Get access

The Underwriter Persistence Phenomenon



    Search for more papers by this author
    • Gerard Hoberg is from the University of Maryland Smith School of Business. I wish to thank Ivo Welch, Arturo Bris, Florencio López-de-Silanes, Tim Loughran, Vojislav Maksimovic, Marcus Opp, Gordon Phillips, N.R. Prabhala, Lily Qiu, Jay Ritter, Bill Schwert, Matt Spiegel, and an anonymous referee for excellent comments and advice. I also thank seminar participants at George Washington University, Indiana University, the University of Maryland, the University of Michigan, the University of Rochester, the Securities Exchange Commission, Towson University, Virginia Tech, and Yale University. All errors are the author's alone.


This study presents new evidence that initial IPO returns have persistent underwriter-specific components. These components cannot be explained by existing measures of underwriter quality, underwriter service, or controls for several known predictors of initial IPO returns. Tests that trace the roots of persistence most broadly support theories of asymmetric information among underwriters. I present such a model, and consistent with its predictions, I find that high underpricing underwriters (1) are responsible for a majority of the partial adjustment phenomenon, (2) make more informed analyst revisions, (3) experience superior market share growth, and (4) are more likely to serve an institutional clientele.