall, missing girls at birth and excess female mortality under age 60 totaled an estimated 3.9 million women in 2008—85 percent of them were in China, India, and Sub-Saharan Africa.\”

A first main cause: Preference for Sons
\”The disadvantage against unborn girls is widespread in many parts of Asia and in some countries in the Caucasus (such as Armenia and Azerbaijan), where the intersection of a preference for sons, declining fertility, and new technology increases the missing girls at birth. In China and India, sex ratios at birth point to a heavily skewed pattern in favor of boys. Where parents continue to favor sons over daughters, a gender bias in sex-selective abortions, female infanticide, and neglect is believed to account for millions of missing girls at birth. In 2008 alone, an estimated 1 million girls in China and 250,000 girls in India were missing at birth. The abuse of new technologies for sex-selective abortions—such as cheap mobile ultrasound
clinics—accounted for much of this shortfall, despite laws against such practices in many nations, such as India and China. Economic prosperity will continue to increase amniocentesis and ultrasound services throughout the developing world, possibly enabling the diffusion of sex-selective abortions where son-preferences exist. … This does not imply that change is impossible: The Republic of Korea’s male-female sex ratio under age five was once the highest in Asia, but it peaked in the mid-1990s and then
reversed—a link to societal shifts in normative values stemming from industrialization and urbanization.\”
A second main cause: Maternal Mortality
\”The female disadvantage in mortality during the reproductive ages is in part driven by the risk of death in pregnancy and childbirth and associated long-term disabilities. Although maternal mortality ratios have fallen by 34 percent since 1990, they remain high in many parts of the world: Sub-Saharan Africa had the highest ratio in 2008 at 640 maternal deaths per 100,000 live births, followed by South Asia (280), Oceania (230), and Southeast Asia (160). Bangladesh, Cambodia, India, and Indonesia have maternal mortality ratios comparable to Sweden’s around 1900, and Afghanistan’s is similar to Sweden’s in the 17th century. … Driving the high maternal mortality rates in many countries are poor obstetric health services and high fertility rates. Income growth and changes in household behavior alone appear insufficient to reduce maternal mortality; public investments are key to improving maternal health care services.\”

Worldwide Gender Equality in Education and Health: WDR #1

The most recent World Development Report from the World Bank is centered on the theme: \”Gender Equality and  Development.\” The first chapter of the report focuses on gains that have been made; the second chapter focuses on dimensions of inequality that persist. In this post, I\’ll focus on two patterns from the first chapter that I had not known about–on average, women the world around seem to have near-parity, and in some cases better than parity, with men in education and in health care. In a later post, I\’ll focus on what seems to me the most appalling widespread gender inequality that remains in the world today.

Global gender parity in education

\”In the past decade, female enrollments have grown faster than male enrollments in the Middle East and North Africa, South Asia, and Sub-Saharan Africa. Gender parity has been reached in 117 of 173 countries with data (figure 1.1). Even in regions with the largest gender gaps—South Asia and Sub-Saharan Africa (particularly West Africa)—gains have been considerable. In 2008, in Sub-Saharan Africa, there were about 91 girls for every 100 boys in primary school, up from 85 girls in 1999; in South Asia, the ratio was 95 girls for every 100 boys.

\”The patterns are similar in secondary education, with one notable difference. In roughly one-third of developing countries (45), girls outnumbered boys in secondary education in 2008 (see figure 1.1). Although the female gender gap tends to be higher in poorer countries, boys were in the minority in a wide range of nations including Bangladesh, Brazil, Honduras, Lesotho, Malaysia, Mongolia, and South Africa.

Tertiary enrollment growth is stronger for women than for men across the world. The number of male tertiary students globally more than quadrupled, from 17.7 million to 77.8 million between 1970 and 2008, but the number of female tertiary students rose more than sevenfold, from 10.8 million to 80.9 million, overtaking men. Female tertiary enrollment rates in 2008 lagged behind in only 36 developing countries of 96 with data (see figure 1.1). …

\”Although boys are more likely than girls to be enrolled in primary school, girls make better progress—lower repetition and lower dropout rates—than boys in all developing regions. …  Gender now explains very little of the remaining inequality in school enrollment …  In a large number of countries, a decomposition of school enrollments suggests that wealth is the constraining factor for most, and in only a very limited number will a narrow focus on gender (rather than poverty) reduce inequalities further …\”

Gender parity in health


\”In most world regions, life expectancy for both men and women has consistently risen, with women on average living longer than men. The gap between male and female life expectancy, while still rising in some regions, stabilized in others. On average, life expectancy at birth for females in low-income countries rose from 48 years in 1960 to 69 years in 2008, and for males, from 46 years to 65. Mirroring the worldwide increase in life expectancy, every region except Sub-Saharan Africa added between 20 and 25 years
of life between 1960 and today. …. And since 1980, every region has had a female advantage
in life expectancy.

In most developing countries, fertility rates fell sharply in a fairly short period. These declines were much faster than earlier declines in today’s rich countries. In the United States, fertility rates fell gradually in the 1800s through 1940, increased during the baby boom, and then leveled off at just above replacement. In
India, fertility was high and stable through 1960 and then sharply declined from 6 births per woman to 2.3 by 2009. What took the United States more than 100 years took India 40 (figure 1.4). Similarly, in Morocco, the fertility rate fell from 4 children per woman to 2.5 between 1992 and 2004.\”

\”On various other aspects of health status and health care, differences by sex are small. In many low-income countries, the proportion of children stunted, wasted, or underweight remains high, but girls are no worse off than boys. In fact, data from the Demographic and Health Surveys show that boys are at a slight disadvantage. …  Similarly, there is little evidence of systematic gender discrimination in the use of health services or in health spending. Out-of-pocket spending on health in the 1990s was higher for women than for men in Brazil, the Dominican Republic, Paraguay, and Peru. Evidence from South Africa reveals the same pro-female pattern, as does that for lower income countries. … Evidence from India, Indonesia, and Kenya tells a similar story. … For preventive health services such as vaccination, poverty rather than gender appears to be the major constraining factor …\”