Economic damages in many consumer class actions — including false advertising and product defect class actions — are estimated using a “price premium” approach. In these matters, plaintiffs assert that putative class members overpaid for certain products because the price they paid was inflated due to a misrepresented or defective product feature.1
For example, in Sinatro v. Barilla America Inc., the plaintiffs allege that the labeling on certain Barilla pasta products misled consumers into believing the products were made in Italy and thus caused consumers to pay a price premium. Class certification was granted by the U.S. District Court for the Northern District of California on May 28.
And in Diesel v. Mariani Packing Co. Inc., the plaintiffs allege that Mariani misled consumers by underfilling their vanilla yogurt raisin product, leading to a price premium. Class certification was granted by the U.S. District Court for the Eastern District of Missouri on April 18.2
A price premium is measured as the difference between (1) the actual market price of products at issue, and (2) the market price of the products but for the alleged conduct. Many damages experts propose or implement methodologies to estimate a price premium that results in a single unique value, expressed either in absolute dollar terms or as a percentage of the consumer purchase price.
Such experts also tend to claim that this single price premium estimate can be applied classwide. A single price premium is not per se inappropriate. However, in many, if not most, instances, material differences in product market conditions lead to different price premiums paid by distinct categories of consumers within a putative class.
Methodologies that estimate a single price premium risk obfuscating economically significant differences in the actual price premiums paid by consumers. Failure to account for these differences will lead to damages estimates that do not reflect the actual damages incurred by consumers — and may even, in some circumstances, ascribe price premium damages to certain groups of consumers who paid no price premium and suffered no economic injury.
Two of the most common methodologies proposed or implemented by experts to quantify price premium damages are conjoint survey analysis and hedonic regression analysis.3 Experts may conduct a conjoint survey analysis and choose to estimate a single price premium pertaining to their selected survey population or may choose to implement a hedonic regression model that estimates a single price premium pertaining to their entire data sample.
These two types of models are frequently proposed or used to estimate price premiums that correspond to an average difference in prices paid across the entire data sample, rather than the price premiums actually paid by putative class members.
Empirical economic research establishes that the price of a product or the prices of a product's characteristics can vary across multiple dimensions, including across time,4 geography,5 retailers,6 or product type.7 There is no a priori reason to expect that a market will determine a single price for a product of the characteristic of that product across a putative class.
Differences in supply or demand factors inherent in the relevant transactions may affect either the existence or the magnitude of any price premiums — and therefore, the existence or the magnitude of any economic damages to certain categories of consumers within a putative class.8 Constructing a reliable price premium damages methodology that can be applied classwide requires an economically sound analysis of these differing supply and demand factors, to ensure that any price premium calculations do in fact apply classwide.
As a hypothetical, consider dairy products sold throughout the U.S. that were falsely labeled as organic from 2014 to 2024. Plaintiffs may assert that prices paid for these products were inflated relative to the prices for these same products sold without an organic label.
Suppose an expert were to conduct a conjoint survey analysis or a hedonic regression analysis, and opined that all U.S. consumers paid a 10% higher price for the falsely advertised products — and thus, all U.S. purchasers were entitled to damages measured according to this estimated price premium.
In the scenario described, it is unlikely that an economically appropriate analysis would find a classwide 10% price premium.9 Instead, such an evaluation would more likely find substantive differences in price premiums across different putative class members.
Some consumers could have paid more than a 10% price premium, while others paid less than a 10% price premium — and potentially no price premium attributable to the organic label. This range in applicable price premiums would be expected because both supply and demand factors affecting the implicit prices for dairy product characteristics vary across the putative class.
For example, suppose additional competitor organic dairy producers and producers of other competitive organic substitutes entered the market during the putative class period. Additional competitive organic substitutes might increase price competition, and diminish the ability for the defendant to charge a price premium for its organic labeling over time. It could therefore be the case that consumers paid a price premium early in the class period, but paid a lower price premium, or no price premium, later in the class period.
