Oxford University Press's
Academic Insights for the Thinking World

Do school food programs improve child dietary quality?

Each school day, more than 30 million children in the United States receive a meal through the National School Lunch Program (NSLP); many of those students also rely on the School Breakfast Program (SBP) for their morning-time meal. These two long-standing federally subsidized meal programs were enacted on the premise of providing school children with “adequate nutrition” and “an adequate supply of food.” As the second largest food assistance program in the US, cash payments to participating schools were over $16 billion in 2014, or about 10% of the US Department of Agriculture’s total spending.

But, as the adage goes, “There ain’t no such thing as a free lunch”–participating schools must meet minimal nutritional standards in order to receive federal cash assistance. Over the past 70 years, school meal standards have become increasingly focused on raising the quality of school food rather than simply supplying food. But exactly how does the quality of a school meal compare to a brown-bag meal from home? Turns out, the answer isn’t as simple as comparing the average school lunch to the average sack lunch; we must dig deeper, far below and above the average child, where very-low and very-high quality diets exist.

When I began researching this topic, the literature was a mixed bag with regards to how school meals stacked up against a home-prepared meal: some found a negative effect within the general population, others a positive effect for low-income children, and yet others found no impact at all. There are a variety of reasons why one might expect conflicting results, such as when and where the data was collected and how exactly one measures diet quality, but the one glaring oversight to me was that all studies focused on the average impact. This one detail is particularly important in the context of school meals when we consider the vast heterogeneity in the quality of US children’s at-home diets as compared to the rather homogenous, federally-regulated school meal.

As an example, consider a child who typically consumes a low-quality diet, possibly due to parental or environmental factors. It is reasonable to believe this child may benefit from a school meal. On the other hand, a child prone to a high-quality diet may be impacted negatively when switching from a sack lunch to a school meal. In fact, on average these countervailing impacts can wash out, leaving policymakers with the conclusion that school meal programs aren’t performing as we expect them to.

The above line of thinking led me to reframe the policy question at hand and ask, how do school meal programs impact children prone to low-quality diets separately from those prone to high-quality diets? Moreover, rather than take a one-at-a-time approach to analyzing single nutrients or food groups, I took a holistic approach using a scoring system developed and validated by nutritionists, the Health Eating Index (HEI). The HEI is an algorithm that gives a score from 0 to 100 based on all foods consumed, in this case over two nonconsecutive 24-hour periods in a nationally representative dataset collected over 2005 to 2010.

The results from my analysis are best shown in the figure below. The x-axis represents an ordering of children from the lowest quality diet to the highest quality diet. The y-axis plots out the impact of substituting a home-prepared meal for a school meal (about 33% of daily calories). For one-half of all children with relatively low-quality diets (i.e., below the 50th percentile on the x-axis) a school meal boosts diet quality; for the other half of all children we see negative and insignificant impacts.

Impact of a school meal relative to a home meal by Travis A. Smith. Used with permission.

To further give perspective, consider the bottom quartile below the 25th percentile mark. Here, we see substituting a home-prepared meal for a school meal boosts dietary quality by over 2 HEI points. This is roughly 2.5 times higher than the average impact. Further, in results not show here, the quality of the at-home diet in the bottom quartile is no different than the quality of meals consumed at full-service and fast-food restaurants. Put simply, a school meal is the highest quality food consumed by children that rank the lowest on the diet quality spectrum.

How “large” is a one-point increase in HEI? Several medical studies have tracked adults over time, monitoring their HEI scores and recording adverse health outcomes. One study finds a 7% decrease in risk for any major chronic disease when moving from the first quintile (i.e., bottom 20%) to the second quintile. A quick back-of-the-envelope calculation implies a one-point increase in HEI equates to a nearly 1% decrease in risk. It remains to be studied how school meals impact diets in adulthood, but when we consider the two-point increase is daily and early on in life, the impact could be substantial.

These findings demonstrate the importance of school meal programs, especially for those children who need them the most. School food programs have undergone many changes since the 1960s, and there is good reason to believe they will continue to reform. As the new administration examines the existing school meal programs, they should consider the heterogeneous nature of the policy effects.

Featured image credit: Healthy choices of fresh fruit, salads and vegetables at Washington-Lee High School in Arlington, Virginia for lunch service on Wednesday, October 19, 2011 by U.S. Department of Agriculture. CC-BY-2.0 via Wikimedia Commons.

Recent Comments

  1. John

    Have food economists given up on economics? Cost effectiveness and benefit cost analysis are standard tools. What is the cost of raising the HEI for food actually consumed? What are the benefits and costs?

    Also, the data are interesting, “all foods consumed…2005 to 2010”. All foods actually eaten by students? Or is it just prepared and distributed? Please cite.

  2. LC 17

    When we look at nutritional value of student meals, has anyone studied what is actually consumed by the student versus what was made available? I hear the amount of food waste from food which is not consumed in the schools (discarded by students) is significant and therefore not providing any nutritional value.

Comments are closed.