The nature of the American family has changed dramatically since the 1950s, but these changes explain only a modest amount of the rise in income inequality in the United States over the past five decades.

In 1950, almost 65 percent of families included a working husband with a nonworking wife. By 1995, this fraction had dropped to only 15 percent. As more married women joined the labor force, dual-worker couples became the predominant family type—but such couples make up less than 50 percent of all families today. Higher divorce rates (which peaked in the 1980s but have remained high, especially for people with low-education levels) and increasing non-marital birth rates have increased the proportion of single-parent working families. And rising life expectancy and the retirement of boomers has swelled the share of nonworking families, reaching over 25 percent of families today, compared with 10 percent in 1950.

These changes could make income inequality look worse than it actually is. What if total family income is falling at the low end of the spectrum not due to lower wages or unemployment but, in fact, because families at the lower end are simply smaller? In that scenario, family income per person would not actually be lower because, in smaller families, total family income is divided among fewer people.

To see if these changes in family structure are driving increases in income inequality, we adjust family income by family size and composition. That way we can compare apples to apples, comparing families based on family needs. We compare the distributions of adjusted and unadjusted family incomes from 1965 to 2014 to see how they have changed. We adjust family income using the Census Bureau’s Supplemental Poverty equivalence factors, which recognize that larger families need more income than smaller families, children need less income than adults, and subsequent children need less income than first children.

Both adjusted and unadjusted family incomes have risen over time, but they have not risen equally for families across the income distribution. Between 1965 and 2013, families in the bottom 10 percent of the income distribution have had virtually no growth in inflation-adjusted income, while families in the top 10 percent have had sharply rising incomes.

Inequality, as measured by the ratio of the 90th percentile to the 10th percentile, is lower using adjusted family income than using unadjusted family income, but it is still high. Adjusting for family composition, families at the top have 10 times the income of families at the bottom (as opposed to 12 times using the unadjusted incomes) in 2013. Income inequality has increased less over the past 50 years when adjusting for family composition than without this adjustment. Income inequality between the 90th and 10th percentile had increased 79 percent since 1965 without the adjustment. With the adjustment, it increased 60 percent—still a remarkable increase.

Changes in family composition, therefore, only partially explain the increase in family income inequality seen over the past five decades.

It is also important to note that income here is measured as private income (e.g., earnings and dividends) plus cash government benefits. Income differences narrow when all taxes and transfers—such as health insurance and in-kind government benefits—are included.

For information about wealth inequality, see Nine Charts about Wealth Inequality in America.

Photo by Eric Gay/AP