2 econ Nobel prizes, 1 error

This came up before on the blog but it’s always worth remembering. From Larry White, quoted by Don Boudreaux:

As late as the 1989 edition [of his textbook, Paul Samuelson] and coauthor William Nordhaus wrote: “The Soviet economy is proof that, contrary to what many skeptics had earlier believed, a socialist command economy can function and even thrive.”

Paul Samuelson and William Nordhaus won Nobel prizes in economics. We’ve talked about this example before; I learned about it from a post by Alex Tabarrok recounting some research by David Levy and Sandra Peart.

There are lots of cases of academics, Nobel-prize-winning and otherwise, falling for fringe theories as they get older. Google Josephson or Shockley, for example. And even the most distinguished scholars have been known to make technical errors and then be too stubborn to correct them.

But the above-quoted example is different: Samuelson and Nordhaus are economists. They’re writing about their area of expertise—or, I should say, purported expertise. They should know better, or at least know that they don’t know enough to claim to know, right?

My point is not that Samuelson and Nordhaus are fools. I’ve not met or read much by either them. They might both be brilliant (although it’s hard to tell because sometimes it seems that economists like to say how brilliant other economists are) and I assume they’ve both done excellent, important work. The above quote from their book . . . make of it what you will.

I guess the problem is that social science is so damn political. Not just politics politics, also academic politics. Last year we talked about the echo chamber within the subfield of climate economics. So much of the academic world seems to be about people promoting each others’ careers.

P.S. I complain a lot about academic economics and psychology. I’m sure lots of other fields have problems that are just as big. Economics and psychology are just easy to talk about because they are relatively non-technical topics. If we want to argue about whether a particular drug works the way it’s claimed, we might need to know a lot of biology. But you don’t need any particular technical knowledge to recognize that the Soviet economy was not thriving and that the claims in those pizzagate papers were not coherent.

50 thoughts on “2 econ Nobel prizes, 1 error

  1. If one lives by the sword, one can die by the sword. It is always dangerous to criticize famous individuals (e.g.,Samuelson, Nordhaus or Gelman) who occasionally make an error. Some errors are more forgivable, some are not. What about the following:

    “I’ve not met or read much by either them.”

    Was this sentence purposefully placed by Andrew in order to engender comments concerning hierarchy of errors or was it an innocent typo?

  2. Thing is, many of us were alive and knew this was nonsense at the time, and yet I remember that nonsense being passed off as true. How could they trust data from the Soviets? Even if they demoted the data, they didnt approach the reality.

    And this happens a lot. We were, it seems, almost entirely surprised by the state of Iraq’s state-owned companies, though huge numbers of people believed correctly that they had to be papered frauds.

  3. Seems that Economics and Psychology are especially politicized because government (broadly defined) is the primary buyer of their products. Government is dominated by politics and subjective ideology.

    American government also dominates all levels of formal education/academics within which these loose disciplines of Economics and Psychology thrive far beyond their actual contributions to society.

  4. Thank God I do math. It’s a lot harder to politisize. Proofs tend to be convincing or not and while in theory their could be edge cases they aren’t common in practice.

      • (For those who are not familiar with “the Mochizuki abc brouhaha”, here is a pretty good discussion:
        https://www.quantamagazine.org/titans-of-mathematics-clash-over-epic-proof-of-abc-conjecture-20180920/)

        Peter said, “Proofs tend to be convincing or not and while in theory there could be edge cases they aren’t common in practice.”

        The Mochizuki case appears to be (currently) what Peter calls an “edge case”.

        “Edge cases” are fairly rare, but sometimes do occur. In mathematics, when someone believes that they have proved a conjecture, they write up their proof and publish (either formally or informally) their write-up for others to read. Often the proof is correct. Sometimes it is hard to follow (whether or not it is correct). Sometimes it has “holes” in it — which sometimes can be patched up, and sometimes can’t.

