Household Food Waste and Inefficiencies in Food Production. Issue 1 (5th October 2020)

• Record Type: Journal Article
• Authors: Smith, Travis A.; Landry, Craig E.
• Is Part Of: American journal of agricultural economics. Volume 103:Issue 1(2021)
• Page Start: 4
• Page End: 21
• Publication Date: 2020-10-05
• Subjects: finite mixture model -- food waste -- household production -- stochastic frontier; D12 -- D13 -- Q18; Agriculture -- Economic aspects -- Periodicals; Agriculture -- Periodicals; 338.105
• DOI: 10.1111/ajae.12145
• Languages: English
• ISSNs: 0002-9092

Abstract

Abstract : This paper investigates household food waste in the context of inefficient food production. Food waste is typically defined as the proportion of edible food that goes unconsumed, whereas inefficiency is measured by a household's inability to reach some technically efficient production frontier. Because food waste can be avoided through improved efficiency, we focus on the determinants of inefficiency using unique household‐level data on food stock usage and food consumption. We simultaneously estimate the productivity of input usage and the corresponding household‐specific inefficiencies. Inefficiencies are further specified to be a function of observable characteristics. For the general population, we find inefficiencies to be associated with race, age, education, household size, joint preparation of meals, distance traveled to the food store, and food shopping frequency. A latent class model clearly distinguishes two groups of households: a relatively more efficient group (about 16% of the population) and a relatively inefficient group (the remaining 84%). The inefficient group is driving most of the aforementioned results for the general population, and we find relatively little explanatory power of covariates for the efficient group. We find the equivalence of observed food waste, which averages 43% in our data, to estimated input‐oriented inefficiency, which ranges 30–45%, to be dependent upon the estimated reference technology and underlying assumptions. However, we find food waste and input‐oriented inefficiency are highly correlated, such that using either outcome leads to similar policy recommendations in terms of associated factors.