Income instability is a potentially important and understudied dimension of the established empirical relationship between family income and children’s healthy development. After a spate of research on poverty and welfare spells during the 1990s (Bane & Ellwood, 1986, 1996; Duncan, 1988; Stevens, 1994), most research on socioeconomic status and child development continues to use static measures of income (Wagmiller et al., 2006). Even prominent studies of the timing of poverty during childhood do not model income volatility explicitly. This gap prevents social science from speaking to the salience of increasing economic instability for children and to the wisdom of social policies designed to stabilize income.

I use the Panel Study of Income Dynamics (PSID) cross-year individual file from 1968 to 2009 to answer two research questions: How does yearly income vary across childhood and across time (historically)? What are the primary causes or “triggers” of income instability and how have they changed over time? The study is a replication and extension of Duncan (1988), which examined these questions using the PSID data from 1968 to 1979. In addition to average income and income-to-needs ratio, I use the coefficient of variation (standard deviation divided by mean) to capture volatility and an average growth rate to capture trend. I also code for six types of events in the survey year prior to the measures of income dynamics: parental job entry; parental job exit; parental job transition between part- and full-time hours (or the reverse); parental marital or cohabitation status change, and change in household composition. Descriptive graphs and tables show income dynamics across four child age sub-groups—0-4 years; 5-9 years; 10-14 years; and 15-19 years—which correspond roughly to early childhood, middle childhood, early adolescence, and late adolescence. I also examine income dynamics across childhood by family income level. Finally, OLS regression models are used to predict the measures of income volatility as a function of family events and a set of time invariant covariates. This research provides a much-needed update to past studies of income dynamics, with a focus on differences by income level and child age.