Licensing and innovation with imperfect contract enforcement

• Author: Richard Gilbert,Eirik Gaard Kristiansen
• Published date: 01 June 2018
• DOI: http://doi.org/10.1111/jems.12240
• Date: 01 June 2018

Received: 7 December 2015 Revised: 5 December 2017 Accepted: 11 December 2017

DOI: 10.1111/jems.12240

ORIGINAL ARTICLE

Licensing and innovation with imperfect contract enforcement

Richard Gilbert1Eirik Gaard Kristiansen2

1Department of Economics, Univer-

sity of California, Berkeley,CA, USA

(Email: rjgilbert@berkeley.edu)

2Department of Economics, Norwegian

School of Economics, Bergen,Norway

(Email: eirik.kristiansen@nhh.no)

Abstract

Licensing promotes technology transfer and innovation, but enforcement of licens-

ing contracts is often imperfect. We model contract enforcement as a game with per-

fect information but probabilistic enforcement and explore the implications of weak

enforcement on the design of licensing contracts, the conduct of firms, and market

performance. An upstream firm develops a technology that it can license to down-

stream firms using a fixed fee and a per-unit royalty. Strictly positive per-unit royal-

ties maximize the licensor's profit if competition among licensees limits joint profits.

With imperfect enforcement, the licensor lowers variable royalties to avoid cheating.

Although imperfect contract enforcement reduces the profits of the licensor, weak

enforcement lowers prices, increases downstream innovation, and in some circum-

stances can increase total economic welfare.

1INTRODUCTION

A predicate for a strong intellectual property regime is the ability to enforce licensing contracts. Technology licensing contracts

commonly require the licensee to pay the licensor a combination of fixed and variable fees. Variable fees, also called “running

royaties,” promote efficient technology transferbut are vulnerable to underrepor ting. According to the Association of University

Technology Managers, royalties that vary with licensee sales accounted for about three-quarters of the revenues collected from

licensing technologies developed bymajor universities, hospitals, and research organizations over the period 2009–2012 (AUTM

Reports). Running royalties allow the licensor and licensee to share the risk of uncertain demand for the licensed technology

(Bousquet, Cremer, Ivaldi, & Wolkowicz, 1998), address managerial incentives (Gallini & Lutz, 1992; Saracho, 2002), and

provide a means for a cash-strapped licensee to finance the cost of the technology. Relative to using only fixed or variable

fees, a combination of fixed fees and running royalties allows a licensor to extract more revenue from licensees that differ in

their willingness to pay for the technology (Schmalensee, 1981). Royalties that depend on sales also allow a licensor to soften

competition for products that employ the licensed technology,potentially increasing the total profits available to the licensor and

licensees (Hernández-Murillo & Llobet, 2006; Kamien, 1992).

We study the implications of weak contract enforcement for prices and investment incentives when licenses employ per-unit

fees to soften competition, although our analysis is applicable to other licensing arrangements in which the use of variable fees

increases joint profits. The form of the intellectual property right is not crucial for the analysis. We assume that it is a patent,

although the implications of weak enforcement for the design of licensing contracts are relevant to other vertical relationships,

such as contracts between a franchisor and a franchisee and contracts between a manufacturer and downstream distributors of

the manufacturer's products.

High per-unit royalties maximize joint profits when licensees supply products that are close substitutes, but also reduce

incentives for downstreaminnovation if higher royalties reduce the marginal profit from investing. Weakenforcement constrains

We are grateful for helpful advice from ReikoAoki, Helmuth Cremer, Geoffrey Heal, Tapas Kundu, Sadao Nagaoka, Patrick Rey, Pekka Sääskilahti, William

Tucker,two anonymous referees, and participants in the Peder Sather Workshop on Industrial Organization, the Hitotsubashi UniversityWorkshop on Evolution

of Standards and Technology,the Nordic Conference on Industrial Organization IX, the IO Workshop at Korea University,and the Eighth Bi-annual Conference

on the Economics of Intellectual Property, Software and Internet at the Toulouse School of Economics.

