In this chapter we describe what is known about the complex relationships between gender, math, and labor market outcomes. We perform a simulation showing that U.S. women would fare somewhat better if they knew as much math as men. We then describe how occupational structures have changed over the past 30 years, as gender differences in mathematics education have decreased, and as other barriers to the career development of women have fallen. We conclude that continued improvement of gender equity in math education is likely to have modest effects on income but that other policies have a greater potential to improve the economic well-being of U.S. women.

This thesis explores the reasons behind child labor in Italian immigrant families. It uses the Integrated Public Use Microdata Series (IPUMS) for the 1910 and 1920 U.S. Censuses to examine ethnic, gender, age, and generational differences among child workers in the United States. The data indicate that the rate of child labor in the Italian group generally resembled child labor rates in other groups: employment rates were higher for boys, for older children, and for first-generation children. The data also show that child labor declined sharply during the early decades of the twentieth century and across successive generations of Italian-Americans. Both primary and secondary sources illustrate that Italians came to the United States with a utilitarian view of children, seeing them as a component in a family economy oriented first toward survival and then toward accumulation. Interviews of Italian child workers suggest that many preferred the rewards of work to the rigors of the classroom. Oral histories taken later in life indicate that, in hindsight, those who had worked as children felt proud that they had contributed to their families' well-being, but many also looked back with regret over missed opportunities.

Analysis of long-run changes in the living arrangements and economic well-being of the aged is limited by the consistent data sources across time and space. Some fragmentary evidence on the living arrangements of the aged in several European and North American countries before the mid-twentieth century is summarized in Table 1. The numbers should be interpreted cautiously. The earliest estimates are especially suspect, since we generally lack information about the enumeration procedures or completeness of the surviving pre-nineteenth century listings of inhabitants. Even in the nineteenth century, there was significant variation in census concepts and definitions among countries and across time (Ruggles and Brower, forthcoming). Moreover, the processing of the existing historical data has not followed standardized procedures from study to study, and we have little information of the representativeness of the local studies. Therefore it would be premature to make too much of the apparent trends and differences shown in Table 1. Despite all of these qualifications, however, we can be confident that prior to the twentieth century most elderly persons in Europe and North America resided with their children and that residing alone was extremely rare.

This study of race stratification was undertaken in three articles concerning three distinct outcomes: wage differentials, area- and population-specific homicide rate differentials, and physiological distress and cardiovascular disease. The articles, although substantively distinct, fit together to explore stratification by race and some detrimental consequences of racism on these elements of well-being for African Americans. The first paper lays the foundation for the theme by establishing the hierarchy of racial stratification among five race/ethnic groups in the American labor force by examining wage discrimination in two periods. The second chapter explores how residential segregation leads to place stratification, the ranking of neighborhoods according to the race of their predominant residential population. I suggest that place stratification is directly associated with ecological SES and neighborhood violence rates. The last chapter extends the relationship between race stratification and ecology even further by arguing that racial segregation and income inequality lead to physically and psychologically hazardous environments in disadvantaged neighborhoods. Those who are residents of these types of areas are hypothesized to be more susceptible to distress and at higher risk of physiological ailments that are induced or exacerbated by physiological responses to stress. In short, race stratification not only affects group differences in SES, it impinges on neighborhood stability and even psychological and physical health.

Children live in a variety of family arrangements, some of which are complex, as a consequence of the marriage, divorce, and remarriage patterns of their parents. In addition, one-third of children today are born out-of-wedlock and may grow-up in single-parent families or spend significant portions of their lives with other relatives or stepparents. This report examines the diversity of childrens living arrangements in American households. The data are from the household relationship module of the Survey of Income and Program Participation (SIPP), collected in the fall of 1996,and update a 1994 study that presented estimates from the 1991 SIPP panel of the number of children growing up in various family situations.1As in the earlier survey, detailed information was obtained on each persons relationship to every other person in the household, permit-ting the identification of many types of relatives, and parent-child and sibling relation-ships. This report describes family situations beyond the traditional nuclear family of parents and their children and includes discussions of extended family households with relatives and non relatives who may contribute substantially to a childs development and to the households economic well-being.

