Personalized Pricing Complicates Anti Discrimination Efforts

As companies increasingly tailor prices using detailed consumer data, a central policy concern is how to prevent pricing discrimination—especially against groups defined by protected characteristics such as race or gender. But anti-discrimination pricing laws may not always produce the distribution of benefits regulators intend.

Yale economists Philipp Strack and Kai Hao Yang show that even when a monopoly firm complies fully with laws requiring equal price treatment across protected groups, the resulting pricing structure can still yield uneven outcomes. In some cases, the group considered “disadvantaged” may not benefit more, and some of its members may be excluded from the market entirely, even though the law is being followed.

The authors find that the seller’s problem becomes a version of optimal transport: the firm must “pair” consumers from the two groups so each pair is offered the same distribution of prices. The firm then selects the profit-maximizing price for each pair. This pairing requirement is what shapes all of the resulting pricing—and the consequences for different groups.

“Although anti-discrimination regulations may strictly benefit consumers from both groups, they do not necessarily favor the disadvantaged group the regulations are designed to protect. In some cases, the advantaged group may benefit more, while some consumers in the disadvantaged group may be completely excluded from the market.”

Imagine a company learns, through customer data analysis, that women tend to have a higher willingness-to-pay than men. The company would like to charge all female customers more to extract the highest profit, but that would violate the anti-discrimination pricing law.

Instead, the business creates mixed pairs of men and women and charges each pair a range of prices. Because the firm must equalize the overall distribution of prices across genders, some low–willingness-to-pay consumers (in this example, some men) may be priced out when pairs face higher prices. Meanwhile, consumers with higher willingness-to-pay (here, some women) may still buy the product but keep less surplus. The main beneficiaries are consumers with intermediate willingness-to-pay, who often end up buying at prices below what they would have been charged under unrestricted personalized pricing.

The researchers also study a stricter policy, one that requires not just equal prices but equal outcomes—that is, the same pattern of who actually buys at what price. Under this stronger rule, the shift in surplus becomes even more pronounced: consumers with higher willingness-to-pay gain more, while consumers with lower willingness-to-pay become even more likely to be excluded.