Book Review: Red Plenty

I.

I decided to read Red Plenty because my biggest gripe after reading Singer’s book on Marx was that Marx refused to plan how communism would actually work, instead preferring to leave the entire matter for the World-Spirit to sort out. But almost everything that interests me about Communism falls under the category of “how communism would actually work”. Red Plenty, a semi-fictionalized account of the history of socialist economic planning, seemed like a natural follow-up.

But I’d had it on my List Of Things To Read for even longer than that, ever after stumbling across a quote from it on some blog or other:

Marx had drawn a nightmare picture of what happened to human life under capitalism, when everything was produced only in order to be exchanged; when true qualities and uses dropped away, and the human power of making and doing itself became only an object to be traded.

Then the makers and the things made turned alike into commodities, and the motion of society turned into a kind of zombie dance, a grim cavorting whirl in which objects and people blurred together till the objects were half alive and the people were half dead. Stock-market prices acted back upon the world as if they were independent powers, requiring factories to be opened or closed, real human beings to work or rest, hurry or dawdle; and they, having given the transfusion that made the stock prices come alive, felt their flesh go cold and impersonal on them, mere mechanisms for chunking out the man-hours. Living money and dying humans, metal as tender as skin and skin as hard as metal, taking hands, and dancing round, and round, and round, with no way ever of stopping; the quickened and the deadened, whirling on.

And what would be the alternative? The consciously arranged alternative? A dance of another nature. A dance to the music of use, where every step fulfilled some real need, did some tangible good, and no matter how fast the dancers spun, they moved easily, because they moved to a human measure, intelligible to all, chosen by all.

Needless to say, this is Relevant To My Interests, which include among them poetic allegories for coordination problems. And I was not disappointed.

The book begins:

Strange as it may seem, the gray, oppressive USSR was founded on a fairy tale. It was built on the twentieth-century magic called “the planned economy,” which was going to gush forth an abundance of good things that the lands of capitalism could never match. And just for a little while, in the heady years of the late 1950s, the magic seemed to be working. Red Plenty is about that moment in history, and how it came, and how it went away; about the brief era when, under the rash leadership of Khrushchev, the Soviet Union looked forward to a future of rich communists and envious capitalists, when Moscow would out-glitter Manhattan and every Lada would be better engineered than a Porsche. It’s about the scientists who did their genuinely brilliant best to make the dream come true, to give the tyranny its happy ending.

And this was the first interesting thing I learned.

There’s a very settled modern explanation of the conflict between capitalism and communism. Capitalism is good at growing the economy and making countries rich. Communism is good at caring for the poor and promoting equality. So your choice between capitalism and communism is a trade-off between those two things.

But for at least the first fifty years of the Cold War, the Soviets would not have come close to granting you that these are the premises on which the battle must be fought. They were officially quite certain that any day now Communism was going to prove itself better at economic growth, better at making people rich quickly, than capitalism. Even unofficially, most of their leaders and economists were pretty certain of it. And for a little while, even their capitalist enemies secretly worried they were right.

The arguments are easy to understand. Under capitalism, plutocrats use the profits of industry to buy giant yachts for themselves. Under communism, the profits can be reinvested back into the industry to build more factories or to make production more efficient, increasing growth rate.

Under capitalism, everyone is competing with each other, and much of your budget is spent on zero-sum games like advertising and marketing and sales to give you a leg up over your competition. Under communism, there is no need to play these zero-sum games and that part of the budget can be reinvested to grow the industry more quickly.

Under capitalism, everyone is working against everyone else. If Ford discovers a clever new car-manufacturing technique, their first impulse is to patent it so GM can’t use it, and GM’s first impulse is to hire thousands of lawyers to try to thwart that attempt. Under communism, everyone is working together, so if one car-manufacturing collective discovers a new technique they send their blueprints to all the other car-manufacturing collectives in order to help them out. So in capitalism, each companies will possess a few individual advances, but under communism every collective will have every advance, and so be more productive.

These arguments make a lot of sense to me, and they definitely made sense to the Communists of the first half of the 20th century. As a result, they were confident of overtaking capitalism. They realized that they’d started with a handicap – czarist Russia had been dirt poor and almost without an industrial base – and that they’d faced a further handicap in having the Nazis burn half their country during World War II – but they figured as soon as they overcame these handicaps their natural advantages would let them leap ahead of the West in only a couple of decades. The great Russian advances of the 50s – Sputnik, Gagarin, etc – were seen as evidence that this was already starting to come true in certain fields.

And then it all went wrong.

II.

Grant that communism really does have the above advantages over capitalism. What advantage does capitalism have?

The classic answer is that during communism no one wants to work hard. They do as little as they can get away with, then slack off because they don’t reap the rewards of their own labor.

Red Plenty doesn’t really have theses. In fact, it’s not really a non-fiction work at all. It’s a dramatized series of episodes in the lives of Russian workers, politicians, and academics, intended to come together to paint a picture of how the Soviet economy worked.

