Litigation risk: Measurement and impact on M&A transaction terms

• Published date: 01 July 2018
• Author: Hubert La Bruslerie,Julien Le Maux
• Date: 01 July 2018
• DOI: http://doi.org/10.1111/jbfa.12318

DOI: 10.1111/jbfa.12318

Litigation risk: Measurement and impact on M&A

transaction terms

Hubert de La Bruslerie1Julien Le Maux2

1DRMFinance, University Paris Dauphine, Paris,

France

2Departmentof Accounting Studies, HEC

Montreal,Montreal, Canada

Correspondence

Hubertde La Bruslerie, DRM Finance, University

ParisDauphine, Paris, France.

Email:hlb@dauphine.fr

JELClassification: G3, K2, G34

Abstract

The purpose of the paper is to propose an original proprietary proxy

of a firm's litigation risk. We extend the scope of litigation risk out-

side of the conflicts with shareholders and the domain of security

litigation. We demonstrate that the source of the risk of litigation

can be found in the firm's policies and in its management's oper-

ational or strategic decisions, even if a sector conditioning effect

exists.Based on a sample of 465 US M&A transactions between 2000

and 2014, we provide evidence that the level of litigation risk, at the

acquirer's level, has a positive and significant impact on the takeover

premium. We also provide evidence that a significant relationship

exists between the acquirer's litigation risk and the means of pay-

ment. An extension of the sample to international transactions is

used as a robustness check; it confirms the previous results.

KEYWORDS

acquisition premium, idiosyncratic risk, litigation proxy, litigation

risk, means of payment

1INTRODUCTION

Litigation risk is an important dimension of the various categories of risk to which a firm is exposed. Litigation risks

cannot be defined only with regard to class action lawsuits. The scope of litigation risk is now larger than the narrow

definition limited to the conflicts with shareholders. The recent exampleof Volkswagen AG, which was deeply involved

in the so-called “dieselgate”in September 2015, offers a strong illustration of the huge financial consequences of this

case. The firm's financial risk was estimated to be between US$20 and 45 billion. The provision accrued by the Volk-

swagen group in its 2015 annual report was €16.2 billion, which was considered too low even though its management

claimedit was sufficient. Exclusively in the United States, an out-of-court agreement for US$23 billion was concluded in

2017 with the Department of Justice. Disputes and proceedings are still pending with other countries’ administrations

and with consumers’ associations for tens of billions in US dollars. These discrepancies among the different evaluations

of the litigation risk in the VW case highlight the issue of measuring litigation risk. The financial valuation of litigation

risk is based on the assessment of future costs, not ex-post costs, as recorded and disclosed in annual reports. However,

the double question of the precise definition and measure of litigation risk has only partially been addressed in the

literature.

952 c

2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/jbfa JBus Fin Acc. 2018;45:952–996.

DE LA BRUSLERIE ANDLE MAUX 953

Traditional measures of litigation risk are indirect measures. For instance, misreporting and financial statement

errors are used as predictors of class action by shareholders (Bhagat & Romano, 2005; Wiedman & Hendricks, 2013).

Audit fees are also an indirect measure of the auditors’ responsibility and the litigation risk to which they are exposed

(Simunic & Stein, 1996; Taylor, Simon, & Burton, 1999). Another way to identify the firm's litigation risk is to consider

the directors’ and officers’ personal responsibilities. Directors and officers are exposed to litigation risk, which may be

covered by personal insurance.The characteristics of the directors’ and officers’ insurance contracts and their pricing

is an indirect measure of the firm's litigation risk (Baker & Griffith, 2007; Cao & Narayanamoorthy,2014). IPO under-

pricing has also been related to litigation risk as a way to limit the probability of triggering class action by disappointed

buyers (Lowry & Shu 2002). These measures are only partial since they follow the risk as appraisedby a specific party

economically and contractually linkedto the firm: auditors, directors, officers, and class action-involved shareholders.

