1666250 - A statistical analysis of shifts in grocery pick-up behavior due to COVID-19 and its implications

The COVID-19 pandemic reshaped how people live, travel, and shop. This, in return, had a significant impact on freight movement and provided opportunities for greener supply chains if smaller truck trips could be decreased/eliminated. While most areas of life have started trending toward pre-COVID-19 conditions, it is not yet known how far behaviors will return to pre-pandemic conditions. The case of grocery-related shopping behaviors, specifically ordering groceries for pickup rather than delivery by truck, is one interesting example. To study this, a web-based stated preference survey was developed and disseminated to a national panel of United States adults. Using the data collected from 2,432 respondents, five user profiles were identified: non-user by choice, non-user by design, new user, skeptical user, and regular user. Descriptive statistics revealed that over 50 percent of the respondents were likely to be non-users by choice, and a further 18 percent were non-users by design (did not have the option). Two binary logit models with the possibility of heterogeneity in the means and variances of random parameters were estimated to evaluate the factors playing a role in the willingness to order groceries for pick online regularly and to become a new user of the service after the pandemic. The model estimation results indicate that socio-demographic, work- and transportation-related factors play a role in customer behavior. Based on the results, several policy recommendations are made to further support equitable development and uptake of grocery-related e-commerce.