But if I can impose a thesis upon the text, I don’t think it agreed with this. In certain cases, Russians were very well-incentivized by things like “We will kill you unless you meet the production target”. Later, when the state became less murder-happy, the threat of death faded to threats of demotions, ruined careers, and transfer to backwater provinces. And there were equal incentives, in the form of promotion or transfer to a desirable location such as Moscow, for overperformance. There were even monetary bonuses, although money bought a lot less than it did in capitalist countries and was universally considered inferior to status in terms of purchasing power. Yes, there were Goodhart’s Law type issues going on – if you’re being judged per product, better produce ten million defective products than 9,999,999 excellent products – but that wasn’t the crux of the problem.

Red Plenty presented the problem with the Soviet economy primarily as one of allocation. You could have a perfectly good factory that could be producing lots of useful things if only you had one extra eensy-weensy part, but unless the higher-ups had allocated you that part, you were out of luck. If that part happened to break, getting a new one would depend on how much clout you (and your superiors) pulled versus how much clout other people who wanted parts (and their superiors) held.

The book illustrated this reality with a series of stories (I’m not sure how many of these were true, versus useful dramatizations). In one, a pig farmer in Siberia needed wood in order to build sties for his pigs so they wouldn’t freeze – if they froze, he would fail to meet his production target and his career would be ruined. The government, which mostly dealt with pig farming in more temperate areas, hadn’t accounted for this and so hadn’t allocated him any wood, and he didn’t have enough clout with officials to request some. A factory nearby had extra wood they weren’t using and were going to burn because it was too much trouble to figure out how to get it back to the government for re-allocation. The farmer bought the wood from the factory in an under-the-table deal. He was caught, which usually wouldn’t have been a problem because everybody did this sort of thing and it was kind of the “smoking marijuana while white” of Soviet offenses. But at that particular moment the Party higher-ups in the area wanted to make an example of someone in order to look like they were on top of their game to their higher-ups. The pig farmer was sentenced to years of hard labor.

A tire factory had been assigned a tire-making machine that could make 100,000 tires a year, but the government had gotten confused and assigned them a production quota of 150,000 tires a year. The factory leaders were stuck, because if they tried to correct the government they would look like they were challenging their superiors and get in trouble, but if they failed to meet the impossible quota, they would all get demoted and their careers would come to an end. They learned that the tire-making-machine-making company had recently invented a new model that really could make 150,000 tires a year. In the spirit of Chen Sheng, they decided that since the penalty for missing their quota was something terrible and the penalty for sabotage was also something terrible, they might as well take their chances and destroy their own machinery in the hopes the government sent them the new improved machine as a replacement. To their delight, the government believed their story about an “accident” and allotted them a new tire-making machine. However, the tire-making-machine-making company had decided to cancel production of their new model. You see, the new model, although more powerful, weighed less than the old machine, and the government was measuring their production by kilogram of machine. So it was easier for them to just continue making the old less powerful machine. The tire factory was allocated another machine that could only make 100,000 tires a year and was back in the same quandary they’d started with.

It’s easy to see how all of these problems could have been solved (or would never have come up) in a capitalist economy, with its use of prices set by supply and demand as an allocation mechanism. And it’s easy to see how thoroughly the Soviet economy was sabotaging itself by avoiding such prices.

III.

The “hero” of Red Plenty – although most of the vignettes didn’t involve him directly – was Leonid Kantorovich, a Soviet mathematician who thought he could solve the problem. He invented the technique of linear programming, a method of solving optimization problems perfectly suited to allocating resources throughout an economy. He immediately realized its potential and wrote a nice letter to Stalin politely suggesting his current method of doing economics was wrong and he could do better – this during a time when everyone else in Russia was desperately trying to avoid having Stalin notice them because he tended to kill anyone he noticed. Luckily the letter was intercepted by a kindly mid-level official, who kept it away from Stalin and warehoused Kantorovich in a university somewhere.

During the “Khruschev thaw”, Kantorovich started getting some more politically adept followers, the higher-ups started taking note, and there was a real movement to get his ideas implemented. A few industries were run on Kantorovichian principles as a test case and seemed to do pretty well. There was an inevitable backlash. Opponents accused the linear programmers of being capitalists-in-disguise, which wasn’t helped by their use of something called “shadow prices”. But the combination of their own political adeptness and some high-level support from Khruschev – who alone of all the Soviet leaders seemed to really believe in his own cause and be a pretty okay guy – put them within arm’s reach of getting their plans implemented.

But when elements of linear programming were adopted, they were adopted piecemeal and toothless. The book places the blame on Alexei Kosygen, who implemented a bunch of economic reforms that failed, in a chapter that makes it clear exactly how constrained the Soviet leadership really was. You hear about Stalin, you imagine these guys having total power, but in reality they walked a narrow line, and all these “shadow prices” required more political capital than they were willing to mobilize, even when they thought Kantorovich might have a point.

