Editor’s Note: This story is part of an occasional series on research projects currently in the works at the Law School.

It’s long been presumed that having a unique brand name provides value to an enterprise. That’s why trademark law exists and why companies spend so much money and expend so much effort protecting their brands.

The reality, however, is that there’s historically been very little empirical evidence to prove this underlying assumption, since intellectual property laws have been so successful at preventing the proliferation of brands with similar names. Because brand owners tend to take aggressive legal measures to protect their mark from a competitor seeking to poach it, there are few markets with a critical mass of similarly sounding brands. This has made it difficult to study the impact of similar- and dissimilar-sounding brand names and generate a better understanding of the benefits that brand distinctiveness actually provides.

Until now, that is. In a paper published last fall in the Journal of Empirical Legal Studies, UChicago Law Professor Jonathan Masur presents new evidence-based findings after studying the wine market in the Bordeaux region of France. This market is unique because it features thousands of similar wines with linguistically similar-sounding names adopted long before modern trademark law took effect, providing a robust data set to analyze how wine producers benefit from distinctive marks.

In the paper, Masur and coauthors Christopher Buccafusco of Duke University School of Law and Ryan Whalen of the University of Hong Kong Faculty of Law conclude that having a name that is linguistically dissimilar from other wines indeed allows a wine producer to charge higher prices. Their findings also indicate that this holds true for cheap, mid-price and highly expensive wines alike.

“This shows that there is real value to a brand in maintaining a linguistically dissimilar trademark,” said Masur, the John P. Wilson Professor of Law and a wine enthusiast. “It also implies that when we think about whether a trademark is dissimilar or not for purposes of trademark law, we should think not only about whether it’s similar to the nearest competitor, but how it compares to multiple other linguistically similar products.”

A Fundamental Question in Trademark Law

The paper stemmed from a prior research project in which Masur and his coauthors examined linguistic similarities among the names of wines in different regions. In comparing Bordeaux wines to other regions, they found that the names of Bordeaux wines overlapped linguistically to a much greater extent than did wine names from other areas.

For instance, more than 100 different wine-producing chateaux (wine estates) in the Bordeaux region use the word “croix” (cross) in their name, including three named “Châteaux la Croix,” two named “Château de La Croix,” and one named “Château Lacroix.” In fact, only about a quarter of producers have names that don’t include words found in the names of other producers. In some cases, these names date back hundreds of years and cannot be challenged under current trademark law.

Wondering what kind of effect this had on the marketplace, the authors sought to obtain data on wine prices and wine ratings and examine how linguistic similarity among wine names might impact their prices.

“This is a fundamental question in trademark law, and in marketing, because the point of trademarks is to allow customers to distinguish between different products,” said Masur. “So, what happens if products have very similar names and consumers can’t distinguish between them?”

A Statistical Test on the Bordeaux Wine Market

While they had the data on wine names, the authors faced the bigger challenge of obtaining data on wine prices and scores. Fortunately, WineSearcher.com, a search engine for wine purchasers, proved to be a valuable resource. 

WineSearcher provided a dataset of Bordeaux wines that included the wine name, the producer’s name, the vintage, the initial sale price, and the average rating by a professional critic on a 100-point scale. Pulling from this data, the authors generated a computational measure of linguistic similarity, based on the number of consecutive characters any given wine shared with the ten other wines that had the most similar names. 

Because the price of a wine depends substantially on its perceived quality—the scores it receives from critics—the authors sought to determine the effect of name distinctiveness on price while controlling for a wine’s quality. 

In other words, among wines that possessed a given rating from critics (for example, wines rated at 93 points), the question was whether wines with more distinctive names would command higher, lower, or equivalent prices to wines with more common names. “Access to the data on prices, critic ratings, and name similarities was critical in allowing us to perform this statistical test,” noted Masur.

Their results indicated that brand name distinctiveness among Bordeaux wines was indeed associated with a price premium: Wines with more distinct names commanded higher prices, even after controlling for the wines’ quality. Interestingly, this effect was true for both lower-rated (and cheaper) wines and higher-rated (and more expensive) wines, and everything in between. 

While the authors’ empirical methods didn’t allow them to establish causation, the results provided at least suggestive evidence of a marketing advantage for wines with distinctive names.

“We didn’t expect that the positive relationship between distinctiveness and price would be consistent across the entire spectrum of quality,” said Masur. “We thought it was possible that lower-priced wines might benefit from having similar names to more highly-rated wines. But that turned out not to be the case. That finding was particularly surprising and particularly interesting.”

Legal Implications

Masur and his coauthors suggested in their paper that these results have important implications for trademark law and policy. In particular, their evidence supported the idea that trademark congestion is harmful because brands with similarly named competitors tend to face a pricing penalty.

While trademark law already seeks to address this by making it harder for companies to register marks in “crowded fields” of linguistically similar brands, the authors said trademark law could go further by making it more difficult and expensive to register descriptive and suggestive marks that are more likely to have similarities to existing ones.

Additionally, the results suggested that courts should rethink the type of evidence used to prove infringement.

“We hope [the paper] causes people to think more comprehensively about linguistic congestion in an entire market,” Masur said.