Working Paper: CEPR ID: DP4609
Authors: Magnus Dahlquist; Frank de Jong
Abstract: The average firm going public or issuing new equity underperforms the market in the long run. A potential explanation of this long-run underperformance has to do with the endogeneity of the number of new issues. That is, due to the clustering of events after periods of high abnormal returns in issues, ex post measures of average abnormal returns may be negative on average despite zero ex ante abnormal returns. This could lead one to incorrectly infer underperformance. We provide a thorough evaluation of the endogeneity problem in event studies as it relates to long-run underperformance and undertake both theoretical and simulation analyses. We argue that it is unlikely that the endogeneity of the number of new issues explains the long-run underperformance in equity issuances.
Keywords: abnormal return measures; endogenous events; event studies; initial public offerings; long-run underperformance; pseudo market timing
JEL Codes: C33; G14; G32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| higher past returns (G12) | increase in the number of IPOs (G24) |
| long-run underperformance (P17) | clustering of IPOs after periods of high abnormal returns (G14) |
| pseudo market timing (G14) | long-run underperformance of IPOs (G24) |
| timing of IPOs (G24) | prior market performance (G14) |
| traditional event study methods (G14) | biased estimations of abnormal returns (C51) |
| dependence of IPO volume (G24) | past market returns (G17) |