IV.

In the end, I was left with two contradictory impressions from the book.

First, amazement that the Soviet economy got as far as it did, given how incredibly screwed up it was. You hear about how many stupid things were going on at every level, and you think: This was the country that built Sputnik and Mir? This was the country that almost buried us beneath the tide of history? It is a credit to the Russian people that they were able to build so much as a screwdriver in such conditions, let alone a space station.

But second, a sense of what could have been. What if Stalin hadn’t murdered most of the competent people? What if entire fields of science hadn’t been banned for silly reasons? What if Kantorovich had been able to make the Soviet leadership base its economic planning around linear programming? How might history have turned out differently?

One of the book’s most frequently-hammered-in points was that there was was a brief moment, back during the 1950s, when everything seemed to be going right for Russia. Its year-on-year GDP growth (as estimated by impartial outside observers) was somewhere between 7 to 10%. Starvation was going down. Luxuries were going up. Kantorovich was fixing entire industries with his linear programming methods. Then Khruschev made a serious of crazy loose cannon decisions, he was ousted by Brezhnev, Kantorovich was pushed aside and ignored, the “Khruschev thaw” was reversed and tightened up again, and everything stagnated for the next twenty years.

If Khruschev had stuck around, if Kantorovich had succeeded, might the common knowledge that Communism is terrible at producing material prosperity look a little different?

The book very briefly mentioned a competing theory of resource allocation promoted by Victor Glushkov, a cyberneticist in Ukraine. He thought he could use computers – then a very new technology – to calculate optimal allocation for everyone. He failed to navigate the political seas as adroitly as Kantorovich’s faction, and the killing blow was a paper that pointed out that for him to do everything really correctly would take a hundred million years of computing time.

That was in 1960. If computing power doubles every two years, we’ve undergone about 25 doubling times since then, suggesting that we ought to be able to perform Glushkov’s calculations in three years – or three days, if we give him a lab of three hundred sixty five computers to work with. There could have been this entire field of centralized economic planning. Maybe it would have continued to underperform prices. Or maybe after decades of trial and error across the entire Soviet Union, it could have caught up. We’ll never know. Glushkov and Kantorovich were marginalized and left to play around with toy problems until their deaths in the 80s, and as far as I know their ideas were never developed further in the context of a national planned economy.

V.

One of the ways people like insulting smart people, or rational people, or scientists, is by telling them they’re the type of people who are attracted to Communism. “Oh, you think you can control and understand everything, just like the Communists did.”

And I had always thought this was a pretty awful insult. The people I know who most identify as rationalists, or scientifically/technically minded, are also most likely to be libertarian. So there, case dismissed, everybody go home.

This book was the first time that I, as a person who considers himself rationally/technically minded, realized that I was super attracted to Communism.

Here were people who had a clear view of the problems of human civilization – all the greed, all the waste, all the zero-sum games. Who had the entire population united around a vision of a better future, whose backers could direct the entire state to better serve the goal. All they needed was to solve the engineering challenges, to solve the equations, and there they were, at the golden future. And they were smart enough to be worthy of the problem – Glushkov invented cybernetics, Kantorovich won a Nobel Prize in Economics.

And in the end, they never got the chance. There’s an interpretation of Communism as a refutation of social science, here were these people who probably knew some social science, but did it help them run a state, no it didn’t. But from the little I learned about Soviet history from this book, this seems diametrically wrong. The Soviets had practically no social science. They hated social science. You would think they would at least have some good Marxists, but apparently Stalin killed all of them just in case they might come up with versions of Marxism he didn’t like, and in terms of a vibrant scholarly field it never recovered. Economics was tainted with its association with capitalism from the very beginning, and when it happened at all it was done by non-professionals. Kantorovich was a mathematician by training; Glushkov a computer scientist.

Soviet Communism isn’t what happens when you let nerds run a country, it’s what happens when you kill all the nerds who are experts in country-running, bring in nerds from unrelated fields to replace them, then make nice noises at those nerds in principle while completely ignoring them in practice. Also, you ban all Jews from positions of importance, because fuck you.

Baggy two-piece suits are not the obvious costume for philosopher kings: but that, in theory, was what the apparatchiks who rule the Soviet Union in the 1960s were supposed to be. Lenin’s state made the same bet that Plato had twenty-five centuries earlier, when he proposed that enlightened intelligence gives absolute powers would serve the public good better than the grubby politicking of republics.

On paper, the USSR was a republic, a grand multi-ethnic federation of republics indeed and its constitutions (there were several) guaranteed its citizens all manner of civil rights. But in truth the Soviet system was utterly unsympathetic to the idea of rights, if you meant by them any suggestion that the two hundred million men, women and children who inhabited the Soviet Union should be autonomously fixing on two hundred million separate directions in which to pursue happiness. This was a society with just one programme for happiness, which had been declared to be scientific and therefore was as factual as gravity.

