Consumer Willingness to Pay for Local Designations: Brand Effects and Heterogeneity at the Retail Level

Abstract

This study examines consumer willingness to pay (WTP) for packaged ice cream with local, private label, and national brands at the retail level. Data were collected through in-person surveys conducted at retail grocery outlets in Utah in 2012. Study results suggest that shoppers are willing to pay significantly more for the national and local brands over the retailer’s private label brand. In fact, the local brand with the state-sponsored designation (Utah’s Own) had a WTP equal to that of the national brand. The use of local designations or labels is important as shoppers were not willing to pay a premium for the local brands in absence of the locally produced or state-sponsored designation label. Additionally, consumer WTP for different brands and the degree of response to local designations varies substantially by consumer attributes. These differences suggest that locally designated products do not compete directly with private label products in the market. This study provides clear evidence of the value of state-sponsored designations, as well as locally produced labeling for food products marketed in a traditional retail setting.

KEYWORDS:

• brands
• choice experiment
• consumer willingness to pay
• ice cream
• state-sponsored designations

Notes

¹ The Utah’s Own program was launched in 2001 and is similar to many other SSDs; see Onken and Bernard (2010) for a comprehensive listing of SSDs. Currently, 300 Utah-based growers and firms are Utah’s Own members (Utah’s Own, 2013). See http://utahsown.utah.gov/join/index.php for specific requirements for entities to participate in the Utah’s Own program.

² For example, Rose and Bliemer (2009) noted that “the enforcement of orthogonality as a design principle does not ensure against the production of behaviorally implausible choice situations” because the levels of the attributes may be such that one alternative dominates the other alternatives. In such a situation, no information is gleaned from the choice, and the observations do not help estimate the relevant utility parameters. We also note that optimal or efficient survey design is largely an issue of statistical power. As further noted by Rose and Bliemer (2009, p. 612), “Any design, even one with a poor or even random allocation of the attribute levels, will theoretically return statistically significant population parameters given a sufficiently large enough sample.” Our large sample size and the strongly statistically significant results indicate sufficient statistical power for the purposes of our study. For more on optimal and efficient survey design, see Rose and Bliemer (2009) or Scarpa and Rose (2008).

³ An alternative estimation strategy would be to omit the constant and include an indicator variable for each brand. This produces identical results.

Additional information

Funding

This research was supported by the Utah Department of Agriculture and Food and the Utah Agricultural Experiment Station (UAES), Utah State University, and approved as journal paper number 8454.