Abstract

The location of firms within supply chains may profoundly affect the long-term efficiency, vulnerability and sustainability of the US food system. As alternative food systems emerge to compete with traditional ones, understanding the configuration of establishments along existing food supply chains is important for assessing their strengths and weaknesses, as well as for determining potential complementarities. We use descriptive and correlation analysis to examine geographic locations of establishments in the meat and fruit & vegetables (F&V) supply chains at the county-level. Our results show dissimilarities in the supply chains considered; for example, while cattle, hog and poultry farms are each found in nearly all continental US counties (>95%) and vegetable operations in the open are present in 90% of counties, citrus operations exist in only 7% of all counties, as they are limited by climatic factors. Asymmetries along the chains include that poultry operations are found in nearly all (98%) counties whereas poultry processors are limited to only 11% of counties. We also find different levels of co-location among establishments making up supply chains: early-stage producers are attracted to farm sites (supply) while retailers are drawn to population centers (demand). For example, cattle and hog (but not poultry) production tend to occur farther away from population centers (r2<0.06) while non-citrus fruit production is less deterred by nearby populations (r2=0.24).

Focusing on the consumer-end of the food system, we find that while supermarkets and grocery stores are present in 98% of all counties, warehouse clubs and supercenters experienced explosive growth in geographic expansion, from 26.7 to 52.7% of all counties, with the number of stores increasing from 1,748 to 4,435 establishments. However, the correlation between supercenters and populations (0.76) is lower than that with supermarkets (0.89) suggesting that, as the former store type starts to serve a larger population, shopping may involve greater transportation costs. In general, our results reveal how different establishment types that make up two key supply chains in the US food system attract or repel one another, where they tend to locate along the producer-consumer continuum, and how this is changing over time. This information can be used by policymakers and community leaders to assess viable configurations of local and regional food systems, as well as the limits they face to expanding such systems given the current geographically localized structure of food supply chains.

Authors