But the Soviet experiment had run into exactly the difficulty that Plato’s admirers encountered, back in the fifth century BC, when they attempted to mould philosophical monarchies for Syracuse and Macedonia. The recipe called for rule by heavily-armed virtue—or in the Leninist case, not exactly virtue, but a sort of intentionally post-ethical counterpart to it, self-righteously brutal. Wisdom was to be set where it could be ruthless. Once such a system existed, though, the qualities required to rise in it had much more to do with ruthlessness than wisdom. Lenin’s core of Bolsheviks, and the socialists like Trotsky who joined them, were many of them highly educated people, literate in multiple European languages, learned in the scholastic traditions of Marxism; and they preserved these attributes even as they murdered and lied and tortured and terrorized. They were social scientists who thought principle required them to behave like gangsters. But their successors – the vydvizhentsy who refilled the CEntral Committee in the thirties – were not the most selfless people in Soviet society, or the most principled, or the most scrupulous. They were the most ambitious, the most domineering, the most manipulative, the most greedy, the most sycophantic: people whose adherence to Bolshevik ideas was inseparable from the power that came with them. Gradually their loyalty to the ideas became more and more instrumental, more and more a matter of what the ideas would let them grip in their two hands…

Stalin had been a gangster who really believed he was a social scientist. Khruschev was a gangster who hoped he was a social scientist. But the moment was drawing irresistibly closer when the idealism would rot away by one more degree, and the Soviet Union would be governed by gangsters who were only pretending to be social scientists.

And in the end it all failed miserably:

The Soviet economy did not move on from coal and steel and cement to plastics and microelectronics and software design, except in a very few military applications. It continued to compete with what capitalism had been doing in the 1930s, not with what it was doing now. It continued to suck resources and human labour in vast quantities into a heavy-industrial sector which had once been intended to exist as a springboard for something else, but which by now had become its own justification. Soviet industry in its last decades existed because it existed, an empire of inertia expanding ever more slowly, yet attaining the wretched distinction of absorbing more of the total effort of the economy that hosted it than heavy industry has ever done anywhere else in human history, before or since. Every year it produced goods that less and less corresponded to human needs, and whatever it once started producing, it tended to go on producing ad infinitum, since it possessed no effective stop signals except ruthless commands from above, and the people at the top no longer did ruthless, in the economic sphere. The control system for industry grew more and more erratic, the information flowing back to the planners grew more and more corrupt. And the activity of industry , all that human time and machine time it used up, added less and less value to the raw materials it sucked in. Maybe no value. Maybe less than none. One economist has argued that, by the end, it was actively destroying value; it had become a system for spoiling perfectly good materials by turning them into objects no one wanted.

I don’t know if this paragraph was intentionally written to contrast with the paragraph at the top, the one about the zombie dance of capitalism. But it is certainly instructive to make such a contrast. The Soviets had originally been inspired by this fear of economics going out of control, abandoning the human beings whose lives it was supposed to improve. In capitalist countries, people existed for the sake of the economy, but under Soviet communism, the economy was going to exist only for the sake of the people.

(accidental Russian reversal: the best kind of Russian reversal!)

And instead, they ended up taking “people existing for the sake of the economy” to entirely new and tragic extremes, people being sent to the gulags or killed because they didn’t meet the targets for some product nobody wanted that was listed on a Five-Year Plan. Spoiling good raw materials for the sake of being able to tell Party bosses and the world “Look at us! We are doing Industry!” Moloch had done some weird judo move on the Soviets’ attempt to destroy him, and he had ended up stronger than ever.

The book’s greatest flaw is that it never did get into the details of the math – or even more than a few-sentence summary of the math – and so I was left confused as to whether anything else had been possible, whether Kantorovich and Glushkov really could have saved the vision of prosperity if they’d been allowed to do so. Nevertheless, the Soviets earned my sympathy and respect in a way Marx so far has not, merely by acknowledging that the problem existed and through the existence of a few good people who tried their best to solve it.

This entry was posted in Uncategorized and tagged , , , , . Bookmark the permalink.

462 Responses to Book Review: Red Plenty

  1. RJMeyers says:

    Power and Prosperity by Mancur Olson contains some additional insight here, though mostly about the breakdown of information flows and the absolutely amazingly devious taxation scheme that the Stalinists came up with to keep people working as hard as possible. It’s a short read and covers some other ground as well.

  2. Pingback: In Praise of Maximizing | wallowinmaya

  3. monolith94 says:

    This book review/analysis is marvelous, wonderful, thought-provoking, great.

  4. Pingback: Thots | RWCG

  5. Pingback: deeply valuable experiences, meditation zombies, and planning preview (2700 words) | Meditation Stuff (@meditationstuff)

  6. Illuminati Initiate says:

    I can’t form a confident view on the desirability of a full planned economy because any discussion of complex economics ends up with me having the suspicion that I might be getting Eulered by both sides.

