November 2023

IZA DP No. 16637: Reciprocity and Learning Effects in Price Competition

One disputed topic in Organization and Management economics is how leadership and collusive agreements are set and maintained in industries where firms are characterised by similar technological opportunities and structures. This topic is particularly important to analyse online and digital markets, which can be regarded as networks where managers share information and where there are no structural differences among firms. In this paper we claim that strategic advantages may be the outcome of repeated interaction among managers and can be driven by two (in some cases) competing forces, information and reciprocity. In fact, on one side, full information on all firms' strategies, help agents to coordinate their decisions and drive the final outcomes towards more profitable solutions. On the other side, when information is limited only to their direct opponents, competitive advantages are maintained when each competitor views the individuals' share of profits as a "fair" allocation. Thus, pricing behaviour is affected both by the willingness to reciprocate the opponent behaviour and the willingness to imitate best strategies observed in other markets. Both pricing behaviours lead to different profit outcomes. We test our hypotheses with a lab experiment on a sequential pricing game. We find a striking difference in pricing behaviour across treatments, and a significant difference also in the ability of the second movers to establish and keep their leadership. Specifically, individuals are highly competitive when information on other players' prices is limited, and only in few markets we observe second movers' advantages. When information on prices on all markets is provided, the picture is entirely different, and prices are very close to the sub-game equilibrium level. Overall, reciprocity can explain the results, however, full information reduces negative reciprocity and competition.