Why “Learn to Code” Failed

0:00 Between 2011 and 21, the number of Computer Science graduates

0:05 at UC Berkeley increased by one thousand, one hundred, and six percent.

0:11 If this trend continues— as one professor observes— all 30,000

0:16 of its undergrads will study Computer Science in just seven years.

0:22 By 2059, the school will churn out more

0:25 “Computer Scientists” than there are people in California!

0:29 Meanwhile, get this: there are fewer software developers employed

0:34 in the United States today than there were six years ago.

0:39 Clearly, something has gone terribly wrong.

0:43 And Berkeley isn’t even the most Computer Science-obsessed university.

0:49 Not even close.

0:51 This is UCB, this is CalTech, and this is MIT.

0:57 You’re not reading this wrong.

0:59 A full forty percent of the school that once helped invent the radar,

1:04 spreadsheet, and lithium-ion battery studies the same one subject.

1:10 Last year, a grand total of seven students graduated

1:14 from MIT with a Bachelor of Science in Chemistry.

1:17 Two-hundred and sixty-six majored in Computer Science and Engineering.

1:22 And that’s not even its largest CS major— Electrical

1:28 Engineering and Computer Science is over twice as popular.

1:32 Over the past 15 years, U.S.

1:35 colleges have transformed from multidisciplinary

1:37 institutions of higher education to Microsoft

1:41 training programs that also dabble in philosophy and physics on the side.

1:47 In 2023, Berkeley joined MIT and Cornell in even creating an entire

1:52 College of Computing— its first new school since Eisenhower was president.

1:59 It’s not hard to see how we ended up here.

2:02 The iPhone was released in 2007; Uber, AirBnB, and Instagram soon after.

2:09 America’s cultural center of gravity was shifting from stuffy

2:14 Wall Street board rooms to youthful Bay Area garages.

2:18 Shows like The Social Network and Silicon Valley

2:21 mythologized the tech founder— the 21st century’s more meritocratic,

2:27 more visionary, and certainly more eccentric answer

2:30 to the industrial tycoons of the past.

2:34 Then, our nation’s leaders endorsed programming

2:37 with their official seal of approval,

2:40 elevating it from a promising new career to “the future.” While in office,

2:46 President Obama called coding a “ticket to the middle

2:49 class.” The White House celebrated Computer Science Education Week,

2:54 promoted “Hour of Code,” and warned of a dire

2:58 shortfall of 1 million STEM graduates by 2022.

3:02 New York Mayor Michael Bloomberg even announced he, personally,

3:06 would “learn to code” in 2012 (although there’s no evidence he ever did).

3:11 There was something for everyone in that 3-word-mantra,

3:15 “Learn to code…” For Republicans,

3:18 this directive neatly aligned with their vision for more vocational,

3:23 career-oriented education.

3:25 For corporations, it promised a steady supply of skilled

3:28 labor— labor— I might add— that came pre-trained, at taxpayer expense.

3:35 For national security hawks, coding helped strengthen America’s global power,

3:40 enabling it to “out-compete” China.

3:44 If diversity was more your cup of tea,

3:46 one could tell a pretty convincing story about

3:49 how this was really all about expanding opportunity.

3:52 And for Democrats steering the economy through the Great Recession,

3:56 “learn to code” sounded empowering, hopeful,

3:58 and— most importantly— vague enough to absorb virtually any

4:03 of the difficult questions posed

4:05 by globalization at an especially fraught moment.

4:09 Layoffs?

4:09 Automation?

4:10 Outsourcing?

4:10 …Not to worry if we all just rolled up our sleeves and, say it with me:

4:18 “learned to code!” Red and blue states alike

4:21 rushed to teach Python and Javascript in classrooms.

4:25 Today, nearly 15,000 U.S.

4:27 high schools offer Computer Science classes.

4:31 Even 37% of middle schools do.

4:34 Eleven states go even further:

4:37 requiring all students take a computer science class to graduate.

4:42 Absent any dissenting voices, there was every incentive to exaggerate the ease

4:47 of coding and romanticize the life of a programmer.

4:51 Obama told us “just about anyone can become” a computer scientist, quote,

4:55 “with a little hard work…” Years later,

4:58 President Biden echoed this sentiment in his signature no-nonsense tone:

5:03 quote “anybody who can throw coal into a furnace can learn how to program,

5:08 for god’s sake.” One popular article from 2012— when U.S.