    I think I am leaning towards a mixed economy with a very extensive welfare state because it seems like it could solve most of the problems a full planned economy aims to solve anyways. Note that the kind of mixed economy I am thinking of would be much more “socialist” than any that exist today.

    • Samuel Skinner says:

      Full planned hasn’t exactly been implemented too often- most communist states accepted some level of minor independent artisans and the like.

      “I think I am leaning towards a mixed economy with a very extensive welfare state because it seems like it could solve most of the problems a full planned economy aims to solve anyways. Note that the kind of mixed economy I am thinking of would be much more “socialist” than any that exist today.”

      It doesn’t solve the problem socialism is meant to solve. Having the government run firms generally doesn’t function too well (there are some cases that do better than average, but those involve being specifically insulated from government control; they were optimized for growth, not social benefit).

      Welfare states only work when the money is coming in and the population is homogeneous. You need the first to get them to exist and the latter to insure political support- if people feel “us” is supporting “them” they will be less willing to pay for it.

      If you want to see the limits of socialist style planning, just look at internal corporate systems- it shows you how far you can efficiently go without price mechanics.

    • cassander says:

      I mean this as a serious question, what makes you prefer that to a near total laissez faire economy with a big minimum income? that seems to me to solve any theoretical poverty problem without introducing any of the problems of central planning. seems to me to be the best of both worlds.

  7. JB says:

    A phrase I keep an eye out for in publications is “aging soviet-era ___”. For example, “Russia is burdened by the failure of aging, soviet-era infrastructure”. Or “crashes blamed on use of aging soviet-era aircraft”. It’s surprising how many times that phrase arises as a trope.

    The initial impression I always get from reading that is: Hmm, the stuff they built in the soviet era is sure falling apart.

    The impression I should be getting is: Hmm, they built all that stuff in the Soviet era, and they haven’t built any replacements since then.

    • Cauê says:

      As long as you also make note of all the things they’ve been producing and didn’t use to…

    • cassander says:

      they haven’t built so many impressive projects, sure, but they are exporting grain (without starving those who grow it) for the first time in about 80 years. I’d say that’s a much better metric for the success of a society.

  8. Johannes says:

    If one wanted to compare developments there are certainly better options than the Soviet Union (or China), because these had so different starting conditions in 1919 or 1945 than the west. (And these countries are HUGE, so there are bound to be effects of distance, more difficult control etc.)
    The best example would be Germany, I guess. Here you have ethnic homogeneity and a very similar education/skill/work ethic level before the 1940ties. The different initial conditions are among others the support by the US (Marshall plan) in the West and some heavy reparations to pay to the Soviets in the East. The East also lacked important resources (the heavy industry/coal mines were concentrated in the West).
    I do not know the numbers (and they are contested!), but some claim that overall Eastern Germany had comparably growth rates to Western Germany. In any case the standard of living was the highest of the Eastern bloc and in the 60s/70s higher than Spain or Ireland. They produced some (almost) high tech in optics, photography etc. and also textiles and furniture for the western markets. Also very nice classical records 😉

    • Multiheaded says:

      On a somewhat unrelated note, they also had considerably less LGBT persecution than Western nations – possibly influenced by the KPD’s Weimar legacy?

      • Anonymous says:

        What leads you to this belief? Simple google searches come up with lists of oppression, but rarely compare them to other times and places. this link claims that the two Germanies were comparable. It also compares to other countries, but isn’t very detailed (and mainly focused on legal rights, rather than persecution), but the conclusion is that it was not out of place given its place on the north-south axis.

        Also, the link emphasizes people before Weimar.

    • Tracy W says:

      And yet, East Germans kept trying to escape to West Germany, so much so the Communists built a wall.

      Vietnam after the Americans left – unified by the Communists, and what happened? Vietnamese boat people risked (and often lost) their lives fleeing for the next decade.

      North Korea and South Korea is another comparison.

      • somnicule says:

        Devil’s advocacy: Those who will attempt to cross borders illegally like that are high risk-takers. The preference they have for capitalist nations is the high variance in outcomes, not an overall higher quality of life.

        • Tracy W says:

          Problem with your theory: The East Germans built the Berlin War because people were just walking across into West Berlin. As illegal crossing attempts go, that’s a very safe way of doing it. Particularly into a bit of land that was only split off from your country a few years ago.

          As for North and South Korea, the difference in quality of life is visible from space, and not to the North’s favour.

      • JB says:

        North Korea and South Korea is another comparison. But you can’t just compare the two countries as they are today to get a picture. How did they get to be as they are? It’s an interesting history, and not quite as clean a story as democracy and capitalism leading to Samsung and Gangnam Style, while a communist monarchy leads to a nation-sized military prison.

