Cultural transmission with incomplete information

This paper introduces incomplete information into the standard cultural transmission framework (Bisin and Verdier, 2001). We consider parents having incomplete information about population shares and about the efficiency of their transmission technology. We show that conjectures about population shares are the key determinants of long-run population configurations. Namely, if these conjectures are positively or mildly negatively biased, there is always long-run cultural heterogeneity. If, instead, they are strongly negatively biased, long-run cultural homogeneity is displayed. We also find that, depending on the properties of conjectures about efficiency of parental transmission technology, standard cultural substitution may not hold. Notably, differently from the literature, cultural substitution, when displayed, does not guarantee long-run cultural heterogeneity. Then, considering parents who, before socializing children, experiment to acquire information, we show that they may not be able to disentangle the impact of the efficiency of their transmission technology from that of population share. Thus, parents generally fail to learn about the unknowns. We conclude the paper with a brief discussion about how cultural leaders may negatively bias conjectures about population shares and foster cultural homogeneity.