Diversity in Tech: The Endpoint Problem
By Kyla Windley & Edith Pan
This article is part of the PIT Policy Team’s broader series on Racial Justice and Diversity in the Tech Ecosystem. Find our introduction to the series here.
The Endpoint Problem
I tiptoed into the kitchen for a snack, thinking “Who takes a Zoom call in the kitchen?” My parents and I were all adjusting to life in quarantine; my parents were adjusting to having a daughter back at home who needs to be fed at least twenty times a day. The kitchen was not a safe place to take a Zoom call, but alas, I got the stern finger-over-lips, furrowed eyebrow look from my mother as she readied her laptop at our dining table.
As I waited for my toast, I heard the chimes and dings of people hopping into the Zoom call. Male voices slipped seamlessly into office banter — mainly centered around the excruciating black hole coronavirus had created in the NBA playoffs series. I cringed when my mother tried to join in with a joke. She doesn’t know anything about the NBA. Predictably, it landed flat. Then, a full three second pause. My mom’s polite smile faltered. The voices picked up again, pursuing another topic and leaving my mom in the dust. On my way back to my room, toast in mouth, I peeked over my mom’s shoulder. Faces were laid out across the Zoom screen, and I was astounded to see that her six colleagues were all young, white men.
I’d never thought to ask about the demographics of my mother’s workplace. I’d never thought it mattered. But in that moment, watching my mother try wholeheartedly to get along with her team, I felt a distinct pang of sympathy. I imagined the years of being the only woman, the one with a thick accent, the one who had to leave early to pick up her kids from school. She had never said a word about it to me. I wish she had, but I also realize that the “no-one-looks-like-me-in-my-workplace-and-it-makes-it-hard-to-come-to-work-every-day talk” barely edges out the “sex talk” as the last conversation a mother wants to have with her teenage daughter.
Since then, Kyla and I have been investigating the subject of diversity in tech. The statistics present a stark reality for minority employees in Big Tech and candidates looking to break in. The personal experiences and narratives we uncovered were often accompanied by disappointment, disbelief, and indignation. Realizing that we had missed one of the biggest struggles in diversifying Big Tech — what happens after you make it — this article sheds light on what we call “the endpoint problem.”
Past the Pipeline
The lack of diversity in technology is often framed as a “pipeline” problem. Disparities in educational quality start from a young age and only widen as children continue through schooling and are given varying access to technology and digital tools. Teachers, professors, and parents are guilty for nudging certain students towards STEM and others away. By the time students graduate, the pipeline problem has fully come to fruition. The pool of candidates qualified for jobs at Facebook, Apple, and Google is overwhelmingly white and male, and there is little opportunity for breaking-in. This is the “pipeline problem,” and it attributes the lack of diversity in Big Tech to factors external to the companies themselves. In future articles, we will be looking at how early education, access to technology, and recruiting homogenizes the pool of Big Tech recruits. Today, though, we’re looking past the pipeline.
While it is true that the severe underrepresentation of certain minority groups is incredibly visible by the time people reach full-time employment, the problem is only more pronounced as a person advances through their career; the lack of diversity in tech can be seen as an “endpoint problem” as well. Problems within Big Tech cause minorities to leave at higher rates, so the already homogenous workforce only becomes more and more so. This article will focus on race and gender disparities in the world’s largest technology companies. Illustrating the struggles of the largest, most well-resourced players in the tech industry will highlight how much progress remains to be made at all levels; when an underrepresented candidate has arrived for her first day of work at Google, the struggles of being underrepresented are only starting. The lack of diversity in tech should equally be recognized as an “endpoint problem.”
The Numbers
In early 2014, the flurry of bad press about Silicon Valley and its most prominent players seemed to be coming to a crescendo. Although the media was notably absent of significant commentary on tech’s lack of racial diversity, many had picked up on the fact that shiny, new Silicon Valley was hostile to minority groups. Heeding to the demand that each platform’s own workforces mirror the diversity of their user bases increasingly seemed to be a social and business imperative.
By the end of spring, the topic of diversity — and the pressure exerted by employees and outsiders alike — could no longer be ignored by tech’s biggest players. On May 28th, 2014, a little over a month after the Times article was published, Google released diversity statistics publicly for the first time. On June 25th, Facebook followed suit in a press release. On August 12th, Apple launched a webpage dedicated to the topic and released diversity statistics for the first time. On October 3rd, Microsoft did the same.
Each report followed the same rough outline: a few lofty statements about the importance of diversity, some vague promises to address the problem, and then the long-awaited statistics presented in aesthetic pie charts and pastel colors. The aesthetic graphics did little to disguise the bleak reality they illustrated.