The possibility that a price premium on organic labeling decreases over time is illustrated in Figure 1 below. In that figure, demand for an organic label on the defendant's products is assumed to decline in the period from 2020 to 2024 relative to demand in the period from 2014 to 2019. This leads to a smaller price premium, measured by the difference in prices between the defendant's products with and without the organic label.10
Figure 1: Variation in the Price Premium on Organic Labeling Over Time
Similarly, consumer demand for organic goods may differ substantially between states. Such geographic differences in demand could result in distinct retailer incentives to charge a price premium for organic labels.
This may lead one to find, for example, that consumers in more health-conscious states paid a price premium for the organic label, while consumers in less health-conscious states paid a lower price premium, or no price premium, for the organic label.
There may also be differences in the price-setting practices of individual retailers. Some retailers might respond to a change in organic labeling by changing prices, while others would not choose to change their prices in a but-for world absent organic labeling.11
Depending on the nature of each retailer's price-setting decisions, it might be the case, for example, that consumers purchasing through a local or specialty grocer paid a price premium for organic labeling, while consumers purchasing through a national supermarket chain did not.
Finally, there may be differences in supply or demand factors between different types of products that could lead to different price premiums attributable to organic labeling. Suppose, for instance, that consumers demand organic labeling on milk products, but are indifferent to organic labeling on ice cream products. It may therefore be the case that consumers who purchased milk paid a price premium, while those who purchased ice cream products did not.
There are a multitude of supply or demand factors that may cause price premiums to vary across a putative class, including across time, geography, retailer or product type. Whether an expert chooses to estimate price premium damages using conjoint survey analysis, hedonic regression analysis or another methodology, they should analyze and account for these potential differences to ensure that their damages methodologies apply classwide.
Failure to analyze and account for material differences in any price premiums across a putative class may lead to damages awards that do not reflect the actual damages incurred by consumers — possibly including damages being awarded to consumers who paid no price premium and incurred no damages due to the alleged conduct.
This article was originally published by Law360 on August 14, 2024.
1. Measuring price premium damages requires isolating the impact of the alleged conduct separate from all other market factors affecting prices for the products at issue. Experts may employ economic models analyzing a product's individual characteristics to measure a change in market prices absent some misrepresented or defective product characteristics. For example, consider an allegation that a dairy producer falsely advertised its products as organic. An expert might estimate a model that quantifies the impact of this organic label on prices for dairy products separate from other product characteristics affecting prices including, e.g., brand, nutritional profile, and taste.
2. Sinatro v. Barilla America Inc., No. 22-CV-03460-DMR, 2024 WL 2750018 (N.D. Cal. May 28, 2024); Diesel v. Mariani Packing Co. Inc., No. 4:22-CV-01368-AGF, 2024 WL 1674520 (E.D. Mo. April 18, 2024).
3. Conjoint surveys are surveys that elicit consumer choices among hypothetical product alternatives and permit analysts to quantify consumer demand for various product characteristics. When combined with an economically sound model of producer supply, conjoint survey analysis may permit analysts to quantify how market prices might change, given a change in product characteristics. A hedonic regression analysis involves the use of actual market data to conduct a regression model that measures the relationship between product prices and product characteristics to determine the market prices consumers implicitly paid for these product characteristics. See, e.g., Rosen, Sherwin, "Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition," The Journal of Political Economy, 1974. For additional information regarding challenges and other considerations for implementing a conjoint survey analysis or hedonic regression analysis, see, e.g., Cameron, Lisa, Daniel McFadden and Pablo Robles, "Price Premium Damages in Product Market Litigation: Issues in Survey-Based Market Simulations," International Comparative Legal Guides, Product Liability 2022, 20th Edition, 2022; Tomlin, Jon, "Reliability of 'Price Premium' Calculations in Class Actions," Law360, Oct. 10, 2017.
4. See, e.g., U.S. Bureau of Labor Statistics, "A Review of Hedonic Price Adjustment Techniques for Products Experiencing Rapid and Complex Quality Change," Sept. 15, 2022, available at https://www.bls.gov/cpi/quality-adjustment/hedonic-price-adjustment-techniques.htm, Table 1, identifying U.S. Bureau of Labor Statistics research on smartphone prices that indicates that prices of smartphone characteristics can vary substantially over time as a percentage of smartphone prices in the span of a few months. See also Berndt, Ernst R., and Neal J. Rappaport, "Price and Quality of Desktop and Mobile Personal Computers: A Quarter-Century Historical Overview," The American Economic Review, 2001, p. 269, describing an analysis of computer prices that considered different estimates for the impact of product characteristics on prices by year. See also de Haan, Jan, and Erwin Diewert, "Hedonic Regression Methods," in Handbook on Residential Property Prices Indices (RPPIs), Eurostat Methodologies & Working Papers, 2013, p. 50, identifying that the implicit prices of housing characteristics may change over time due to changing demand or supply conditions.
