Dong, Diansheng

January, 2023

By: Stewart, Hayden; Dong, Diansheng
US household purchases of fluid dairy milk and plant-based milk alternatives are investigated using household-level data from the 2018 National Consumer Panel. About 58.5% of all households bought only dairy milk and 4.4% bought only plant-based products over the course of that year. Another 37.1% switched between the two types of products, buying dairy milk on most shopping occasions and plant-based products on a smaller number of occasions. Dairy milk will not likely lose out to plant-based products anytime soon given these purchase patterns; rather, the two products may coexist, with plant-based options playing a minor role.

January, 2019

By: Zheng, Yuqing; Dong, Diansheng; Burney, Shaheer; Kaiser, Harry M.
Sales taxes on either grocery food or restaurant food exist in almost every U.S. county. By combining county-level sales tax data with the USDA’s recent national household food acquisition and purchase survey, we examine how a food sales tax affects consumers’ expenditures on grocery and restaurant food.We find that a grocery tax reduces consumers’ grocery food expenditures and increases restaurant food expenditure and a restaurant food sales tax increases consumers’ grocery food expenditures. We also find no differential impacts from food sales taxes based on consumers’ income or participation status in the Supplemental Nutrition Assistance Program.

April, 2014

By: Carlson, Andrea; Dong, Diansheng; Lino, Mark
There is a common perception that it costs more to eat a healthy diet than a less healthy one. We derive a panel data model that accounts for unobserved specific individual effects to estimate the relationship between diet quality and total daily food expenditure. Since total daily diet cost and diet quality are both calculated from the foods chosen in our data, we account for the fact that there is an endogenous relationship between diet quality and cost. We find that while total daily food expenditure is statistically significant in relation to diet quality, the degree of association is very small.

December, 2012

By: Stewart, Hayden; Dong, Diansheng; Carlson, Andrea
U.S. per capita fluid milk consumption has decreased since the 1940s. This study uses data collected between 1977 and 2008 from USDA surveys to investigate whether generational change is a contributing factor. More recent generations are found to consume less whole milk and less lower-fat milk, controlling for their age at the time of the survey and other consumption determinants. These findings underscore the importance of checkoff programs, the National School Lunch Program, and other initiatives that encourage children to consume milk. Our methodology may also be adapted to analyze long-run trends in the consumption of other foods.

July, 2002

By: Schmit, Todd M.; Dong, Diansheng; Chung, Chanjin; Kaiser, Harry M.; Gould, Brian W.
A two-step model with sample selection is applied to panel data of U.S. households to estimate at-home demand for fluid milk and cheese, incorporating advertising expenditures. The model consistently accounts for sample-selection bias, unobserved household heterogeneity, and temporal correlation. Generic advertising programs for fluid milk and cheese were effective at increasing conditional purchase quantities, with very little effect on the probability of purchase. In contrast to aggregate studies, the long-run generic advertising elasticities for cheese were larger than for those of fluid milk. Advertising response varied considerably across sub-product classes, while branded advertising expenditures were largely insignificant.

December, 2000

By: Gould, Brian W.; Dong, Diansheng
Increased availability of scanner-based panel data has enabled researchers to better understand nondurable commodity purchase dynamics. In this study, we focus on one component of the purchase process--when to buy. The relationship between the discrete purchase decision and a set of household and purchase characteristics is quantified using a simulated maximum-likelihood procedure. Given the longitudinal nature of our data, unobserved heterogeneity is addressed by adopting an auto-correlated error structure. Our empirical application is household purchases of cheese. We find evidence of significant persistent unobservable household heterogeneity, which is not eliminated by the inclusion of lagged exogenous variables.