Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
New relationships are often plagued with uncertainty because one of the players has some private information about her "type." The reputation literature has shown that equilibria that reveal this private information typically involve breach of trust and conflict. But are these inevitable for equilibrium learning? I analyze self-enforcing relationships where one party is privately informed about her time preferences. I show that there always exist honest reputation equilibria, which fully reveal information and support cooperation without breach or conflict. I compare these to dishonest reputation equilibria from several perspectives. My results are applicable to a broad class of repeated games.