When you shops at Amazon, you would probably assume that the price of the same product from the same seller would be the same for everyone. Of course, certain groups of customers, such as Amazon Prime members may be able to buy things at discounted prices, but if there is no clear reason for being offered a discount, it seems natural that everyone should pay the same price for the same product. This is, at least, how pricing works in most stores we know.
However, since companies are now able to gather more and more data on customers, each buyer may be charged a different price for the same product based on his or her browsing history or the number of items in his or her cart. Amazon conducted a price experiment in 2000 by offering random prices for the same product and suffered from bad publicity because some customers discovered that they were being charged different prices for the same product. Although Amazon claimed to have not used demographic information to determine the amount of discounts, a study in 2013 about Netflix by Benjamin Shiller showed that factoring in demographic information and web browsing data may be an extremely attractive option for companies looking for ways to increase profit.  He found that using demographic information to approximate a buyer’s willingness to pay and offering a price based on that calculation raised profits by about 0.14% and using web browsing data increased profits by as much as 1.4% relative to the profit generated from offering discounts to customers who buy a large quantity of items, which is already a standard practice in almost all industries. His research also demonstrated that using many kinds of data about a customer can accurately predict his or her behavior. For example, he noted that the probability of a generic customer subscribing to Netflix is 16%, but when he used variables related to detailed web behavior such as whether a user previously visited Wikipedia or IMDb, he could predict a user’s probability of subscribing from nearly zero to 91%.
To be sure, no existing law explicitly prohibits organizations from engaging in such price discrimination. The Robinson-Patman Act, enforced by the Federal Trade Commission, forbids organizations from engaging in differential pricing when the practice harms healthy competition. Although the spirit of the law seems to be that companies should not charge different prices to different customers for the same product unless there are legitimate and verifiable reasons to do so, such as the difference in cost dealing with different kinds of customers, in practice the law is evaluated “consistent with broader antitrust policies.”  Because charging different prices to different customers based on their demographic or browsing information does not necessarily hurt other businesses and because price discrimination based on large amounts of data usually occurs in the competitive online market, such pricing is not automatically outlawed by the Robinson-Patman Act.  Anti-discrimination laws prohibiting price discrimination based on race, gender, religion, or other personal characteristics also do not apply to price discrimination based on consumer data because companies are charging customers based on their perceived ability to pay, not on some personal characteristic. 
However, it may be difficult to draw a clear line between violation of and compliance with the laws. For instance, it is very costly to develop an algorithm that closely approximates a customer’s willingness to pay. If having the ability to engage in differential pricing plays a critical role in sustaining a business, smaller companies that do not have the resources to implement price discrimination strategies may be driven out from the market.  In such cases, it would be difficult to determine if the very practice of price discrimination eliminated competition or a general disparity between the companies did so. Furthermore, algorithms that calculate a customer’s willingness to pay may inadvertently incorporate stereotypes based on personal attributes. Because computer algorithms are designed to resemble people’s behavior, they may reinforce human prejudices. For example, a study by Carnegie Mellon researchers found that Google tended to show more advertisements for high-income jobs to men than to women.  Considering that demographic data are likely to be used in conjunction with web browsing behavior to more accurately predict a consumer’s willingness to pay, the algorithm may take personal attributes into account when deciding which items to recommend or what price to charge. A misuse of algorithms may result in effects similar to those caused from discriminating based on personal characteristics, and whether to outlaw such practices remains an open question.
Price discrimination based on increasingly sophisticated consumer data is a relatively new concept and it seems that the practice is still in its embryonic stage. Practical challenges such as developing more accurate algorithms and preventing buyers to engage in arbitrage remain. But since price discrimination has a potential to bring increased profit to companies and benefit more customers, there is no doubt that the practice would continue to be developed and explored. Now may be the time to reexamine existing antitrust and anti-discrimination laws so that no legal quagmire emerges when price discrimination based on consumer data becomes prevalent.
 Shiller, Benjamin. “First Degree Price Discrimination Using Big Data.” Working Paper. 2014. http://www.brandeis.edu/departments/economics/RePEc/brd/doc/Brandeis_WP58.pdf
 Federal Trade Commission, “Price Discrimination: Robinson-Patman Violations.” https://www.ftc.gov/tips-advice/competition-guidance/guide-antitrust-laws/price-discrimination-robinson-patman
 Shpanya, Arie. “What is price discrimination and is it ethical?” Econsultancy, January 6, 2014. https://econsultancy.com/blog/64068-what-is-price-discrimination-and-is-it-ethical/
 Friel, Alan. “Take Care in Using Consumer Data to Drive Dynamic Pricing of E-Commerce.” Data Privacy Monitor, May 5, 2015. http://www.dataprivacymonitor.com/marketing/take-care-in-using-consumer-data-to-drive-dynamic-pricing-of-e-commerce/
 Srinivasan, Rags. “Amazon Price Discrimination Done Well.” Iterative Path, January 17, 2013. https://iterativepath.wordpress.com/2013/01/17/amazon-price-discrimination-done-well/
 Miller, Claire. “When Algorithms Discriminate.” The New York Times, July 9, 2015. http://www.nytimes.com/2015/07/10/upshot/when-algorithms-discriminate.html?_r=0
Photo Credit: Flickr User Soumit
The opinions and views expressed through this publication are the opinions of the designated authors and do not reflect the opinions or views of the Penn Undergraduate Law Journal, our staff, or our clients.