Association for Public Policy Analysis & Management

Despite the recent decline in the U.S. poverty rate to 12.7% in 2016, evidence suggests that the poor face severe and persistent hardship. That same year, more than 45% of households in poverty were categorized as deeply poor, with incomes at less than one-half the federal poverty level (US Census Bureau, 2017). Between 1996 and 2011, the number of households living in extreme poverty (defined as less than $2 cash income a day per person) increased an estimated 159% to 1.65 million households (Shaefer & Edin, 2013).

One fundamental household daily need is food, and how the deep poor meet their nutritional demands is not well understood. The research on food management strategies identifies a range of challenges and tradeoffs confronting low-income households – for example, between food and energy expenses (Nord & Kantor, 2006) – but there is little research on the food-related tradeoffs confronting very low-income households. Similarly, while prior research indicates that vulnerable populations employ multiple food coping strategies – from relying on informal supports to consuming food past its expiration (Kempson et al., 2003; Anater, 2011; Feeding America, 2014) – less is known about how their use varies by income level and how coping relates to food tradeoffs. The use of trade-offs and coping strategies may vary by participation in the Supplemental Nutrition Assistance Program (SNAP), which served 44.2 million people each month in 2016 (USDA, 2017).

Using data from Feeding America’s 2014 Hunger in America survey, which sampled 60,825 individuals using nonprofit food programs, this study addresses two research questions:

1. How do deeply poor and higher-income households differ in the type and number of monthly food management tradeoffs faced, the food coping strategies employed, the pantry/meal programs used, and food insecurity?
2. Do the deep poor who participate in SNAP differ from those not participating in SNAP in terms of tradeoffs, coping strategies, pantry/meal program usage, and food insecurity?

We used Linear Probability Models to examine how outcomes vary by income category after controlling for respondent characteristics. Preliminary findings suggest the deep poor were less likely to use coping strategies or to have made tradeoffs at least once in the past year compared with respondents with slightly higher incomes. However, the deep poor were significantly more likely to experience tradeoffs every month than respondents from higher income categories, suggesting that the deep poor frequently had to choose between paying for food and other basic necessities. We also find that the deep poor were less likely to be food insecure than respondents with slightly higher incomes. There was no evidence of income differences in the use of community food programs. Among the deep poor, SNAP participation appears unrelated to trade-offs, coping strategies, frequency of use of pantries and meal programs, or the probability of being food insecure. Results shed light on the most vulnerable populations struggling to meet their nutritional demands.