Association for Public Policy Analysis & Management

Although it is well documented that children of immigrants are more likely to be food insecure than children of native-born families, no study has examined longitudinal patterns of child food insecurity (CFI) in immigrant households. CFI is a particularly severe measure of material hardship, because parents often seek to shield their children from food insecurity. Higher rates of CFI among children of immigrants indicate an urgent need for targeted policies and programs, but the precise pattern of need for families over time is poorly understood.

Method

We use panel data from the Early Childhood Longitudinal Study, Kindergarten Class of 1998, which contains large numbers of children of immigrants and native-born children. We categorize children’s families as: (1) native-born, for native-born children with native-born parents; (2) second generation, for native-born children with one or two foreign-born parent(s); and, (3) first generation, for foreign-born children with foreign-born parents. We use growth curve modeling to assess differences in initial levels of food insecurity and food insecurity over time for children in all three family types. We take the novel step of measuring time as the number of years that a family has lived with a child in the U.S. Thus, we use a sequential cohort design to assess both initial rates of CFI in the year when parents first have a child in this country and over time. We assess CFI using a dichotomous response based on a standard 8-item measure developed by the USDA. We control for multiple covariates.

Results

In our results, we find that compared to children of native-born families, initial probability of CFI was significantly higher for children in first (0.071) and second (0.020) generation families. Each additional year of living with a child was associated with significant increases in the probability of CFI for children in native-born and second generation families; the rate of CFI decreased significantly but incrementally over time in first generation families. Additional control for participation in food and nutrition assistance program (WIC, SNAP, National School Lunch Program) did not affect these results. However, after controlling for parents’ English language proficiency, only the initial difference in probability of CFI between native-born and first generation families was significant.

Conclusion

This study is the first to investigate longitudinal patterns of CFI among children of immigrants. We find that first- and second-generation immigrant families have higher rates of CFI upon first having a child in the US, and that rates remain higher in perpetuity for second-generation families or for many years for first-generation families. We find that participation in nutrition assistance programs does not alter the nature of these disparities, suggesting that limited access or post-Welfare reform “chilling” effects were not responsible. Rather, parents’ English proficiency explained many differences in CFI between immigrant and non-immigrant families, though first generation families had significantly higher levels of initial CFI even after controlling for language proficiency. Our results suggest the need for policies and programs to reduce food insecurity among children of immigrants and particularly among first-generation families.