        It appears (from the article cited above) that all but one step in the proof is convincing to all who have commented, but there is one part whose proof hasn’t been convincing to some people. So Mochizuki’s proposed proof has not (at this point) been universally accepted as sound. It may be that the problematical part just needs better explanation, or it may be false. Even if the problematical part is false, there might be a way to get around it, producing a proof that has many things in common with the proposed one — or there might be an entirely different proof. Bottom line: The jury is still out.

        • Yes. But it very much has been politicized. Mochizuki’s group at Kyoto has gotten significant funding from the Japanese government on the basis of the proof being correct, and the Japanese press has covered it as the proof being correct being the last word (with big-print headlines, multiple times). But (the last I heard) was that the math types who thought the proof was wrong went to Kyoto to discuss it with Mochizuki, and were not satisfied.

  5. I do think that Samuelson’s ridiculous position on the USSR’s economy highlights the biggest problem with the kind of economics that he championed. He wanted a quantitative macro-economics driven by data. That required taking the GDP data very seriously, which he did. I can’t find the quote, but he even belittled Arrow’s impossibility theorem. The fact that he would take the USSR’s data on GDP more seriously than first hand accounts of what life was like for people living in the USSR was a feature not a bug of Samuelson’s thought. And, it still infects economics and other social sciences. It’s the “numbers don’t lie” crowd. Sometimes non-quantitative data is better, and at a minimum it shouldn’t be ignored. In the 1980s, we had a teacher exchange in my high school with a school in Poland while Poland was still under communist control. The teacher (hand picked by the communists) insisted that life was great in Poland. However, when one of the teachers took him to a grocery store, he started to cry. He had never seen so much food in his life. That was one non-quantitative data point that was worth more than all of Samuelson’s forecasts.

    • This is a great point, Steve, but it is far from unique to economists. It is at the root of all kinds of controversies. The uber-myopic sabermetrician who thinks that baseball need not be watched to be understood. The pollster who sees only numbers, not issues. The epidemiologist who sees ordinary differential equations instead of lives. The lure of expert medical diagnosis systems over in person evaluation by doctors.

      None of these perspectives is wrong if approached with humility. Sabermetrics has immensely helped us understand baseball; polls have made elections far more efficient; epidemiologists’ models have their uses (I guess, though I’m a little more shaky on this one lately); and individual doctor decisionmaking is markedly improved by expert systems.

      But the pioneers are often zealots, understandable given the resistance they had to surmount. The quantification of everything is great, and valuable. But mistaking the numbers for the ultimate objects of interest is surely not limited to economists, although it is a common failing among us.

      • “But the pioneers are often zealots, understandable given the resistance they had to surmount.“

        RIP Dennis Lindley, who single-handedly (it seemed) fought the battle for Bayesian statistics against a frequentist statistical world throughout the 60s and 70s. He became more and more extreme as time went on and enjoyed provoking reaction – when commenting on an invited paper in JASA he started by announcing that the USA was the only one-party state in the world that had two parties.

      • “polls have made elections far more efficient” – This makes me curious actually. What is meant by “efficient”, and how would elections be “inefficient” without polls?

      • Jonathan:

        Are there any actual uber-myopic sabermetricians who thinks that baseball need not be watched to be understood? My impression from reading sabermetrics is that these guys watch baseball all the time.

        • I dunno. But I have certainly read stuff complaining that watching is inferior to looking at numbers if what you want to do is evaluate players or strategy. I have seen arguments that biases creep in when you watch that the numbers filter out. That doesn’t mean those guys refuse to watch games. It just means that they don’t use the evidence from watching games to inform their judgment. If your biases are big enough, that actually makes sense. The problem is where that is elevated to a universal assumption that *nobody* can do better seeing than reading the numbers. That sort of unjustified overreach is surely false. There can be lots of people who are better informed by watching than by reading stats. And, of course, the optimum is to take both, and make best use of the parts of each that are mutually orthogonal.

