After watching several video lectures made by Richard D. Wolff I noticed he frequently draws historical events as evidence to base his arguments on. One thing he mentions frequently (in several videos) that I was unable to verify on the web was the USA labor shortage.

I don't remember the exact details but he says that since around 1866 (after the civil war) and up until 1970 there was a labor shortage in the USA that caused, among other things, a steady increase in real wages throughout the whole century.

Was there really a labor shortage in the US? Was it as long as he claims?

  • This is a question about economics, not history. May 27, 2014 at 17:43

3 Answers 3


Can this question be resolved through historical sources and methods?

I don't think I can answer the question without more research, that I'm not going to do unless paid to do so. But I would argue that we can answer the question, and that the methods are similar to those required by historical scholarship.

We need to do two things. First, we need to define our terms and clarify what we mean by a labor shortage. I'm sure Dr. Wolff defines "labor shortage", but I'm equally sure that since wikipedia terms him America's prominent Marxist economist, that I'd find his definition incomprehensible and unworkable. (Aside: Strictly speaking, I'm not discarding a man's academic career merely because he is a Marxist; I think it would be more accurate to say that Marxist economic analysis requires assumptions that I don't share, and that I have to tie my mind into knots to even begin to follow. )

America has a labor shortage over the period if the price of labor rises (indicating that the supply is insufficient). America does not have a labor shortage if the (a) the inflation adjusted price of labor remains steady or falls, or (b) the rate of unemployment rises over the long term.

I'm using this definition because it avoids the fundamental problems of the academic definition proposed by Mr. Durden. Labor supply is affected by a variety of factors including birthrate, immigration, changes in employment needs, regulatory pressures, public choice economics, etc.

I'm also going to introduce a major caveat before I go any further. The period described by Dr. Wolff is curious - from the civil war to the present day. If I recall correctly Tyler Cowan of George Mason has estimated that in the post bellum period, 90% of the employment was agricultural, and 10% industrial. By 1910 those percentages had reversed and 90% of the employment was urban/industrial. (The numbers may be off, but the rough magnitude of the shift is, I believe, generally agreed.) My understanding is that agricultural unemployment is very rare - in point of fact almost all agricultural labor includes surges of greater than 100% employment. So I'll grant Dr. Wolff the period 1865 to 1910, but I'm not sure that "labor shortage" in a primarily agricultural economy leads us to any useful conclusions. Certainly no meaningful public policy conclusion can be reached from this period that has any meaning for later periods.

The period 1910 to 1940 is interesting - I think we all agree that the economy was affected by two world wars that distort the picture. War distorts the allocation of labor in ways that are not useful for formulation or analysis of public policy. I've argued that the Great Depression was the result of poor public policy. As @T.E.D. states, it is impossible to argue that the unemployment rate during the great depression supports Dr. Wolff's theory.

So let's start the analysis with the period 1950->present, see if there is a trend, and then go back and see if the trend helps us to understand events in the more difficult periods. Is there a rising price of labor? Is the unemployment rate rising or falling?

@T.E.D. has provided the beginning of an answer - there is data on historical unemployment levels from 1940, and estimated values for the period prior to 1940. (no source on the chart AFAICT, which prevents us from understanding the context, assumptions and bias). If I were to estimate a line, I'd say the period 1940 forward is a rising line. Based on the assumptions above, that weakens support for Dr. Wolff's theory.

I would want to test this hypothesis further by:

  • Checking the average wage level over the period - is there the correlation between unemployment & the price of labor that I'm implying? If not, why not? (I suspect that if there is not a correlation, then it is a lag due to structural components, and that the most likely places to look are (a) sector employment - if for example, the price of skilled industrial labor rises dramatically, while the price of unskilled labor falls, that would create both a rising cost and rising unemployment). Alternatively it could be specific industries - if the price of buggy whip manufacturers falls dramatically, but the price of computer technicians rises, then that could result in the same phenomenon. We'd have to make sure that the study compares prices over a timeframe long enough to provide for the possibility of retraining and skills aging out.

  • Age based unemployment.

  • Changes in the regulatory environment. The price of labor for college professors and librarians falls throughout history because there supply far exceeds demand. In Maryland at least, the price of barbers and morticians rises consistently because these two groups have obtained a legal monopoly. Auto dealerships have a legally protected oligopoly in the USA - it is illegal to disrupt their trade. This can result in rising unemployment and rising prices. I think Mr. Durden is in error to ignore this, but I think it is also an error to treat it simplistically. If we could measure the efforts to achieve legally protected oligopoly, and the effect on the price of labor and the supply of labor, I think that might tell us something significant. (This is closely tied to what I believe @Samuel Russell calls "class warfare".)

