Friday, 12 January 2018

Stephen Hawkings on Robots: It's the Distribution, Stupid.

I’ve seldom experienced déjà vu as strongly as I did reading Alexander C. Kaufman’s recent note on Stephen Hawkings’ brief Reddit AMA (h/t David Ruccio).

A member of the public asked:
  1. "Have you thought about the possibility of technological unemployment, where we develop automated processes that ultimately cause large unemployment by performing jobs faster and/or cheaper than people can perform them?
  2. "In particular, do you foresee a world where people work less because so much work is automated?
  3. "Do you think people will always either find work or manufacture more work to be done?"
See “technological unemployment” there?

Hawkings’ answer:
“If machines produce everything we need, the outcome will depend on how things are distributed. Everyone can enjoy a life of luxurious leisure if the machine-produced wealth is shared, or most people can end up miserably poor if the machine-owners successfully lobby against wealth redistribution. So far, the trend seems to be toward the second option, with technology driving ever-increasing inequality.”
Judging by Kaufman’s story, that answer didn't soothe Hawkings' public. After adding a reference to Thomas Piketty’s work, Kaufman explicitly just joined the dots that were only implicit in Hawkings' answer: robots, by themselves, are not the problem; how benefits and costs of automation are distributed is.

Added productivity doesn’t necessarily benefit workers.



Astute readers may wonder how all this talk, so 21st century, so cutting-edge, can remind one of anything. This may surprise them, but it can and, to top it all off, it’s something rather old, which has troubled the good and wise for a while.

In fairness I’m not the only one to notice it. Credit must go to David R. Henderson, too, from the Hoover Institution.

He has been wondering about automation and employment (he doesn’t care much about wages, though). Frankly, it ain’t an inspired piece (the chart above is enough to dismiss two points Henderson attempted: the effect of women in the jobs market and … “pay is closely tied to productivity”).

But my interest is not what Henderson got wrong, but what he almost got right.


This is what.

A year after the September 1929 crash of NYSE, the Great Depression had gone global.

(source h/t Calgacus)

The world then (as today) was suffering “from a bad attack of economic pessimism”, as Keynes wrote in October 1930. People were talking of a lost decade.

The pessimists, Keynes thought, were wrong. So, to cheer the brooding public Keynes published a version of a short essay originally composed in 1928, before the depression. No depression could possibly shake his faith in capitalism. The result was “The Economic Possibilities for Our Grandchildren”.


It took twelve years, the New Deal, and the Second World War mobilisation to return the US unemployment rate to pre-September 1929 levels: only in May 1942 it fell to pre-Depression levels. One may argue why it took that long until the cows come home: that it took that long is the fact.

All that sounds pretty modern, too, doesn't it?

Keynes’ pessimists weren’t too wrong, I’d say.


“Economic Possibilities” must be one of Keynes’ most strange essays ever.

The man whom, according to Skidelsky, “rejected or ignored explanations of events in terms of vested interests and technology”, barely mentioned unemployment and when he finally did, he explained it thus:
“We are being afflicted with a new disease of which some readers may not yet have heard the name, but of which they will hear a great deal in the years to come--namely, technological unemployment.” (Original emphasis)
One year into the Depression and scarcely six before the publication of his “General Theory” and one finds not even the subtlest hint Keynes thought aggregate demand had anything to do with unemployment. Hell, one doesn't find the term "aggregate demand", period. It's all about supply side.

Not only that. He added: “mankind is solving its economic problem (Keynes' emphasis, too). Which is? Scarcity, not distribution. He just assumed people would earn a comfortable living with little effort.

That may sound like Utopia, but Keynes was no vulgar utopian, no siree: Those cashed-up involuntary slackers would feel awfully bored, he explained, just like “the wives of the well-to-do classes” in Britain and the US (whose wealth deprived them of cooking, cleaning, and mending and yet, couldn't “find anything more amusing” to do). They'd risk a “nervous breakdown”.

Keynes arrived at that "startling" conclusion after a historical survey beginning “two thousand years before Christ”. Within 100 years, he reckoned, we’ll live in paradisiacal boredom. In other words, he did in 1930 exactly what he condemned in 1923:
“This long run is a misleading guide to current affairs. In the long run we are all dead. Economists set themselves too easy, too useless a task, if in tempestuous seasons they can only tell us, that when the storm is long past, the ocean is flat again.”
A historian of economic thought of the future could be forgiven to think Keynes never wrote that essay.