Company to company, the numbers were so similar that they’re not worth quoting individually. Across the board, the companies’ workforces were around 70 percent male and 85 to 91 percent White and Asian, with those percentages even higher for tech-focused jobs and leadership roles. Tim Cook of Apple reiterated the company’s belief that “inclusion inspires innovation.” Facebook’s press release declared that “diversity is something that we’re treating as everyone’s responsibility at Facebook.” For Microsoft, diversity was “a business imperative” as well as “a journey that requires constant self-assessment and recommitment.” The high-minded corporatespeak did little to to address the plight of current minority employees or the role of formal company policies in creating a hostile work environment.
Big Tech Leadership
The endpoint problem is even more pronounced by the fact that very free people of color and women stay with a Big Tech company long enough to progress to a leadership position. Facebook defines leadership as “the Director level and above.” Within this definition, in 2020, 3.4 percent of FB leadership is Black compared to 2 percent in 2014. Again, given the definition, this is likely a very low increase when considering the number of individuals who were given leadership roles. And imagine if the definition of leadership were above the director role — the numbers would be even lower. When looking at the specific leadership roles filled underrepresented minorities — Facebook’s Global Head of Diversity, Google’s Chief Diversity Officer, and Microsoft’s Chief Diversity Officer — all of whom are women. But when looking at the executives pages for all of these companies, levels above Facebook’s definition of leadership, the roles filled by women or women of color are with the exception of Sheryl Sandberg tend to have titles such as Chief Diversity Officer, Chief Privacy Officer, Executive VP of HR, and VP of Environment, Policy, and Social Initiatives. Women at such large companies seem to only have a place at the top if it is non-technical at nature.
Where is the money going?
In 2014, Facebook’s Global Head of Diversity, Maxine Williams, a Black woman, outlined several steps Facebook would be taking to increase diversity at the company. One of the first statements shared by Williams is that “diversity is essential to achieving Facebook’s mission.” After sharing the company’s diversity statistics for the first time, a few highlights being women comprising only 15 percent of tech roles but 47 percent of non tech roles and having Black people across the board be two percent or less of the workforce, Facebook said in addition to sharing their diversity statistics they would be implementing a host of initiatives to work towards more diversity in the company. Despite partnering with pipeline programs such as Girls Who Code, providing unconscious bias training for current employees, and expanding the Facebook University, an internship at Facebook focused on college freshmen from underrepresented groups, Facebook still struggles to increase diversity in certain areas.
In addition to the initiatives briefly outlined above, Facebook and other big tech companies have made monetary pledges towards diversity initiatives. In 2014, Google said they had a plan to donate $150 million to a combination of outside organizations and communities as well as internal efforts. In 2015, Apple made a statement saying they would donate $50 million to tech workforce diversity efforts, $40 million going towards the Thurgood Marshall College Fund to create a database of computer science majors at historically Black colleges and universities, and $10 million to the National Center for Women and Information Technology to create a more robust pipeline of female technology workers. This is the only Big Tech company that specifically publicized where their money towards these efforts would be going. It was only in 2020 following the killing of George Floyd and resurgence of the movement against racial injustice that companies began publicly denouncing racism and re-acknowledge the lack of diversity in their companies.
And the monetary commitment that was made from Big Tech towards racial injustice did not focus much on internal efforts to increase diversity — as a few examples, Twitter’s CEO Jack Dorsey pledged $3 million to Colin Kaepernick’s Know Your Rights, Amazon pledged $10 million towards social justice and Black communities, and Google pledged $12 million to social rights groups. While these organizations are certainly important and need money to continue their operations, there are similar trends to the money that these companies devoted to internal efforts: a lot of the time it is not publicised where exactly the money goes, there is no follow-up, and the money often is not put towards internal efforts for diversity. Internship programs such as Facebook’s Facebook University for undergraduates and multiple companies’ commitment to unconscious bias training are the only continuous efforts that seem to be pointing inwards.
Racial Bias Shines Through
Companies such as Google and Facebook have implemented unconscious bias training programs for their employees, but that training does little to make the employees who are experiencing the detrimental effects of their coworkers’ unconscious biases feel at home in their places of employment.
In September of 2019, Leslie Miley, an engineering manager with decades of experience in the tech industry, was joining his colleagues in Google’s San Francisco offices. His badge was visible, clipped to his belt. He explained that a fellow Google employee, not on the security team, ran in front of him and physically stopped him, demanding to see his badge. And this is not the first time this happened to Miley, a Black man at the company, before: this “bias in badging” implicitly and insidiously sends the message to Miley and other people of color that they do not belong in this industry dominated by white men.
