Recipient Organization
ARIZONA STATE UNIVERSITY
660 S MILL AVE STE 312
TEMPE,AZ 85281-3670
Performing Department
(N/A)
Non Technical Summary
Arguably the most important development in grocery retailing over the last 20 years, private label expansion has had a profound impact on the entire food chain, affecting supplier relations, product development processes, pricing, and ultimately consumer welfare. According to The Nielsen Company, private labels reached an all-time high of 23.7% of total grocery units sold in 2009. Moreover, roughly 90 percent of supermarket dollar sales growth from 2008 to 2009 was due to a $1.5 billion increase in store brand sales. Another way to look at private label growth over the years is in terms of category expansion. In 1997 there were positive sales of store branded products in some 339 grocery categories. 1998 saw this number increase 51% to 513. Over the next five years category expansion slowed dramatically, averaging just 1.1% annual growth, resulting in positive private label sales in 543 categories by the end of 2003. Not surprisingly, a recent search of Mintel's Global New Products Database (GNPD) revealed that three of the top five developers of new food products over the last 10 years were retailers - Target, Kroger, and Supervalu. Private label growth has significant implications for U.S. agricultural producers and end consumers. First, as retailers play an ever larger role in the development and manufacture of new products farmers and growers may find themselves negotiation agricultural prices directly with supermarket chains. Second, retailers' increasing channel power may lead to the elimination of industry middle-men and thus could lead to farmers and growers earning a higher share of channel profit margins than in the past. Third, private label growth has been shown to provide benefits to consumers via lower prices, either through greater production efficiencies or retailers' increasing ability to demand lower wholesale prices or higher trade promotion dollars from manufacturers. Research is needed that will lend understanding to how private label expansion will likely continue and the subsequent implications for agricultural producers and end consumers. Thus, we propose a research agenda aimed at improving our understanding of interrelationships among store branded products across multiple categories and the subsequent implications for agricultural producers and end consumers. Our study seeks to estimate the extent to which pricing and promotional activity of store branded food in one category affects demand for store branded products in other categories through the store. We also investigate the extent to which private label branding contributes to store loyalty and consumer welfare. These effects will be estimated using a model of consumer choice and a unique dataset of store and household behavior. Results from our research will be disseminated through academic and trade publications may affect the strategic decisions of retailers and manufacturers regarding products and pricing. For example, producers and manufacturers will be interested to know more about the power of store brand price effects across categories since retailers may use these effects to strengthen their position relative to leading national brands.
Animal Health Component
(N/A)
Research Effort Categories
Basic
100%
Applied
(N/A)
Developmental
(N/A)
Goals / Objectives
The primary objective of our proposed research is to estimate the extent of demand interrelationships among private label products across categories and, thereby, identify incentives for private label proliferation that have previously been ignored. Completion of this objective is contingent on our ability to adequately model consumer choices across multiple product categories and brands within each category, so achieving this overall objective involves a number of operational objectives, namely: 1. to develop a multi-product model of private label demand that encompasses several categories that may be unrelated from a usage standpoint; 2. to identify and develop a dataset that is amenable to estimating the proposed model at a consumer-choice level that is not subject to the limitations of existing household panel datasets; 3. to simulate the implications of the estimated model for consumer welfare, both as dependent upon variety and prices; 4. to draw implications for agricultural supply chains from the expected proliferation of private label production into new categories. 5. To extend the core cross-category model to include inter-store competitive effects of private labels, and to estimate the relative strength of intra- and inter-store substitution and complementarity on the incentives to offer private labels. Timeline of Events: Stage 1. 01/11-4/11 - Formulate and refine empirical model of cross-category and inter-store effects of private label pricing and promotion; Stage 2. 5/11-10/11 - Household and store-level data from iSpendWise will be collected, cleaned, and prepared for analysis within our established empirical framework; Stage 3. 11/11-04/12 - Estimate, test and refine the econometric specification (implementation phase); Stage 4. 5/12 - 7/12 - Extend econometrics model to include store choice and estimate; Stage 5. 8/12-10/12 - Conduct a series of welfare simulations in order to estimate the impact on consumer surplus of the intra- and inter-store effects of private label pricing. Fulfillment of our proposed research will generate outputs that will be of use to various academic, government, and industry groups. Specific activities will include the assessment of supermarket data on the purchase of private label food products and the analysis of this data to better understand inter-category effects of private label branding. Events will include the presentation of our findings at numerous academic conferences. Service outputs will accrue through our interaction with the proposed data vendor for this project. Our analysis will result in numerous discussions and consultations with the data provider that are likely to enhance the future marketing and distribution of their products and services. Our primary product output will consist of new empirical data regarding private label pricing and demand interactions across multiple food categories.
Project Methods
The proposed project will be carried out using an econometric model of cross-category private label demand that accounts for three essential features that are absent from existing modeling approaches. First, our model is able to estimate the effect of private label introduction on the substitution among stores, or the ability of private labels to attract customers from one store to another. Second, we account for the impact of cross-category demand interrelationships within the chosen store. Third, we explicitly account for the extent to which private labels are differentiated from other products in attribute space by using a distance metric market basket (DM-MB) econometric model of demand. In a distance metric model the substitutability between brands is a direct function of their distance in attribute space, where distance is defined as Euclidean distance in nutrient, brand and flavor attributes. With these three features, we make a significant contribution to the literature on private label demand. Because we use a maximum likelihood estimation method, we will test the core hypotheses of our model, namely the strength and direction of inter-category effects of private label pricing and promotion, using likelihood ratio (LR) tests. As is well understood, the primary disadvantage of using LR tests lies in the fact that the model must be estimated under both restricted and unrestricted assumptions. The benefit, however, is that the test is easily interpreted and well-suited to cases where the null and alternative hypotheses are nested within the same model. Efforts used to disseminate the knowledge generated by this study will include informal outreach to retailers and other organizations who serve the industry (e.g. Arizona Food Marketing Alliance). Project success will be determined in large part by the degree to which our methods and findings are disseminated throughout highly reputable academic and industry venues. Thus, the most quantifiable measure of success will be the number of presentations, publications, and future citations associated with the proposed research. Once the first three stages of our objective are complete we will begin submitting our research for presentation at several marketing and economics conferences. Acceptance of our research for presentation will provide an important indicator of the value of our project. Thus, key milestones will correspond to the presentation of our initial findings at several conferences during spring and summer of 2012 as well as publication of our research later that year. Prior to these milestones, project success will be determined by the completion of each objective stage according to our proposed schedule.