Over decades many have attempted to make sense of “Economic Possibilities”. The focus generally is on Keynes’ quantitative predictions (“the standard of life in progressive countries one hundred years hence will be between four and eight times as high as it is to-day”, “three-hour shifts or a fifteen-hour week”), things like how that Essay in Eccentricity fits within his overall work are never considered, at least in the popular literature.

So, let’s go with the flow. It's easier.

As we’ve seen, the shorter-term predictions of Keynes’ opponents (the pessimists) didn’t miss the mark by much.

How did his own longer-term predictions fare? Although per capita GDP has serious shortcomings, if one measured the “standard of life” in those terms, his first prediction may have been already fulfilled. The second prophecy, however -- it’s the general consensus -- went badly wrong and seems to be getting worse.


In a light-hearted piece, Elizabeth Kolbert searched everywhere for answers: from the absurd (according to a feminist author, it’s all men’s fault: they don’t do their fair share of domestic chores, dammit) to the even more absurd (she asked Gary Becker and Luis Rayo).

The Hoover Institution promotes a non-Keynesian interpretation of Keynes. Henderson tries to explain that failure, too: he does that within that interpretation.

The cake for the oddest, most absurd balloney explanation, however, must go to Tim Worstall: people work longer than Keynes predicted because women don’t need to spend as much time doing domestic chores as they used to. They aren’t bored any longer, nor they risk a nervous breakdown, because they’ve bought Ha Joon Chang’s Washing Machine ®, which deprives them of cooking, cleaning, and mending!


A little thinking outside of the box could have saved all parties involved much effort. A couple of years after “Economic Possibilities” appeared, Bertrand Russell (“In Praise of Idleness”, Harper’s Magazine, October 1932 issue), a man personally, socially, culturally, and maybe even politically close to Keynes, surprisingly knowledgeable about economics, provided a good explanation, in the same playful spirit Kolbert employed:
“Suppose that at a given moment a certain number of people are engaged in the manufacture of pins. They make as many pins as the world needs, working (say) eight hours a day. Someone makes an invention by which the same number of men can make twice as many pins as before. But the world does not need twice as many pins: pins are already so cheap that hardly any more will be bought at a lower price. In a sensible world everybody concerned in the manufacture of pins would take to working four hours instead of eight, and everything else would go on as before. But in the actual world this would be thought demoralizing. The men still work eight hours, there are too many pins, some employers go bankrupt, and half the men previously concerned in making pins are thrown out of work. There is, in the end, just as much leisure as on the other plan, but half the men are totally idle while half are still overworked. In this way it is insured that the unavoidable leisure shall cause misery all round instead of being a universal source of happiness. Can anything more insane be imagined?”
A doubling in productivity, in other words, would have to translate into a halving of working time, for the same wages, for a Keynesian scenario to materialise. Nobody would have been worse off and workers would have been positively better off. That was the rosy scenario Keynes was peddling, with the kind of problems the “economically pessimistic” unemployed of 1930 would have loved to have.

That isn’t gonna happen, was Russell’s conclusion. Half the workers, he argued, will work as long as before for the same wages, producing twice as much, the other half will lose their jobs. Surviving bosses may double their profits.

That simple answer, obtained from a simple reasoning, seems to fit the relevant facts. Productivity has increased (tick), but wages remain stagnant (tick), overwork coexists with underemployment (tick) and the capital share went gangbuster (tick).

To make that easier to understand: it’s not the scarcity, but the distribution, stupid.

That sounds pretty much like what Hawkings said, only more detailed, yes? That would be enough to explain the déjà vu feeling, but in truth, there’s more.

Here’s Brad DeLong channeling Keynes. The problem with the alleged end of scarcity, DeLong believes, is not that people are poor but that they have no jobs. Because, you see, people want to work for work’s sake, pay be damned. But he ain't just repeating Keynes' odd essay. He replaced being bored by being "phished for phools"! (Why he kept the middle F is just another mystery).

Oh God, have mercy on us. Give us strength.

14-01-2018. To avoid putting words unduly -- and unintentionally -- in Hawkings' mouth, I made it more explicit who wrote that it was capitalism that one should worry about.


