Recent studies suggest that the underperformance of IPOs in the post-1970 sample may be a small sample effect or “Peso problem.” That is, IPO underperformance may result from observing too few star performers ex post than were expected ex ante. We develop a model of IPO performance that captures this intuition by allowing returns to be drawn from mixtures of outstanding, benchmark, or poor performing states. We estimate the model under the null of no ex ante average IPO underperformance and construct small sample distributions of various statistics measuring IPO relative performance. We find that small sample biases are extremely unlikely to account for the magnitude of the post-1970 IPO underperformance observed in data.

We examine the efficiency of using individual stocks or portfolios as base assets to test asset pricing models using cross-sectional data. The literature has argued that creating portfolios reduces idiosyncratic volatility and allows more precise estimates of factor loadings, and consequently risk premia. We show analytically and empirically that smaller standard errors of portfolio beta estimates do not lead to smaller standard errors of cross-sectional coefficient estimates. Factor risk premia standard errors are determined by the cross-sectional distributions of factor loadings and residual risk. Portfolios destroy information by shrinking the dispersion of betas, leading to larger standard errors.