5. See, e.g., Chang, Jae Bong, Jayson L. Lusk and F. Bailey Norwood, "The Price of Happy Hens: A Hedonic Analysis of Retail Egg Prices," Journal of Agricultural and Resource Economics, 2010, finding significant geographic variation in the implicit prices of egg product characteristics.
6. See, e.g., Stanley, Linda R., and John Tschirhart, "Hedonic Prices for a Nondurable Good: The Case of Breakfast Cereals," The Review of Economics and Statistics, 1991, finding that the implicit prices of breakfast cereal characteristics varied across different grocery retailers.
7. See, e.g., Leschewski, Andrea, Dave D. Weatherspoon and Annemarie Kuhns, "A Segmented Hedonic Analysis of the Nutritional Composition of Fruit Beverages," International Food and Agribusiness Management Review, Volume 19 Issue 3, 2016, finding differences in the implicit prices of product characteristics between fruit juice and fruit drink products.
8. Some experts may argue that their price premium estimates apply classwide because they estimate a price premium as a percentage of product prices, and therefore can account for variation in product prices across a putative class when quantifying economic damages. However, economic research establishes that the prices of product characteristics can vary across multiple dimensions including time and geography, both in absolute dollar terms and as a percentage of product prices. Quantifying a price premium in percentage terms therefore does not address the possibility that a single price premium calculation risks obfuscating economically significant differences in any price premiums paid by putative class members.
9. The results in Chang et al. (2010) are illustrative of the fact that a single calculated price premium can obfuscate economically significant differences in the actual price premiums paid by putative class members. In their analysis of egg prices, Chang et al. estimate a price premium of over 20% for eggs labeled as natural, using data on egg purchases throughout the U.S. However, the authors determined that this price premium was less than 7% of the product price in Dallas/Fort Worth, and that this price premium was not statistically significant in San Francisco/Oakland. Suppose consumers alleged that certain eggs had been falsely advertised as natural, and that consumers paid a price premium for this natural label. Consistent with the results of Chang et al. (2010), an expert might opine that all U.S. consumers paid over a 20% price premium classwide. This conclusion would fail to acknowledge that putative class members in Dallas/Fort Worth may have paid no more than a 7% price premium for their natural eggs, and the expert might attribute damages to these class members according to a price premium that is roughly three times the actual price premium they paid. Moreover, the expert might attribute over a 20% price premium to putative class members in San Francisco/Oakland, when these consumers may have paid no price premium for natural eggs and therefore incurred zero economic damages. See Chang, Jae Bong, Jayson L. Lusk, and F. Bailey Norwood, "The Price of Happy Hens: A Hedonic Analysis of Retail Egg Prices," Journal of Agricultural and Resource Economics, 2010, Table 4.
10. This figure presents the standard demand and supply framework taught in introductory economic courses, and incorporates a variety of simplifying assumptions that may not be applicable in markets for differentiated products, like the market for organic dairy products. This standard supply and demand model is presented only for illustrative purposes to provide insight into why a price premium could change over time.
11. For example, retailers might implement simple cost-plus pricing policies, such that product prices only change given a change in costs. In this instance, a retailer may not respond to any change in consumer demand due to a change in organic labeling by changing product prices. See, e.g., Arcidiacono, Peter, Paul B. Ellickson, Carl F. Mela and John D. Singleton, "The Competitive Effects of Entry: Evidence from Supercenter Expansion," American Economic Journal: Applied Economics, 2020.
© Copyright 2024. The views expressed herein are those of the author(s) and not necessarily the views of Ankura Consulting Group, LLC., its management, its subsidiaries, its affiliates, or its other professionals. Ankura is not a law firm and cannot provide legal advice.
