What this covers

Byron Auguste makes the case that the US labor market is organized around a fundamental and self-imposed mistake: employers screen workers out based on whether they hold a bachelor's degree rather than screening them in based on what they can actually do. This distinction matters enormously. Most human capital develops on the job rather than in college, Auguste argues, which makes degree requirements poor proxies for ability. Roughly 70 million American workers have built skills through alternative routes—apprenticeships, on-the-job training, military service, self-teaching—yet face systematic exclusion from jobs they are equipped to perform. The result is a Collective Action Problem of massive scale: if 70 percent of new jobs demand a degree while only 40 percent of American workers hold one, the arithmetic cannot work, and the system harms itself.

Auguste traces how this lock-in happened. Path Dependency and institutional incentive structures created the trap. James Conant and the Ivies introduced the SAT in the 1930s as a genuine alternative route to elite education; rankings systems like U.S. News & World Report later rewarded colleges for exclusion rather than inclusion; and Google, despite possessing the world's most sophisticated people-analytics tools, discovered that bachelor's degrees add zero predictive power once actual skill data exists. The conversation also examines why the college wage premium persists: Auguste argues it reflects not just what colleges teach but a Two-Tiered Labor Market in which degree holders enter sales, management, and tech roles where employers invest heavily in development, while equally skilled non-degree holders land in retail and warehouse positions trained only for compliance—creating wage gaps that stem from opportunity, not ability.

The discussion ranges across education policy, labor economics, organizational behavior, and the Expectations vs Reality Gap driving deaths of despair. Auguste contrasts the Screen-in vs Screen-out framework with how German vocational systems work and why they cannot simply transfer to America. He addresses why the information problem facing workers mirrors that facing executives, why Skill-Set Overlap means many low-wage jobs share skills with far better-paying ones, and how opaque AI systems would merely replicate the biases embedded in existing hiring patterns. Throughout, he distinguishes between what makes sense for individuals operating within today's system and what would make sense for society to redesign.

Sharpest takeaway

Auguste argues that the US labor market suffers a massive, self-inflicted market and collective-action failure: employers screen workers OUT based on the shallow signal of a bachelor's degree rather than screening them IN based on demonstrable skills, locking out tens of millions of capable 'STARs' who learned through alternative routes.

  • Most human capital is developed on the job, not in college, so degree requirements are poor proxies for ability
  • 70 million workers are skilled through alternative routes and overlapping skill sets let them move into higher-wage jobs
  • The signaling bottleneck and college rankings incentives are institutional path-dependencies, not efficient market outcomes

The argument · threads10 threads · 38 claims
0.78

Degree requirements screen people out without evaluating skills despite millions doing jobs without degrees.

6 pointscentrality 5/5
  • Although transaction costs require some screening when hundreds apply per job, there are now far better signals than a bachelor's degree—for example posting the company's own training modules and seeing who can complete the first three—so employers should screen people IN based on skills rather than screen them OUT based on pedigree.

    “you should screen in based on skills, not screen out based on pedigree. That's the fundamental issue.”

  • When millions of people already perform a job without a bachelor's degree (e.g. two-thirds of admin assistants lack one despite 75% of those jobs 'requiring' it), the job does not truly require a degree; employers are preemptively pre-screening out non-degree holders, which is not the same as the job requiring a degree.

    “By that definition, 75 percent of admin assistant jobs require a bachelor's degree even though two-thirds of admin assistants don't have a bachelor's degree.”

  • There is a massive market failure in the actual workflow of hiring: workers who 'applied four times and were rejected' were in fact never evaluated—they were screened out without their skills being considered—and when a skills-and-attributes-based system that doesn't screen people out was applied through a partnering company, they got the job.

    “they hadn't rejected them. They had screened them out. They had not evaluated their skills and found them wanting. They had screened them out without considering them.”

  • Excluding millions of capable people because, on average, degree holders are more likely to have a skill set is a form of prejudice—prejudging an individual based on a statistical distribution—because the distributions of actual skill heavily overlap, so many non-degree holders quite likely have the required skills.

    “It's prejudging somebody based on a statistical distribution, even though they quite likely could have those skills.”

  • Requiring bachelor's degrees for jobs in sales, general management, and IT operations where skills are easily observable means employers privately overvalue degrees, but at a system level it is incredibly destructive math and a collective-action problem: if 70% of new jobs in the decade to COVID demanded a degree while only 40% of American workers hold one, the math cannot work, making it system-level self-harm.

    “If 70 percent of new jobs in the 10 years to COVID were ones that employers said, "Oh, you need a bachelor's degree," but only 40 percent of American workers have a bachelor's degree, well, then obviously, that math doesn't work.”

  • The sheepskin effect—where someone who quits just short of finishing a degree earns much less than someone who finishes—is an enormous market failure, but arbitraging it by hiring almost-finishers is happening only at small scale because it's a quicker path to profits to market pedigree-credentialed people that employers already want.

