Association for Public Policy Analysis & Management

The income of individuals and households has become much more volatile since the 1970s. In part, the nature of the economy has changed from one characterized by long-term employment with the same employer to one with a greater reliance on informal jobs, multiple jobs, and inconsistent work schedules. In addition, changes in who is living in a household can affect the total resources that household members can access. A number of studies have examined this trend and documented the effects of income volatility on the financial security of individuals and households, but most studies look at volatility across years rather than within a year.

This paper uses data for 2009-2012 from the Survey of Income and Program Participation (SIPP), a nationally representative survey, to describe and analyze how household income fluctuates month to month within a year. We combine that information with 2012-2013 data from the US Financial Diaries (USFD) project, a study of 235 low- and moderate-income households, which provides more detailed information on people who experience volatility.

Based on SIPP data, we find more than half of all prime working-age adults will have at least one month over the course of a year when they experience a change in their household taxable income that is at least 25 percent above or below their average monthly household income for the year. For many, these relatively large month-to-month income swings are the norm. One-quarter of all prime working age adults had at least six months in a single year where their income deviated by at least 25 percent.

Prime working-age adults in lower income families appear more vulnerable to income fluctuations than those in higher income families. Other factors correlated with income volatility include being younger, living in a single-person headed household, having someone in the household be self-employed, working part time, starting or ending a job, having someone move in or out of the household, and having no children in the household. These relationships hold, even controlling for demographic and other related characteristics using regression analyses.

In addition, we examined the volatility of income of the individual members in the household. In general, we found that similar covariates were associated with volatility in individual incomes as with volatility in household income. We estimated the covariance in household members’ incomes and found that, on average, incomes are slightly positively correlated. Specifically, household members’ incomes were positively correlated for 55 percent of households, which could happen if individuals with similar risk factors tend to live together. However, there are still a substantial number of households (45%) where incomes are negatively correlated and the household acts to diversity risk.