Editor’s Note: Weekly Wrap has been pre-empted by spring conference travel season. It will return soon.
“It is a good thing to follow the First Law of Holes: if you are in one, stop digging.” — Denis Healey
A great cry has risen up across the nation from technology companies large and small and an assortment of education and civic leaders. “The tech talent pipeline is BROKEN!”
Under mounting pressure last year, major players in the industry of the future released their “diversity” stats. The numbers from Google, Facebook, Apple and others ranged from grim to embarrassing. The Silicon Valley Business Journal’s “Diversity Scorecard” sums it up.
A skewed talent mix
You don’t need an algorithm to interpret these numbers:
- On gender — Male tech employees at rates as high as 80 percent dwarfed women 4-to-1.
- By race — White tech employees exceeded 50 percent in the largest employers.
- Asians (considered white) were at least 30-40 percent and a majority in three firms;
- Hispanics held a narrow slice of tech positions, ranging from 2-3 percent;
- Blacks were 1 percent of tech employees, with Apple’s 6 percent a towering exception
Many have asked “How can a huge industry in the largest and most diverse state in the country have such a skewed talent mix?”
The dominant view seems to be that the tech talent pipeline is terribly inadequate. The analysis: there is a severe supply-side problem of too little STEM education for women and minorities from elementary schools and up through the Ph.D. level
The Business Journal article cited above also covers Intel’s January announcement of its decision to spend $300 million to diversify its workforce. (That mature firm’s workforce is described as “76 percent male and 57 percent white. Only 8 percent of employees are Hispanic and just 4 percent are black.”)
If too little pipeline is the problem, buying more must be the answer
Speaking at the Consumer Electronic Show, Intel CEO Brian Krzanich announced that its initial down payment on fixing the numbers would be spent on “nonprofits and educators that support the pipeline of diverse job candidates with a goal of reaching ‘full representation’ of women and minorities by 2020.”
Intel’s reaction is the tip of the iceberg. As the issue of tech demographics has become more visible and unacceptable, the expansion of the STEM pipeline threatens to surpass the Keystone pipeline as an urgent national priority in some circles. Hardly a week goes by without an announcement of a new corporate investment, government initiative, foundation program or academic track to prepare tomorrow’s workforce for STEM proficiency and preference.
There can be little doubt that enhanced STEM education is needed. The industries of the future – information technology, biosciences, hard sciences – need a competent and well-educated workforce.
And importantly, participants in the new economy should represent all segments of the society: men, women, white, Asian. Black and Hispanic. And in far better ratios.
Let’s check the pipeline for holes
But should equity await the education of a new generation? Or are there more immediate, direct and forceful steps to be taken now? And more importantly, to meet today’s multiple needs and make the most of any new flow of talent to come, tech companies and the rest of us should mount a major effort to identify and address the considerable talent leaks that are coming to light.
In synch with the “build the pipeline” chorus has been the growing call from women and people of color to recognize and address the attitudes, behaviors and prejudices that have created hostile, unwelcoming environments. Such workplaces have driven significant numbers of tech-proficient women and minorities from the field.
A tiny sample of dozens of articles includes:
- Kieran Snyder in Fortune: Why women leave tech: It’s the culture, not because ‘math is hard. This tech veteran and CEO’s research is sub-titled “Stories from 716 women who left tech show that the industry’s culture is the primary culprit, not any issues related to science education.”
- Lauren Williams, Think Progress: How Silicon Valley Can Get Less White. This overview of solutions to the demographic problem mentions Google’s initial response ($40 million for STEM education and anti-bias training.
- Denise Cummins, PBS Newshour: Why the STEM gender gap is overblown.]
This data-rich survey of women in STEM fields argues that many aspects of this issue are over-stated and built on unproven – and often untrue — assumptions
There are shared – and often familiar — challenges
When you review this research, interview refugees from tech, as we have, and study the companies’ views expressed in their own words you will see a compelling picture of a deeply compromised pipeline. Among the factors driving some people out of the industry are:
- Dominant cultures described as white “adolescent” (aka “geek”) and exclusive;
- Environments that tolerate and can sometimes thrive on prejudice;
- “Workaholic” standards that leave little room for families or personal interests;
- An aversion to most forms of flexible work, including part-time and telework;
- A belief that as an innovative industry, tech doesn’t need to adopt external practices.
Similar challenges means similar – and proven — solutions
This is not a list of minor challenges to overcome. But the good news is that similar problems have been identified and worked through in other settings. I am reminded of work I did in the late 1990s with a major accounting firm.
Deloitte & Touche was dominated by technically proficient and well-educated men. Rather than being masters of code and automation, they were interpreters of the tax code and organization. While the search for talent was always central, they did not have an external pipeline problem per se. Their talent challenge was more internal, one they described as “a hole in our pipeline.”
A brief summary of their problem and solution:
- In 1994, women were 50 percent of new hires and 5 percent of partners; annual turnover was 22 percent;
- Strong, creative leaders (CEO, national HR and Women’s Partner leaders) acted firmly;
- Initiative for Advancement of Women required men and women as colleagues training;
- Intensive research and training led to reduced and flexible schedules across the firm;
- Internal networks, training, mentoring maximized growth of internal resources;
- Appropriate child care supports were provided for a dispersed, mobile workforce.
Among the initiative’s gains in the first four years were significant effects on the internal and external pipelines:
- Turnover was reduced from 22 percent to 16 percent for an estimated saving of $25 million.
- The percentage of women partners rose from 5 percent to 14 percent; the trend continued thereafter.
- In year four, 200 valuable, experienced hires joined the firm, crediting Deloitte’s inclusive, flexible climate for their decision.
Leadership, structure and talent management – as well as bias
A change process in accounting is hardly a thoroughgoing template for achieving greater equity in tech. But it does suggest some characteristics that a successful effort will have to meet.
It needs a deep analysis of the problem; a comprehensive response to the complex set of requirements for a diverse population; and a sustained effort that will grow and last for as long as it takes for a cohort of STEM students to move through middle and high school.
In a future post I will take up the issue of “hidden bias” training that is emerging at Google and elsewhere. One can hardly oppose the identification and elimination of bias. But the attitudes and actions described by the refugees from tech in the earlier articles suggest that bias may just be the tip of the iceberg.
As with the STEM pipeline analysis, it is in all of our interests to analyze the nature of the problem and the adequacy of the solutions lest we waste major time and resources.
This originally appeared on The Collaboration@Work.org blog.