Reply to Comments

Abstract

First, He suggests that the assumption of fully rational, autarkic agents can be misleading in view of growing experimental evidence on the role played by social norms and reciprocity in human behavior. In particular, it is likely that, as, e.g., in Cantono and Silverberg (2009), consumers' willingness to pay (WTP) for green products is affected by the WTPs of their neighbors. Including peer effects in consumption would certainly have relevant implications for our analysis giving an active role both to local (i.e. spatial) and aggregate income inequality. The spatial distribution of agents endowed with different income levels would affect the distribution of preferences for a given aggregate level of inequality. In spite of its relevance, this extension would deliver quite intuitive implications. Consider, for sake of simplicity, two populations charac- terized by the same level of aggregate inequality and different levels of segregation by incomes, which is here a sufficient statistics for broader socio-economic conditions. It is clear then that the first population would display a stronger pioneer consumer effect, while the second a larger mass of potential adopters, i.e. larger market size effect. From a purely theoretical perspective, all our results for technology diffusion apply to this more general case. Relevant implications would emerge, instead, by allowing the government to intervene in both the sorting process and in the determination of the income distribution. However, such analysis would lack realism: policies explicitly affecting sorting are not feasible in market economies where the house market determines the level of segregation. Finally, the empirical evidence in support of peer effects in consumption is still scant, see footnote 4 in the paper, so this assump-