Seminar (Dept. of General Management)

Theme：Chasing Immortality: Artist Reputation after Death

Speaker：Letian Zhang（Harvard University）

Time：2021-12-23 09:00

Address：Zoom Meeting

Language：Chinese

Venue：Zoom Meeting https://zoom.us/j/82467002229?pwd=TEhZOExIbWZhemxQUXdMdFVqWTRhZz09 

Meeting ID：824 6700 2229

Password：375083

ABSTRACT：

This study explores the longevity of individual reputation. We empirically examine the popular belief that artists who are long dead are more revered. To do so, we constructed a large historical corpus database spanning nearly two centuries and applied the latest machine learning technique on it. Specifically, we built separate word embedding models for each time period and used them to examine how the description of 3,400 famous artists---including painters, architects, musicians, and writers---change after their death. We find that most artists reach the highest point in their reputation right before their death, after which point their reputation starts to decline, losing nearly one standard deviation every century. This posthumous decline applies to all types of artists, including artists who died young or unexpectedly. Contrary to the Matthew effect, the decline is the steepest for those who held the highest reputation in their lifetime. To understand the mechanism, we showed that changing genre taste explains a major part of the reputation decline.

SHORT BIOGRAPHY：

Letian (LT) Zhang is an assistant professor of business administration in the Organizational Behavior Unit, Harvard University. He studies social inequality in firms and markets. His research draws on sociological theories to understand disparities involving social class, race, and gender, and status.  Professor Zhang’s work covers a wide range of empirical settings including financial markets, mergers and acquisitions, sports, entrepreneurship, and Chinese labor markets. He has published in Administrative Science Quarterly, American Journal of Sociology, Organization Science, and in peer-reviewed mathematics journals.