Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We study a dynamic game of advice where the sender's preferences are unknown to the receiver. The novel feature of the model is that there is more than one type of biased sender. We show that the more equal the proportions of different biases in the sender population, the greater the credibility of the information transmitted. Somewhat surprisingly, however, we also find that the receiver does not benefit from this equality. We discuss our results in the context of political lobbying and show that institutions that increase transparency lower lobbyists' incentives for truthtelling, but unambiguously promote the policymaker's welfare.