A Bayesian's Bubble


  • C. WEI LI,


    Search for more papers by this author
    • Li is with Ourso College of Business, Louisiana State University. Xue, who was an assistant professor at Kansas State University, passed away on September 14, 2006. We thank two anonymous referees and the editor (Robert Stambaugh), whose suggestions have led to substantial improvements of the paper. Throughout the process of developing this paper, we have benefited tremendously from discussions with Matt Billett, Jon Garfinkel, Erik Lie, Ashish Tiwari, and Paul Weller. We also thank Eric Higgins; Ji-Chai Lin; Yiming Qian; Jay Sa-Aadu; and seminar participants at the Hong Kong University, Louisiana State University, McMaster University, National University of Singapore, University of Iowa, and University of South Carolina for suggestions and comments. Li is thankful to Lihe Wang and Dongshen Li for their hospitality when he visited Xian Jiaotong University. Ping-Wen Sun provided research assistance. All errors are our own.


The acceleration of the U.S. productivity growth in the late 1990s suggests a significant advance in technological innovation, making the perceived probability of entering a “new economy” ever increasing. Based on macroeconomic data, we identify a Bayesian investor's belief evolution when facing a possible structural break in the economy. We show that such belief evolution plays a significant role in explaining both the stock market boom and crash during 1998 to 2001. We conclude that a rational investor's uncertainty about the future of the U.S. economy provides an alternative explanation for the late 1990s stock market “bubble.”