GDP per capita, unemployment, and per capita caloric intake in the United States: Long-Run and short-run evidence from an ARDL model, 1961–2018

This study examines the long-run and short-run associations of GDP per capita and unemployment with per capita dietary energy supply in the United States over 1961–2018. Annual data are analyzed using an autoregressive distributed lag (ARDL) framework that distinguishes long-run equilibrium relationships from short-run adjustment. Unit-root tests confirm that no variable is integrated of order two, and the bounds test supports cointegration. The baseline estimates show a positive long-run association between GDP per capita and caloric supply and a negative association between unemployment and caloric supply. These findings remain qualitatively unchanged after a global multiple-break search and break-dummy re-estimation of the unchanged baseline model. Breusch–Godfrey, Breusch–Pagan–Godfrey, White, ARCH, Jarque–Bera, CUSUM, and CUSUMSQ diagnostics are reported using empirical test statistics and relevant critical values. The results are consistent with a purchasing-power channel, but they should not be interpreted as evidence that higher calories necessarily improve nutrition. Engel's Law and the nutrition-transition literature imply that economic expansion may increase dietary diversity while also facilitating greater availability of energy-dense, nutrient-poor foods. The findings therefore support counter-cyclical nutrition policies that protect food access during labor-market downturns while emphasizing diet quality during economic expansions.