7.1 Mechanics of event studies
An event study looks at an event that was unexpected and quantifies the impact of that event on the stock prices. The event will be value relevant only if it moves the stock price more than what was expected by the market. In order to determine whether an event is value relevant, we need to have a benchmark against which we can measure the stock price movement.
7.1.1 The benchmark
The most common benchmark used in finance is the Fama-French (FF) 3-factor asset pricing model. The model is a modified version of more famous and theory-rich Capital Asset Pricing Model (CAPM).42 FF model asserts that the stock prices are determined by 3 risks: market risk, size risk, and value risk. FF model for a stock \(i\) in time \(t\) is given by
\[ExcessRet_{it} = \beta_{MKT}*MktRF_{it} + \beta_{SMB}*SMB_{it} + \beta_{HML}*HML_{it}\] where, \(ExcessRet_{it}\) is the stock’s return above the risk-free return, \(MktRF_{it}\) is the market return above risk-free return, \(SMB_{it}\) is the “size factor”, and \(HML_{it}\) is the “value factor.”43
\(\beta_{MKT}\), \(\beta_{SMB}\), and \(\beta_{HML}\) are the measures of market, size, and value risks, respectively. As long as we have estimates of these risks44, we can use them to obtain expected stock returns in time \(t+k\) by observing \(MktRF_{it+k}\), \(SMB_{it+k}\), and \(HML_{it+k}\) as follows:
\[\widehat{ExcessRet}_{it+k} = \hat{\beta}_{MKT}*MktRF_{it+k} + \hat{\beta}_{SMB}*SMB_{it+k} + \hat{\beta}_{HML}*HML_{it+k} \]
Thus, \(\widehat{ExcessRet}_{it+k}\) is our benchmark against which we will compare the realized stock returns. If the realized returns are statistically not distinct from the expected returns, we conclude that the event was not value relevant.
William Sharpe won the Nobel Memorial Prize in Economics for CAPM.↩
We will not get into the details of these factors but you can read more about them from Kenneth French’s website.↩
Call them \(\hat{\beta}_{MKT}\), \(\hat{\beta}_{SMB}\), and \(\hat{\beta}_{HML}\)↩