  6. I don’t think that statement was political (not in the sense of trying to do propaganda for one faction). William Easterly, which is one of the major accuser of false expertise in Development Economics, wrote in “The tiranny of experts” (maybe it was in “The Elusive Quest for Growth”) that in the ’60ties even he started believeng that the Soviets were actually managing to grow through a mix of industrial investments and 5-year plans. And these beliefs survived even after he was sent in Russia as a delegate and found himself in a run-down hotel with no AC and scarce resources (and he knew that that was the hotel for important people, therefore other hotels must’ve been even worse). He explains that the misinformation about URSS’s superior economic performance coming from both the Soviets and US intelligence was so ubiquitous that even skeptical economists started believing in it and I suppose Samuelson was one of those.
    On a side note, Easterly says that one of the fundamental reason why Economists obtained their position as relevant figures and not just academic figures was exactly this misinformation about socialist nations’ performance. Suddenly the Economists were the only experts that could help the market-based economies to defeat the planning-based ones.

  7. “a socialist command economy can function and even thrive”

    I think reasonable people can disagree on this. It is actually pretty complicated:

    https://nintil.com/the-soviet-union-gdp-growth/

    What if some Soviet central planners did a good job and others not so much? If you cannot rule that out from the data, then you cannot refute the quoted claim. In fact, you only need a single good planner at some point in history to support the claim.

    • I hope people are actually clicking on the link above. It’s a nice overview of the question whether the Soviet command economy could thrive.

      To reply to Libet — one can probably get a good idea of how the gdp’s of market economies compare by judging their hotels. But in a command economy one simply cannot. The Soviets chose to use their gdp to do things like build an enormous military and subsidize the Eastern bloc. If they choose to not do things that a market economy would do, like install air conditioning in hotels and provide enough food for the population, that does not have much bearing on the question of what sort of economic growth a command economy can produce.

      • “If they choose to not do things that a market economy would do, like install air conditioning in hotels and provide enough food for the population”

        ???

        The point is they were forced to make that choice because their economic resources were horrendously underutilized.

  8. It is useful to consider how things looked when Samuelson was developing his ideas. The Soviet Union survived massive destruction in WWII, and then rebuilt to be a great power and beat us to space. I was in high school when Sputnik launched, and there was a widespread feeling that we were falling behind. That context makes it easier to see how Samuelson missed the danger signs.

    • I agree, and was about to make that point. However, the quote above is from the 1980s. Samuelson continued to forecast strong GDP even as late as the 80s
      The quote above is defensible out of context, but not in context. Samuelson wasn’t just saying planned economies can survive. He said the USSR was thriving in the 1980s. That’s Baghdad Bob territory.

      • However, it may well be a quote that originally appeared in the textbook many years before, and simply survived multiple editions… The first edition was in 1948, and the 1997 edition was the 16th.

        Samuelson is considered by many to be the greatest economist who ever lived. The Wikipedia page on him gives some hint of this.

  9. Relevant here is a wonderful novel about the management of the
    Soviet economy — “Red Plenty”

    Totally convincing about how a centrally planned command economy will fail.

  10. This discussion reminds me of a book “Measuring the World” by Daniel Kehlman. The book is a fictionalized biography of two intellectuals, Carl Friedrich Gauss and Alexander von Humboldt, who were contemporaries. Gauss was a true genius who believed that a good mind, a pencil, and a comfortable desk were all that was needed to make scientific discoveries. Humboldt trekked all over the world across dense jungles and climbed the highest mountains his body could withstand. We need both thinkers and jungle-sloggers, and our thinkers should spend time in the jungles of real world data, and our collectors of flowers and data from the far corners of the world need to sit and think quietly at times. Dr. Samuelson should have spent some time drinking vodka with steelworkers in Magnitogorsk on a couple of occasions or firefighters from Chernobyl.

    • This book is a great read. As you say it is based on the lives of these two great scientists but is a work of fiction.

      So you may or may not care to believe the author when he describes Gauss deriving the theory of least squares while making love to his wife.

