Advances in digital computing technologies have allowed firms access to a vast amount of consumer data at the granular level. Consumer data analyzed by powerful machine-learning tools can be a source of competitive advantage. One important way firms gather consumer data is through customers’ purchase histories, based on which firms can exercise behavior-based, or history-dependent, price discrimination. For example, a seller may reward its repeat customers by issuing them targeted discount coupons. Conversely, a seller may discriminate against its repeat customers in order to poach its rivals’ customers or attract new customers.
As the quality of data improves, firms can exercise price discrimination with finer market segmentation, leading to personalized pricing in the limit. While personalized pricing may not yet be common in practice, it is becoming more relevant in some industries, prompting ongoing policy debates on personalized pricing (e.g., CEA, 2015; OECD, 2018).
Ultimately, the seller’s pricing strategies depend on the quantity and quality of data it has access to, as well as the strategic interaction with its competitors. In addition, given that competing firms often have opportunities to share information with one another, an important policy question is whether such information sharing should be allowed when firms compete based on history-based personalized pricing. In the paper awarded the 2023 International Journal of Industrial Organization Best Paper Award, Behavior-Based Personalized Pricing: When Firms Can Share Customer Information, Chongwoo Choe, Noriaki Matsushima, and Mark J. Tremblay theoretically analyze this question.
The key results from their study can be summarized as follows:
First, when allowed to share customer data with each other, competing firms have incentives to do so. Although sharing customer data intensifies competition when the data is used for price discrimination, it softens up-front competition when the data is gathered. The benefits from the latter outweigh the costs from the former.
Second, customer data sharing between competing firms hurts consumers as a whole. This is because of the softened up-front competition and more effective price discrimination exercised by firms. But some consumers may benefit because data sharing allows better matching between consumers and firms.
Third, while customer data sharing between competing firms unambiguously benefits firms and hurts consumers, its effect on social welfare, i.e., consumer surplus plus profits, can be positive or negative depending on how large the benefits are from better matches.
The above results remain valid even when firms are asymmetric in their cost of production or when consumers’ preferences change over time.
The main policy-relevant takeaway from our study is that the pro- or anti-competitive effect of information sharing between competing firms needs to be understood in a dynamic context. For example, open banking is mandated in many countries around the world, the main rationale of which is to foster competition and innovation. But the anticipation of information sharing can dampen banks’ incentives to invest in and gather customer information in the first place. Therefore, we need to carefully consider the dynamic interaction between information sharing and competition. Thus, as more and more firms and industries establish data collection technologies, we expect the consideration of information sharing to impact how data collection occurs and how it impacts consumer prices.
CEA (2015). Big data and differential pricing. Council of Economic Advisers, Executive Office of the President of the United States.
OECD (2018). Personalised pricing in the digital era. Organisation for Economic Cooperation and Development. DAF/COMP(2018)13.