The 2011 America Invents Act: Does it Undermine Innovation?

• Published date: 01 June 2015
• DOI: http://doi.org/10.1111/jems.12092
• Author: Kaz Miyagiwa
• Date: 01 June 2015

The 2011 America Invents Act: Does it

Undermine Innovation?

KAZ MIYAGIWA

Department of Economics

Florida International University

Miami, FL 33199; and

ISER, Osaka University

Osaka, Japan

kmiyagiw@fiu.edu

With the 2011 America Invents Act, the United States discarded its century-old first-to-invent

patent-awarding system in favor of a first-to-file rule. Critics have argued that the first-to-file rule

rewards speed in patent applications rather than creativity, therebyundermining innovation. We

evaluate this concern within a dynamic model of a patent race, and find first-to-invent (weakly)

more conducive to innovation than first-to-file. Defending prior users’ rights can promote both

pro- and anti-R&D effect of a switch to first-to-file.

1. Introduction

On September 16, 2011, President Obama signed into law the Leahy-Smith America

Invents Act (AIA), creating the most significant change in the U.S. patent system since

1953. The main feature of this new legislation is a change in the rule that establishes

priority of invention between competing inventors. Now, in a case of competing patent

claims, the United States awards the patent to the person who first files a patent appli-

cation. Historically,the United States, unlike the rest of the world, issued a patent to the

person who could prove to have made the initial discovery.1The U.S. government had

been under considerable pressure to adopt a first-to-file rule, but all previous efforts to

comply with the rest of the world failed in the face of powerful opposition.2

The principal argument against first-to-file has been that it undermines innovation

by rewarding the person who quickly files a patent application instead of the initial

discoverer.To bolster this argument, critics of first-to-file adduce the fact that the United

States has led the world in innovation for more than a century and attribute that feat

to the first-to-invent feature of the U.S. patent law.3This leads us to our first research

question: does a transition from first-to-invent to first-to-file undermine innovation?

The author is grateful to an anonymous referee and the co-editor for helpful comments and suggestions for

revising the paper, and thanks Yuka and seminar participants for their comments on earlier versions. The

author also thanks CORE and the Universit´

e catholique de Louvain for their hospitality during his visit.

A research grant from Microsoft’s Academic Research Program is gratefully acknowledged. Any opinions

expressed or remaining errors are the author’s.

1. Canada switched to a first-to-file system in 1989, and the Philippines in 1998, making the United States

the last country with a first-to-invent system.

2. A first-to-file rule was contained in the Patent Act of 2005 introduced by Rep. Lamar Smith, R-Texas.

Also, at the 2007 G8 Summit meeting in Germany, PresidentGeorge W. Bush formally stated his support for

a first-to-file rule.

3. The following statement typifies this view: “It should be understood that it is because the United States

has a first-to-invent structure and the rest of the world has a first-to-file structure that the United States

is the production and employment machine that it is.” (http//www.piausa.org/layout/set/print/patent_

reform_issue.)

C2015 Wiley Periodicals, Inc.

Journal of Economics & Management Strategy, Volume24, Number 2, Summer 2015, 211–227

212 Journal of Economics & Management Strategy

In fact the AIA addresses critics’ concern, as it contains a provision enhancing the

concept of prior user rights defense, which gives the original inventor immunity for

patent infringement, in case other individuals gain the patent. Previously,the prior user

rights defense was applicable to innovations in business methods only under the 1999

American Inventors Protection Act. Now, with the cessation of first-to-invent, the AIA

has expanded the scope of prior user rights to cover all innovations. This leads us to a

second research question: does the prior user rights defense encourage R&D under the

AIA?

The main objective of this paper is to address these two questions. Tothat end, we

draw on the classical Loury (1979) model of a patent race, in which invention occurs with

a probability that depends on the size of a firm’s initial investment in an R&D facility.

This framework reflects the finding of Cohen and Ishii (2005) that (1) most patent races

are among major corporate research laboratories chasing well-defined research topics;

and (2) in such races, the initial setups in R&D are of utmost importance.

We however depart from the Loury model in that upon discovery a firm must

choose one of two filing strategies: file a patent application immediately or improve on

the invention. Improving on invention is time-consuming but makes a better product,

and is considered the preferred choice. In the presence of R&D rivalry, however, the

decision to improve exposes the inventor to the risk of losing the patent to a rival that

files an application first.

Two sources of informational asymmetry influence the timing of a firm’s decision

to file, given the risk of preemption. First, when a firm makes a discovery, it does not

inform the rival (and the general public). Second, firms do not observe each other’s filing

strategy. As a result, neither firm knows whether the rival has discovered an invention

until it gets patented or a product embodying it appears in the market.

Our results show that a transition from first-to-invent to first-to-file decreases (or

leaves unchanged) the levels of investment in R&D except in special cases. Thus, we

conclude that first-to-invent is in general (weakly) more conducive to innovation than

first-to-file. This result is invariant to the presence of prior user rights. Interestingly

enough, though, prior user rights can exacerbate the negative impact of a switch to

first-to-file.

The theoretical prediction of this paper enjoys some empirical support. In 1989,

Canada made a switch similar to the AIA. Studying this natural experiment, Lo and

Sutthiphisal (2009) have found that the switch to first-to-file failed to stimulate Canadian

R&D efforts. More recently, Abrams and Wagner(2012) have shown that Canada’s switch

resulted in significant drops in patent applications from individual and small-scale

corporate inventors.

Despite the heated debate concerning the AIA, there exists only a scanty literature

that formally compares the effectsof first-to-file and first-to-invent on R&D investments. 4

Scotchmer and Green (1990) examine this issue in a model, where two firms compete

for two sequential innovations. Two assumptions they make set their model apart from

ours. First, the innovations become public knowledge without being patented; that is,

information is perfect and complete. Second, innovations are discovered with fixed

probabilities, which leaves unexamined the impact on R&D incentives from a switch to

4. Aoki and Prusa (1996) compare the information disclosure rules between the United States and the

Japanese. This issue was resolved when, with the 1999 Intellectual Property and Communications Omnibus

Reform Act, the United States adopted the international norm of disclosing patent applications after 18 months

from a filing.