  • If we spent 20 minutes thinking about the problem, we'd probably come up with other hypothesis to test.

So the summary - historical methods tell us to collect data, create and test hypothesis, and search for bias that might affect the sources and the conclusions. Can we respond to Dr. Wolff's theory through historical methods? Yes. I think they overlap heavily with economic sources & methods in this case.

Am I doing to answer Dr. Wolff's thesis? Hell no. For one thing, that would require me to understand modern Marxist economic theory at a far deeper level than I'm willing to do. The second and equally important reason is that IF you can disprove the thesis of a scholar of Dr. Wolff's reputation, then you've probably got at least a book, if not a PhD thesis on your hands. I'm not interested in that level of effort.


Any such statement clearly seems to be ignoring several recessions. Most notably the Great Depression, during which anywhere from 15-20% of the US workforce was unemployed for an entire decade. During this time, people were quite literally starving to death for lack of work. I suspect you would have had a really tough time convincing much of anyone there was a labor shortage in 1934.

enter image description here

Seeing as the Economist in question held down a job in that profession for several decades, I seriously doubt he's just ignoring this. So I suspect he either has a definition of "labor shortage" that's a bit divergent than the typical "insufficient qualified candidates (employees) to fill the market-place demands for employment at any price", or he's got some really good reason why all the various market panics, recessions, and depressions of the era didn't count.

But in a strict pedantic sense he is clearly wrong: there were certainly periods in the century after the Civil War, including an entire decade at one point, where there was a large surplus of labor.

  • This answers the question perfectly. What i really wanted to know is whether the "US labor shortage" is something historians refer to as well. I take it that's not the case. May 27, 2014 at 18:12
  • Well, I think there is probably a case to be made that in the antebellum period the US was operating on a kind of opportunity-rich "frontier" economy, and it isn't anymore. That's significantly more nuanced of an argument than saying there was a "labor shortage" that lasted a whole century though (and one that probably would need an economics stack to dive into properly). Labor, like any other market, can fluctuate wildly.
    – T.E.D.
    May 27, 2014 at 18:17
  • Given that there are six measures of US unemployment, can you indicate which measure the chart represents? May 27, 2014 at 21:20
  • 1
    @SamuelRussell - It was taken straight out of the linked website if you want to go investigate that. IIRC the US government used a different measure back then than any of the measures it uses today. I forget which U# it is closest to (not the U3 normally reported by the press). But just eyeballing the chart, you can see part is using a different "estimated" measure too. Probably its best to just take it as-is (a relative measure for the time shown, with a clear big jump), and not try to compare the absolute %'s to today's numbers.
    – T.E.D.
    May 28, 2014 at 14:37

There is no such thing as a "labor shortage". "Labor Shortage" is just a propaganda term used by employers who are trying to find some excuse to pay less. For example, many manufacturers complain that there is a "shortage" of machinists. What this means is that they would like to pay machinists $10 an hour and, surprise, surprise, no machinist wants to work for $10 an hour. If the manufacturer offered $100 an hour, he would have machinists coming out of his ears. He would have machinists lining up at his front door wanting to work for him. He would have machinists flying from all over the world to work at his factory.

Likewise, employees use the same political language. They say there is a "job shortage". Of course, there is no job shortage. If you are willing to work for $5 an hour you will find hundreds of employers willing to hire you. In fact, for $5 an hour * I * will hire you.

There is no such thing "labor shortages" and "job shortages". They are just made up terms used by people for political purposes.

  • I think he meant labour shortage in the sense that there were actually more available jobs than workers. May 27, 2014 at 17:49
  • 2
    You don't seem to get it. There are an infinite "number of jobs". If you and your friends agree to work for me for 25 cents an hour I will hire you all, even if you have a thousand friends. I just "created" 1000 jobs instantly. You seem to have the erroneous idea that "jobs" are some kind of fixed commodity. A "job" is just some guy willing to pay some other guy to do something. May 27, 2014 at 18:08
  • 4
    In a pure theoretical market, this may be a valid opinion. However we work in a world where we need to consider things like minimum wages, professional certification (there may be an infinite number of laborers, but the number qualified to do brain surgery is much less), training (The demand for computer scientists in the post civil war era was small; the demand for steelworkers declined precipitously in the latter half of the 20th century), immigration law and a hundred other things.
    – MCW
    May 27, 2014 at 18:11
  • 1
    Actually, I'm very sympathetic with this answer. For unskilled labor, it is 100% correct. For labor in general...well, it depends on what exactly Mr. Wolff was talking about (which we still don't know).
    – T.E.D.
    May 27, 2014 at 18:21
  • 2
    Per @MarkC.Wallace this answer fails to account for the balance of forces in class warfare. May 27, 2014 at 21:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.