This specific incidence chronicled by Leslie Miley is only one of many incidences of racial discrimination towards big tech employees who are of underrepresented minorities, and leads to another problem that makes the lack of diversity in tech clearly an endpoint problem as well: retention of diverse employees.
High Attrition Rates and the Strategic Presentation of Data
Generally speaking, tech has one of the lowest retention rates of all industries: Facebook sits at the highest of Big Tech with employees spending an average of 2.02 years at the company, and Uber is at the low end with an average of 1.23 years before an employee leaves the company. While tech companies do not publicly release stratified retention rates based on the identities of their employees, the outpour of the firsthand accounts of sexual harassment, gender and race discrimination, as well as bullying and racial bias publicized through blogs, social media posts, and even lawsuits would suggest there are even higher than average turnover rates for the very groups these companies are struggling to keep.
In 2019, Apple reported that 53% of new hires in the U.S. were from historically underrepresented minorities in tech. In the same year, Google also showed their largest recorded increase in hiring Black tech employees in the U.S. . While these statistics are surely impressive, it is also important to think about the groups included in “historically underrepresented minorities”: women as well as people who identify as Black, Hispanic, Native American, or Native Hawaiian & Other Pacific Islander. What would the number have been if we pulled out women? What about Native Americans?
Regardless of the stratification of data, it is evident that the increase of diverse hires at these companies is not reflected in the overall increase of diversity. To take Google as an example, there was an unspecified number of record new Black tech hires, however, when looking at their diversity report, the attrition index for Black employees at Google is 112, 12% above overall U.S. attrition. The figures are 131, 117, 97, and 80 for Native Americans, White people, Hispanics, and Asians, respectively. And while there was a 2.7 percent decrease in White employees this past year, 2.1 percent of that decrease was towards the increase of Asian employees. Data points such as the aforementioned strongly suggest that retention of underrepresented minorities is a large part of the endpoint problem, potentially just as much of a problem as the pipeline to Big Tech.
Conclusion: The Pipeline and Endpoint Problem
The most recent batch of reports showed promising inroads towards increased diversity. At Facebook, women composed 37% of the workforce, up from 31% in 2014. The organization increased from 2% Black employees in 2014 to 3.8% in 2020, with the addendum that a majority of these new employees occupy non-technical roles. Google showcased lowering attrition rates for women and racial minorities, despite the male-to-female ratio hovering around the same point as 2014. While it can be difficult to move these percentage points significantly when companies have swelled to tens of thousands of employees, it must be acknowledged that a ton of work still remains to be done. Facebook should be applauded for almost doubling the proportion of Black employees within the organization, but 3.8% still falls short of the diversity needed to mirror the platform’s user base. Black employees compose 1.7% of technical roles at the organization, while the percentage of computer science degrees awarded to Black students each year hovers around 6%. The modest gains should be framed in perspective of the progress that still needs to be made..
As all of these companies have publicly stated some way or another, diversity is a crucial part of any organization’s success across all markets. And with tech roles in particular, a lack of diversity can be detrimental, as the biases held by those in tech roles can deeply influence the biases of the cutting-edge technologies they build. Diversity is also necessary for Big Tech company employees as well; there is harm not only to the underrepresented minority employees but also the overrepresented employees — everyone’s performance is enhanced with the introduction of talent, skills, and experience brought from a diversity of backgrounds.
The subsequent articles in this series will move backwards from the endpoint and discuss tech recruiting, VC funding, college education, and access to tech during high school and earlier. All of these touchpoints on the pipeline have a profound effect on the lack of the diversity in the tech industry.
About the Authors
Kyla Windley is a junior at Stanford studying Product Design and Creative Writing. In addition to the PIT lab, Kyla is part of the Stanford Concert Network and is working with a group on a mobile application to make social tech more social. In her free time, Kyla likes drawing, going on bike rides, and trying new restaurants.
Edith Pan is a junior at Stanford University studying Economics and Computer Science. At Stanford, Edith is a peer advisor for the Haas Center Cardinal Quarter program and Economics Department and a member of the women’s Ultimate Frisbee team. In her free time, Edith likes baking and reading murder mysteries.
Works Cited
Bock, Laszlo. “Getting to Work on Diversity at Google.” Google, Google, 28 May 2014, blog.google/topics/diversity/getting-to-work-on-diversity-at-google/.
Bui, Quoctrung, and Claire Cain Miller. “Why Tech Degrees Are Not Putting More Blacks and Hispanics Into Tech Jobs.” The New York Times, The New York Times, 25 Feb. 2016, www.nytimes.com/2016/02/26/upshot/dont-blame-recruiting-pipeline-for-lack-of-diversity-in-tech.html#:~:text=Among young computer science and,to American Community Survey data.
