Four essays on social interaction models

The present PhD dissertation explores the effects of social interactions on individual’s behaviour (at the micro level) and the collective equilibrium emerging (at the macro level). Social interactions refer to situations where the payoff or utility an individual receives from a given action depends directly on the choices of others. More precisely, the first chapter explores the role that social interactions have on natives’ attitudes towards migrants in Latin American countries. In addition, the effects of other economic motives are also explored. The results found evidence that conformity behaviour is a strong determinant of natives views towards migrants. The second chapter studies the effect that multiple equilibria, due to social interactions, have in the context of duopoly competition. More precisely, we model agents with heterogeneous preferences who decide to buy from one of the two firms on the market. The firms are differentiated by the strength of their social recognition. In this sense, we speak about firm-specific network effect: the firm with the higher parameter is perceived by consumers to have a higher social recognition (exert higher social pressure). The model formalizes a Bertrand economy, where in a two-stage game, firms simultaneously choose optimal prices, and then potential adopters decide upon their preferred technology. The market equilibrium obtained confirms intuition: the firm with the highest network effect exerts a higher social pressure on potential adopters and obtains a higher market share. Surprisingly, this outcome is not always observed: an excessive network effect may be detrimental for the strongest firm, which, eventually, can be thrown out of the market with a positive probability. This is due to the reinforcing effect of consumers’ behaviour that could deviate the competition in favour of the apparently weakest firm. The last part of this thesis consist in two chapters containing two independent extensions of the model developed in chapter 2. Especially, in chapter 3 firms are allowed to decide their level of differentiation (location). Therefore, on the first stage firm’s decide upon their differentiation, then their price, and finally consumers decide between the two products. For low levels of social interactions, we show that it is optimal for the firms to be highly differentiated. On the opposite, for high levels of social interactions, the opposite situation applies: firms offer the same standard product, although in this case, only one firm survives and monopolizes the market. Finally, chapter 4 focuses on the demand side of the market where an outside (not buying) option is offered to the consumers. Moreover, we benefit from this generalization to show an application of our model to the smartphone industry. Here we associate the concept of brand awareness to the definition of firm-specific network effects. As a case study, the model is calibrated with real data from the smartphone industry obtaining an estimate of the value of the brand awareness of two dominant brands.