An Empirical Study of Patent Grant Rates as a Function of Race and Gender

• Author: W. Michael Schuster,R. Evan Davis,Julie Ravenscraft,Kourtenay Schley
• Published date: 01 June 2020
• Date: 01 June 2020
• DOI: http://doi.org/10.1111/ablj.12159

American Business Law Journal

Volume 57, Issue 2, 281–319, Summer 2020

An Empirical Study of Patent Grant

Rates as a Function of Race

and Gender

W. Michael Schuster,*R. Evan Davis,** Kourtenay Schley***

and Julie Ravenscraft

****

In this article we examine the rate at which patent applications are granted as

a function of the inventor’s race and gender. Empirical analysis of more than

3.9 million U.S. applications finds minority and women applicants are signifi-

cantly less likely to secure a patent relative to the balance of inventors. Further

analysis indicates that a portion of this bias is introduced during prosecution

at the Patent Office, independent of the quality of the application. Mechanisms

underlying these disparities are explored. The article concludes with a discus-

sion of our results and their interaction with patent law, innovation policy,

and employment trends.

INTRODUCTION

Society benefits when inventors come from all walks of life.

1

Unfortu-

nately, female and minority (nonwhite) inventors have historically been

*Assistant Professor at the University of Georgia’s Terry College of Business. The authors

would like to thank Lex Washington, John Holden, Greg Day, Aaron Hill, Bryan Edwards,

and Marten Hillen for their help on this project. The authors also thank the participants at

the 2019 Race + IP Conference at the New York University School of Law and the 2018

Academy of Legal Studies in Business Annual Conference.

**Assistant Dean and Professor of Professional Practice at Oklahoma State University’s

Spears School of Business.

***PhD candidate at Oklahoma State University’s Spears School of Business.

****

PhD candidate at Oklahoma State University’s Spears School of Business; Instructor of

Accountancy, Missouri State University.

1

Alex Bell et al., The Lifecycle of Inventors 13 (June 5, 2016), (unpublished manuscript)

https://scholar.harvard.edu/files/xavier/files/inventors.pdf.

©2020 The Authors

American Business Law Journal ©2020 Academy of Legal Studies in Business

281

underrepresented among U.S. patent holders.

2

While research on aggre-

gate patents granted to underrepresented groups finds this trend to be

improving,

3

a related question remains underexplored. Rather than sim-

ply counting the number of patents granted, we investigate the likeli-

hood that a minority or female patent applicant succeeds in receiving

their requested patent.

To conduct this study, the named inventors from approximately 3.9

million patent applications were coded as having particular demographic

attributes. This analysis required comparison of inventors’ names to gen-

der and race databases because the U.S. Patent and Trademark Office

(USPTO or Patent Office) does not collect this information. With that

data, and controlling for other relevant variables, we find minority appli-

cants had their patent applications granted significantly less often than

the balance of applicants.

4

The current study is the first to address patent

grant rates as a function of the inventor’s race. We additionally observe

that female applicants were less likely to secure a patent

5

—a finding con-

sistent with past work, which we confirm using different databases.

6

The second stage of the analysis compares grant rates for identifiably

female inventors (those with a feminine name) versus female inventors

with androgynous names. We find women whose gender was identifiable

by patent examiners were significantly less likely to secure a patent rela-

tive to women whose gender was not apparent.

7

In contrast, minority

inventors with race-neutral or white-sounding names were no more

likely to secure a patent compared to minority inventors with nonwhite,

2

Holly Fechner & Matthew S. Shapanka, Closing Diversity Gaps in Innovation: Gender, Race,

and Income Disparities in Patenting and Commercialization of Inventions,19T

ECH.&INNOVATION,

727, 729 (2018); K.J. Greene, Intellectual Property at the Intersection of Race and Gender: Lady

Sings the Blues, 2 J. GENDER,SOC.POL’Y& L. 365, 365 (2008); Cassidy R. Sugimoto et al., The

Academic Advantage: Gender Disparities in Patenting,10P

LOS ONE 1, 5 (2015).

3

U.S. PATENT &TRADEMARKOFFICE,PROGRESS AND POTENTIAL:APROFILE OF WOMEN INVENTORS

ON U.S. PATENTS, https://www.uspto.gov/learning-and-resources/ip-policy/economic-research/

progress-potential (last visited Mar. 31, 2020).

4

See infra Table 1.

5

Id.

6

Kyle Jensen et al., Gender Differences in Obtaining and Maintaining Patent Rights,36NATURE

BIOTECH. 307, 307 (2018).

7

See infra Table 3 in Appendix A.

282 Vol. 57 / American Business Law Journal

race-specific names.

8

This analysis indicates at least some of the bias

against female inventors’ applications is introduced at the Patent Office,

though it does not support the same conclusion for minority applicants.

Our article proceeds in five parts. Part I conducts a literature

review of the interplay between race, gender, and patent activity. It

additionally describes a potential subconscious mechanism underly-

ing biases against traditionally underrepresented inventors—

implicit bias. Part II presents our expectations. Premised upon the

literature, we expect lower grant rates for female and minority

inventors’ applications, including some interaction effect between

these attributes. Likewise, we hypothesize that a negative bias is

introduced during examination, if the patent examiner is able to

identify minority or female inventors. That recognition may trigger

certain subconscious, implicit biases. Part III tests our hypotheses

using data from more than 3.9 million patent applications from

2000 to 2015. Through our analysis we find support for our primary

hypotheses related to grant rates for minority and women inventors.

Our analysis does not, however, find additional negative biases (nega-

tive interaction effects) for inventors that are both female and non-

white. Part IV summarizes our results, finding lower patent grant

rates for female and minority inventors, and presenting evidence

that some of the bias against women is introduced at the Patent

Office. Lastly, Part V describes social welfare concerns implicated by

our findings. We conclude with modest suggestions on how to rem-

edy these problems.

I. BACKGROUND

This part reviews current research on the interaction of patent activity,

gender, and race. Particular emphasis is given to historical underrepre-

sentation of particular segments of society to set the stage for our analysis

of application grant rates as a function of race and gender. Further dis-

cussion is given to potential subconscious partialities against nonwhite

and female applicants during patent examination—namely, implicit

biases.

8

See infra Table 2.

2020 / Function of Race and Gender 283