We develop a new methodology to estimate the importance of herd behavior in financial markets. We build a structural model of informa tional herding that can be estimated with financial transaction data. In the model, rational herd behavior arises because of information event uncertainty. We estimate the model using transaction data on a NYSE stock (Ashland Inc.) during 1995. We then detect the periods in the trading days in which herd behavior occurs. Herd behavior often arises and is particularly pervasive in some days. The propor tion of herd buyers (sellers) is 2% (4%) and is greater than 10% in 7% (11%) of information event days. Herd behavior causes important informational inefficiencies: on average, the distance of the price from the one that would have prevailed had traders not herded is 4% of the expected asset value.