In 1966, the School Breakfast Program (SBP) was established to provide a nutritious breakfast to children who may otherwise not receive one. However, research on the SBP has been inconclusive regarding whether this program actually increases the likelihood of eating breakfast for participating children. Results vary according to the definition of “breakfast” with positive results only when “breakfast” is defined more strictly. Analyses of the nutritional impact of the program also suggest that provision of breakfasts in school has a positive impact on the nutritional quality of children’s diets.
While it is reassuring that the SBP may improve the quality of breakfast for participants who eat breakfast, the apparent lack of a program effect on its basic goal of providing breakfast for those who may otherwise not get one is disappointing, especially given the associations between breakfast eating and cognitive outcomes, short-term school performance, and even obesity. The nonresult may be driven by unobserved differences between program participants and nonparticipants that are also correlated with eating patterns. Such endogeneity can bias estimated program effects in cross-sectional studies. Recently, there have been efforts to adopt a universal free breakfast program in school to include all children, regardless of family income. Data on whether the SBP meets its basic goal of promoting breakfast consumption would be helpful in this context.
This study analyses the effect of participation in the SBP on breakfast consumption by using time diary data from the Child Development Supplement of the Panel Study of Income Dynamics. To control for unobserved differences between program participants and nonparticipants that may be related to their food intake, participation effects are identified by comparing differences in breakfast patterns between weekdays (when children are in school) and weekends (when they are not) for program participants versus nonparticipants. Thus, the two days’ of diary data on each child are used to control for unobserved differences between program participants and nonparticipants that may be related to their food intake.
Results show that stated “participation” in the SBP reduces the likelihood of breakfast consumption. The results are robust to alternative definitions of “breakfast,” and checks for the quality of the time-diary data. The strongest negative effects appear for children with family incomes between 135 percent and 185 percent of the Federal poverty line. A plausible interpretation of this counterintuitive result is that, contrary to parents’ expectations, children claiming to participate in the SBP may not actually be eating the meals provided at school (and consequently skipping breakfast entirely). This interpretation is consistent with observational studies of SBP participants that find a significant difference between parent reports of program participation and the child’s actual breakfast consumption. The results may indicate that the current delivery of school breakfasts needs improvement, supporting the case for such innovations as classroom provision, grab-and-go breakfasts, and other methods to promote school breakfast consumption among program participants.
Direct inquiries about this study to the Project Contact listed above.