5:13 unemployment was still well over 8%— declared,

5:16 “to anyone out there who says you can’t get a job: You can have one.

5:21 A fun one.

5:22 Learning code is not about numbers and mathematics.

5:26 It’s more like architecture…” Well,

5:28 young Americans didn’t need to be told twice!

5:32 Younger Millennials and older Gen Z looked around,

5:35 saw their siblings struggle to find work after the Great Recession,

5:39 and decided they’d take the ‘cool’ job that lets you earn

5:43 six figures working from a bean bag while wearing a hoodie,

5:46 requires a bachelor’s degree at most,

5:49 and is apparently remarkably “quick” and “easy” to learn, please.

5:54 But although Microsoft and Facebook and Apple

5:57 were ready for this massive influx of programmers,

6:01 universities most certainly were not.

6:04 The popularity of college majors, of course, fluctuates all the time.

6:09 Seasoned law school admissions officers still remember the sudden

6:12 rise in applications following the hit 80s drama “L.A.

6:16 Law,” just as pharmacy school briefly became

6:19 popular during the early 2000s pharmacist shortage.

6:23 But the Computer Science boom was something else entirely.

6:27 The basic problem for schools was this: while the total number

6:32 of undergraduate CS students tripled in just 10 years, the number of Ph.D.

6:37 students— who become professors— has stayed more-or-less the same.

6:42 The reason for this is no mystery: a Ph.D.

6:46 student can expect to receive a $40,000/year stipend— if they’re lucky.

6:51 That same student can easily earn $200,000 or more at Amazon or Netflix.

6:58 In other words, there isn’t anyone to teach all those undergraduates.

7:03 The result, in many Computer Science departments,

7:07 is an impersonal, factory-like experience.

7:10 Professors are perpetually stressed and invariably overworked.

7:14 Class sizes are massive— 4, 5, 6 hundred students in introductory courses.

7:20 Undergrads are even deputized as TAs— meaning the person who took

7:26 this class last semester might be the only person you can ask questions.

7:30 And yet schools still turn eager students away.

7:34 It’s become common at many universities

7:36 to make students re-apply to their Computer

7:39 Science program at the end of their first or second year, for example.

7:43 At Swarthmore, the University of Maryland,

7:46 and UCSD, students are entered into a lottery!

7:50 Meaning, you can apply and be accepted to Swarthmore (not to mention pay just

7:54 over $90,000/year) only to later be denied

7:57 access to your preferred major by random number.

8:02 Needless to say, this leaves many

8:04 students feeling… disillusioned with the college experience.

8:07 They graduate with debt, with little career help from their school,

8:11 whose resources are spread thin,

8:13 and with minimal— if any— contact with their professors along the way.

8:18 On top of all that, many of these students

8:21 feel they were inadequately prepared for the job market.

8:24 Because many CS programs have their origins in math and engineering departments,

8:29 their curricula often focus more on the how— the general algorithms,

8:34 ways of thinking, and processes— than the what— the specific syntax,

8:39 programming languages that are currently marketable,

8:41 or skills tested in modern job interviews.

8:45 Graduates will often have dabbled in half

8:48 a dozen languages but truly mastered none, forcing them to learn on their own.

8:53 It wasn’t hard to imagine there might be a better way.

8:57 And this being the era of quote “disrupting” “outdated” industries like taxis,

9:02 hotels, and DVD stores,

9:04 the proposed “solution” was wrapped in lofty, emancipatory language:

9:09 education was about to be revolutionized; universities, out-innovated.

9:15 Enter: the coding bootcamp.

9:18 The idea was simple: if you already knew what you wanted— a six figure

9:23 job at Amazon— you could forgo the 4-year Computer Science degree,

9:28 pay just 10 or 20, or 30 thousand dollars,

9:31 and “learn” to code in a fraction of the time— usually 12 weeks or less.

9:37 By stripping out general education requirements and teaching to the test

9:42 (tech interviews often ask the same set of questions),

9:45 they promised a streamlined, more efficient,

9:48 and more accessible route to the promised land.

9:51 At their peak, hundreds of bootcamps graduated nearly half as many students

9:56 as 4-year Computer Science programs and took in north of $200 million a year.

10:02 But as they grew, as they began appearing on billboards

10:06 and in TV commercials and attracting a wider and wider group of students,

10:11 they began feeling many of the same strains as universities.