    “I think it's an enormous market failure, yes.”

0.70

Credentialism closes pathways that once allowed hardworking people without degrees to advance.

4 pointscentrality 5/5
  • There is a crucial distinction between how it makes sense to act within the system as it exists today (where graduating from a quality college is genuinely beneficial and advisable) and how the system should be designed; it is a terrible way to organize a society to make something costing tens of thousands of dollars and many years a prerequisite for a decent job, even though within that society an individual should still choose college.

    “it's very different to say, how does it make sense to act in the system as it exists today, and how should the system be? That's different because, for an individual person, they can't change the system.”

  • Falling US life expectancy (paralleled only by Russia pre-COVID) is driven by deaths of despair—suicide, drinking—rooted in the gap between how people thought their life would be and how it actually is; whereas over 90% of white Americans born in 1940 lived better than their parents, for those born in 1980 it's a coin flip, making this the first generation where more than half live materially worse than their parents, a profound shock that hits men harder given gender expectations.

    “I think there's a lot of evidence that it's how you thought your life would be relative to what it is.”

  • Insane levels of pedigree bias—attributing huge value to a bachelor's degree rather than to actually doing the job—have closed off the pathways by which hardworking, conscientious white men without college degrees used to make life progress, and this shift in reality relative to expectations is a significant driver of declining outcomes (though pedigree bias harms many groups, not just white men).

    “A big part of it is just insane levels of pedigree bias, where we attribute so much value to having a bachelor's degree as opposed to actually doing the job.”

  • The US labor market is not a spot market but one where decades of human-capital investment are monetized, where self-concept plays a big role, and where credentialism spreads through cut-and-paste path dependency—e.g. HR went from a field staffed by diverse operational backgrounds to the most credentialized field in three years because managers reason 'my job is important and I have a degree, so this must require one'—so hysteresis effects cannot be ignored.

    “There's a tremendous amount of cut and pasting going on, just without thinking about what the role requires. I think you can't ignore these hysteresis effects”

0.70

College wage gaps reflect opportunity gaps, not skill gaps, through differential human-capital investment by employers.

4 pointscentrality 5/5
  • The college wage premium reflects not just signaling and college learning but a third factor: a two-tiered labor market in which degree holders get placed into entry-level sales/management/tech where employers invest in developing their human capital, while equally-skilled non-degree holders get placed into retail/warehouse/factory roles trained only for compliance and efficiency, producing huge wage gaps ten years later that don't stem from what they learned—so the skills gap is a result of an imposed opportunity gap.

    “After 10 years of that, well, then you're going to have a massive difference in wage outcomes that, in the particular scenario I'm talking about, really don't have to do with what they learned. They have to do with the initial signaling.”

  • Of the 70 million workers skilled through alternative routes, about 30 million have the skills for a job that pays 50% more than their current one, and a database of 130 million job transitions shows that STARs (and degree holders) typically move into higher-wage jobs with highly overlapping skill sets—so instead of these movements happening at random we should put 'street lights' around them and treat accidental pathways as intentional ones.

    “The 70 million STARs — 30 million of them have skills for a job that pays 50 percent more than the jobs they're in today.”

  • The most direct way to identify talent is not genomics but observing what jobs people already do, because there is huge overlap between the skills required for many low-wage 'gateway' jobs (about 50 of them) and many middle-wage jobs—for example a second-level call center rep has many of the skills for enterprise sales that may pay twice as much.

    “you can see what skills people have most easily by what jobs they do today. One of the things that's really powerful is that there's a huge overlap in the skill set required for many low-wage jobs and those for many middle-wage jobs.”

  • Per Gallup international engagement ratings, Singapore has the highest percentage of workers in the middle who are neither lit up nor disgruntled—an observation tied to the trade-off Singaporeans describe between maximizing conscientiousness and sacrificing creativity—though Singapore's rise from extreme poverty to one of the world's richest countries does not speak poorly of conscientiousness.

    “Singapore has the highest percentage of people in the middle that are neither lit up nor are they disgruntled.”

0.49

College completion signals affluence and family stability rather than conscientiousness or merit.

2 pointscentrality 4/5
  • College completion is a huge signal of affluence and generational wealth rather than a clean measure of conscientiousness; non-completers often drop out due to economic and social conditions—needing to work schedules that conflict with classes, running out of money, family instability—not 'Animal House' behavior, and completers simply have more cushion, fallback, and shots on goal when they make mistakes.

    “College right now is a signal of many things, but I think the statistics would say it's a huge signal of affluence and generational wealth”

  • There is no such thing as a typical bachelor's degree because a college degree is not one thing but 18,000 different things, varying by topic and by school, so studies measuring an income gap ten years later conflate vastly heterogeneous credentials.

