The Leverage Ratchet Effect

• Date: 01 February 2018
• Published date: 01 February 2018
• DOI: http://doi.org/10.1111/jofi.12588
• Author: PETER M. DEMARZO,PAUL PFLEIDERER,MARTIN F. HELLWIG,ANAT R. ADMATI

THE JOURNAL OF FINANCE •VOL. LXXIII, NO. 1 •FEBRUARY 2018

The Leverage Ratchet Effect

ANAT R. ADMATI, PETER M. DEMARZO, MARTIN F. HELLWIG,

and PAUL PFLEIDERER∗

ABSTRACT

Firms’ inability to commit to future funding choices has profound consequences for

capital structure dynamics. With debt in place, shareholders pervasively resist lever-

age reductions no matter how much such reductions may enhance firm value. Share-

holders would instead choose to increase leverage even if the new debt is junior and

would reduce firm value. These asymmetric forces in leverage adjustments, which

we call the leverage ratchet effect, cause equilibrium leverage outcomes to be history-

dependent. If forced to reduce leverage, shareholders are biased toward selling assets

relative to potentially more efficient alternatives such as pure recapitalizations.

IN THIS PAPER,WE SHOW THAT THE INABILITY of firms to commit to future funding

choices has profound and previously unexplored consequences for understand-

ing capital structure outcomes and dynamics. Once debt is in place, sharehold-

ers will resist any form of leverage reduction no matter how much the leverage

reduction may increase total firm value. At the same time, shareholders would

generally choose to increase leverage even if any new debt must be junior to

existing debt. The resistance to leverage reductions, together with the desire

∗Admati, DeMarzo, and Pfleiderer are from the Graduate School of Business, Stanford Univer-

sity. Hellwig is from the Max Planck Institute for Research on Collective Goods, Bonn. A previous

version of this paper was circulated in March 2012 under the title “Debt Overhang and Capital

Regulation.” We are grateful to Jonathan Berk, Eli Berkovitch, Bruno Biais, Jules van Binsber-

gen, Rebel Cole, Doug Diamond, Christoph Engel, Michael Fishman, Francisco Perez Gonzalez,

Oliver Hart, Florian Heider, G´

erard Hertig, Oliver Himmler, Arthur Korteweg, Peter Koudijs,

Alexander Ljungqvist, Alexander Morell, Adriano Rampini, Michael Roberts, Mark Roe, Steve

Ross, Klaus Schmidt, David Skeel, Chester Spatt, Ilya Strebulaev, Jeff Zwiebel, several anony-

mous referees, and seminar participants at the Bank of England, Boston College, Carnegie Mellon

University,FDIC, Federal Reserve Bank of New York, Harvard Law, Imperial College, INET 2012

Annual Conference, MIT, NYU, NBER, Princeton University, Oxford University, Stanford Univer-

sity,Tel Aviv University Finance Conference, University of North Carolina, University of Pennsyl-

vania, University of Southern California, University of Utah, Vienna Graduate School of Finance,

Washington University, Wharton, and the 2016 WFA annual conference for useful discussions and

comments. Anat Admati is a member of the FDIC systemic resolution advisory committee. Dur-

ing the time when this research was done, Martin Hellwig served as Vice Chair of the Advisory

Scientific Committee of the European Systemic Risk Board. His research has in part been funded

by the 2012 Max Planck Research Award from the German Federal Ministry of Education and

Research. The authors do not have any potential conflicts of interest to disclose, as identified in

the JF Disclosure Policy. Corresponding Author: Peter DeMarzo.

DOI: 10.1111/jofi.12588

145

146 The Journal of Finance R

to increase leverage, creates asymmetric forces in leverage adjustments that

we call the leverage ratchet effect.

We first study shareholders’ attitudes toward one-time changes in lever-

age achieved by buying back or issuing debt of various seniorities. We

show that the leverage ratchet effect is always present, even in the pres-

ence of other frictions that make the current level of debt excessive. Next,

in a model with standard trade-offs, we examine the equilibrium dynam-

ics that result when creditors anticipate and take into account the leverage

ratchet effect. Finally, we explore how the leverage ratchet effect plays out

when shareholders can change leverage by buying and selling assets in ad-

dition to the firm’s debt or equity, and show that the leverage ratchet ef-

fect has important implications for the interaction of investment and funding

decisions.

