Catering to Characteristics
Executive Summary — Can patterns of corporate net stock issuance help identify times when particular characteristics, such as industry, size, or book-to-market ratio, are mispriced? The authors of this study argue that differences between the characteristics of issuers and repurchasers can shed light on characteristic related stock returns. Consider the case in which analysts were interested in forecasting the returns of Google. The standard approach would be to collect Google's characteristics (e.g., large, technology, non-dividend paying, etc) and associate these characteristics with an average return in the cross-section. The authors argue that if other stocks with these characteristics are issuing stock, this bodes poorly for Google's future returns, even if Google is itself not issuing. This research by HBS professor Robin Greenwood and Harvard doctoral student Samuel Hanson has implications for studying the stock market performance of seasoned equity offerings (SEOs), initial public offerings (IPOs), and recent acquirers. Key concepts include:
- The approach in this paper helps forecast returns to portfolios based on book-to-market, size, share price, distress, payout policy, profitability, and industry.
- The issuer-repurchaser spreads are informative for future returns, even controlling for firms' own issuance and repurchase decisions.
(Formerly titled "A Corporate Arbitrage Approach to the Cross-section of Stock Returns.") When investors overvalue a particular firm characteristic, corporations endowed with that characteristic can absorb some of the demand by issuing equity. We use time-series variation in differences between the attributes of stock issuers and repurchasers to shed light on characteristic-related mispricing. During years when issuing firms are large relative to repurchasing firms, for example, we show that large firms subsequently underperform. This holds true even when we restrict attention to the returns of firms that do not issue at all, suggesting that issuance is partly an attempt to cater to broad time-varying patterns in characteristics mispricing. Our approach helps forecast returns to portfolios based on book-to-market (HML), size (SMB), price, distress, payout policy, profitability, and industry. Our results are consistent with the view that firms play an important role as arbitrageurs in the stock market.