10:15 First, they realized their students were lending at least as much

10:19 of their reputation to them as they were to students.

10:23 In other words, if one of them failed an interview or was fired,

10:27 that employer or recruiter would remember and associate

10:30 their performance with the bootcamp on their resume,

10:34 making it harder for future students to land jobs there.

10:38 Silicon Valley, after all, is a small world.

10:41 Most students are looking for work from one of just five companies.

10:45 Thus, they began carefully guarding their reputations,

10:49 selecting only those applicants who enhanced their brand.

10:53 One bootcamp, Hack Reactor, for instance,

10:56 has an acceptance rate of just 3%— about the same as Harvard.

11:02 Another problem was a lack of teachers.

11:05 Also like universities,

11:06 bootcamps struggled to compete with the salaries on offer by Big Tech,

11:11 often resorting to hiring their own

11:13 recent graduates who couldn’t find work elsewhere,

11:16 inflating their employment figures in the process.

11:20 Finally, they faced pressure to raise tuition.

11:24 Initially, bootcamps pitched themselves to investors much like WeWork did:

11:28 as software companies.

11:29 They could develop their curriculum once and then

11:33 churn through students with zero marginal cost.

11:36 But, like WeWork, reality turned out to be more complex.

11:41 In practice, bootcamps are either little more

11:44 than classrooms filled with textbooks— in which case,

11:47 they don’t offer much that those same textbooks alone

11:50 can’t for a fraction of the cost— or they’re highly-structured,

11:55 personable, guided experiences in small,

11:58 intimate settings— in which case, that’s expensive!

12:01 As the number of students grows, you’ll need to keep hiring new teachers,

12:07 by definition, to maintain the same student-to-teacher ratio.

12:11 If anything, the marginal cost went up.

12:14 Later students, attracted by their increasingly “get rich quick-syle” marketing,

12:18 had less familiarity with and less interest in programming

12:22 and therefore required more guidance than their earlier,

12:26 more intrinsically motivated peers.

12:29 Just to maintain the same outcomes, therefore, bootcamps had to raise prices.

12:34 And that’s a problem when students are paying out of pocket!

12:39 Bootcamps wanted in on those sweet, taxpayer-funded federal student loans.

12:44 But, as unaccredited institutions, they weren’t eligible.

12:49 Meanwhile, for reasons we’ve covered in previous videos,

12:52 colleges found themselves short on cash.

12:55 They wanted— needed— a slice of that sweet bootcamp revenue.

13:00 But they aren’t very good at recruiting non-traditional students.

13:05 And the Higher Education Act of 1965 prevented them from paying third-parties

13:10 for bringing them students after those incentives

13:13 led to aggressive and misleading marketing.

13:16 …Except that in 2011, the Department of Education created a loophole:

13:22 schools could pay third parties to bring them students if those payments were

13:26 part of a larger “bundle of services,” of which recruiting was only one part.

13:32 In other words, as long as a school didn’t

13:35 call these payments “kickbacks,” they now had a green light.

13:40 A university would “partner” with what they called

13:43 an OPM— an “Online Program Manager”— this was the bootcamp.

13:48 The OPM would create the curriculum,

13:50 hire the teachers, and recruit the students,

13:54 which the university would then slap its name on, usually for a 40% cut.

14:00 As far as the government was concerned,

14:02 these programs were accredited, so bootcamps got access to student loans.

14:07 Students saw a trusted 100-year-old brand name,

14:10 not some random fly-by-night bootcamp.

14:13 And universities could get away with what was otherwise illegal.

14:19 Put differently: bootcamps became more and more

14:22 like the “unwieldy,” “inefficient” 4-year universities they

14:25 originally sought to “disrupt,” until they were

14:29 swallowed whole by those very same universities.

14:32 Now, this more-or-less worked through the pandemic,

14:35 when growth in the tech industry was off the charts

14:39 and companies raced to hoard as much labor as possible.

14:42 In one 2023 Wall Street Journal profile,

14:45 a woman said she was paid $190,000/year by Meta to do… almost

14:50 nothing— presumably because aggressive hiring

14:54 at the time was rewarded by investors, and competition for talent fierce.

15:00 Then came the inevitable crash.

15:03 Nearly half a million tech workers were laid off in 2023 alone.

15:09 Another quarter million lost their jobs once in 2022 and again in 2024.

15:16 In total, that’s roughly the number of U.S.

