Habitual behavior of household food expenditure by store type in the United States.

We examine how socio-demographic factors, spending habits, and characteristics of the retail food environment affect household expenditure across all food and beverage categories by store outlet in the United States. The six outlets considered are grocery stores, convenience stores, discount stores, club stores, drug stores, and dollar stores. The source of data for this analysis is the Nielsen Homescan Panel over the period between 2011 and 2015. We employ a dynamic correlated random effect Tobit model to incorporate habitual purchasing behavior as well as a novel method to deal with zero observations using the inverse hyperbolic sine transformation. The results suggest that habitual spending behavior undoubtedly is a key factor in affecting food and beverage expenditures across all store outlets. Additionally, household size, age, urbanization, education, race, ethnicity, and region are drivers of household food and beverage expenditures across the six store outlets.