The market response to implied debt covenant violations

• Published date: 01 October 2018
• Author: Derrald Stice
• DOI: http://doi.org/10.1111/jbfa.12321
• Date: 01 October 2018

DOI: 10.1111/jbfa.12321

The market response to implied debt covenant

violations

Derrald Stice

HongKong University of Science and Technology,

HongKong

Correspondence

DerraldStice, Hong Kong University of Science

andTechnology,6051 LSK, HKUST, Clear Water

Bay,Kowloon, Hong Kong.

Email:acstice@ust.hk

Abstract

Previous research documents a negative stock price reaction to the

announcement of a debt covenant violation (DCV). However, man-

agers of firms that violate a covenant often obtain waivers and rene-

gotiate debt contracts with lenders before the SEC requires them to

disclose a violation. Firms therefore may not report some covenant

violations, and prior research has not documented their cost to

shareholders.Exploiting the fact that over half of all private debt con-

tracts contain a debt covenant reliant on some variation of account-

ing earnings, I construct a sample of firms with debt contracts that

contain at least one earnings-based covenant. Combining earnings-

based-covenant contract values from debt agreements with infor-

mation publicly available at the earnings announcement date, I pre-

dict firms in violation of a debt covenant and provide evidence that

equity investors react negatively to these implied violations, regard-

less of whether managers ever disclose that a violation occurred. In

additional tests, I find no evidence of a negative stock price reaction

to a firm disclosure of a DCV that market participants could infer

from previously reported earnings, but I demonstrate that equity

investors do react to the disclosure of a violation of a balance-sheet

covenant that would not have been inferable. This study comple-

ments previous research on DCVs bydocumenting the costliness to

shareholders of violations subsequently resolved with lenders but

not disclosed.

KEYWORDS

debt contracts, debt covenant violation, implied violation, private

lending agreements, technical default

1INTRODUCTION

Prior research provides evidence that debt covenant violations (DCV) are costly for borrowers and lead to increased

interestrates, renegotiation fees and a decreased ability to access debt markets (Beneish & Press, 1993; Roberts & Sufi,

J Bus Fin Acc. 2018;45:1195–1223. wileyonlinelibrary.com/journal/jbfa c

2018 John Wiley & Sons Ltd 1195

1196 STICE

2009a; Sufi, 2009).1Additionally, Beneish and Press (1995a) document a negative stock price reaction to the public

disclosure of a DCV in quarterly (Form 10-Q) and annual (Form 10-K) financial reports. However, managers of firms

that violate a covenant are frequently able to obtain waivers and renegotiate debt contracts with lenders before the

SEC requires them to disclose a violation (Dichev & Skinner,2002).2Consequently, many covenant violations are not

reported, and prior research has not documented their cost to shareholders. Tomore fully understand the total costs

to shareholders of violating a debt covenant, this study empirically investigatesviolations that arguably occur but are

resolved with lenders and not disclosed.

Because over one-half of all private debt contracts contain an earnings-based debt covenant, I restrict my sample

to firms whose debt contracts contain at least one earnings-based covenant.Combining covenant contract terms from

debt agreements with information publicly available at the earnings announcement date, I predict firms likely to have

violateda debt covenant and provide evidence that equity investors react negatively to these implied violations, regard-

less of whether managers ever disclose that a violation occurred. Additionally,I provide evidence that there is no price

response to the disclosure of a DCV in an SEC filing for firms that previously reported earnings that implied a high like-

lihood of a violation. These results complement and extend the findings of prior studies that document negative stock

price reactions to the announcement of DCVsby verifying the cost to shareholders of violations subsequently resolved

with lenders but not disclosed. In contrast to prior studies that have implicitly assumed that the cost of a violation not

reported was zero (see, e.g., Beneish and Press, 1995a, I document that the three-day abnormal stock price reaction to

implied debt covenant violations is negative0.62%.

Prior work by Beneish and Press (1993) estimates that the average cost of a DCV that is attributable to increased

interest rates and renegotiation fees is between 1% and 2% of the marketvalue of equity for their sample of firms that

disclose a covenant violation. In a subsequent study, Beneish and Press (1995a) investigate the stock price reaction

to a DCV disclosure. They find that announcements of technical default of debt covenants are associated with a sig-

nificantly negative 3.52% return in the three-day period surrounding announcements of DCVsand that 60% of these

announcements occur on the SEC filing date of the 10-K or NT 10-K. The fact that most DCV violations are disclosed

on SEC filing dates suggests that managers tend to wait until the latest possible date under SEC regulations to reveal

the existence of an unresolvedtechnical default. 3

Because earnings are announced, on average, several weeks before firms officially submit financial statements

to the SEC (see, Alford, Jones, & Zmijewski, 1994; Griffin, Lont, & McClune, 2014), new information about possible

earnings-based DCVsmay be available to investors well before the firm officially acknowledges the violations in an SEC

filing. If new information about the likelihood of a DCV is revealed in announced earnings, then I predict that returns

will be decreasing in this likelihoodon the date of an earnings announcement. I also predict that themarket will respond

positively to impliedcovenant violations that reverse in the next period. Furthermore, if equity market participants can

accurately assess the likelihood that a violation took place after observing the information in the earnings announce-

ment, then I predict that the negative price response to subsequent DCV disclosures typically observed on the SEC

filing date will be attenuated.

Totest these predictions, I construct an implied covenant violation measure using the reported earnings from the

announcement, prior-period financial statement information available at the earnings announcement date, and debt-

covenant-specific information from Dealscan. Because covenants based on earnings are the most commonly included

covenants in debt contracts and prior research has documented their increased use over time, I conduct my analy-

ses using the two most prevalent earnings-based covenants – debt-to-EBITDAand interest coverage.4Using earnings

1I use the term covenant violation throughout the paper to refer to one form of violation – technical default (or technical violation). This form of violation,

basedon accounting numbers, is distinct from other forms of violation (e.g., payment default, borrowing or paying out dividends in excess of base limits).

2SEC Regulation S-Xstates that ‘any breach of covenant ... which default or breach existed at the date of the most recent balance sheet being filed and which

hasnot been subsequently cured, shall be stated in the notes to the financial statements’ (SEC Regulation S-X Rule 4-08, Paragraph c, emphasis added).

3Usinga sample of recent covenant violations, Griffin, Lont, and McClune (2014) report that managers persist in delaying disclosure of DCV until the date of

theSEC filing. Of the 1,718 covenant violations in their sample, 94.2% of covenant violations were disclosed in the 10-K or 10-Q, with the rest appearing in an

8-K.I discuss the robustness of my findings to 8-K disclosures in Section 4.

4Demerjian (2011, Table1) reports that debt-to-earnings covenants are the most commonly included financial covenants in private debt contracts, appear-

ing in 53% of contracts in his sample. Additionally,he documents that their inclusion has increased over time – from 38% in 1996 to 65% in 2007 – though