15:18 manufacturing jobs lost from the early 2000s “China Shock.” The unemployment

15:24 rate in tech is now higher than the national average.

15:28 Recent graduates have had their offers rescinded.

15:31 And many professors,

15:33 including this one at Berkeley— one of the most well-known Computer Science

15:36 programs in the nation— report that even their best students can’t find jobs.

15:42 Bootcamps are doing even worse— if they still exist at all.

15:48 One, called “2U,” declared bankruptcy last year.

15:51 Launch Academy announced it would “pause enrollment.” And Dev Bootcamp,

15:56 which was acquired by education giant Kaplan, closed its doors forever.

16:02 It all happened remarkably fast.

16:05 It seems like only yesterday that tech was “the future.” Not so long ago,

16:11 the White House predicted a shortfall of 1 million STEM workers by 2022.

16:17 Instead, there are fewer software developers in the United

16:20 States today than there were six years ago.

16:24 How did this happen?

16:27 Well, it’s not rocket science: “Learn to code,” meet another 3-word mantra:

16:34 “supply and demand.” Recall that in just ten years,

16:37 the number of Computer Science grads

16:39 at Berkeley increased by over a thousand percent.

16:44 This was always unsustainable.

16:47 Now, to be clear: “Learn to code” didn’t cause this recent downturn.

16:52 In 2022, U.S.

16:53 interest rates rose to their highest level in 15

16:57 years— from nearly 0% to well over 5.

17:01 And the tech industry is unusually exposed

17:04 to this number— making software requires massive up-front investment.

17:08 When the cost of borrowing goes up,

17:12 making software becomes a lot more expensive.

17:16 But there’s no doubt that the over-supply of programmers— driven by “Learn

17:20 to Code”— made workers more expendable in the eyes of employers.

17:25 When programmers are so abundant,

17:27 it’s a whole lot easier to fire them on demand— treating them

17:32 as a spigot they can quickly turn on when needed and off when not,

17:37 rather than a resource to invest in and retain through thick and thin.

17:43 Every industry, of course, has ups and downs.

17:46 And brutal as this downturn has been,

17:49 growth will eventually return— the Bureau of Labor Statistics

17:54 still expects computer-related occupations to grow 11% by 2033.

17:58 The same, to be fair, cannot be said of all fields.

18:04 Still, a starry-eyed 18-year-old back in 2016 could

18:08 be forgiven for thinking that tech was somehow… different.

18:14 For years, common sense had been thoroughly drowned

18:17 out by the increasingly religious zeal of “Learn

18:21 to code.” “Learn to code” was more than

18:24 a gentle encouragement to consider programming as a potential career.

18:28 It was an all-encompassing vision of “the future,”

18:32 unbounded by the laws of economics.

18:35 That 2012 article from earlier, for example,

18:38 compared coding to reading and writing—

18:41 a surprisingly common sentiment at the time.

18:44 The message wasn’t that programming was a useful skill,

18:48 it was that soon, everyone would need to program.

18:52 Otherwise serious people argued that everyone from electricians to teachers

18:57 to insurance adjusters would be coding in the not-so-distant future.

19:01 If you accept that premise, one thousand percent growth in Computer

19:06 Science graduates doesn’t sound so crazy.

19:09 But if programming is not a universal skill like reading or writing,

19:13 if it’s still just one, admittedly often lucrative career among many,

19:18 then it just means a thousand percent

19:21 more competition for a limited number of jobs.

19:24 Today, there are something like 1.9 million people employed in tech.

19:29 Even if we double that estimate, it would still only represent 2.3% of the U.S.

19:36 labor force.

19:37 Programming simply could never have

19:39 absorbed all seven million unemployed Americans.

19:42 Never mind that someone still needs to manufacture the keyboard

19:46 you program with, or feed the person who manufactures that keyboard,

19:51 or repair that person’s internet.

19:53 Tech is also at least as volatile as any other industry.

19:58 The sudden and stratospheric rise of AI is a prime example.

20:03 These ups and downs become clear when we zoom back in time.

20:07 This recent boom has just been so unusually long that, until recently,

20:12 many young people had never experienced anything else.

20:16 But “learn to code” wasn’t just at odds with Econ 101,

20:21 it was also ignorant of, or at least

20:24 indifferent to, the wide diversity of human interests,

20:27 talents, personalities, and life circumstances.

20:31 Even the most accomplished programmer will tell you:

20:34 building software can be quite difficult.

