Race and Gender in the USPTO: Schuster’s Hard Data for Hard Issues

[I asked some of my RAs to write guest posts this summer, lightly edited by me.  This one is by Jennifer Black, a 3L at Villanova University Charles Widger School of Law]

Intellectual property rights are just that: rights.

Much like other rights, however, they have been unequally granted to people based on factors outside of their control throughout our country’s history. Intellectual property is a means for upward mobility of individuals who, through their own ingenuity, creativity, or otherwise, contribute something of value to our society. It is this exchange of benefits that the patent system is built upon. However, when certain individuals are less likely to reap the rewards of their inventions, they are both disincentivized from creating as well as from engaging with the patent system. Although the extent of these biases is yet unknown, research regarding the subject has been conducted with the intent of identifying and remedying inequity.

The scope of this inequity is difficult to comprehend except by collecting, analyzing, and comprehending the data. Mike Schuster and his coauthors did just that in his article, An Empirical Study of Patent Grant Rates as a Function of Race and Gender (published version in the American Business Law Journal), which examines the patent granting rates as a function of inventors’ races and genders. As scientists and engineers, patent practitioners and examiners will undoubtedly appreciate the amount and quality of his data.

Schuster’s article first focuses on the patent system’s bias against women. While women have come far in their representation in the patent system—from 0.3% of patents in the first 100 years of the United States to 12% in 2016—this is a far cry from equality in a country that is 50.8% female.

Schuster’s study regarding female inventors was twofold: first, he hypothesized that female inventors would be granted patents at lower rates and second, he hypothesized that this disparity would decrease for female inventors with gender nonobvious names. The former hypothesis was supported, yielding a disconcerting result. Women were found to be 62% as likely as male inventors to have their patents granted. This gap narrowed for female inventors without gender identifying names.

Much like female inventors, certain racial minorities were also found to receive patents at lower rates than white counterparts. These numbers, however, were less thoroughly presented and discussed. There were some indications that different racial minorities have different experiences at the USPTO. For example, Asian applicants were indicated to have better outcomes than Black and Hispanic applicants.

Women and minorities’ struggles to obtain patents are a product of aggregating levels of barriers.

First, female and minority students are systematically underrated by teachers and they are discouraged from pursuing STEM and engaging in invention. This has been demonstrated in a number of studies that have teachers rate students’ academic performance generally, academic performance in math and science, and overall intelligence.

Next, women and minorities experience discrimination in employment situations. This is seen both at the forefront, where women and minorities are less likely to get jobs than white male counterparts despite equal credentials. Additionally, Schuster proposed the theory that employers are aware of the USPTO’s implicit biases against these affinity groups, therefore would avoid having these inventors being named first on patent application as a way of increasing odds that the company would receive its patents.

Then, there is the additional layer of implicit bias in the patent system. This was the step analyzed in Schuster’s article.

Minority and female inventors are not the only ones who should care about encouraging and demanding equality in the patent system. It has been proven time and time again that diversity promotes the “progress of science and the useful arts.” If inventors are less successful based on their race and gender, society will lose out on the benefit of diversity in invention. Harm has been aggregating since the founding of the United States—now that we are aware of the still remaining biases, we must focus on remedying them. There have been far too many “Lost Einsteins” for us to remain inactive in the face of bias.

Schuster offered potential solutions to promote equality in the face of systemic bias.

First, he suggested that patent examination should proceed anonymously. While the benefit of this change is clear—it is impossible to discriminate against an inventor you can’t identify—the drawbacks must be mentioned. During examination, inventors are entitled to file their own applications. Transitioning to an anonymous system would not benefit these inventors, as the examiners would be aware of both their pro se status as well as their affiliate groups. This could also result in inventors feeling pressure to seek a patent attorney or agent despite potential financial barriers. Additionally, inventors are permitted to attend examiner interviews, which would negate the benefit of an anonymous system. While this would not affect threshold discrimination (interviews are rarely granted before a rejection has issued), it would not prevent biases from marring the rest of the review of the patent application.

Second, he suggested education as a way to mitigate implicit bias. These biases have been identified in health care, criminal justice, academia, employment, and the judicial as well as the patent system. Each of these systems have proven to benefit from education aimed at the identification, acknowledgment, and mitigation of these biases. However, these trainings must be used in conjunction with real, concerted efforts aimed at preventing such biases from seeping into the patent review process.

Therefore, I believe that the education of patent examiners should be bolstered by two additional factors: maintaining records of inventors’ information and monitoring examiners’ grant rate in light of this information. Much like employment demographic data, applicants maintain the right to refuse to share these details, however, the presence of use of these data would be invaluable. Rather than commentators such as Schuster having to extrapolate the race and gender of various inventors, having this information readily available for internal use and monitoring would allow the Patent Office to prevent such issues from continuing.

While it may cause discomfort among the examiners to know that their grant rates are being monitored, that discomfort is necessary in light of the data. When weighing examiner discomfort against rights granted to all people in the Constitution, the latter more than tips the scale. Growth does not come without difficulty and our system is in dire need of change. I do believe that the increasing depth and regularity of data-based studies such as Schuster’s provide some much-needed accountability in the system and for that, I cannot recommend reading this study enough.