Christopher Jencks

The Spread of Single-Parent Families in the United States since 1960

About half of all American children can expect to live with both of their biological parents at age fifteen, compared to two-thirds of children born in Sweden, Germany, and France, and nine-tenths of those born in Italy. This form of American exceptionalism reflects both higher rates of divorce and higher rates of breakup among cohabiting couples in the US. The increase in divorce, which began in the early 1960s but leveled off in the early 1980s, affected women at all educational levels. The increase in nonmarital childbearing, which was concentrated between the early 1960s and early 1990s, mainly affected non-white women and white women without college degrees. These changes appear to be a product of changes in sexual mores, which reduced the role of sexual attraction and increased the importance of economic calculations in decisions about whether to marry. The increased importance of economic factors coincided with a decline in non-college men’s ability to support a family and perhaps also with an increase in conflict over men and women’s roles.

Would Equal Opportunity Mean More Mobility

Adult economic status is positively correlated with parental economic status in every society for which we have data, but no democratic society is entirely comfortable with this fact. As a result, all democratic societies have adopted policies aimed at reducing the effect of family background on life chances, and most left-of-center political parties think that governments should do even more. This paper makes two main arguments. First, equal opportunity does not imply eliminating all sources of economic resemblance between parents and children. Specifically, equal opportunity does not require that society eliminate the effects of all inherited differences in ability. Nor does it require that society prevent parents from transmitting different values to their children regarding the importance of economic success relative to other goals. Second, the size of the correlation between the economic status of parents and their children is not a good indicator of how close a society has come to equalizing opportunity. Measuring equality of opportunity requires data on why successful parents tend to have successful children. In particular, it requires data on the degree to which a society has minimized obstacles to economic success that we know how to alter, such as parental neglect and ineptitude, inequitable distribution of effective teachers, and labor market practices that favor the well-born.

Changing Attitudes toward Premarital Sex: Cohort, Period, and Aging Effects

Previous research has suggested that attitudes toward premarital sex in the United States did not change much between the 1930s and 1960s, that they became more liberal (permissive) during the 1960s and 1970s, and that they remained fairly stable during the 1980s and 1990s (Glenn and Weaver 1979; Scott 1998; Singh 1980; Smith 1990, 1994; Thornton 1989; Thornton and Young-DeMarco 2001). Although both Scott (1998) and Smith (1990, 1994) argue that changes since 1960 should not be characterized as a sexual revolution, the changes were quite large by conventional standards. In 1969, more than 75 percent of American adults with an opinion on premarital sex said that it was wrong. By the 1980s only 33-37 percent of American adults said that premarital sex was either always or almost always wrong. Recent work (notably Scott 1998 and Smith 1994) has considered three possible explanations for changes in sexual attitudes: replacement of more conservative birth cohorts born early in the twentieth century by more liberal cohorts born later in the century (cohort effects), age-related changes in the views of each cohort (age effects), and cultural changes that alter the views of all cohorts simultaneously (period effects). Explanations of this kind pose a well-known identification problem. A respondents age at the time of a survey is by definition equal to the difference between the survey year and the respondents year of birth. It follows that age effects can always be expressed as some combination of period and cohort effects, that period effects can always be expressed as some combination of age and cohort effects, and

The Changing Effect of Family Background on the Incomes of American Adults

We analyze changes in the determinants of family income between 1961 and 1999, focusing on the effect of parental education, occupational rank, income, marital status, family size, region of residence, race, and ethnicity. Our data, which cover respondents between the ages of thirty and fifty-nine, come from two Occupational Changes in a Generation surveys, the General Social Survey, and the Panel Study of Income Dynamics. The multiple correlation between respondents’ family income and their parents’ characteristics fell between 1961 to 1999. During the 1960s the overall dispersion of respondents’ family incomes also fell, so the income gap between respondents from advantaged and disadvantaged families narrowed dramatically. During the 1970s, 1980s, and 1990s the overall dispersion of respondents’ family income rose again. But because the correlation between respondents’ family income and their parents’ characteristics was still falling, the income gap between respondents from advantaged and disadvantaged families showed no consistent trend. All else equal, the economic cost of being Black, Hispanic, or born in the South fell between 1961 and 1999. The cost of having a parent who worked in an unskilled rather than a skilled occupation fell between 1961 and 1972 but not after that. Indeed, occupational inequality among parents has probably become more important since 1972. Neither the effect of parental education nor the effect of parental income changed significantly during the years for which we have data. Daughters were considerably less mobile than sons in the 1970s, but this difference diminished in the 1980s and 1990s. Respondents with parents in the bottom quarter of the socioeconomic distribution were more likely to remain in their quartile of origin than respondents with parents in the top quarter of the distribution. We conclude by arguing that while both justice and economic efficiency require a significant amount of exchange mobility, neither justice nor efficiency implies that the correlation between family income and parental advantages ought to be zero. The case for programs that seek to reduce intergenerational inheritance depends on whether they reduce poverty and inequality.