  11. Where do you get the idea the Soviet economy was that bad? It clearly wasn’t competitive with capitalist economies, but that doesn’t mean it was failing. I always thought the collapse was because of social issues more than economic ones; living in a police state sucked, and it was becoming increasingly obvious economic life was better in the west. The transition to a market economy was what crashed Russia’s GDP, which was stable until then. Again, not being an apologist for the Soviet Union, which sucked; I just don’t see how the quote is terribly inaccurate from an economic point of view.

  12. I always taught my students that the USSR economy worked, badly. The USSR was a world military superpower with a respectable life expectancy and an outstanding literacy rate. Consumption levels were low but still better than most developing countries. (All of these measures of “thriving” declined when the command economy collapsed.)

    East Germany had a command economy and had a very respectable material standard of living.

    One of the reasons the USSR’s economy suffered was the boycott of the Western powers. A market economy facing a comparable boycott would have also faced formidable obstacles.

    I never taught from Samuelson’s textbook so I don’t what he meant by “thrive”, but if the counterfactual is that a command economy can’t have first-world life expectancy, a high level of literacy, and a powerful military then the USSR does indeed prove that a command economy works, albeit badly.

    Janos Kornai had many fascinating insights into what a planned economy can and cannot accomplish.

    • Asher is right. Samuelson and Nordhaus were making a political statement, in an effort to sound sagely moderate between the extremes of marxism and capitalism, which was unscholarly, since the truth often *is* one extreme or the other. In everyday life, if half the newspapers says 2+2=4 and half say 2+2=5, most people think saying 2+2=4.5 is the praiseworthy, moderate, position. But their “thriving” statement, while conveying falsehood and blameworthy, has a correct interpretation too. The Soviet Union was much richer in 1989 than it was in 1919. It had industrialized. It had radios and TVs and airliners. It showed an important fact that economists should emphasize: no matter how bad your economic policies, it’s really hard to keep technical change and regular investment from making your country richer over time. You may only get half the growth of your neighbors, but it’s still growth.

      That was actually the big point of Schumpeter, Samuelson’s advisor. In the long run, all that matters is growth, not year-to-year macro policy, and all that matters to growth is innovation, so fostering innovation (broadly construed to include opening up new restaurants, etc.) is all that really matters. And, we can add as an appendix, unless you’re the US, it’s not scientific and technical innovation that matters— you can import that from the US– it’s “soft” innovation (the new restaurant) and use of imported techniques (wireless telephones in Indian villages).

  13. I don’t believe this mistake required an unusual amount of self-deception. You can imagine a world in which circumstances were regular enough that centralized planning could be made to work. Even at the current time China is clearly not a free-market democracy, but Party interference is not yet killing their economy – in fact some still believe that they are getting a mercantilist advantage. If you are prepared to believe that such a thing could exist, look around before the Berlin Wall fell, when even hawks and Science Fiction writers were predicting the survival of Soviet Communism into the indefinite future – and you have what seems to be an example of this theory playing out in front of you.

  14. I don’t think the quote is a good illustration of your point. Samuelson was clearly guilty of overrating the Soviet economy, but “The Soviet economy is proof that, contrary to what many skeptics had earlier believed, a socialist command economy can function and even thrive.” is an OK statement. Socialist command economies functioned for several decades and in some cases even (very briefly) thrived. And it was a long-held belief by economists that this was utterly impossible, command economies should just collapse and being gone forever (historically, they indeed collapsed until the commanders found the right mixes). The quote does not imply what it seems to be in your mind and the mind of most commenters: an assertion of some kind of parity in the performance of command economies vs market economies. You can easily find better quotes for that in Samuelson’s older texts, but I doubt you can find them in the most recent ones where, as in this case, he or his coauthors were more cautious.

  15. When I think of Nordhaus I think of those DICE/RICE models of climate impacts with hilarious methodology for damage functions. They claim a global loss of income potential of 8.5% for 6C of warming. LOL. They have no idea how to think about risk, non linearity or higher order effects. Looking around, not hard to see us come undone at 2C of warming.
    https://www.pnas.org/content/early/2017/01/30/1609244114
    To the extent that policy makers and other elites take this kind of exercise seriously, the damage done is tremendous.

