It’s Crowded at the Top: A New Marketplace Podcast

(Photo: Metro Centric)

Our latest podcast, “Crowded at the Top,” presents a surprising explanation for why the U.S. unemployment rate is still relatively high. (You can download/subscribe at iTunes, get the RSS feed, listen via the media player above, or read the transcript.)

It features a conversation with the University of British Columbia economist Paul Beaudry, one of the authors (along with David Green and Benjamin Sand) of a new paper called “The Great Reversal in the Demand for Skill and Cognitive Tasks“:

What explains the current low rate of employment in the U.S.? While there has substantial debate over this question in recent years, we believe that considerable added insight can be derived by focusing on changes in the labour market at the turn of the century. In particular, we argue that in about the year 2000, the demand for skill (or, more specifically, for cognitive tasks often associated with high educational skill) underwent a reversal. Many researchers have documented a strong, ongoing increase in the demand for skills in the decades leading up to 2000. In this paper, we document a decline in that demand in the years since 2000, even as the supply of high education workers continues to grow. We go on to show that, in response to this demand reversal, high-skilled workers have moved down the occupational ladder and have begun to perform jobs traditionally performed by lower-skilled workers. This de-skilling process, in turn, results in high-skilled workers pushing low-skilled workers even further down the occupational ladder and, to some degree, out of the labor force all together. In order to understand these patterns, we offer a simple extension to the standard skill biased technical change model that views cognitive tasks as a stock rather than a flow. We show how such a model can explain the trends in the data that we present, and offers a novel interpretation of the current employment situation in the U.S.

So it appears that, while returns to education remain strong, there are far too many highly educated workers for the available jobs. We also make note in the podcast of a new paper by Hal Salzman, Daniel Kuehn, and B. Lindsay Lowell which argues that, for all the hand-wringing about the U.S.’s inability to educate (or import) a sufficient number of STEM workers, there is in fact no shortage of such workers and that only half of U.S. STEM graduates end up with a STEM job.

We also make note of one potential silver lining in this “de-skilling” process, as Beaudry calls it: more highly educated workers may flow into fields like teaching. It’s worth noting that decades ago, when educated women had fewer job options than today, many of the best and brightest were teachers. Their outflow from the teaching ranks put a hurt on education in general. For further reading, see “Do Alternative Opportunities Matter? The Role Of Female Labor Markets In The Decline Of Teacher Supply And Teacher Quality, 1940-1990,” by Marigee P. Bacolod; and “Why the Best Don’t Teach,” by then-New York City schools chancellor Harold O. Levy.

Audio Transcript

Kai RYSSDAL: Time now for a little Freakonomics Radio. It’s that moment every couple of weeks we talk to Stephen Dubner, the co-author of the books and the blog of the same name. It is “the hidden side of everything.”  Dubner, how are you man?


Stephen J. DUBNER: Hey Kai. I’m great.  I bet you can’t wait to see the new unemployment numbers on Friday.  That’s kinda like catnip for you, huh?


RYSSDAL: It makes my day, pal!  Come on.  I live for that stuff.


DUBNER: Let me say this: whatever the numbers say on Friday, the general picture is pretty clear, which is that unemployment remains relatively high.  Everyone has a pet theory, but I’m here to tell you today about a trio of economists who have a new research paper out with a pretty interesting angle to explain the unemployment.


RYSSDAL: Three economists, three times as much fun, Dubner.  What do you got?


DUBNER: Let’s start with something we know to be true – that, on average, the more education you get, the better you do in the labor market.  There’s no question about that, right?


RYSSDAL: Yeah.  It’s “ROE,” baby -- it’s “Return On Education,” much like Return On Investment.


DUBNER: Exactly right. And that’s thought to be especially true for “knowledge workers,” people who primarily use their brains for their job -- managers and tech jobs, things like that. But one author of this new paper, Paul Beaudry, looked at 30 years worth of hiring data and he found that demand for knowledge workers actually stopped growing quite a while ago.