“Building a More Diverse Facebook.” About Facebook, 26 June 2014, about.fb.com/news/2014/06/building-a-more-diverse-facebook/.
“DIVERSITY IN HIGH TECH.” U.S. Equal Employment Opportunity Commission, www.eeoc.gov/special-report/diversity-high-tech.
“Facebook Diversity.” Diversity, diversity.fb.com/read-report/.
Foley, Mary Jo. “Microsoft: 29 Percent of Our Global Workforce Is Now Women.” ZDNet, ZDNet, 3 Oct. 2014, www.zdnet.com/article/microsoft-29-percent-of-our-global-workforce-is-now-women/.
Foley, Mary Jo. “Microsoft: 29 Percent of Our Global Workforce Is Now Women.” ZDNet, ZDNet, 3 Oct. 2014, www.zdnet.com/article/microsoft-29-percent-of-our-global-workforce-is-now-women/.
Glass, J. L., et al. “What’s So Special about STEM? A Comparison of Women’s Retention in STEM and Professional Occupations.” Social Forces, vol. 92, no. 2, 2013, pp. 723–756., doi:10.1093/sf/sot092.
Harrison, Sara. “Five Years of Tech Diversity Reports-and Little Progress.” Wired, Conde Nast, www.wired.com/story/five-years-tech-diversity-reports-little-progress/.
Hewlett, Sylvia Ann. “What’s Holding Women Back in Science and Technology Industries.” Harvard Business Review, 2 Nov. 2014, hbr.org/2014/03/whats-holding-women-back-in-science-and-technology-industries.
“How Family Life Is Held against Women in Tech.” Atavist, 21 Mar. 2016, features.crosscut.com/can-a-tech-career-and-family-life-coexist-.
Kapin, Allyson. “Sexual Harassment In Silicon Valley: Still Rampant As Ever.” Forbes, Forbes Magazine, 15 Sept. 2020, www.forbes.com/sites/allysonkapin/2020/09/15/sexual-harassment-in-silicon-valley-still-rampant-as-ever/#a4d9c732cc48.
Levy, Karyne. “Google Finally Discloses Its Data On Diversity, And The Numbers Aren’t Great.” Business Insider, Business Insider, 28 May 2014, www.businessinsider.com/google-diversity-numbers-2014-5.
Lowensohn, Josh. “Microsoft Says 29 Percent of Its Employees Are Female.” The Verge, The Verge, 3 Oct. 2014, www.theverge.com/2014/10/3/6905545/microsoft-says-29-percent-of-its-employees-are-female.
Lowensohn, Josh. “Microsoft Says 29 Percent of Its Employees Are Female.” The Verge, The Verge, 3 Oct. 2014, www.theverge.com/2014/10/3/6905545/microsoft-says-29-percent-of-its-employees-are-female.
“Microsoft’s 2020 Diversity & Inclusion Report: A Commitment to Accelerate Progress amidst Global Change.” The Official Microsoft Blog, 23 Oct. 2020, blogs.microsoft.com/blog/2020/10/21/microsofts-2020-diversity-inclusion-report-a-commitment-to-accelerate-progress-amidst-global-change/.
“Microsoft’s 2020 Diversity & Inclusion Report: A Commitment to Accelerate Progress amidst Global Change.” The Official Microsoft Blog, 23 Oct. 2020, blogs.microsoft.com/blog/2020/10/21/microsofts-2020-diversity-inclusion-report-a-commitment-to-accelerate-progress-amidst-global-change/.
Miller, Claire Cain. “Technology’s Man Problem.” The New York Times, The New York Times, 5 Apr. 2014, www.nytimes.com/2014/04/06/technology/technologys-man-problem.html?searchResultPosition=7.
Quirk, Kathy. “Why Women Leave Engineering.” UWM REPORT, 17 Jan. 2020, uwm.edu/news/why-women-leave-engineering/.
“Sexual Harassment in Tech: Women Tell Their Stories.” CNNMoney, Cable News Network, money.cnn.com/technology/sexual-harassment-tech/.
Shelley J. Correll and Lori Mackenzie. “To Succeed in Tech, Women Need More Visibility.” Harvard Business Review, 21 Apr. 2017, hbr.org/2016/09/to-succeed-in-tech-women-need-more-visibility.
time, Getting there will take, and money. But in the end. “Silicon Valley’s Plan to Be More Diverse.” CNNMoney, Cable News Network, money.cnn.com/2015/06/19/technology/tech-diversity-roundup/index.html.