A direct measure of litigation risk originates from Francis, Philbrick, and Schipper (1994), who identify an indus-

try causal effect on litigation risks. Firms belonging to the biotechnology, computer,electronics, and retail sectors are

exposed to litigation in the very restricted sense of security and reporting litigation. This classification is based on the

simple fact, evidenced by the authors, that these sectors were exposed to “a high incidence of litigation during 1988–

1992” (p. 144). This analysis is quite old and is restricted to US firms and to class action suits, so is it still valid? We

can extend the concern and question whether the industry membership method is a sufficient proxy. The sectorial

approach seems poor, as it ignores the firm's characteristics and introduces a type ofmacro-determinism at the firm

level. Clustering lawsuits within an industry may be explainedby serial behaviors and peer behaviors (Aktas, De Bodt,

& Roll, 2011). Kim and Skinner (2012) question the validity of a pure industry-based proxy. Theyfind that this proxy

should definitely be supplemented by firm-specific variables, such as turnover and stock volatility.The latter is a well-

known (market) measure of a firm's idiosyncraticrisk.

However, the VW example highlights the need to investigate litigation risk away from the narrow scope of class

action and disputes with shareholders. Legal or judicial disputes, breaches of regulation or conflict for any reason with

any type of stakeholder, anywherein the world in which the firm has interest or its products are sold contribute to

ligation risk.

The first research question in this article deals with defining a direct measure based on a proxy considering an

extendedview of litigation risk. We investigate if the number of publications in the media may be a relevant new proxy.

Looking back to the Volkswagen case, the total number of citations of VW in the litigation and crime section of the

Factiva database amounted to 23,649 in the year 2015. Before the “dieselgate”outburst, approximately 80 citations

were recorded each month. From September until the end of the year,the average number of monthly citations was

5,700, implicating VW with litigation, corporate fraud, and legal proceedings. The VW story also seems relevant for

following the ex-ante litigation risk for other competitors in the industry. Looking at General Motors, the total number

of citations in the corporate crime and litigation section in the year 2015 was 1,524. However,the average number of

publications linkedto litigation was approximately 110 per month until September; interestingly, it surged to 165 pub-

lications per month after September. The rise of litigation risk due to corporate fraudor legal proceedings increased

globally throughout the industry.

From a financial point of view,we focus more on the consequences of litigation risk than on its sources because lit-

igation risk appears and develops through many different conflict possibilities between and within stakeholders. We

define litigation risk as the legal or contractual costs linkedto any type of dispute with any type of stakeholders. Litiga-

tion risk refers to uncertain cash flows resulting from litigation or disputes for any source or cause. Litigation risk (like

any risk) is ex ante by nature. For instance, litigation risk is uncertain as long as a court decision is pending. Using our

proxy,we can question whether litigation risk is random, for instance, if it is linked to a complaint from a plaintiff who

sues the firm for any reason. Alternatively,firms belonging to given sectors are systematically exposed to complaints

and disputes. If the management's decisions are systemically borderline or repeat themselves, litigation risk shows a

stable component. This is also the case if a sectorial effect conditions decisions and business practices.

Our second research question deals withc rossing the enlarged definition of litigation risk with a specific eventin the

firm'slife, i.e., an M&A transaction. When a firm is acquired, it may cause additional or reveal existing litigation risks. We

can follow it through important events such as M&As. Acquisitions of target firms are shocks, and we anticipate that