    “A college degree isn't one thing. It's 18,000 different things.”

0.44

Machine learning on biased hiring data replicates and amplifies existing discrimination; AI should solve skill gaps.

8 pointscentrality 2/5
  • Advancing information technology should make training easier rather than harder because digital tools offer more ways to deliver and engage with information, are designed for humans to use, and can be learned from anywhere unlike physical tools that require being co-located with other artisans.

    “I don't think technology makes training harder. It should make it easier. There are so many more ways to deliver information, to engage with information.”

  • An opaque AI that accurately predicts labor outcomes is a terrible idea because machine learning predicts by detecting patterns in actual outcomes; if those outcomes are generated by degree bias and occupational segregation, the AI will simply replicate and turbocharge the failures of the biased human system (garbage in, garbage out)—but AI applied to the right problem (e.g. teaching a worker the missing 15% of skills) could be extremely helpful.

    “What AI is doing — the machine learning AI is turbocharging the failures of the human system.”

  • A single school's high rejection rate is no mark of greatness; the schools most worthy of respect are those that develop the most people, and the deeper problem is the monoculture of a single metric for judging both people and universities, given the genuine diversity of people, jobs, and regions.

    “I do not consider a school so great on the basis of its huge rejection rate. I think that's terrible. The schools I respect the most are the ones that actually do the most to develop the most people.”

  • The most important thing about applying AI is the problem statement and the constraints imposed: an AI will 'go to town' on whatever problem you give it, so solving the wrong problem with AI is a disaster while pointing it at the right problem (helping people who have most but not all of the needed skills) makes it very powerful.

    “This is the biggest thing about AI. What is the problem you're trying to solve? Secondarily, what are the constraints you're imposing?”

  • Maryland has 1.3 million STARs—half of whom have the skills for jobs that pay more, cutting across all races and regions and especially powerful for veterans, rural residents, African Americans and Latinos—and the state under Governor Hogan removed degree requirements where they didn't belong, conducted outreach to STARs, posted hundreds of jobs on Stellarworx, and the example has prompted other states, counties, and employers to follow.

    “There are 1.3 million STARs in Maryland, half of whom have the skills for jobs that pay more, and they cut across all races and all regions.”

  • The core higher-education problem is supply: quality seats have grown very little while demand has soared and businesses 'weaponize' degrees by gating decent jobs behind them, so with mostly fixed supply you create a horrible zero-sum bottleneck in the talent-development and signaling systems—making affirmative action and testing into life-or-death fights—when what's actually needed is 50% to 100% more high-quality college seats plus valid alternative routes for people of all ages.

    “if you keep the number of college seats constant, and you demand that everybody try to go to college... you create this horrible zero-sum bottleneck in the talent-development system and in the signaling system.”

  • When the SAT began in the 1930s under James Conant and the Ivies, it was an alternative route into college that—though a terrible test—was better than nothing, allowing students outside the elite prep-school feeder system to distinguish themselves after the high school movement (9% to 79% of teenagers in high school between 1909 and 1939) created applicants colleges had no other way to evaluate.

    “If you think about something like the SAT, when it first started... it was an alternative route into a college.”

  • Universities respond rationally to a terrible incentive structure: businesses made the selective degree the magnetic signal for a decent life, and rankings (U.S. News & World Report) reward colleges for excluding as many people as possible, so it is unsurprising that colleges chase elite status by maximizing applications and rejections and resist adding seats—meaning colleges should instead be judged by how many people they successfully include and how much progress they help them make.

    “you get to the top of the rankings in U.S. News & World Report by excluding as many people as you can. That is a terrible incentive structure.”

0.39

Effective hiring probes what candidates learned, how, and at what cost through role-appropriate interviewing.

3 pointscentrality 3/5
  • Effective interviewing avoids a single uniform question and instead probes what people have done and learned; a smooth progression with no mistakes or learning is unimpressive, so one should drill into what a candidate learned, how, at what cost, and how they apply it now, while weighting interview expectations to the role (a salesperson who can't sell you is concerning; a careful researcher who doesn't is fine).

    “If someone describes a smooth progression without ever making a mistake or ever learning a thing, I'm not that impressed.”

  • Interviews are valuable for senior executive hires where you weight the track record but use the interview to probe the production function—how they achieved results and whether at the expense of or to the benefit of their team—looking for anomalies that don't jive with the record; but a whole series of interviews may be overrated for standardized, high-volume roles, where people should instead be able to demonstrate readiness and be pulled into similar jobs without another full round.

    “a whole series of interviews might be overrated for jobs where there's a huge number of positions to be filled, and the role is relatively standardized and understood.”

  • Organizational cultures tend strongly to replicate themselves: you can enter with a very different attitude and aspiration and still end up norming to the organization's existing norms, which probably applies to police departments and explains why culture is central to police quality.

