Contracting on litigation

• Author: J.J. Prescott,Kathryn E. Spier
• Date: 01 June 2019
• Published date: 01 June 2019
• DOI: http://doi.org/10.1111/1756-2171.12274

RAND Journal of Economics

Vol.50, No. 2, Summer 2019

pp. 391–417

Contracting on litigation

Kathryn E. Spier∗

and

J.J. Prescott∗∗

Two risk-averse litigants with different subjective beliefs negotiate in the shadow of a pending

trial. Through contingent contracts, the litigants can mitigate risk and/or speculate on the trial

outcome. Contingent contractingdecreases the settlement rate and increases the volume and costs

of litigation. These contingent contracts mimic the services provided by third-party investors, in-

cluding litigation funders and insurancecompanies. The litigants (weakly) preferto contract with

risk-neutral third parties when the capital market is transaction-cost free. However, contracting

with third parties further decreases the settlement rate, increases the costs of litigation, and may

increase the aggregate cost of risk bearing.

1. Introduction

This article studies contingent settlement contracts in litigation, exploring both the deals that

are struck between the litigating parties themselves and their agreements with outside investors.

Traditionally, scholars have viewed settlement as a simple transfer payment from a defendant to

a plaintiff in exchange for the plaintiff abandoning a claim.1However, in reality, parties can and

often do write detailed contracts before trial that turn on the future trial outcome. We explicitly

account for this by allowing litigating parties to write general contracts with each other that

are contingent on the outcome of litigation. Then, placing lawsuits into a market context, we

compare these “inside” contracts to the “outside” contracts offered by competitive third-party

investors. Although inside and outside contracts create value in similar ways, we show that

contingent contracts between the litigants themselves may lead to relatively fewer trials, less

wasteful litigation spending, and less aggregate risk.

∗Harvard University and NBER; kspier@law.harvard.edu.

∗∗University of Michigan; jprescott@umich.edu.

Wethank Philippe Aghion, Lucian Bebchuk, James Dana, Andrew Daughety, John Goldberg, Oliver Hart, James Hosek,

Louis Kaplow, Claudia Landeo, Alex Lee, Jonathan Molot, Susan Norton, Eric Rasmusen, Jennifer Reinganum, Bill

Rubenstein, Martin Schmalz, Anthony Sebok, Holger Spamann, Glenn Weyl, and two anonymous referees for helpful

comments and suggestions. We also thank seminar audiences at the 2015 NBER Winter Meetings, the 2015 American

Lawand EconomicsAssociation Conference at Columbia University, the 2014 Law and Economics Theory Conference at

U.C.Berkeley, the Harvard Law School in 2015, and the University of Amsterdam in 2015. Spier acknowledgesfinancial

support from NSF grant no. SES-1155761.

1Surveys include Spier (2007) and Daughety and Reinganum (2012).

C2019, The RAND Corporation. 391

392 / THE RAND JOURNAL OF ECONOMICS

Contingent settlement contracts appear in many different legal contexts and take a varietyof

forms. Consider the following examples. In an automobile liability case, a $125,000 jury award

was reduced to just under $94,000 because the parties agreed in advance to a 75%/25% split of

any court-awarded damages.2In a high-stakesmedical malpractice case, a $30 million jury award

was reduced to $5.3 million pursuant to a “high-low” contract signed by the parties before trial.3

In yet another lawsuit, the parties agreed to a damages payment of $6,000 if the jury found the

defendant to be less than 50% at fault, $11,250 if she were found to be exactly 50% at fault, and

$22,500 if she were found to be more than 51% at fault.4Contingent contracts with third-party

financial service providers, including insurance companies and litigation funders, have become

increasingly common as well.

This article explores the positive and normative implications of contingent settlement agree-

ments in a model with two risk-averse parties, a plaintiff and a defendant. At trial, the factfinder

(who may be a judge, a jury, or an arbitrator) will award damages. Trials are costly and risky,

and the parties have potentially different subjectivebeliefs about the likely outcome. The parties’

subjective beliefs, preferences, and litigation costs are assumed to be common knowledge, so

negotiations take place under complete information. The parties may decide to completely settle

out of court, thereby ending the dispute and avoiding the risks and costs of trial. Through a simple

out-of-court settlement, the defendant is effectively purchasing 100% of the plaintiff’srisky legal

claim. Alternatively, the parties may “agree to disagree” and bring the dispute to trial. In this

environment, the litigating parties may enter into contingent agreements with each other and/or

with outside investors.

First, ignoring the external capital market, we show that the parties will write an inside

contract that specifies a lump-sum payment and a contingent payment that is monotonic in the

likelihood ratio of their subjective beliefs. If the parties have constant absolute risk aversion

(CARA) expected utility and their beliefs are normally distributed with divergent means, then

the defendant pays the plaintiff a guaranteed lump sum and a fixed proportion of the court-

determined damages. These contingent settlement contracts bear a striking resemblance to the

financial contracts traditionally offeredby third-party investors. Through the contingent settlement

contract, the defendant is in effect buying a partial equity stake in the plaintiff’s claim. Similarly,

through the contract, the plaintiff is effectively selling an insurance policy to the defendant.

Finally, we allow the litigating parties to write contingent contracts with outside investors.

These investors are risk neutral, share common beliefs, and operate in a competitiveenvironment.

In these idealized circumstances, the litigating parties jointly prefer to write financial contracts

with third-party investors rather than with each other (although this preference is weak). Because

the parties perceive themselves to be better off with the backing of outside investors, some

cases that would otherwise have settled will go to trial instead. Thus, with outside investors, the

settlement rate falls and the litigation rate rises. Interestingly, we show that the optimal contracts

with outside investors may actually expose the litigating parties to more risk rather than less.

Insofar as these contracts increase both the aggregate cost and risk of litigation, third-party

involvement in litigation reduces social welfare.

Litigation literature. This article takes the literature on the economics of litigation in a

new direction. Many scholars have argued that settlement negotiations may fail when the parties

have divergentbeliefs or noncommon priors about what will happen at trial (Landes, 1971; Posner,

2Palimere v. Supermarkets Gen., No. 05186, 1989 WL 395822 (Pa.Com.Pl.Dec. 1989) (Verdict and Settlement

Summary).

3Andersen (2013). With a high-low agreement, the “defendant agrees to pay the plaintiff a minimum recoveryin

return for the plaintiff’sagreement to accept a maximum amount regardless of the outcome of trial” (Garner, 2004).

4Claudia Clemente v.Lisa Duran, No. L-003405-04, 2006 WL 4643243 (N.J.Super.L.2006) (Verdictand Settlement

Summary).

C

The RAND Corporation 2019.