        When Park Chung-hee seized power in South Korea, North Korea’s economy was humming smoothly, guided by technical expertise and trade from other Soviet-bloc countries, and was dominating the impoverished South. Park declared martial law, established a repressive and authoritarian military dictatorship in South Korea, and through a series of five-year plans bent the economy towards industrial production, government-directed heavy industry, and exports. South Korea’s mega-companies such as Hyundai, Samsung, and LG were fostered as government-protected and heavily subsidized industries. South Korea’s economy began to grow rapidly during this time.

        South Korea’s government grew more and more authoritarian as Park continued to centralize power for nearly 20 years, until he was assassinated by his security chief. Afterward, the country transitioned to democracy. Its current competitive economic status is widely attributed to Park’s economic restructuring.

        Meanwhile, during the same period, the Kim regime in the North was directing an increasing amount of the nation’s economic output toward feeding and its own excesses, rather than national development. Its once-dominant economy gradually began to falter as the funds were directed to the Kim family personality cult, but was supported by a steady flow of aid and economic support from the Soviet Union, which needed to maintain it as an ally. When the Soviet Union fell, it lost its main trading partner.

        And now we have the situation of Korea today. To my mind, it doesn’t really give a resounding endorsement of any ideology, although it is a fascinating case study.

        • Tracy W says:

          The South Korean growth policy was built on exports, and on the government supporting successful exporters: ie those companies that were showing they could compete on the world market. South Korea got the information that way.

          The policy of government support to successful exporters meant that the government was getting money from everyone else and directing it to said exporting companies. Which is quite different from the normal social democratic tendency to tax the wealth producers and redistribute the income to the poorest. Park Chung-hee reversed Adam Smith’s advice that the the government should only attend to the interests of producers in as much as that contributed to the interests of consumers.

          The replacement democratic government moved to a more natural approach, and individuals’ economic prosperity took off.

          As for North Korea, your summary seems to be “it did fine when the Soviets were chucking money at it.” You do point out another problem with a centrally-planned economy: that the central planners might easily decide to start diverting more and more resources to themselves. (This is not unique to centrally-planned economies of course, there are plenty of monarchs and other dictators who did similar things).

          Obviously any one example doesn’t tell us much as there are so many different things going on in any economy. But there’s a shortage of examples of communist countries going great, and overall that is telling.

          • JB says:

            Obviously I’m very glad I don’t live in North Korea (although maybe in 1970 it would have been a harder choice).

            Chucking money at your allies was a standard strategy during the cold war on both sides of the iron curtain, but of course it required a wealthy enough superpower to afford such foreign aid.

            My main point is that South Korea is far from a textbook case of a capitalist democracy, which people unaware of its history fail to recognize when they look at its success and freedoms today. And that North Korea’s growth was pretty much on par with it until the Kim regime decided to turn the country into their personal playground. Which is absolutely a fair point against a centrally-planned economy, but also a common failure seen in other totalitarian states which don’t have communist ideology, as you mentioned, and South Korea could easily have gone the same way if Park had been so inclined.

            If it had, I am sure that America would still have propped it up as an ally to counterbalance the Soviet and Chinese presence in East Asia, just as it did with many other totalitarian regimes throughout the cold war.

  9. AlphaCeph says:

    Just a mathematical comment: linear programming seems like a stupid way to try to plan an economy. It would seem that building a simulation and using monte-carlo methods would be more appropriate. I mean to start with it would be trivial to parallelize. Also easier to sanity check the model by running it against known history. And you could add into the model an epsilon amount of people deliberately disobeying the plan and see how robust it was to them.

    Can someone who knows more about the tradeoff between these two please comment?

    • Tracy W says:

      How do you build a sensible simulation to use the monte-carlo methods on? A monte-carlo simulation only produces a useful range of output if its simulation method has something to do with the underlying country.

      You have to take account of all sorts of constraints like:
      – You can’t use more of something than you have of it. Eg if your plan calls for production of 100 new wheelbarrows in year 1, and for the use of 200 new wheelbarrows in year 1, you’re stuffed).
      – It takes time to move things around. So you need to build in transport times.
      – Some things can’t be moved around, or can only be moved to particular locations.
      – Some things are inputs into further production. Eg wheelbarrows used in horticulture.
      – Some things are inputs into their own production (eg steel smelters use steel in their operations)
      – the ability to substitute alternative things in a system of production differs depending on what you’re producing and how. And where, as in the case of the pig farmer Scott writes about.

      This sorts of problems are why people are talking about linear programming.

      • AlphaCeph says:

        I think monte carlo is better than LP because in both cases you have to do all the work of inputting the constraints, but in the LP method you also have to solve the resulting mathematical problem, which may never happen.

        In the monte carlo family of methods, once you have the constraints inputted you randomly choose production decisions and you have an almost instant solution. Of course it’s likely to be laughably bad, but that’s where heuristics and clever techniques come in: you bias your random choices according to heuristics, and you use techniques to try and make subsequent guesses converge on the correct solution.

