We analyze institutional allocation in initial public offerings (IPOs) using a new data set of U.S. offerings between 1997 and 1998. We document a positive relationship between institutional allocation and day one IPO returns. This is partly explained by the practice of giving institutions more shares in IPOs with strong premarket demand, consistent with book-building theories. However, institutional allocation also contains private information about first-day IPO returns not reflected in premarket demand and other public information. Our evidence supports book-building theories of IPO underpricing, but suggests that institutional allocation in underpriced issues is in excess of that explained by book-building alone.