Because capital structure decisions are made in environments with uncer-

tainty, the asymmetries associated with the leverage ratchet effect can have

significant consequences for the dynamics of leverage and firm value. To see

this, consider a firm with some debt in place. A negative shock to the value

of that firm’s assets increases its leverage, but because of the leverage ratchet

effect shareholders will not voluntarily reduce leverage to its original level. In

contrast, a positive shock to the value of the firm’s assets reduces leverage, but

in this case shareholders will generally have incentives to increase leverage,

possibly even beyond its original level. Changes in tax rates, bankruptcy costs,

and other parameters that affect funding will also tend to induce asymmet-

ric responses when shareholders have control over leverage decisions.1Active

leverage reductions will generally not take place even if they would increase

firm value, while leverage increases are to be expected. Understanding how the

asymmetries implied by the ratchet effect play out requires a dynamic model

in which creditors understand shareholder incentives and anticipate their ac-

tions. Analyzing these equilibrium dynamics is one of the key contributions of

this paper.

The simplest manifestation of the leverage ratchet effect, as noted in Black

and Scholes (1973), arises in the frictionless setting of Modigliani and Miller

(1958). To see the effect in this simple setting, consider a firm that has one

class of debt with total face value DH, and assume that this debt is risky, that

is, there is a positive probability of default. Will shareholders be willing to

use cash or issue equity to buy back some of this debt and reduce the total

amount owed to DL? In perfect markets the total value of the firm, V, will

not change with the change in capital structure. Thus, letting EHand ELbe

the values of equity in the high and low debt capital structures, respectively,

and similarly letting qHand qLbe the per-unit market values of debt, we

have

V=EH+qHDH=EL+qLDL.

1Evidence of such an asymmetric response in the context of tax rate changes is provided in

recent work by Heider and Ljungqvist (2015).

Leverage Ratchet Effect 147

In buying back debt, the shareholders must pay qL(DH−DL), which means

they end up with

EL−qL(DH−DL)=(EL+qLDL)−qLDH

=V−qLDH=EH−(qL−qH)DH.

Hence, shareholders are strictly worse off as long as qL>qH, which we should

expect because with less debt the likelihood of default typically decreases and

the amount that each creditor recovers in the event of default generally in-

creases. Intuitively, by reducing leverage, shareholders transfer wealth to ex-

isting creditors.2Conversely,if shareholders can raise new debt of equal senior-

ity to fund a payout for themselves, wealth is transferred in the other direction

and shareholders benefit at the expense of existing creditors.

While this simple version of the leverage ratchet effect is well known from

Black and Scholes (1973), it may appear to depend on there being (i) no benefit

to firm value from a leverage reduction, and (ii) a single class of debt (which

is diluted by a leverage increase). We show that the effect does not depend on

these assumptions, but rather is quite pervasive and is even strengthened by

the introduction of frictions. In particular, we show that, even when a leverage

reduction would alleviate frictions such as bankruptcy or agency costs, share-

holders resist leverage reductions no matter how large the potential gain to the

total value of the firm. Intuitively,the benefits from lower bankruptcy or agency

costs accrue to creditors, and shareholders are unable to capture these bene-

fits because they must pay the higher post-recapitalization price for the debt

they buy. Moreover, we find that the leverage ratchet effect and other agency

problems of debt mutually reinforce each other. Specifically, higher leverage

intensifies shareholders’ desire to choose excessively risky investments, and at

the same time the shareholders’ ability to shift risk strengthens their resistance

to leverage reductions.

With respect to shareholders’ incentives to increase leverage, we show that,

unless the tax benefit of debt has been fully exhausted, shareholders can gain

from a one-time debt issuance even when new debt must be junior (and thus

there is no mechanical dilution of existing creditors). These incentives arise

even when, due to other frictions, the new debt would destroy firm value, as

would happen, for example, when deadweight bankruptcy costs deplete the

assets significantly. Intuitively, by increasing debt, shareholders impose an ex-

ternality on existing senior creditors, who face a higher likelihood of incurring

bankruptcy costs as well as intensified shareholder-creditor conflicts. Leverage

choices based on static trade-off theory are therefore inherently unstable.

2We assume throughout our analysis that creditors are small and dispersed so that conflicts of

interest cannot be resolved by collective bargaining. Therefore, the price at which debt is repur-

chased is at least its marginal value if retained by a creditor. Note, however,that in this example

equity holders would still lose even if debt could be repurchased at the original price qH.In that

case, the loss to shareholders is (qL–qH)DL, the gain to the remaining creditors from the reduced

dilution of their claim in default.