    • Comparing how physical sciences do modelling vs Nordhaus & co would be funny if it didn’t contribute to delayed action on climate change. The “cause-free curve-fitting” that is often used in economics would be laughed at in a “real” science but is sadly used as a policy-shaping input by governments and global bodies like IPCC
      see
      This Little Known Nerd Fight Will Shape The Fate of Our Way of Life
      https://bigthink.com/ipcc-climate-science
      and new paper by Steve Keen
      The appallingly bad neoclassical economics of climate change
      https://www.tandfonline.com/doi/full/10.1080/14747731.2020.1807856
      Here’s the abstract

      Forecasts by economists of the economic damage from climate change have been notably sanguine, compared to warnings by scientists about damage to the biosphere. This is because economists made their own predictions of damages, using three spurious methods: assuming that about 90% of GDP will be unaffected by climate change, because it happens indoors; using the relationship between temperature and GDP today as a proxy for the impact of global warming over time; and using surveys that diluted extreme warnings from scientists with optimistic expectations from economists. Nordhaus has misrepresented the scientific literature to justify the using a smooth function to describe the damage to GDP from climate change. Correcting for these errors makes it feasible that the economic damages from climate change are at least an order of magnitude worse than forecast by economists, and may be so great as to threaten the survival of human civilization

      • I have no opinion about Steve Keen, but yep totally agree otherwise. Andrew Gelman did a good job debunking the apparently state of the art effort by Richard Tol to devise an economic damage function for climate change here:

        https://statmodeling.stat.columbia.edu/2014/05/27/whole-fleet-gremlins-looking-carefully-richard-tols-twice-corrected-paper-economic-effects-climate-change/

        To his credit, in that paper I posted Nordhaus appears to no longer use Tol’s 2013 paper in his model, but that saga does not build much confidence in that whole sub-discipline!

        Ignoring tail risks of climate change arising from our uncertainty plus the complex feedbacks and higher order effects that happen in coupled human-natural systems, and pretending that point estimates like 5% or 8% or whatever are meaningful… it’s all dangerous charlatanism. We don’t have time to fool around with that kind of thing anymore.

        • ” it’s all dangerous charlatanism. We don’t have time to fool around with that kind of thing anymore.”

          Hyping and misidentifying the sources of risk isn’t any less dangerous. While Newsome and Inslee run around hyping climate change as the cause of fires today, just last year Inslee was promoting controlled burns to restore natural forest conditions as the solution to fire risk and even NRP is running pieces about that today.

          It’s more reasonable to assign today’s fire danger to environmentalism than to oil companies. It’s the massive fuel load and overabundance of young trees that heats up the fire and drives it into the crown – that’s aside from the fact that most of these fires are started by people.

        • It’s kind of all-of-the-above, I think.

          Like flooding. Climate change makes hurricanes more likely to have unusually high rainfall, but development patterns mean the landscape can’t absorb as much rainfall as it once could.

          Hurricane Harvey would have been much less likely to produce as much rain as it did in 1877 rather than 2017 – but coastal East Texas in 1877 would *also* have been less vulnerable. (To the rain; the historical really destructive Gulf Coast hurricanes, like the ones that essentially destroyed Indianola [the county seat was actually moved north!] and Galveston, were mostly storm-surge damage.)

          I don’t see how economic effects of climate change can be predicted meaningfully. In 1930 current development patterns wouldn’t have been predictable. And technological advancement, if anything, might mean the economy changes *more* in the next 80 years than the last 80.

        • I think that’s more land subsidence, though. Sea level hasn’t risen very much.

          Of course, locally, sinking land and rising sea levels have the same effect. (And I know at least one neighborhood in the greater Houston area – Brownwood in Baytown – was condemned & abandoned due to subsidence, combined with hurricane damage; it’s the Baytown Nature Center now.)

          But subsidence doesn’t hit everywhere at once.

