Recipient Organization
UNIV OF CONNECTICUT
438 WHITNEY RD EXTENSION UNIT 1133
STORRS,CT 06269
Performing Department
Agri & Resource Economics
Non Technical Summary
NOTE: Project change request to terminate project because PD left employment at UCONN.Food marketing involves a broad set of activities intended to deliver food from producers to consumers. As such, it has a significant effect on the choices that households make. To address the nationwide obesity epidemic, it is essential to examine how food marketing affects healthful food choices made by households. This research will examine the effect of one type of food marketing, retail grocery coupons, on household purchase behavior. Given the recent economic crisis in the US, household coupon use has increased at a tremendous rate. While coupons can provide economic benefits to households, they may also alter their purchasing behavior which could ultimately lead to more unhealthful food purchases. Alternatively, they could help subsidize household purchases of more healthful items. This research will rely on a unique household level data set of grocery store purchases from 2006-2008, which corresponds to the peak of the latest economic crisis. This research will seek to identify what types of households use grocery store coupons and what types of products households buy with the coupons; how coupon usage impacts the nutritional quality of household purchases; and what effect income changes have on coupon use. Understanding the impact of coupons on household consumption holds important policy implications. Public interest groups currently scrutinize television advertising of unhealthful food products and many companies have subsequently agreed to reduce or restrict their advertising of unhealthful foods to young children. Coupons provide a form of advertising that can have a larger impact on purchases than advertising to children, however. For one, coupons generally target adults who are the household decision makers. If parents make unhealthful purchases, it impacts the entire household. Further, because coupons offer a price discount in addition to an advertisement, they have more of an effect on adult purchases than advertising by itself. Finally, coupons have become as prevalent as standard product advertising, appearing not only in newspapers but now on the internet as well. There are important demographic disparities to consider as well. For example, low income households receive greater benefit from using coupons as a percentage of their household budgets. As such, the effect of coupons on healthful consumption will be higher for low income families. Additionally, households with more leisure time such as older, retired households have more ability to acquire coupons. There may also be racial and ethnic differences which impact coupon use. The results of this research will have important implications for developing ways to improve household food purchases and they will inform food marketing policy in general.
Animal Health Component
100%
Research Effort Categories
Basic
(N/A)
Applied
100%
Developmental
(N/A)
Goals / Objectives
The overall objective of this research is to examine the effect of one type of food marketing promotion, retail grocery coupons, on household purchase behavior. To this end, this research has three specific objectives: 1. Identify the characteristics of households that frequently use grocery store coupons and the types of coupons that they use, 2. Estimate the impact of coupon usage on the nutritional quality of household purchases and 3. Determine how changes in household income impact household coupon use.
Project Methods
Objective 1. Identify the characteristics of households that frequently use grocery store coupons and the types of coupons that they use. It is important to understand how coupon usage varies among household types. The impact of coupons is likely to vary across various demographic and income groups. We will compare coupon usage across different income and household member combinations. The types of coupons chosen are likely to be an important consideration as well. It is also important to know what types of coupons certain households use, where coupons are categorized as being healthful or unhealthful depending on the cereal they promote. We will compare the healthfulness of cereals using several methods. One approach will be to use a version of the United Kingdom's Nutrition Profiling model which categorizes the quality of food products based on healthful nutrients and unhealthful nutrients. We will also make comparisons based on measures of sugar, sodium, fiber and protein. Objective 2. Estimate the impact of coupon usage on the nutritional quality of household purchases. To study the effect of household coupon use on the nutritional quality of cereal purchases made my households, we will estimate the nutritional quality of household cereal purchases, where nutritional quality is measured using macronutrients or the UK's NP model, as a function of prices, coupon use and other household demographics. The coupon parameter will measure to what extent coupon usage effects the nutritional quality of cereal purchases relative to purchases made without coupons. We will disaggregate the effects of manufacturer and retail coupons. Households in our data make infrequent purchases of cereal. Even with aggregation to the bi-weekly level, we still observe zero purchase periods in our data set. We will employ methods to account for such selection bias. The decision to use coupons is endogenous to the choice of nutritional quality that is purchased. To control for endogeneity we will use data on coupon usage in other markets as instrumental variables. Objective 3. Determine how changes in household income impact household use of coupons. Households across the United States have faced economic uncertainty over the past 5 years. Consequently, the use of coupons has seen a dramatic increase. We will estimate a reduced form model of the relationship between coupon usage and income. Household income level may be endogenous to the decision to use coupons. Further, we are interested in how income change impacts coupon use, rather than the effect of static income. We can make use of natural income variation occurring within the data set and estimate the first difference of average income on the first difference of average coupon usage. Income changes can be either positive, negative or zero. Observed and unobserved effects that are correlated with income will be differenced out, thereby reducing the likelihood of endogeneity. Any time varying attributes that are related to income will be captured by the time varying demographic characteristics. In addition to income change, we can examine changes in household employment which is provided in our data set.