    “how much you can come in with a very different attitude and aspiration, and you end up norming to the norms of the organization.”

0.39

Information failures prevent workers and executives from coordinating on skill development and hiring.

3 pointscentrality 3/5
  • A major barrier to digital transformation is not worker reluctance but an information problem: a Harvard Business School/Accenture survey found senior executives blamed frontline workers for not wanting to learn new skills, yet surveyed frontline workers overwhelmingly wanted to learn but said they were given no guidance on which skills to learn—mirroring the information vacuum facing a low-wage worker trying to advance.

    “senior executives said that one of the biggest barriers to digital transformation is that their frontline workers did not want to learn new skills. But they also surveyed frontline workers who, off the charts, wanted to learn new skills”

  • Popular culture and especially elite narratives over-index on college as the only path to success, and while going to college is not bad advice given the university system's value, the message that the only path to success runs through college becomes self-reinforcing and harmful.

    “elite narratives, if you will, I think really over-index on college... But the idea that the only path to success is through college, I think, becomes self-reinforcing.”

  • No single entity, person, organization, or sector can solve the credentialism problem alone because it operates through hiring platforms and many institutions; a for-profit would just skim the obvious cases for arbitrage, so the missing piece is a shared set of valid screen-in signals, and the employers with the most power to change the system (poaching winners) need it least while those losing the poaching wars lack the power to change it.

    “I really think it's a multi-sector problem. The question isn't, how can a nonprofit solve it? Because a nonprofit can't solve it. The question is, what's missing?”

0.39

Education pathways should blend college and vocational learning with competency-based validation throughout.

3 pointscentrality 3/5
  • The German/Northern European vocational model is strong at multiple pathways from education to employment, but it cannot simply be replicated in the US because German companies don't just 'do apprenticeships'—they plug into a centuries-old system (rooted in the Hanseatic League) where upper-secondary education, labor councils, and industry councils do the heavy lifting; moreover the German system is weak at retooling mid-career adults, so an education-to-employment model alone is insufficient for a dynamic economy.

    “German companies plug into an apprenticeship system, and they play their part in it. Upper secondary education, labor councils, industry councils — these kinds of collective entities do the heavy lifting”

  • College and vocational education should not be nearly as separate as they are; vocational should be understood as applied learning with far more of it embedded in college (as Northeastern University does), and college should be seen as competency-based skills acquisition along the way (as Western Governors University does) where the degree is just the ribbon and bow at the end of a process in which every level of learning can be validated.

    “the college and vocational should not be nearly as separate as they are. We should think about vocational as applied learning, and there should be way more of it in college.”

  • High schools in the US were originally developed as a way to get into good jobs of the future, not as a way to get into college, and only in 1983 (with 'A Nation at Risk') did the country decide the point of high school was for everyone to go to college—without anyone having told the high schools that until then.

    “High schools were not developed as a way to get into college. High schools were developed as a way to get into a good job, jobs of the future at that time.”

0.35

Conscientiousness matters more than IQ for adult retraining because learning requires emotional resilience.

2 pointscentrality 3/5
  • For retraining adults over 25, conscientiousness matters more than IQ because people can learn new things but learning is uncomfortable—it requires admitting you don't know something and attending to fine points—so IQ is overrated and conscientiousness underrated for the most part.

    “I think IQ is overrated and conscientiousness is underrated for the most part because people can learn new things, but as you know, learning can be uncomfortable.”

  • To get more young men into nursing and eldercare, a good start is paying people more since eldercare typically pays very poorly, combined with creating more pathways, because people move through the job market based on existing skills and experiences—so just as few women in early-stage construction makes entry harder, few men in early-stage care work makes their entry harder.

    “in eldercare, paying people more would be a good start. That typically pays very poorly”

0.25

Hybrid skill sets and diverse talent generate creativity that homogeneous credentialed cohorts cannot.

3 pointscentrality 2/5
  • Much of creativity comes at the margin between disciplines, so combining orthogonal or hybrid skill sets (e.g. software development plus art or music) is incredibly valuable because it lets people notice things others don't, and more diversity of talent and combined life experiences would generate more creativity.

    “I think so much of creativity comes at the margin between disciplines.”

  • Google, which has the most sophisticated people-analytics team in the world, concluded that when you have all the data the bachelor's degree has zero additional explanatory power, and that the best teams are not aggregations of the 'best individuals' but have their own dynamics including team players—so AI applied to attributing individual profit contribution is misguided since most jobs are collaborative team products.

    “they've concluded that when you have all the data, the bachelor's degree has zero additional explanatory power.”

  • A job is not a gift given to someone; work is solving problems, STARs are problem-solvers, and since society has many unsolved problems, keeping out people who can help solve them is ridiculous.

    “A job is not a gift. Work is solving problems, and STARs are problem-solvers.”