INDUCING VARIETY: A THEORY OF INNOVATION CONTESTS

• Date: 01 November 2019
• Author: Igor Letina,Armin Schmutzler
• Published date: 01 November 2019
• DOI: http://doi.org/10.1111/iere.12403

INTERNATIONAL ECONOMIC REVIEW

Vol. 60, No. 4, November 2019 DOI: 10.1111/iere.12403

INDUCING VARIETY: A THEORY OF INNOVATION CONTESTS∗

BYIGOR LETINA AND ARMIN SCHMUTZLER1

University of Bern, Switzerland; University of Z¨

urich, Switzerland

This article analyzes the design of innovation contests when the quality of an innovation depends on the

research approach, but the best approach is unknown. Inducing a variety of research approaches generates an

option value. We show that suitable contests can induce such variety. The buyer-optimal contest is a bonus

tournament, where suppliers can choose only between a low bid and a high bid. This contest implements the

socially optimal variety for a suitable parameter range. Finally, we compare the optimal contest to scoring

auctions and fixed-prize tournaments.

1. INTRODUCTION

The use of contests to procure innovations has a long history, and it is becoming ever more

popular. Recently, private buyers have awarded the Netflix Prize, the Ansari X Prize, and the

InnoCentive prizes. Public agencies have organized, for instance, the DARPA Grand Chal-

lenges, the Lunar Lander Challenge, and the EU Vaccine Prize.2Reflecting the increasing

importance of these prizes, a literature on contest design has developed. This literature focuses

almost exclusively on how incentives for costly innovation effort can best be provided. How-

ever, effort is not the only important requirement for a successful innovation. A case in point

is the 2012 EU Vaccine Prize to improve what is known as the cold-chain vaccine technology.

The ultimate goal of the prize was to prevent vaccines from spoiling at higher temperatures,

which is particularly challenging in developing countries. The rules of the competition contain

the following statement:

It is important to note that approaches to be taken by the participants in the competition are not

prescribed and may include alternate formulations, novel packaging and/or transportation techniques,

or significant improvements over existing technologies, amongst others.3

This statement explicitly recognizes the fundamental uncertainty of the innovation process:

Even when the buyer communicates a well-specified objective (such as finding a way to prevent

vaccine spoilage), neither she nor the suppliers will necessarily know the best approach to

achieving this goal. This uncertainty about the quality of innovation resulting from a particular

approach will only be resolved by the act of innovation itself. The innovator will therefore have

to choose between several conceivable approaches without being sure whether they lead to

the goal. If innovators pursue different approaches, chances are higher that the best of these

∗Manuscript received June 2017; revised January 2019.

1We are grateful to the associate editor and three anonymous referees for helpful comments and to Raphael

Anthamatten, Jos´

e Apesteguia, Jean-Michel Benkert, Dirk Bergemann, Alessandro Bonatti, Stefanie Bossard,

Bernhard Ganglmair, Claudia Geiser, Ulrich Kamecke, Navin Kartik, Alessandro Lizzeri, Christian Michel, Joao

Montez, Georg N¨

oldeke, Nick Netzer, Marco Ottaviani, Konrad Stahl, and David Wettstein and to various seminar

audiences for helpful discussions. Shuo Liu provided excellent research assistance. Letina would like to thank Stanford

University for its hospitality and the Swiss National Science Foundation for financial support (projects P1ZHP1_155283

and 100014_131854). Please address correspondence to: Armin Schmutzler, Department of Economics, University of

Z¨

urich, Bl ¨

umlisalpstrasse 10, 8006 Z ¨

urich, Switzerland. E-mail: armin.schmutzler@econ.uzh.ch.

2See “Innovation: And the winner is ...,” The Economist, August 5, 2010.

3EuropeanCommission (2012), “Prize Competition Rules.” August 28, 2012. http://ec.europa.eu/research/health/pdf/

prize-competition-rules_en.pdf (accessed on April 3, 2015).