Do Rising Top Incomes Lift All Boats

Pooling data for 1905 to 2000, we find no systematic relationship between top income shares and economic growth in a panel of 12 developed nations observed for between 22 and 85 years. After 1960, however, a one percentage point rise in the top deciles income share is associated with a statistically significant 0.12 point rise in GDP growth during the following year. This relationship is not driven by changes in either educational attainment or top tax rates. If the increase in inequality is permanent, the increase in growth appears to be permanent. However, our estimates imply that it would take 13 years for the cumulative positive effect of faster growth on the mean income of the bottom nine deciles to offset the negative effect of reducing their share of total income.

Who Has Benefited from Economic Growth in the United States Since 1969? The Case of Children

One can use the Census Bureaus income statistics to show either that low-income children were considerably worse off or considerably better off in 1999 than in 1969. Likewise, one can use Census statistics to show that middle-income children gained very little or a great deal between 1969 and 1999. Resolving these disagreements requires agreeing on the best price index, the best adjustment for changes in household size, and the best treatment of noncash benefits. In addition, one must reconcile discrepancies between trends in income and consumption. Since there is no consensus on any of these matters, we investigate trends in childrens well-being using more direct measures of material well-being, such as housing conditions, neighborhood safety, motor vehicle ownership, telephone service, regular medical checkups, and food consumption. Almost all these measures suggest that low-income childrens material well-being rose between the early 1970s and the late 1990s. This finding implies that traditional price indices such as the CPI-U overstated inflation.

How Did the Social Policy Changes of the 1990s Affect Material Hardship among Single Mothers? Evidence from the CPS Food Security Supplement

Although many opponents of welfare reform predicted that it would increase hardship, the official poverty rate for female headed families with children fell from 42 percent in 1996 to 34 percent in 2002. Skeptics have nonetheless argued that declines in official poverty rates may have been accompanied by increases in material hardship, since single mothers who entered the labor market often incurred new expenses and lost valuable noncash benefits. We investigate this possibility using the Current Population Survey’s Food Security Supplement. Food-related problems declined among mother-only families between 1995 and 2000 and rose between 2000 and 2002, but the decline was far larger than the subsequent increase. These changes parallel changes in the official poverty rate during the same years. In contrast to previous economic expansions, the proportional decline in poverty during the late-1990s was at least as large among mother-only families as among two-parent families. We argue that this change was linked to welfare reform and other social policy changes that encouraged single mothers to enter the labor force. As a result, single mothers’ material standard of living probably improved more during this economic expansion than during earlier ones.

Mind the Gap: Compositional, Cultural and Institutional Explanations for Numeracy Skills Disparities between Adult Immigrants and Natives in Western Countries

This paper empirically tests diverse theoretical explanations for observed skills disparities between adult immigrants and non-immigrants. Using skills data from 100,000 adults (16-65) in 18 Western countries, we show that in almost all countries, adult immigrants are less numerically skilled than non-immigrants, but that the size of the skills gap varies strongly cross-nationally. Multilevel models reveal that differences related to immigrant populations composition on demographic and socioeconomic characteristics, employment and countries of birth largely explain these regularities. In addition, countries religious diversity, immigrants social and educational integration are associated with smaller skills gaps, while labor market protectionism and educational systems vocational orientation are related to larger gaps. Theoretical and policy implications are discussed.