Researchers at Northeastern University confirmed last month what most of us suspected to be true: that online retailers use price discrimination to maximize profits. The study shows that a wide variety of firms, from travel sites to large retailers, adjust prices based on a wide variety of factors. In addition, the study shows that firms engage in price steering, which is arranging the order of results to encourage the user to purchase a specific product. For example, for one month Orbitz had higher priced hotels sorted near the top of results for Apple users.
The researchers tested whether prices and results varied based on the operating system, browser, and whether a person was logged into an account. The control for the research was an automated computer that simultaneously made the same searches without cookies. Of course, these factors do not cover other commonly cited forms of price discrimination such as geography and time.
The study found that some firms engage in a wide variety of price steering and price discrimination methods. Travel sites tend to offer lower prices to members, but then again it is believed that members will purchase more often. Home Depot and Travelocity both used price steering on consumers that accessed their site through a mobile device, Priceline adjusted prices based on the user’s search history. In many cases the difference in price was minimal; for example Home Depot charged Android users an average of $.24 more on just a couple products. When asked how consumers can find the best price, the researchers pointed out that online firms use many methods to price discriminate which makes avoiding it difficult, but they encouraged consumers to try multiple search methods (different devices or browsers) to compare prices.
2 Comments
I did not experiment when I bought tickets to fly to my supplier visits on the same site, but I did compare across sites. Identical prices. But airline tickets do vary by time of day, ca. 12:30 am or 1:00 am is the cheapest. Not by a lot, but prices tend to be lowered and then get reset around 2 am, there’s still a once-a-day “clear the system” cycle.
I suspect book purchases may vary more. However, there’s also price dispersion: if you check the web often you surely note that the same item can have quite different prices. See the text for more on that issue.
But how do these prices compare to posted in-store prices for retailers like Home Depot? I think companies will continue to discriminate until the risk of arbitrage increases. A company that could collect data on how companies discriminate, mimic those behaviors, purchase bargains and sell them, the whole process could be abruptly stopped in those companies that rely heavily on discrimination. Until then, I suppose we can appreciate the minimal dead weight loss.
Comments are closed.