Paul BEAUDRY “Then you start noticing that it has plateaued in 2000 -- even though more and more people are getting educated.  It should have kept on going.”


RYSSDAL: So Dubner, you know 2000 was like 13 years ago, right?  That’s before the recession.  The whole deal.  What happened?


DUBNER: Well, what happened is the tech boom. In the 1990’s, the tech boom led to much higher demand for knowledge workers.  So more and more people started getting the appropriate college degrees. But in 2000, remember, we had the tech bust.  Which meant that demand for knowledge jobs fell, and fast. Beaudry and his colleagues call it “The Great Reversal.”


RYSSDAL: So wait -- let me try this one more time.  2000 was 13 years ago, man.  Make this make sense in the unemployment situation today.


DUBNER: Here’s the thing: all those highly-educated workers who educated themselves up from what was supposed to be the everlasting tech boom, they didn’t get the jobs that they thought.  But those workers don’t go away, and then there are new graduates in the pipeline every year. But there still aren’t nearly enough high-end jobs to suit them. So what happens? Here’s Paul Beaudry again:


BEAUDRY: “If these educated people aren’t getting their jobs in that sector, they must be pushing down.  That’s when we started noticing all this cascading.  I wouldn’t want to exaggerate.  It’s not like everyone is getting a barista job.  But that’s exactly the feeling.  It’s kind of like this pushing down.


DUBNER: So that’s a pretty word – “cascading” – to describe a pretty ugly effect: too many highly-educated workers aren’t able to find the jobs they’re expecting.  As a result, they get shoved down the labor ladder a couple rungs. Here’s a fellow named Clayton Thomas.  He’s a software consultant in Salt Lake City.  When he graduated from Duke a few years back -- he studied life sciences -- he was expecting to have no trouble finding a good research job.


Clayton THOMAS “I went to job fairs where I would discuss openings with companies and I found that I was competing with people who had masters degree, even PhD’s for basically entry level positions, which was pretty scary.”


RYSSDAL: All right, so Dubner, what’s that saying?  I heard it in the service a lot.  “Stuff,” shall we say, “rolls downhill.”  What happens to the people on the bottom rungs of those ladders?


DUBNER: Exactly right.  As the top pushes the middle, the middle pushes the bottom.  The crowding at the top pushes everybody down. But let me stay on the top for a minute.  As much as we hear about needing to educate more  STEM workers – “STEM” stands for science, technology, engineering, and math – there’s another new research paper out which argues that there is, in fact, no shortage of STEM workers and that only half of U.S. STEM graduates end up with a STEM job already.


RYSSDAL: This is the most depressed you have ever made me, my friend.  There must be some silver lining here.  You gotta hit me with that.


DUBNER: Let me try: if you go back several decades, you’ll find that a huge share of the college-educated women in the U.S. labor force worked as schoolteachers.  Right?  So even though it didn’t pay that well, a lot of the best and brightest women in the country were teachers, because they didn’t have as many opportunities as they do now.  When the feminist revolution opened up those opportunities, it was great news for women, who became doctors and lawyers and radio show hosts, on and on.  But bad news for the classrooms.  So, see what you think of this idea.


RYSSDAL:  Is this the “hidden side?”  Is that where we’re going?


DUBNER: This is so hidden.  This “cascading” that we’re seeing now, with the top pushing down -- what it means is that it might be good news for a field like education, because more and more highly-educated people -- women and men -- could end up working as, say, teachers, or elsewhere in education. So this year for instance, Teach for America, which recruits from elite college campuses, had the most applications in its 20-year history. It accepted only 17 percent of the applicants – which means that on some campuses, it’s almost as hard to get into Teach for America as it was to get into the college itself.  So, if selectivity is any measure, that means that the corps of potential teachers, at least, is drawn from a very, very talented pool.


RYSSDAL: All right.  I think we can say silver lining found there.  Stephen Dubner.  Freakonomics dot-com is the web site.  He’s back in a couple of weeks.  See ya, man.

Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.



View All Comments »
  1. Luis Augusto Fretes Cuevas says:

    The obvious problem with this hypothesis is that the unemployment for college graduates is much, much, much lower than the unemployment rate of people with less education, you can’t explain the current unemployment rate with that.

    And… Obviously the economy had enough jobs in 2006, so there was enough demand…

    Still interesting stuff.

    Thumb up 1 Thumb down 3
    • Kevin Shmevin says:

      How is that a problem? That’s part of the cascade. The top college educated people take the few cognitive jobs. The other college graduates take more unskilled jobs, and the less educated are squeezed out.

      Thumb up 5 Thumb down 1
      • Luis Augusto Fretes Cuevas says:

        Except, of course, there’s no data that suggests that college graduates are taking low skill jobs out of the cold hands of low skill workers.

        Not to mention it’s obviously flawed, the economy certainly had enough jobs for college graduates between 2002 and 2007.

        Thumb up 1 Thumb down 4
    • Kevin Shmevin says:

      “Except, of course, there’s no data that suggests that college graduates are taking low skill jobs out of the cold hands of low skill workers.”

      Is this not suggested by the fact that more college grads are working in unskilled jobs, and fewer unskilled people are employed?

      Thumb up 1 Thumb down 0
  2. Doug says:

    Also important to remember is that older workers are refusing to retire and vacate positions that would allow middle shred workers to step into. If the middle level workers can’t move up then neither will the rung below then or below them…

    Thumb up 4 Thumb down 0
    • Seminymous Coward says:

      Well, given the retirement situation, I find it hard to blame them.

      Thumb up 3 Thumb down 0
    • tungbo says:

      That’s a fair point. But given the low rate of savings and inadequate preparation for retirement, coupled with the Great Recession, do you blame them for working a few extra years to build up their nest of eggs?

      Thumb up 0 Thumb down 0
      • James says:

        But why retire in the first place? A lot of the boomer generation are far healthier than their parents were at the same age. (Remember that the original social security schemes were founded on the premise that most people wouldn’t live long enough to collect anything.) If you’re in good health, and have work you enjoy and are good at, why stop and devote your remaining years to golf or other forms of premature senility?

        Thumb up 3 Thumb down 0
    • Harriet says:

      Refusing to retire! Ha! Without a pension, plus social, you can’t afford to retire! Not all pensions were created equally, either. I think the non- retirement of old folks, like me, is mostly financial but part … Hmmm … Financial! Bring back pensions and we’ll retire at 55 the way we used to!!

      Thumb up 0 Thumb down 0
  3. Ross Brown says:

    This is a good paper and addresses a real shift in our economy, but it isn’t nuanced enough to tease out what I am seeing as an employer as a real long term shift in how prospective employees are evaluated.

    I think it is really important to distinguish two different skill sets – having the skills to do a cognitive task well (get complex or specialized tasks done) vs. the skills to solve problems and frame problems well to be solved. The first is really an advanced form of a vocational degree – there is little difference between getting an AA in air conditioner repair vs. a BS in Accounting – both educational approaches teach the person HOW to do a task, just the level of abstraction is different (compressors vs. double-entry accounting). You are still teaching someone to do something.

    The second are the degrees that teach people how to frame a problem and approaches to problem-solving where there is no set method or best practice. These degrees are usually considered ‘soft-skill’ education in that you graduate with a strong mind, but no demonstrated ability to ‘do anything’.

    In the past 25 years of my career, I’ve watched what were considered professional, middle class jobs held by people who were trained to do things fall apart because of automation. It’s not just the factory that is being replaced with robots, it’s the CPA (Turbotax and Quickbooks), the realtor (Redfin, Zillow, etc), the stockbroker (E-Trade, Fidelity), the corporate accountant (SAP, Oracle), etc. etc.

    The hollowing out of our middle class is both the wholesale re-engineering of our financial system to benefit the wealthy AND the loss of jobs to automation – technology by definition creates a ‘winner take all’ model – Intuit shifted the CPA business to themselves, online brokers from stock brokers, etc.