1757

C

(2019) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association

1758 LETINA AND SCHMUTZLER

approaches yields a particularly valuable (high-quality) innovation. Thus, variety of research

approaches has an option value. We therefore ask whether innovation contests can be used to

incentivize suppliers to diversify their research approaches so as to generate a high expected

value of the innovation.

In addition to the expected value of the innovation, contest design may also affect distribution.

A contest that induces diversity may yield a high expected value of the innovation and thereby

foster efficiency, but at the same time leave high rents to the suppliers. Thus, the main question of

our article will be: Which contests are optimal for the buyers, when the expected value (reflecting

the induced variety of approaches) as well as the expected payments to the suppliers are taken

into account? In addition, we address the relation between the buyer’s choice and efficiency,

asking under which circumstances the optimal contest implements the socially optimal amount

of diversity.

The diversity of potential approaches, which is highlighted in the guidelines of the Vaccine

Prize cited above, played an important role in many other examples of innovation procurement.

First, the often-cited Longitude Prize of 1714 for a method to determine a ship’s longitude at sea

featured two competing approaches.4The lunar method was an attempt to use the position of the

moon to calculate the position of the ship. The alternative, ultimately successful, approach relied

on a clock that accurately kept Greenwich time at sea, thus allowing estimation of longitude by

comparing with the local time (measured by the position of the sun). Second, when the Yom

Kippur War in 1973 revealed the vulnerability of U.S. aircraft to Soviet-made radar-guided

missiles, General Dynamics sought to resolve the issue through electronic countermeasures,

whereas McDonnell Douglas, Northrop, and eventually Lockheed attempted to build planes

with small radar cross section.5Third, the announcement of the 2015 Horizon Prize for better

use of antibiotics contains a statement similar to that in the announcement of the vaccine prize.6

Architectural contests are similar to innovation contests. A buyer who thinks about procuring

a new building usually does not know what the ideal building would look like, but once she

examines the submitted plans, she can choose the one she prefers. Guidelines for architectural

competitions explicitly recognize the need for diversity. For example, the Royal Institute of

British Architects states: “Competitions enable a wide variety of approaches to be explored

simultaneously with a number of designers.”7

Motivated by these examples, we thus focus on the design of innovation contests, with a

view toward the induced variety of research approaches. We consider a setting where both the

buyer (the contest designer) and the suppliers (contestants) are aware that there are multiple

conceivable approaches to innovation. Furthermore, none of the participants knows the best

approach beforehand. However, after the suppliers have followed a particular approach, it is

often possible to assess the quality of innovations, for instance, by looking at prototypes or

detailed descriptions of research projects. In such settings, can buyers design contests in such a

way that suppliers have incentives to provide variety? And will they benefit from doing so?

The existing literature on innovation contests mainly focuses on incentives for costly inno-

vation effort. To our knowledge, we are the first to analyze the optimal design of innovation

contests with multiple conceivable research approaches. Our baseline model is chosen to isolate

this design problem in a stark way. We assume that there are two homogeneous suppliers who

decide whether to exert costly research effort and which research approach to choose. In the

baseline, the buyer has strong instruments to induce effort: We assume that, once a supplier

4See, for example, Che and Gale (2003) for a discussion of the Longitude Prize.

5See Crickmore (2003).

6“The rules of the contest specify the targets that need to be met but donot prescribe t he methodology orany technical

details of the test, thereby giving applicants total freedom to come up with the most promising and effective solution,

be it from an established scientist in the field or from an innovative newcomer.” European Commission (2015), “Better

Use of Antibiotics.” March 24, 2015. http://ec.europa.eu/research/horizonprize/index.cfm?prize=better-use-antibiotics

(accessed on April 3, 2015).

7See Royal Institute of British Architects (2013), “Design Competitions Guidance for Clients.” http://competitions.

architecture.com/requestform.aspx (accessed on April 3, 2015).