Decomposing Racial and Ethnic Disparities in Food Hardship during the COVID-19 Pandemic

Background & Purpose

Racial and ethnic disparities in food hardship, which is associated with a range of poor health, mental health, and social outcomes, are well-documented in the United States. Households with children are at elevated risk. These issues took on particular importance during the COVID-19 pandemic, which threatened a wave of hardship and placed unique stresses on households with children. While the wave of food shortfalls was less than initially feared, disparities were sustained throughout the period.

Prevalence of risk factors, such as single motherhood, varies by racial and ethnic identity and could contribute to differential risk of food hardship. It is also possible, though, the degree of risk associated with a characteristic varies by group—single motherhood, for example, could penalize members of one group more than another. Disentangling these pathways is important for understanding the dynamics of food hardship and targeting interventions. We examine the role of four key risk factors in disparities in food hardship in families with children during the COVID-19 pandemic: young household head age, educational attainment, single motherhood, and unemployment. First, how were risk factors distributed across racial and ethnic identity groups and, second, how did the strength of the relationship between risk factors and food hardship vary by group?

Methods

We used the Census Bureau Household Pulse Survey, a large survey conducted throughout the pandemic, from April 2020 to July 2022 to build a sample of White, Black, and Latinx households with children (n=784,043). We first examined the prevalence of food insufficiency (inadequate quantity of food) and variables representing the four key risk factors across groups. Next, we used Blinder-Oaxaca decomposition, a technique that separates the relationship between predictor variables and a disparity between two groups into a part explained by differences in the distribution of characteristics and a part explained by differences in the relationship between those characteristics and the outcome, to evaluate the role of risk factors in disparities in food insufficiency.

Results

Approximately 23% of Black households and 19% of Latinx households reported food insufficiency compared to 9% of White households. Decomposition analyses indicated differences in prevalence of risk factors explained approximately 25% of the Black-White and 37% of the Latinx-White disparity. A larger fraction of both disparities arose from differences in degree of risk or protection. In particular, White households received more protection from college education than Black or Latinx households; if Black households received the same returns to college education as White households, for instance, it would reduce the food insufficiency gap by 36%.

Conclusions and Implications

The COVID-19 pandemic highlighted stark racial and ethnic disparities in food hardship in households with children. While there are differences in the prevalence of risk factors across racial and ethnic groups, the degree of risk or protection provided by these characteristics also differs. These findings have broader implications for understanding patterns of food hardship in the U.S. and point to important structural considerations for interventions.