Demographic change and income inequality in the United States, 1976-1989

"The U.S. economy experienced significant increases in the degree of income inequality over the past two decades.... In this paper we consider the effects of race, age, female headship, and college education on the distribution of family income by developing a multivariate methodology that allows us to gauge the influence of one factor while holding other determinants of family incomes constant. Over the period studied we find that race had only a minor effect on the overall size distribution of income. Age had a somewhat greater effect than race. In contrast, the impact of female heads and college education were quite substantial. The multivariate estimates reveal that the effects of female heads and college education both increase the Gini to a much greater extent than the progressivity of federal income taxes decreases it. The effects of college education and female headed families on inequality have grown larger across time, while the influence of age has declined. We find that the effects of race on inequality have changed little over the 1976 to 1989 period."     

The big story

When a mega-policy issue--health care reform--runs up against the Information Age, the result is likely to be increased earnings and career-making opportunities for journalists.     

Patterns of age and coverage errors in Tropical African surveys

The concept of calculating averages of African age distributions is developed in an attempt to distinguish the effect of age and coverage errors in demographic surveys in Tropical Africa from real fluctuations in the age structure. The concept is also used to discern different patterns of age errors and omissions and to indicate the extent of these errors in African age data. The method described is applied to data from 50 surveys undertaken in Tropical Africa between 1950 and 1973.     

A note on the equality of distribution of child mortality: Java-Bali, 1950-1976

Infant mortality is a widely recognized social indicator that is a useful index of the relative socioeconomic condition of subgroups of the population of a developing country. In developed countries, a more appropriate indicator of mortality is the proportion of babies who die before they reach their 5th birthday. 3 birth cohorts are identified from data provided by the Indonesian Fertility Survey conducted in Java and Bali in 1976 in order to examine the trend in mortality: births occurring 15 or more years before the survey, 10-14 years before the survey, and 5-9 years before the survey. 4 characteristics of parents related to education, occupation and urban vs. rural residence are used to analyze child mortality trends. A table shows the estimates for various combinations of these characteristics. Child mortality has decreased, but more importantly, it has decreased most for parents on the low end of the socioeconomic scale. Either parent having secondary education appears to be the key factor in the experience of low child mortality, although as more parents achieve secondary education, another factor may emerge as more important.