Julia DeLong and Dr. Clayne Pope, Economics

One of the more profound and far-reaching social transformations America went through in the latter half of the twentieth century was the growing replacement of married households by single (never married) and divorced households. While the sociological consequences of this transformation are generally debated, we seldom consider the economic consequences. In my honors thesis I established a connection between the changes in household structure in the United States during the period 1960 – 2000 and the simultaneous increase in income inequality.

Initially, my research centered around the question of why income inequality has been increasing. However, as the project took shape, it became clear that the more interesting issue was the economic consequences of the demographic change. After I realized this, it was a lot easier to ask helpful questions, locate appropriate sources, and find useful mathematical models for my research and analysis.

One of my preliminary tasks was to define the demographic trends whose effects I wished to quantify. First, in my reading I came across references to what demographers call the “second demographic transition”: widespread postponement of marriage and parenthood (the first transition being a universal drop in mortality and fertility rates). These have understandably transformed the demographic structure of the population. Second, as decisions about marital status have been changing, decisions that affect income within households have been changing. This is primarily manifest in the dramatic increase in the female labor force participation rate, which nearly doubled in the last half of the twentieth century: the rearrangement of household structures has changed the incentives associated with entering the labor force, which already means there are definite economic consequences of changing household structure.

Several economists and demographers have already approached this issue, and I reviewed the literature as preparation for my thesis. Generally, they considered the impact of female-headed households, family structure, or marital status on some measure of economic well-being, such as wealth, the Gini coefficient, or the poverty level. However, only one study came close to what I was trying to accomplish: a paper that calculated what proportion of inequality can be accounted for by components of family income such as employment of household head or spouse and family structure. This paper found that the proportion of married families accounts for 11% of the widening of the gap between the top and bottom deciles of the family income distribution.

For my own analysis of the data, I used US census data available from IPUMS and extracted samples to use with Stata. (Learning the language of the software was a research project in its own right and required frequent consultation with economics faculty members.) I used two different models to evaluate the effects of family structure on income inequality. The first I adapted from Bishop, Formby, and Smith, using coefficients from a standard regression to “correct” the income distribution to reflect a population where there is no income variation due to household type. Then I calculated a Gini coefficient (a common measure of inequality) of the revised distribution and compared it to the original. What I found was that household type is responsible for 13% of the increase in inequality of household income and 87% of the increase in inequality of income among labor force participants.

My second analysis began with a simple OLS regression, calculating the effect of family structure on the Gini coefficient. I tried six different models, changing which variables I included: in some models I separated all different kinds of households (married, divorced, widowed, etc.), and in others I only tested married households against all others. In some models I also considered the impact of the female labor force participation rate. Fortunately, the results all came out the same, showing that my model was a reasonable test of the situation. I used the values I had calculated to illustrate how changes in household structure might affect Gini coefficients of income inequality. This allowed me to create a scenario of what inequality would look like now if household structure had not changed since 1960, then compare it with the original. In this model, I found that the decrease in married households has caused about 31% of the increase in the Gini coefficient since 1960.

Changing household structure, while not the only cause of increased income inequality, is nevertheless a significant cause, and warrants further study. My findings confirm what others have mentioned and also suggest that changing family structure has had a significant impact on individual income inequality as well. I believe this is the likeliest and most interesting area for future research—the link between changing family structure and household income inequality is fairly intuitive, but the connection between changing family structure and individual income inequality is more tenuous. It definitely exists, however, as evidenced by the 87% finding in my first model above.

This honors research project was a valuable experience for me and a fitting conclusion to my undergraduate education. It encouraged me to work a lot more closely with several faculty members (not just my advisor) than I would have otherwise; I hadn’t ever taken advantage of their resources before, and this mentoring actually helped me in all areas of my education. I also had a chance to use what I had learned about academic research and put it to practical use. What I’m most grateful for is that I learned to consider what others have done and build on their knowledge and accomplishments to find my own voice and interests.

References

• Lesthaeghe, R. J. & Neidert, L. (2006). “The second demographic transition in the United States: Exception or textbook example?” Population and Development Review 32, 669-698.
• Lee, C. (2005). “Rising family income inequality in the United States, 1968 – 2000: Impacts of changing labor supply, wages, and family structure.” National Bureau of Economic Research, Working Paper 11836.
• Bishop, J. A., Formby, J. P., & Smith, W. J. (1997). “Demographic change and income inequality in the United States, 1976 – 1989.” Southern Economic Journal 64, 34-44.