|Abstract or Summary
- Most research on U.S. hunger and food insecurity is at the household level, where it occurs. Because policy to address food insecurity is created and implemented at the national and state level, hunger research at the state level provides important contextual information. By including weather data with state level economic and demographic characteristics, this study attempts to explain the variation in food insecurity rates at the state level. Using ordinary least squares regression analysis, this research estimated the relationships between state-level food insecurity rates and explanatory variables which include demographic characteristics as well as weather-related variables.
Including the July Residential Energy Demand Temperature Index in the model, a “cool or eat” effect was shown to exist at the state level. High energy demand in July, associated with comparatively warmer weather, has a positive effect on food insecurity at the state level. This analysis was unable to capture a definitive “heat or eat” effect at the state level. One model using the deviation in December Heating Degree Days showed that in colder December weather, food insecurity increases, suggesting that a “heat or eat” effect exists at the state level. However, when a “December Shock” variable capturing the effect of the coldest years was included in the model, food security improved with more severe weather. The unexpected negative relationship between a cold shock in December and food security could be due to the success of programs in place to mitigate the effects of extreme weather. Innovative state programs coordinating LIHEAP and SNAP benefits hold promise to improve food security and bring more federal money to low-income people.