The Net Beneﬁts to Leverage
ARTHUR KORTEWEG∗ ABSTRACT
I estimate the market’s valuation of the net beneﬁts to leverage using panel data from 1994 to 2004, identiﬁed from market values and betas of a company’s debt and equity. The median ﬁrm captures net beneﬁts of up to 5.5% of ﬁrm value. Small and proﬁtable ﬁrms have high optimal leverage ratios, as predicted by theory, but in contrast to existing empirical evidence. Companies are on average slightly underlevered relative to the optimal leverage ratio at reﬁnancing. This result is mainly due to zero leverage ﬁrms. I also look at implications for ﬁnancial policy.
THEORIES OF OPTIMAL capital structure typically explain companies’ choice of debt versus equity ﬁnancing by a trade-off: ﬁrms choose a leverage ratio that optimally weighs the beneﬁts of debt such as interest tax shields (Kraus and Litzenberger (1973)) and agency beneﬁts due to reductions in free cash ﬂow (Jensen (1986)) against the costs of debt, which include the direct costs of bankruptcy (Warner (1977) and Weiss (1990)) as well as indirect costs such as debt overhang (Myers (1977)), asset substitution (Jensen and Meckling (1976)), and asset ﬁre-sales (Shleifer and Vishny (1992)). The literature tests this tradeoff by running cross-sectional regressions of leverage on a set of variables that proxy for the beneﬁts and costs (see Harris and Raviv (1991) for an overview, and recent work such as Rajan and Zingales (2003) and Frank and Goyal (2004)). A shortcoming of the regression approach is that it is not possible to detect whether ﬁrms have too much or too little debt on average. This question is important in light of the claim that companies leave a substantial amount of tax beneﬁts on the table and are therefore systematically underlevered (Miller (1977) and Graham (2000)). A second drawback of the regression method is the implicit assumption that ﬁrms are always optimally levered, resulting in misleading effects of ∗ Arthur Korteweg is from the Graduate School of Business, Stanford University. This paper is based on my dissertation entitled “The Costs of Financial Distress across Industries,” completed at the University of Chicago. I thank my committee—Monika Piazzesi, Nick Polson, Morten Sørensen, and Pietro Veronesi—for their guidance and support. This paper has beneﬁted greatly from suggestions by an anonymous referee, an associate editor, and Editor Campbell Harvey. I also thank Mike Barclay, Alan Bester, Hui Chen, Peter DeMarzo, John Heaton, Dirk Jenter, Anil Kashyap, Paul Pﬂeiderer, Michael Roberts, Jay Shanken, Robert Stambaugh, Ilya Strebulaev, Amir Suﬁ, Michael Weisbach, Jeff Zwiebel, and seminar participants at Boston College, Emory, the University of Chicago, Georgia, London Business School, Notre Dame, Rochester, Stanford, Wharton, and the Board of Governors of the Federal Reserve for helpful discussions, comments, and suggestions. All errors remain my own.
The Journal of Finance R
proﬁtability and ﬁrm size on optimal capital structure. In reality, ﬁrms choose their capital structures in a dynamic setting. Companies let leverage ratios drift until the gain from rebalancing outweighs the cost of adjusting (Fischer, Heinkel, and Zechner (1989) and Leary and Roberts (2005)), so that ﬁrms are away from their optimal capital structures most of the time without behaving suboptimally. High proﬁts mechanically lower leverage, so that cross-sectional regressions show a negative relation between proﬁtability and leverage, even if optimal-debt ratios are positively related to proﬁtability. Similarly, small ﬁrms face higher issuance costs and therefore wait longer between reﬁnancings, resulting in lower average leverage than big ﬁrms, even though in theory they may have higher optimal-debt ratios. In light of these issues, this paper addresses four main questions in the capital structure literature: (i) how large are the...