Can the Pecking Order Explain the Costs of Raising Capital?

The pecking order hypothesis predicts that equity costs exceed debt costs when managers require outside funding. Asymmetric information costs motivates this hypothesis. I use an econometric model to estimate issuance costs managers face to test the prediction and motivation of the pecking order. The estimates challenge the existence of a pecking order. First, debt costs increase from about 50% of equity costs in 1973 to 140% of equity costs in 2002. Second, information asymmetry does not appear to increase managers’ reliance on debt relative to equity, although information asymmetry costs are clearly present in debt and equity individually. Firm capital structure is among the most important decisions managers face. Nonetheless, nearly 50 years after Modigliani and Miller (1958) shed light on the factors that can create value in financing choices, theories based on optimal leverage fail to explain observed capital structures. One of the most influential explanations, the pecking order of Myers (1984), is that there is no optimal debt ratio. Instead, information asymmetry creates transaction costs to accessing external equity, which are greater than the costs of accessing external debt, pushing managers to use internal equity financing, then debt financing, and, finally, external equity. In this paper, I show that the fundamental predictions of the pecking order are not borne out over time, and that modifications to the simple pecking order cannot align the theory and evidence. Besides being simple and intuitive, the pecking order can explain a number of empirical regularities. The information asymmetry motivating the pecking order is one of a small number of rational explanations for negative reaction of share prices to new equity issues, and the pecking order is consistent with evidence that firms prefer internal funds to external funds. Fama and French (2002) suggest that the pecking order is supported by a negative relationship between earnings and leverage. Shyam-Sunder and Myers (1999) show further that the internal financing deficit explains more of the variability in capital structures than does a simple measure of optimal capital structure for a set of large firms. Support for the pecking order is not unanimous, however. While Fama and French (2002) find a negative relation between earnings and debt, they also find evidence that leverage reverts to its mean. This is more consistent with a tradeoff or optimal capital structure theory than the pecking order. Frank and Goyal (2003a) present evidence that the pecking order works particularly well in large firms with uninterrupted data, which we would typically assume have low information asymmetry. The authors show further that the explanatory power of the pecking order appears to fade through time. Recently, however, Lemmon and Zender (2003) show that the pecking order See Smith (1986) for a review of share price reactions to new issues. See Fazarri, Hubbard, and Petersen (1988) and Whited (1992), among other financial constraint studies. The negative relationship is observed in almost all studies, and appears across wide methodologies. See, for example, Titman and Wessels (1988) and Rajan and Zingales (1995) for two of the best-known examples.