    As a result, the demand for people with 20 years experience as a CPA is rapidly diminishing. As an employer, I am specifically hiring people graduating from the top of their program with degrees that taught them to think (philosophy, theoretical math, biosciences, etc.) from secondary schools (Whitman College, Harvey Mudd, etc). These students are orthogonal thinkers who can approach problems not with a toolset, but an open mind.

    So, I firmly believe and have experienced that there is a glut of ‘skilled’ workers at the upper end of the employment scale, but there is a strong desert of ‘thinking’ workers who can solve problems that didn’t exist 5 years ago. Looks like the old joke of “What does a philosophy major say after graduation” is changing from “You want fries with that?” to “Here is a new way to think about this problem.”

    Well-loved. Like or Dislike: Thumb up 9 Thumb down 1
    • Joe Dokes says:


      Your analysis is spot on, until the final paragraph. I do think employers THINK they are looking for problem solvers, but based upon my very anecdotal evidence I still think most employers who are maleable, fungible, and expendable.

      Well-loved. Like or Dislike: Thumb up 5 Thumb down 0
      • Joe Dokes says:

        The second sentence is unclear it should read:


        Your analysis is spot on, until the final paragraph. I do think employers THINK they are looking for problem solvers, but based upon my very anecdotal evidence I still think most employers are looking for employees who are maleable, fungible, and expendable.

        Thumb up 4 Thumb down 1
    • tung bo says:

      How about experienced workers who’ve demonstrated that they can ‘think’ and ‘innovate’ on the job? Should you not also consider these people who have demonstrated project management and people skills AND also be creative? Could the higher wage be a barrier?

      I hear repeatedly from engineers who’ve been out of a job due to shifting market conditions. They have a hard time getting hired unless the job descriptions fit them like a glove. Even though their knowledge will make transition to related discipline relatively easy, the companies and recruiters don’t want to or have the knowledge to make the mental leap.

      US is a youth oriented culture. We want our celebrities to be young and we always think new ideas from from the young. Hiring new graduates is cheap and low risk. There is some merit to this, however most young staff don’t have the people skill to execute well.

      While I support increasing immigration to maintain a decent working/retirement ratio, I don’t believe that a quest for STEM workers should be the emphasis. It would be productive for companies to invest in retraining programs for experienced workers and it would reduce the unemployment rate ( not to mention the bad effects of long term unemployment ).

      Thumb up 2 Thumb down 0
    • Veritea says:

      You have nearly got it. I think there are two additional thoughts I would add to what you have posted:

      1 – I have not particularly noticed what you call “thinking workers” to come from a liberal arts background. In my experience they come from top engineering schools, or distinguish themselves by their ability to solve problems in unique ways and make everyone around them better at their jobs. Individuals in this category are worth 10 or more “task-workers”. The small number of engineers who are “thinking workers” mean that there is a chronic shortage of them – this is the STEM shortage you hear about.

      2- There is, at the same time, a huge number of “task-workers” available. They are are a dime a dozen. Now there are still many task workers needed, but when an employer is hiring for a task position they have dozens of well qualified candidates to choose from. If you find yourself being treated like an expendable cog then you are a task worker. You feel this way because your are expendable and easily replaceable.

      Some people might feel that they are “thinking workers” that are just under-appreciated. This is an unlikely situation. In fact, the test for if you are a thinking worker is to ask yourself “Have I transformed the work of my job and of my coworkers around me in nearly every position I have ever held”? If the answer is “No”, then you are almost certainly a “task-worker”. There is nothing wrong with being a task worker, upward of 95% of college graduates are only capable of task work. Just realize that when you read about STEM shortages they are not talking about you.

      The parent poster was spot-on. There is a desert of “thinking-workers.” They are tough to find and extremely valuable when you get them. The lack of “thinking-workers” is one of the key limits on economic growth, which is why we should have an immigration policy that encourages them to come to the US no matter where they are from. My suggestion is to replace the H1B lottery with an auction where the visas go to the highest bidding companies. This way visas will always be available for the truly valuable workers as it will be worth it for their employers to bid enough to guarantee a spot.