  1. Good catch. I never liked Economic Possibilities. It is good to see you found something
    weird in it, but
    I do not understand your whole article.
    1st Hawkins doesnot say the problem is captilaism
    2nd. I can not see the connection between Economic Possibilities and the robots

  2. You are absolutely right. Hawkings didn't say those words. Kaufman did. I thought that the "joining dots" thing would have been enough. I see I was wrong.

    I fixed the text to make that clear. Thanks.


    About "Economic Possibilities" and modern automation. Increased productivity means more output per hour of work. Under capitalism, increased productivity could benefit workers by (1) reducing working times without reducing real wages, or by (2) increasing real wages without reducing working times (or a little bit of both).

    In case (1), you do your job, if you finish early, you can go home early. In case (2), you get a pay rise because you did your whole shift and during that time you did more, produced more stuff.

    Russell argues (1) is what Keynes said it would happen (and what people today hope it will happen with robots). Russell disagrees with Keynes.

    What actually happened is that capitalists hogged all the benefits of increased productivity, pretty much like Russell back then said they would, and how Hawkings-Kaufman seem to be saying now.

    In other words, what really happened is that you finished your job early, but had to do your entire shift without getting any pay rises: your boss got more stuff and you got nothing. You were lucky: others lost their jobs.

    Russell could see that, Keynes couldn't or didn't want to.

    I suspect it's precisely because of this that people like Henderson and Worstall find that essay useful and are happy to quote it. Keynes, after all, has this bogus reputation as Lefty Prophet. People may be induced to believe the conclusions Henderson and Worstall prefer, if Keynes held them.

    Makes sense?


    You don't need to log into Blogger, but if you could add a moniker (initials, a nickname, Internet handle, whatever), it would be great. :-)

    1. Thanks. I understand now.

  3. That graph of US unemployment is not "a fact". Unemployment by any sane definition was rather lower than that. They are defined to make it logically impossible for Roosevelt to have done anything against the Depression's unemployment, for they define unemployment = (people not having jobs) + (people employed by New Deal programs).

    The prevalence of those absurd numbers, decades after they were corrected is part of what is imho a nearly universal lack of scholarly standards right or left concerning the New Deal & FDR, a nearly universal effort to say the grandparents of today's USAn were idiots for voting for him.

    1. Hi Calgacus,

      That data are official, hosted by FRED. I did not get them from a dubious source, unless FRED is dubious. (I noticed I hadn't included the link, so I did it now and copied it below) [*]

      Now, if you are sure that data are faulty, deliberately designed to make the New Deal and FDR look bad, I'm more than happy to consider your argument. Feel free to link to the relevant literature.

      Likewise, if you have better, more reliable data supporting your contention I'm more than willing to examine it.

      Sounds good to you?


  4. Just because they're official doesn't make them not garbage. The really big problem in economics is not lying or bad data, but widespread conceptualizations, incessantly obfuscated by Big Words, that once explained can be seen to be absurdities by 12 year olds.

    This is the difference between the Darby & Lebergott numbers. The latter seem to be the ones posted, they are the ones seen in almost all sources and they omit all the employment from the New Deal programs, simply by calling those employed by the WPA etc "unemployed". I erred a bit on the logical impossibility, because such numbers do reflect the large "multiplier" effect of New Deal employment programs. Marshall Auerback among the MMTers has done good work on this. Here are a couple of links: Time For a New “New Deal” or The Real Lesson from the Great Depression: Fiscal Policy Works!

  5. Calgacus,

    Thanks for the links to Marshall Auerback's papers.

    I sympathise with bona fide MMTers and, therefore, a priori I tend to accept their claims. This time will be no exception, so I'll take Mr. Auerback's claims at face value; moreover I confess I had no idea those figures were subject to controversy.

    Having said all that, let me comment two things only:

    (1) Okay, those figures are crap, well, what are the right figures? We need figures to assess the goodness of the New Deal, don't we?

    (2) Mr. Auerback writes:
    Even pro-Roosevelt historians such as William Leuchtenburg and Doris Kearns Goodwin have meekly accepted that the millions of people in the New Deal workfare programs were unemployed, while comparable millions of Germans and Japanese, and eventually French and British, who were dragooned into the armed forces and defense production industries in the mid-and late 1930s, were considered to be employed.

    Need I say more?