Pricing based on consumers’ behaviour and interconnections

Nowadays, the improvements in obtaining and processing information and the development of big data are increasingly important matters from an ethical as well as an economic viewpoint. On the one hand, such phenomena imply the access to individuals’ sensitive information raising issues concerning their privacy. On the other hand, firms show clear-cut incentives to engage in targeting strategies such as personalised rebates, bring-a-friend rewards, targeted prices in online social networks. The goal of this thesis is to provide some theoretical explanation of the incentives of firms to use the information and how the distribution of surplus is affected by that use. The manuscript is divided into three chapters, each one readable as a distinct paper. The first two chapters investigate the consequences of pricing policies based on the past purchase behaviour of consumers in markets characterised by horizontal and vertical differentiation (Chapter 1) and by cross-group network externalities (Chapter 2), whereas the third one proposes a network-based analysis of referral bonuses (Chapter 3). The first paper demonstrates that firms might converge to asymmetric pricing behaviours leading to one-direction switching. These equilibrium behaviours lead to the exit of the low-quality firm the market and, under certain conditions, will result in a detriment for consumers in relation to uniform pricing, overturning the welfare results of the previous literature. The second paper presents a model of two-sided markets in which, after a first round of purchases, platforms are allowed to price-discriminate in the subscribers’ side. The model shows that stronger cross-group externalities make two-direction switching less probable. Moreover, second-period competition is strengthened compared to the case intra-side uniform pricing, whereas in the first period it is relaxed if subscribers exhibit stronger externalities than firms. The model presented in Chapter 3 shows how the level of referral bonuses strongly depends on the distribution of connections in the social network. While in random networks, the bonus is set in such a way that roughly the most popular half of informed consumers spread the voice, in scale-free networks we expect a bonus such that only the very connected people are incentivased.