How Stable Are Corporate Capital Structures?

In the paper, How Stable Are Corporate Capital Structures? which was recently made publicly available on SSRN, we examine the stability of corporate capital structures. Overall, the evidence indicates that time-series variation in the leverage of individual firms is of first-order importance, with leverage instability reflecting the external funding of company expansion and with mature firms trending away from conservative financial policies. Capital structure stability is the exception rather than the rule at publicly held industrial firms, with the preponderance of firms having leverage ratios that take on a wide range of values and that differ over time in mean value. Leverage stability occurs only infrequently and, when it does, is virtually always temporary, with stability episodes largely occurring when firms have low leverage. Departures from stability are rarely followed by rebalancing to the prior stable leverage regime or by establishment of a new stable regime. Leverage instability is strongly associated with asset growth and external funding to support that growth, and also reflects a trend away from conservative leverage that played out mainly over the 1950s and 1960s and that began in the mid- to late-1940s with the need to fund expansion during the post-World War II boom.

This evidence indicates that any empirically viable theory of capital structure must be able to explain significant time-series variation in leverage, and it offers several important corollary implications for existing or potential theories. First, tradeoff models in which firms have stationary (or nearly stationary) leverage targets and managers rebalance leverage to avoid large deviations from target are far wide of the mark empirically. Second, if firms have target leverage zones—as the CFO survey evidence of Graham and Harvey (2001) indicates that many firms do—then those zones either have very wide boundaries, or managers specify zone boundaries for relatively short financial planning horizons and view the boundaries as “soft” or tentative limits that can be changed markedly as circumstances dictate.

Third, it would seem essential to treat investment policy as endogenous, and to move away from framing capital structure optimization as an exercise in rebalancing leverage while holding investment policy fixed. In other words, our evidence suggests that, as an empirical matter, the capital structure problem is more about access to capital to fund investment than it is about optimizing the debt/equity payout mix (i.e., the interest/principal versus dividend/repurchase mix of distributions). Without a shift in emphasis away from attaining (or rebalancing to remain near) an optimal debt/equity mix, theories founder on their inability to explain why so many firms do not maintain even approximately stable capital structures. And without a shift in emphasis toward access to capital, theories founder on their inability to explain the connection between leverage instability and the external funding of company expansion.

The joint emphasis on the primacy of investment policy and the unimportance of optimizing the debt/equity mix hearkens back, of course, to Modigliani and Miller (1958) and Miller (1977). The point here is not that leverage evolves as a neutral mutation, as Miller conjectured it might, but rather that leverage ratios take on such a wide range of values at so many firms that it is difficult to believe there are significant benefits from adhering even loosely to a leverage target. It seems far more plausible that variation in leverage per se is just not that important for firm valuation, so that considerations other than rebalancing to a target are likely to be the main drivers of the time path of corporate leverage.

Our reading of the evidence is that empirically credible theories will treat variation in investment and the coincident need for funds as important determinants of the evolution of a firm’s capital structure. They will also likely contain other factors that affect marginal external financing decisions and that foster time-series variation in leverage, such as (i) credit market conditions, (ii) security valuation errors due to the difficulties of estimating company prospects, (iii) market-timing opportunities due to capital market inefficiency, (iv) security flotation costs, (v) managerial attitudes toward debt, and (iv) social norms regarding the prudence of borrowing. Credible theories will exclude any and all combinations of tax, bankruptcy, or agency cost conditions that dictate narrow boundaries on a firm’s optimal (or near-optimal) debt/equity mix. The call in a nutshell is to abandon what the data indicate is a fatally flawed emphasis on rebalancing toward a leverage target in favor of theories of capital structure rooted in the importance of access to capital to fund investment, coupled with a variety of ancillary factors—whose weights are yet to be determined—that govern marginal funding decisions and that can help explain the substantial time-series variation in leverage that is pervasive at publicly held industrial firms.

The full paper is available for download here.