      Thumb up 1 Thumb down 0
      • Stu says:

        Eh. Almost. I think I agree with you about the classification of workers. But your auction solution doesn’t, I think, take full account of labor market dynamics and differential resources available to companies in bidding, or indeed the ability of the government to set the number of available H1B visas (or its alternative) at an appropriate level.

        That is, it’s easy to conceive of certain marketplaces that need more workers, but aren’t able to hire them at anything greater than a fairly low wage (I think Yglesias might have discussed this)–say, the market for watermelon pickers. In Yglesias’ example, the employer could hire this individuals competitively at $15 per hour, but not at $15.50 per hour, since the market for watermelons was globally competitive and the individual employer didn’t have a strong enough hand in setting watermelon prices (also availability of substitutes etc. etc. etc.). Americans aren’t willing to do the job at $15 per hour because it is awful, backbreaking work. The employer won’t bid on a visa because that would raise the hourly rate for its workers, which it already can’t afford. And thus watermelon production shifts from where $15 an hour workers aren’t available to where they are.

        I don’t think this is necessarily a skilled/unskilled example, either. The constraint is on the number of workers available domestically to fill the roles required, and the additional cost it would take to bid on a visa (which will also be subject to supply constraints, per your dictum that only the MOST valuable workers should win). I think that that’s an artificial constraint. I think that bringing in watermelon pickers or petroleum engineers who speak portuguese constitute NPV-positive projects that might not get taken up because deep-pocketed NYC law firms and investment banks suck up all the visa spots. And that’s not good.

        Thumb up 0 Thumb down 0
  4. Seminymous Coward says:

    Too many educated workers is a “good problem” from society’s perspective. Adjusting the labor market to suit the workers seems like a good solution to the issues it causes from the individuals’ perspectives.

    If only there were a way to increase the proportion of high-skill jobs in the USA… Perhaps American firms could even export low-skill jobs to other countries with a lack of attractive employment opportunities, thus improving everyone’s fortunes? Wait, no, I heard that was bad because… uh…

    Thumb up 2 Thumb down 0
  5. James says:

    Seems like there are two places where this analysis is at best over-simplified. First, not all education is created equal: a BA or MBA is not at all the same as a BS or MS, and a BS in Sociology is not at all the same as one in engineering.

    Second, there is still a lot of resistance to paying skilled knowledge workers at the same level as semi-skilled managers, so we find a lot of people moving sideways into management just for the money.

    Well-loved. Like or Dislike: Thumb up 7 Thumb down 1
  6. Bob Hammer says:

    If there are so many unemployed or under-employed (below qualifications), highly technically trained people in the US, why do companies claim that they are not enough technically-trained people to hire? Why do companies want so many H1B visa immigrants to fill the plethora of technical jobs? Both things can’t be true. So what is the truth?

    Well-loved. Like or Dislike: Thumb up 5 Thumb down 0
    • James says:

      Yeah, I’m involved with a university research project, and one of our best PhD candidate people (US born, too) hasn’t been around at all this year, ’cause Google offered him close to six figures to start…

      Thumb up 0 Thumb down 1
    • Joe Dokes says:

      The reality is that H1-B visas are used by companies to push down wages. There have been numerous examples, the most famous is on Youtube, where companies deliberately seek out an H1-B, then right the job qualifications tailored to that employee, then place an advertisement for the position and when no American meets the job qualifications AT THAT PAY GRADE, then they bring in the H1-B.

      Joe Dokes

      Well-loved. Like or Dislike: Thumb up 5 Thumb down 0
      • James says:

        I suspect you don’t have much first-hand knowledge of the tech job market. (I can’t force myself to think of YouTube as a reliable source of data.) There may be a few cases in which H1bs are used to push down wages, but those are rare. Most are legitimate needs for highly-skilled people. Nor do I know of any place in the US where tech wages (for people with advanced degrees) could reasonably be said to have been pushed DOWN, when internships at top Silicon Valley companies can run $5-7K/month.