20:36 Like any skill, not everyone will be great at it.

20:41 Nor, for that matter:

20:42 enjoy it— for the same reasons not everyone enjoys working in an office,

20:47 or sitting down all day, or staring at a computer,

20:51 or working on a team, or solving complex problems on a deadline.

20:55 This is all quite obvious when you replace “code” with “repair

20:59 medical equipment,” “study epidemiology,” “become a nurse,” or any other job.

21:05 Overlooking the wide spectrum of human skills,

21:08 interests, and circumstances was convenient.

21:10 And although it sounded empowering as a soundbite,

21:15 it was often much closer to exploitative, in practice.

21:19 “Learn to code” reduced people to interchangeable units of labor.

21:24 Coding bootcamps represented the most

21:26 distilled version of this one-size-fits-all “solution”:

21:30 promising that single mothers, recently laid-off factory workers,

21:34 and well-heeled highly-computer-literate 18 year-olds could all

21:37 be funneled into a bootcamp and emerge,

21:41 12 weeks later, as Computer Scientists making $150,000/year.

21:46 Instead, many single parents working two

21:49 jobs earnestly enrolled in bootcamps only

21:51 to discover they offered minimal support

21:54 for and couldn’t accommodate non-traditional students.

21:57 Former general contractors struggled to keep up

21:59 with their peers who had prior coding experience.

22:02 And since they were told learning to code

22:06 is “so incredibly easy,” they blamed themselves.

22:09 Thousands graduated with nothing more than a $20,000 hole in their bank account,

22:14 3 months of lost wages,

22:16 and a line on their resume that, in the eyes of many employers,

22:21 worked against, not for them.

22:23 Others, of course, were luckier.

22:26 But, perhaps most humbling of all,

22:28 even many of them ultimately fell victim to “Learn

22:32 to Code.” They did exactly as they were told,

22:35 set their true passions aside and pursued coding as a career,

22:39 were admitted into a selective bootcamp or Computer Science program,

22:43 studied hard, were fortunate enough to land a competitive,

22:47 highly-paid job at Amazon, and after all that, were still one of its,

22:53 say, 18,000 employees laid off with the stroke of a pen on January 5th, 2023.

23:00 No matter how many times it's repeated,

23:02 no matter how many high schools it’s taught

23:04 in, “learn to code” is no more a magic solution

23:08 to economic uncertainty or job insecurity than “learn to service

23:12 wind turbines” or “learn occupational therapy”— demand for which,

23:16 by the way, is expected to grow by 60 and 22%,

23:22 respectively— far faster than software developers.

23:26 Everyone, everywhere is at the mercy of the labor market.

23:30 The only solution— the best we can do— is

23:34 be adaptable and invest in non-industry-specific skills like problem solving.

23:39 Today’s sponsor, Brilliant,

23:40 can help you do just that— spread your eggs across many baskets.

23:46 Brilliant is the perfect sponsor for this video because they believe, as I do,

23:50 that what’s truly valuable in this fast-moving era we

23:54 live in are higher-level skills like critical and strategic thinking,

23:58 creativity, and problem solving.

24:00 Programming languages come and go.

24:02 We forget the formulas and facts we read years ago in a textbook.

24:08 But the way you approach and solve problems stays with you forever.

24:12 Brilliant takes that approach and applies it to logic,

24:16 puzzles, data analysis, and geometry.

24:18 They have courses on “Real-World Algebra,” “Thinking in Code,” “Scientific

24:23 Thinking,” “Everyday Math,” “How AI works,” “Geometric thinking,” and much more.

24:30 Sure, you’ll learn to program in Python,

24:32 to calculate probability, and analyze data.

24:35 But you’ll also sharpen your mind, learn to think like a programmer,

24:41 scientist, or mathematician,

24:42 and develop a powerful learning habit along the way.

24:46 Ironically, by focusing a little more on the process

24:49 and a little less on the outcome, you actually get better results:

24:54 by making learning fun and interactive,

24:57 rather than all about “getting the right answer,” Brilliant’s beautifully

25:00 designed lessons make you want to keep coming back for more.

25:04 To check out Brilliant completely free for 30 days, click the link on screen now

25:10 or in the description below— that’s Brilliant.org/PolyMatter.

25:15 Doing so will also get you 20% off an annual premium subscription.

25:21 Go, pick what interests you most, and start learning today!