A Simple-But-Powerful Test for Long-Run Event Studies

Testing for long-run abnormal performance has become an increasingly important part of the finance literature. We propose a test for abnormal performance in long-run event studies using the buy and hold abnormal return (BHAR). We augment the control firm approach of Barber and Lyon (1997) by using multiple control firms to create multiple correlated BHARs for each sample firm. Using the control firm structure allows us to avoid the new listing, rebalancing, and skewness biases. Further, despite the correlation amongst the BHARs, using multiple control firms allows us to increase the power of the test beyond that of existing tests. Finally, we show that our test is well-specified in both random and nonrandom samples.