The impact of analyst recommendations on stock prices in Austria (2000–2014): evidence from a small and thinly traded market

Using a individually collected database for stock recommendations with more than 8,000 research reports issued over the period 2000–2014 on stocks listed in the Austrian Traded Index, we examine short-term market reactions. Besides traditional clustering of recommendation changes (upgrade, and downgrade) we take previous recommendations into account to get a very detailed subset of different event types. In order to consider Austrian market characteristics, we use an ARMA-market-GARCH model in addition to the market model to compare estimation results for abnormal returns. For the most extensive recommendation changes, from SELL to BUY, we document an abnormal return of 1.232 % and from BUY to SELL, - 1.534 % compared to pure BUY or SELL recommendations with 0.436 and - 0.672 %, respectively. Furthermore we are able to show that the magnitude of abnormal returns on the day of the event depends on the firm size and the delta between the target price of the recommendation and the actual price of the stock. Although we detect higher abnormal returns during the crisis, we do not find statistical evidence that investors tend to trust analysts’ opinions more during turbulent times. We conclude that the more sophisticated model does not improve the outcome and therefore a simple market model is sufficient to study events in the Austrian market.