J Econ Manage Strat. 2018;27:297–314. © 2018 WileyPeriodicals, Inc. 297wileyonlinelibrary.com/journal/jems

298 JOURNAL OF ECONOMICS & MANAGEMENTSTRATEGY

the maximum royalty that the licensor can charge without inducing licensees to cheat and, if the constraint is binding, lowerst he

profit of the licensor. Nonetheless, a central conclusion is that weak enforcementincreases incentives for innovation by licensees

in some circumstances and may increase total welfare. However, if contract enforcement is very weak,t he licensor mayabandon

the use of variable royalties to regulate downstream competition and instead choose to license a technology exclusively or

vertically integrate with one or more potential licensees.

Several authors have focused on the economic consequences of potential infringement by unauthorized technology users

and addressed licensing as a means to deter infringement that would occur without a license. Examples include Gallini (1984,

1992), Gallini and Winter (1985), Aoki and Hu (1999), and Schankerman and Scotchmer (2001). In these papers, the purpose

of the license is to offer an alternative to infringing conduct and licensing, when it occurs, is enforced perfectly. In the contexts

studied by these authors, weak patents encourage licensing, which discourages wasteful efforts to invent around the patent. In

our model, the licensed technology is valid and would be infringed without a license. Weak contract enforcement causes the

licensor to reduce royalties, which has the beneficial effects of lowering prices and increasing incentives for productive efforts

to improve upon the licensed technology.

We focus on compliance with licensing contracts and ignore the possibility that firms without a license may infringe or

invent around the licensed product. Situations commonly occur in which a firm that is not a licensee cannot practically compete

by imitating a supplier's product. The firm may require know-how, research tools, or materials that are vital to make or sell

a commercial product or may not have the technological resources to invent around the licensed product. In other settings, a

franchisee or distributor may be unable to sell a product that is a close substitute for the product supplied by a franchisor or a

manufacturer, but may act opportunistically in ways that contravene a contractual arrangement between the parties.

Imperfect enforcement of licensing contracts differs from imperfect enforcementof patents: if a patent tur ns out to be invalid,

other firms can use the technology for free. There is a public good feature involved in challenging the validity of a patent (Farrell

and Shapiro, 2008). Breach of a licensing agreement is different because it does not make the technology freely available to

others. A firm's incentives to challenge a patent are weakened by increased competition because the gains from litigating the

patent are small. In contrast, breach of a licensing contract lowers only the breaching licensee's costs and increases only that

licensee's profit.1

Our analysis is related to the literature on cumulative innovation (Green & Scotchmer, 1995; Scotchmer, 1991) in that a focus

is on innovation by technologyusers. Contract enforcement in our model is similar in some respects to the scope of protection for

upstream innovation in Green and Scotchmer(1995). Our results contrast with those in Spulber (2013), who concludes that strong

intellectual property rights complement competition in creating incentives for innovation. However, Spulber (2013) does not

consider downstream innovation by technology users. We find that when licensing contracts are enforced perfectly, an increase

in competition can cause a rights owner to charge higher royaltiesthat var y with output, whichleads to a reduction in investment

by licensees to improve the licensed products. Weak contract enforcement promotes downstream innovation by lowering the

royalties that the rights owner can profitably sustain without inducing licensees to cheat on their contracted payments. Thus,

in our model, it is not strong patent rights, but rather weak enforcement of patent rights that promotes competitive pricing and

investment.

Competition does not have an equivalent salutary effect on upstream innovation in our model as it does in Spulber (2013)

because the licensor would use variable royalties to soften competition if contracts are perfectly enforceable. Moreover, we

find that perfect contract enforcement is not welfare maximizing if downstream investment is a decreasing function of the per-

unit royalty. In that case, weak contract enforcement can increase economic welfare by promoting downstream innovation and

lowering prices.

2THE STRUCTURE OF PRODUCTION AND CONTRACTING

There is a single upstream firm denoted by 𝑈and two downstream firms denoted by 𝑗=1,2. The upstream firm invests 𝑢>0

to develop a new technology that enables production at constant marginal cost 𝑐. The technology is protected by a valid patent

which we assume, for simplicity,has infinite duration. The upstream fir m does not produce final goods, but insteadlicenses t he

technology to the downstream firms, each of which may invest 𝑒𝑗to increase the value of the technology. Production or sale of

the downstream products infringes the upstream technology and cannot lawfully occur without a license. The incremental cost

of licensing is zero.

With perfect information, the licensor could design a contract that rewards licensees only if they choose the joint-profit-

maximizing outputs and investments. Wereasonably assume that this is not feasible. Such forcing contracts would not be optimal