          Yeah, with modern US (or other developed nation) levels of resources the effects of climate change in any one place should be quite handle-able. (Although the best way to handle it in some cases might be “move out of that place and rebuild the city elsewhere”… but we have the resources for that too.)

          We might very well see a migration of US population from places like Houston and Miami to places like Dallas and Chicago.

          But moving /all/ the Gulf Coast & low-elevation East Coast cities, say, over a span of a few decades would be a much larger issue.

          My larger concern is less-developed nations, though, especially those that have most of their population in low-latitude, low-elevation coastal areas subject to hurricanes/typhoons, other forms of flooding, etc.

        • Some of the above is clearly absurd.

          1. There’s a big difference between a controlled burn and an uncontrolled burn. Controlled burns are accepted policy for lessening fire risk. You may be right in thinking that this policy is currently obsolete in our conditions of drought and high temperatures, but you provide no evidence for this.
          2. A major part of the problem is the decades old policy of the Forest service to quench ANY burns, leading to the growth of very flammable underbrush over the years, as well as ecological disruption. My impression is that sometime back (1980’s?) forest experts advised the Forest service to allow naturally occuring small scale burns. Of course there was pushback from local property owners, but I believe some progress has been made on this issue.
          3. Much of the “massive fuel load and overabundance of young trees” comes from clear cutting by timber companies. Selective cutting, which would tend to avoid those consequences costs a lot more money. How can this be attributed to environmentalism?

  16. “Economics and psychology are just easy to talk about because they are relatively non-technical topics.”

    Obviously economics is known for being over-the-top technical, so this is a surprising claim, to put it mildly. I’ll grant that the more statistical research, as opposed to the formal theory work, is probably not hard for a statistician to get into.

    And then the Soviet economy, I don’t think this is the gotcha quote that you think it is. Obviously the situation was less clear at the time, and you’re looking at a later edition of a textbook mostly written much earlier. And clearly the Soviet economy really did function for decades, albeit without “thriving”, and that contradicted the expectations of many anti-socialists. This is a pretty good example of the kind of clumsy analysis that results if you mistakenly believe economics isn’t technical.

    • Will:

      First, you say “the Soviet economy really did function for decades, albeit without ‘thriving.'” But “thrive” is the word used in that book. The criticism here is of what they actually wrote.

      Second, you say that it was a later edition of a textbook, and that’s right, but that’s the point here, that they didn’t update. Background is here, where Levy and Peart ask, “Why important textbooks of the 1960s and 1970s were so over-confidentabout Soviet economic growth that evidence of model failure was repeatedly blamed on events outside the model’s control?”

  17. Definitions change over time. What functions, what thrives, and what is a socialist command economy. Maybe 31 years in the future, someone may look back to 2020 and say: “Look that brilliant statistician who could not see the thrive of the Chinese socialist economy and the fall of the liberal capitalism, happening just in front of his eyes”.

    • Alex:

      I’m not faulting the authors of that textbook for not seeing something. If they’d just not taken a stance on the Soviet economy, or said this was not their area of expertise, or whatever, that would’ve been fine. The problem was that they made a strong statement which they did not update as new facts came in. And, yeah, if I were to make a strong and authoritative-sounding statement about the Chinese economy, a statement which in the future was clearly wrong, then damn straight people could call me on it—especially if I continued to make that wrong statement year after year without correcting it.

  18. Economic theory uses as a cornerstone of human behavior the fact that people want to to get richer/wealthier. In my opinion, it is a really big criterion that´s often overlooked that the theories that helps society to get wealthier in a more straightforward manner are more valuable than others, regardless of their complexity or simplicity or even regardless of field of study. Economic scientists should stick to society´s goals. Macroeconomic theories that help to increase national income during strong recessions or depressions do that – as an example. Of course there are a lot of works that only will show its income value as they evolve and sprad through society and thererore are tough to measure. But the criterion remains the same.

Leave a Reply to John Williams Cancel reply

Your email address will not be published. Required fields are marked *