Chapter one examines voter rationality. Standard agency theory suggests that rational voters will vote to re-elect competent politicians. Further, rational voters should try to filter signal from noise when assessing the competence of their elected agents. This paper measures the extent to which voters separate signal from noise in deciding whether to re-elect their state governors. I find some evidence of sophistication: voters appear to evaluate their state's economic performance relative to the national economy. Yet I also find evidence of irrationality: voters in oil-producing states tend to re-elect incumbents during oil price rises, and dump then when the oil price drops. Similarly, voters in procyclical states are consistently fooled into re-electing incumbents during national booms. I conclude that voters are best characterized as quasi-rational. Chapters two and three turn to the effects of unilateral divorce laws. Chapter two examines whether these laws caused divorce rates to rise. The Coase Theorem suggests that merely redistributing property rights should not change marriage-market allocation. The existing empirical literature disagrees. I revisit this literature showing that these results reflect a failure to jointly consider both the political endogeneity of these divorce laws and the dynamic response of divorce rates to a shock to the political regime. Taking explicit account of the dynamic response of divorce rates, I find that liberalized divorce laws caused a discernible rise in divorce rates for about a decade, but find little evidence of a persistent effect. While unilateral divorce laws have only small effects on marriage-market allocation, chapter three find profound effects on distribution. Suicide rates provide a quantifiable measure of well-being, and we find that female suicide rates fell by about a fifth when states liberalized access to divorce. Domestic violence against women declined by about a third, and intimate homicide rates declined by a tenth, suggesting that these laws improved outcomes for women. Legal institutions appear to have profound effects on outcomes within families.

Seeking to enrich our understanding of inequality movements, economic historians have used new data to illuminate the earlier darkness that Kuznets urged us to explore. To explore earlier worlds without income tax returns or modern household surveys, they have turned to non-income measures of purchasing power and well-being. The articles in this issue give a good sampling from this new wave of the economic history of inequality.

Analysis of long-run changes in the living arrangements and economic well-being of the aged is limited by the lack of consistent data sources across time and space. Some fragmentary evidence on the living arrangements of the aged in several European and North American countries before the mid-twentieth century is summarized in table 1. The numbers should be interpreted cautiously. The earliest estimates are especially suspect, since we generally lack information about the enumeration procedures or completeness of the surviving pre-nineteenth century listings of inhabitants. Even in the nineteenth century, there was significant variation in census concepts and definitions among countries and across time (Ruggles and Brower, forthcoming). Moreover, the processing of the existing historical data has not followed standardized procedures from study to study, and we have little information of the representativeness of the local studies. Therefore, it would be premature to make too much of the apparent trends and differences shown in table 1. Despite all these qualifications, however, we can be confident that prior to the twentieth century most elderly persons in Europe and North America resided with their children and that residing alone was exceedingly rare. Today, the great majority of the aged populations of North America and Europe reside alone or with only their spouse. Moreover, recent studies suggest that the percentage of the aged who live alone has also begun to rise in many Asian and Latin American countries (Hermalin and Ofstedal, 1996; Uhlenberg, 1996; Martin, 1989; DeVos, 1995). Taken as a whole, the evidence suggests that the shift towards independent residence of the aged is a worldwide phenomenon. According to the consensus of scholarly opinion, the simplification of the living arrangements of the aged during the twentieth century has resulted primarily from an increase in the resources of the aged, which has enabled increasing numbers of elderly to afford independent living. The author's analysis suggests the opposite: he argues that the decline of the multigenerational family occurred mainly because of increasing opportunities for the young and declining parental control over their children.

Objective. African American migrants from the South have long been blamed for many social problems that emerged in northern cities during the second half of the twentieth century. While recent research shows that migrants actually compared favorably with non-migrants on such characteristics as labor force participation, employment, income, and family stability, little effort has been expended to assess the more "macro" effects of migrants on the well-being of the indigenous northern black population. In this paper we investigate the possibility that northern-born black males were less likely to be employed and had lower wages in cities with proportionately larger southern-born populations in 1970. Methods. A multilevel analysis is conducted using individual-level data for 1970 from the Integrated Public Use Microdata Series and data for metropolitan statistical areas from the Summary Statistic File 4C (Fourth Count) for 1970. Results. We find that percentage southern-born in an urban area is negatively related to black male employment, but that the relationship disappears when controls are introduced. Southern-origin concentration in an urban area is not related to annual income from salary and wages for black males. Conclusions. We conclude that the evidence from our analyses raises serious doubt about a negative "collective" impact of southern migrants on the economic well-being of the indigenous northern black population.