        Thumb up 1 Thumb down 3
    • Veritea says:

      They can both be true – read my comment from 2:49pm. It is caused by pretending apples and oranges are all apples. Highly trained engineers are not all equally valuable, and in fact, some are not particularly valuable at all.

      Thumb up 0 Thumb down 0
  7. Mark Feldman says:

    I believe that the paper by Salzman, Kuehn and Lowell make a fundamental mistake by confusing DEGREED with EDUCATED. I cannot find that distinction anywhere in the Economic Policy Institute paper. If I am correct that people are mistaking degreed for educated – which the data and my experience as a math professor lead me to believe I am – then there IS a dearth of well educated STEM students.
    Here are just two of many pieces of evidence. (For those who want to see more, you can go to my blog
    First, here is a direct quote from how one of the country’s top research universities brags in a magazine that it is working hard to produce STEM graduates:
    “In a typical [physics] class, they hear one or more 10-minute lectures over the material, talk about two-minute problems in groups and discuss their answers. Often there is a demonstration that illustrates the material. At last, they go home and rework the original set of homework problems… Students immediately responded to the new format; they were clamoring to get in.”
    Here are some of the things students (with very high SAT’s, probably averaging 740 in math) say about the course.
    “His exams have unlimited time and are so easy!..”
    “if you ask him questions during exams, he’ll answer them for you in full detail. really easy to do well in this class”
    “It was touted as the best at [the university], but I just finished the semester, and I feel like I did not learn much … I am a Physics major and am worried about the voids in my background. Please give more fundamentals”
    Almost 90% of the students in this class have historically expected an A.
    Now, here is some data (from a totally different source) that I believe show that there is a dearth of well trained STEM graduates and that it shows up in salaries reported by the State of Texas for majors in “Computer and Information Sciences, General” and by Carnegie-Mellon University.
    UT, Austin
    and UT, Dallas has students with about the same SAT range
    Salaries are $71,059 and $65,328, respectively.
    Texas State
    at San Marcos and UT, Tyler also have about the same SAT range.
    Salaries are $58,992 and $45,336, respectively.
    Community College Assoc Deg Median is $57, 395 but
    South Texas
    College Median for the same degree is $26,597.
    Finally, at Carnegie-Mellon University, the median was $95,000.
    From all of this, plus much more,I cannot help but conclude that degrees don’t always represent ability or education. I also conclude that not enough universities are providing their students with the appropriate skills – though some are, and those students are in high demand. This lack of training is much to the detriment of students and parents that believe they are getting a good education. There are many talented young people who may not be able to attend MIT or CalTech or Carnegie, but, if given a legitimate education could find employment in these STEM fields They have been robbed of their chance by the attitudes of too many people in higher education.
    Clark Kerr’s statement from 1980 cannot be repeated too much, “…This shift from academic merit to student consumerism is one of the two greatest reversals of direction in all the history of American higher education…” THAT is the problem.

    Thumb up 3 Thumb down 1
    • James says:

      “His exams have unlimited time and are so easy!..”

      Just the opposite of one of my graduate physics teachers. If you had had the answers to all the questions written out in front of you, you could not have copied them down in the allotted time. The guy believed in challenges :-)

      Thumb up 0 Thumb down 0
  8. L.Tisue says:

    As the US has spent less and less on research and given the “Free Market” more tax breaks and less regulation , getting an education is less and less likely to get someone into the middle class. In countries that have not adopted radical captialism such as Sweden, Denmark, and South Korea, one can still get an education and get into the middle class.
    As an 40 over Ex-pat with a master’s degree I can find better paying jobs in other countries than I can in the U.S. For the countries have natioalized health care and disablity, so it is not a burden to hire a worker that statistically will cost more for health care.

    Thumb up 3 Thumb down 0