Arnold Kling  

Priceless Economics

Marco Rubio: Trade Agreements ... Taking Lousy Government for Gr...

A McKinsey Report says,

Based on current trends in population, education, and labor demand, the report projects that by 2020 the global economy could face the following hurdles:
  • 38 million to 40 million fewer workers with tertiary education (college or postgraduate degrees) than employers will need, or 13 percent of the demand for such workers

  • 45 million too few workers with secondary education in developing economies, or 15 percent of the demand for such workers

  • 90 million to 95 million more low-skill workers (those without college training in advanced economies or without even secondary education in developing economies) than employers will need, or 11 percent oversupply of such workers

Pointer from Daniel Lippman, via Tyler Cowen.

In freshman economics, you learn that supply and demand curves cross. If there are "too few" of something, the price goes up. If there is an "oversupply" of something else, the price goes down.

The McKinsey report is an example of priceless economics. (Sadly, there are many other examples, including Keynesian macro as taught in freshman economics.) The quote above is written as if no adjustment in relative wages will ever take place.

If the market system works, the factors that McKinsey is describing should produce lower wages for low-skilled workers and a higher premium for high-skilled workers. To the extent that supplies and demands by skill category are inelastic, these effects will be strong. However, if those elasticities are higher (as one might expect over decades), the effects will not be so strong.

Central planning and prices are substitutes. Since it assumes that prices fail to clear labor markets (even in the long run), the McKinsey report is filled with ideas for central planning.

Instead, one might ask, what factors inhibit adjustment in labor markets? For example, how does "low-skill," which should imply a low wage, become ZMP, which means that the cost of hiring the worker is higher than the marginal product of that worker? What are governments doing to keep the cost of hiring so high (think of minimum-wage laws, the linking of health insurance to employment, etc.)?

To what extent does "high-skill" mean "protected by a credentials cartel?" Something like one-third of all jobs in the United States now require a license. Government plays such a large role in providing education and regulating health care that those New Commanding Heights industries are blocked from changing the mix of labor in response to market forces.

Another question is how these inhibitions to adjustment play out globally. It is one thing to see factors at work in one country. From a global perspective, there is the additional consideration that some of the adjustment is impeded by formal and informal barriers to immigration.

Finally, I want to suggest one more possibility. Perhaps the seemingly low "supply of high-skill" (aka highly-credentialed) workers is a reflection of low demand for the lifestyle of high-skilled workers. Maybe at the margin many people would prefer more leisure to higher cash incomes.

Comments and Sharing

COMMENTS (15 to date)
noiselull writes:

This read like a more sophisticated version of a recent Daniel Kuehn post.

Daniel Kuehn writes:

Ya, that was a pretty bad report.

All the points you make are good... of course the licensure depends on the market. People will make these points even about unlicensed professions. And I'm not always sure how big of a difference licensure makes anyway, in certain professions (I'd take an unlicensed physician or a nurse doing more physicians tasks, but would you want an unlicensed surgeon?? I wouldn't... I'd want some type of official certification for anyone performing surgery on me, and in the end it probably would look about the same as the current licensing).

Aside from the problems you state, the other problem is the attempt to forecast demand and supply (even if they were to treat them as schedules rather than quantities). Supply can be fairly straightforward to forecast - a lot of that is the educational system's capacity to educate and demographic trends. But demand is notoriously difficult to forecast, particularly when we're talking about high-skill professions. We just don't know what skills are going to be in demand in the economy of the future. Richard Freeman has a good analysis of the inaccuracies of BLS occupational forecasts.

I have some chapters coming out in an NBER volume on the engineering workforce that address a lot of these issues, which I'll share whenever it finally gets released. My colleagues and I also just got a Sloan Foundation grant which is going to fund a lot more of work on exactly this area.

Don't believe chicken-little stories about labor shortages - we have decades of research by economists of all sorts of ideological persuasions debunking it. Unfortunately you still hear it from lots of people - and from more than just politicians as well, unfortunately.

Daniel Kuehn writes:

One more thing I found particularly funny about this report - notice that it isn't even some kind of mismatch or structural story. According to the report every class of worker is going to be in short supply. Seems kind of funny, right? How could that come about?

In that kind of situation, what you'd want is higher capital-to-labor ratios and labor-saving innovations, and then all of this would be a moot point, right? If only there were historical precedent for that sort of thing!

Oh wait...