        • Tracy W says:

          You’re missing the point. Monte carlo what?

          Say I model the economy in a monte carlo simulation by tossing two dice and summing the total. What sort of heuristics and clever techniques are going to turn that dice-tossing into a useful model for an entire economy?

          Monte carlo isn’t magic pixie dust.

          • AlphaCeph says:

            You have to build a model of the constraints on and effects of economic allocation decisions, and then roll dice to make those decisions. This leads to an allocation, which is scored according to a consumption function. The “constraints on and effects” part is hard, i.e. you need a model specifying what can be made, used, produced etc, but those same constraints are required for the LP method. Comprendez?

  10. RCF says:

    How much of the Soviet space program was due to the achievements of the USSR, and how much was it due to them having just conquered a country full of rocket scientists? Shouldn’t the fact that their progress stalled once they didn’t have scientifically advanced countries to loot raise suspicions?

  11. Richard says:

    So what is a current example of a large number of people, all working on the same team while doing radically different things and being centrally managed?

    Sounds to me like the US Army/Navy/AirForce would fit.

    They should face the same problems as the communists did with regards to resource optimisation, and frankly from when I was in a different NATO army, the pig farmer and tire factory stories seem incredibly familiar.

    Perhaps linear programming and whatnot could first be applied to avoiding $640 toilet seats to see if it works as intended.

    • cassander says:

      Western militaries are able to buy a lot of goods from the civilian economy, which reduces the problem. However, when you look at the areas more or less completely separate from the civilian economy, e.g. logistical systems for units in heavy combat or certain types of weapons procurement, you absolutely see exactly the same issues.

  12. http://vserver1.cscs.lsa.umich.edu/~crshalizi/weblog/918.html (recommended reading!) influenced my “too computationally intensive” view.

    • AlphaCeph says:

      The idea that running an economy requires 10^70 FLOPs just doesn’t pass the “smell test”. Market economies are run by humans making < 10,000 or so relevant decisions per person per year. Multiply by the number of people and you get about 10^13 operations. Any computational complexity approach that goes above 10^13 – especially by many orders of magnitude – is therefore wrong.

      I suspect the mistake is that planning an economy is NOT a generic optimization problem, rather that there is special structure that can be exploited. For example, locations are not just arbitrary tags to add to your goods – they form a 2-dimensional space so simplifications are possible. Consumer goods are the same: they have a similarity structure so that 10000000 different goods cluster into 100 or so categories that are partially fungible: clothing, transport, entertainment, food, medical care, etc etc.

      • Alex Godofsky says:

        Market economies are run by humans making < 10,000 or so relevant decisions per person per year.

        Humans are not globally optimizing, they are a network of local pseudo-optimizers.

        I suspect the mistake is that planning an economy is NOT a generic optimization problem, rather that there is special structure that can be exploited.

        Shalizi addresses this!

        • AlphaCeph says:

          “Humans are not globally optimizing, they are a network of local pseudo-optimizers”

          Any central planning system could make use of similar work-arounds! If a central planning system could match the performance of the market by having local pseudo-optimizers, that would be impressive.

          And by the way, I doubt that the move from local pseudo-optimizers to global approximate optimization eats *57 ORDERS OF MAGNITUDE* of computational power.

          And even if it does, there is probably a sliding scale of optimality versus computer power. Market economies are probably not better than the best possible computer approximation with present day supercomputers, simply because they don’t have as many FLOPS available, even if the best algorithm is just to simulate a market economy, which I doubt.

          “Shalizi addresses this!” – not to my satisfaction. A lot of what he says is invalidated by my point that human beings in market economies represent a small amount of computer power and yet still solve the problem.

          • Jaskologist says:

            Worrying about computing power is putting the cart before the horse. We don’t even have an algorithm for planning the economy, a way to quantify the many, many inputs involved in production, a way to quantify even the most basic input of “value,” a way to gather (and constantly regather) those inputs even if we knew what they were, and we most certainly don’t have an algorithm for “innovation.”

          • AlphaCeph says:

            @Jaskologist: quantifying inputs to production is already done today, each company devotes management time to identify what resources it needs. I don’t see why this couldn’t be done in a centrally planned system, assuming that you can solve the incentive problems.

            I think the computational and informational problems of techno-communism are solvable.

            The incentive problems seem a lot harder.

          • Alex Godofsky says:

            Any central planning system could make use of similar work-arounds! If a central planning system could match the performance of the market by having local pseudo-optimizers, that would be impressive.

            Except the entire discussion centers around a particular family of algorithms that are resistent to this approach, and Shalizi addresses why he doesn’t believe they would be easily amenable to this sort of simplification.

            And for the record, he’s actually an expert qualified to talk about this stuff. He was one of my university professors.

            And by the way, I doubt that the move from local pseudo-optimizers to global approximate optimization eats *57 ORDERS OF MAGNITUDE* of computational power.