954 DE LA BRUSLERIE ANDLE MAUX

litigation risk is affected after a completed transaction. Shareholders’ situation changes as they face a new economic

project. The situation of any other firm's stakeholders is similar.However, by itself, the acquisition and the merging of

two firms may induce new litigation risks. For instance, the acquisition decision may itself stimulate shareholders to

complain, or regulation bodies or competitors may raisequestions or develop proceedings. However, at the same time,

the acquisition may also reduce the litigation risk if an acquirer merges with a conflicting competitor and settles past

disputes. We recognize that litigation risk is an endogenous variable, which mayexplain an acquisition and the terms of

the acquisitions. However,an M&A transaction may also explain changes, i.e., it may explain an increase or decrease in

litigation risk. We are awarethat litigation risk may contribute to the occurrence of an acquisition. Acquisitions are not

pure exogenousshocks, and they may affect the firm's litigation risk level in a much more complex way,as if we had con-

sidered, for instance, legal rule changes. We do not address questions of how litigation risk influences the occurrence

of an acquisition or how the acquisition terms will explain the probability of a future litigation case after a transaction.

We use M&As as events that allow us to empirically analyze the dynamic of litigation risk. The literature on M&As

and litigation risk is limited. The question has important consequences as the litigation risks may condition the setting

of the transactionterms. The issue of an informational risk linked to possible financial restatements after an acquisition

has been investigatedby Amel-Zadeh and Zhang (2015). We aim to contribute to the literature on this issue. By looking

only at completed transactions, our research question addresses whether and how litigation risk will influence the

setting of the transaction terms, i.e., their premiums and means of payment.

When an M&A transaction occurs, litigation risk is privately assessed by the parties and should impact the terms

of the transaction. The litigation risk may, for instance, lower the price paid. This question is complex:in an acquisi-

tion operation, we need to consider two litigation risks: that of the acquirer and that of the target. We will investigate

whether the risk of litigation is effectively priced in M&A terms. This question has not yet been extensivelyaddressed

in the literature (exceptby Le Maux & Francoeur, 2014). Krishnan, Masulis, Thomas, and Thompson (2012) and Yuan

and Zhang (2015) similarly developedstudies crossing litigation risk and contractual pricing. However,the scope of lit-

igation is limited to class action lawsuits by shareholders, and they refer to effective litigation cases and not to ex-ante

litigation risk. Moreover,Krishnan et al. (2012) focus only on M&A premiums, and Yuan and Zhang (2015) focus on the

interest rate of debts.

The first contribution of this paper is that it proposes an original proprietary proxy for a firm's litigation risk.

We develop a direct measure based on the number of publications of news related to litigation cases or disputes in

the media system. We extend the scope of litigation risk outside of the conflict with shareholders and the domain

of security litigation. The litigation literature commonly refers to class action lawsuits developed in US courts (e.g.,

Rogers & Stocken, 2005; Yuan & Zhang, 2015). Our approach recognizes that litigation risk can find its source in the

firm's policies and in its management's operational or strategic decisions, even if a sector-conditioning effect exists.

We show that our litigation risk proxy based on the number of citations is relevant, as it is highly correlated with

the ex-post litigation costs accounted for in financial statements. Our broadened definition of litigation risk leads to

the consideration of many possible sources of litigation disputes outside of security class action lawsuits. We show

that litigation risk has a stable component at the firm level and is an idiosyncratic feature of the firm that has a rel-

atively stable component. We measure the litigation risk for different given periods of time for a sample of firms

involved in M&A deals. The issue of a continuous and direct proxy of litigation risk exposure is relatively new in the

literature.

Another contribution of the paper is to show that, when a firm is acquired, it may cause additional or revealexisting

litigation risks and that this influences the terms of the transaction.We demonstrate that the drivers of the association

between litigation risk and acquisition terms should be found not only at the sector level but also at the firm level.The

empirical part of the paper refers to a sample of M&A transactionsin North America and Europe over the period 2000–

2014. An M&A transaction may change the level of litigation risk because it per se introduces new risk ensuing either

from the transactionterms or imported from the target firm. We confirm the hypothesis that the litigation risk is priced

in the M&A terms and influences the choice of the means of payment. A practical implication of our paper is that the

litigation risk proxy is known ex ante, and as such, it may help the acquirer and its investment bank fine tune the setting

of the transaction terms.