John Palmer writes:

This type of report reminds me of the CLUB OF ROME computer models of the early 1970s which demonstrated conclusively that the world was going to run out of all sorts of vital natural resources in 15 - 30 years. Their models had NO prices or incentives in them.

Glen Smith writes:

Thing is, besides the first two years where I pretty much learned things that don't help me in a traditional economic way (well, there is that work ethic and discipline thing needed for doing stuff I don't want to do), the rest of college was a waste (except for any signal value I got from getting my credentials) because the things I learned were things I wanted to know, had a strong aptitude already and/or had obvious value to me in terms of future prospects (where work ethic and discipline are unneeded). Don't need discipline or work ethic to do things I would be doing anyway. In any case, what I do now has nothing to do with my college degree. The best people in my current field (computer science) only need degrees to get beyond the gate-keepers (and some don't even need them then).

Adam writes:

Hey Arnold,

I wrote a version of the argument in your last paragraph a little while ago:

Dave Schuler writes:

With respect to professional degrees and doctorates is it reasonable to speak of the operations of the market? Aren't those so dominated by govern jobs or subsidized so highly by the government that we're really talking about a command economy for them?

That's not a rhetorical question. I'd really like to know the answer.

Costard writes:

Dave - I've wondered this, too. For instance, what to make of the demand for MBAs in business, despite so many eyes rolling with regard to the usefulness of business programs. Or the drive by educators - not just in colleges but in secondary schools - towards advanced and terminal degrees, which nevertheless fail to budge test scores or result in any measurable progress.

My gut reaction is that people appreciate the comfort of these acronyms in their business dealings - a sort of touching, vestigial faith. When hiring in large bureaucracies (business or otherwise) is more about CYA than about capturing talent, managers will take every bit of insurance they can get. Oftentimes there doesn't seem to be a whole lot of discretion with respect to salary and benefits, and consequently there's little leeway on the job-qualification side of the equation. Perhaps this has to do with (what I perceive as) the increasingly large gulf between small businesses and large firms. Mid-size companies seem to be in many industries a dying breed; and they would be the group most likely to attract and hire unconventional talent.

Joe Cushing writes:

The report is bad for another reason. It's ridiculous to try to predict how many employees the world will need 8 years from now--at any fixed price. There are a million things that could happen between now and then. We don't even know where we will be in the business cycle at that time. If the world keeps going the way it has been with big government programs to fight this slump, we will still be in the bottom part of our L shaped recovery 8 years from now. The great depression lasted as long.

Mike Rulle writes:

I guess you never worked with Mckinsey before, as this would be less of a surprise to you!

Beckett writes:

But you forget the heterodoxy of decision making. Take me for instance. I have a genius IQ (its 137). I got a 4.0 in high school. I got a 2350 on the SAT. But guess what, I don't want to go to college. I want to try become a musician. How can you explain that? I have no desire to go make money. None. Guess I must be defective, according to the above analysis.

Cyberike writes:

A couple of points: First, the undersupply of skilled workers is a myth. I have seen several studies that debunk it, but instead of quoting them, just go to the local want ads. Example: process engineer, BS Chemical Engineering plus three years experience in chloride processing, specifically titanium or silicon tetrachloride manufacturing.

This is just one (of many) examples. There is no shortage of skilled workers, there is a shortage of workers with specific experience in a specific job, many times in jobs that did not even exist 5 or 10 years ago. If you get that experience, is there any guarantee that the job will exist 5 or 10 years from now?

Second, the supply demand curve needs to take into account automation. There is (still) high demand for low skill workers, but that demand can be filled through automation at a cost far lower than a living (American) wage for human workers. The only places where human labor can compete with automation are those countries where wages are incredibly low, and why those countries are not able to sustain their economic growth. Their wages can't rise above what it would cost for a machine to do the work, and the cost of automation is steadily declining.

djf writes:

"From a global perspective, there is the additional consideration that some of the adjustment is impeded by formal and informal barriers to immigration."

Gee, Arnold - any possibility that the projected oversupply of low-skilled workers may be due to the lack of effective barriers to immigration?

Steve Sailer writes:

Thanks God we imported all those unskilled and highly fertile illegal immigrants!

KnowPD writes:

McKinsey's clients hire college graduates. If the labor cost of college graduates goes up, then their client's profit levels go down. Therefore, McKinsey has incentives to declare insufficient number of college graduates in order to give their clients information to influence policies that reduce labor cost by increasing "investment" in education. We see the exact same phenomenon with tech companies complaining there aren't enough qualified engineers.

Comments for this entry have been closed
Return to top