            That’s your loss. The math on this is long-established. We know the runtime of the simplex algorithm.

            And even if it does, there is probably a sliding scale of optimality versus computer power.

            OMG just read the freaking Shalizi post! Just read it! He talks about how approximation approaches don’t gain you enough performance improvements! The math on this is not up for debate!

            A lot of what he says is invalidated by my point that human beings in market economies represent a small amount of computer power and yet still solve the problem.

            Human beings don’t implement the same algorithm, at all.

            If your contention is merely that IF we could write a computer program to accurately simulate the economic behavior of 7 billion people, THEN it would probably be runnable on a sufficient aggregation of modern computers, then fine. I agree with you. But no one has any freaking idea how to do that! You are basically asking us to be Hari Seldon. Guess what, psychohistory isn’t real.

          • AlphaCeph says:

            @Alex Godofsky:

            I see what our disagreement is: I’m assuming that one is allowed the freedom to choose any algorithm one wishes to solve the problem.

            Yes, the runtime of the simplex algorithm is not up for debate. But if the simplex algorithm takes 10^70 ops to run the economy, that just means it’s a bad algorithm because either

            (a) it isn’t taking advantage of the structure of the problem or

            (b) it isn’t making enough approximations

            I know this because the actual economy is run based up 10^13 ops – probably far fewer in fact.

            This is in perfect agreement with your statement that human beings don’t run the same algorithm. Well I agree! That’s because linear programming is simply the wrong algorithm to run. Now you might say that’s me moving the goalposts since the discussion centred around linear programming, for which I apologise.

            But the question I want to answer is whether efficient central planning is possible at all, not whether it is possible *using linear programming*.

          • Alex Godofsky says:

            But the question I want to answer is whether efficient central planning is possible at all, not whether it is possible *using linear programming*.

            Yes, but the link you were responding to, and also the subject of Scott’s original post, are a very different context. Upon the discovery of linear programming, some people thought “oh hey, THIS is the thing that will finally let us centrally plan the entire economy!” Which goal had been previously unattainable because people hadn’t yet invented good general-purpose algorithms for resource allocation.

            Both Scott and Shalizi discuss the hypothesis that, had Kantorovich’s method actually been applied at the scale of the entire economy, Soviet economic outcomes would have been different. Shalizi rejects this hypothesis on the grounds that it would have been, and still is, computationally infeasible to actually model the entire economy using a linear program, even with the algorithmic advances we have made in the past seven decades.

            Everything about this post and everthing about that link are within the context of the simplex algorithm. If you want to suggest that there exists in principle some other algorithm that would do better, that’s fine. You might be right. But unless you are able to actually identify that algorithm and prove that it works, your existence proof is not actually useful or informative.

          • Sniffnoy says:

            You are basically asking us to be Hari Seldon. Guess what, psychohistory isn’t real.

            I dunno, I keep hearing good things about cliodynamics… 🙂

          • AlphaCeph says:

            @Alex Godofsky:

            Well I will certainly agree with you that linear programming seems like a bad idea.

            Given the complexity of an economy, new techniques would probably be required. I would expect monte carlo techniques would outperform exact methods like LP. Make a model of production facilities, the linkages between them, consumers, etc. The modelling effort required would be vast, however in a market economy it all gets done. Once you have a model, use monte carlo techniques to try to optimise the allocation decisions. They would have to be smart monte-carlo techniques though, you would have to use many heurisrics.

          • Alex Godofsky says:

            Monte Carlo isn’t just a magic wand you can wave at stuff, and “make a model of production facilities, the linkages between them, consumers, etc.” can’t even be handwaved with “the modelling effort required would be vast”. That doesn’t even begin to describe it. No one aggregates a single combined economic model. “The market” does so by storing the information in a decentralized fashion close to where it originally resides – the individuals making economic choices. You somehow need to convince every single person to reveal that information to the central planner.

            Hell, I have difficulty planning my own workflow at the office! I can’t begin to imagine how the Big Program would model my economic output as anything other than a question mark. Especially since my job is performing part of the calculation problem.

          • AlphaCeph says:

            @Alex:

            I agree that it would be a massive effort – again compare that to the current market econony and the trillions of dollars we spend on finance, management, HR, etc etc etc.

            You couldn’t expect to fully succeed for less than $100 billion worth of labour.

          • Alex Godofsky says:

            What are you trying to prove at this point? What is your goal? “Oh, capitalist economies spend a lot of resources coordinating complex endeavors involving millions of people. Therefore it’s also possible to program a computer to do all of the same calculations they are doing, since after all they are doing the calculations which shows that it is possible!”

            There appears to be no remaining useful content to your argument. You do not seem to have any particular reason that your program would be easier to accomplish than the current program. You do not see to have any particular reason to believe it would actually produce a better plan than the current program.

            “X has this flaw!” does not get you anywhere close to “therefore Y, which is superior to X, not only exists but is actually possible to build”.

          • AlphaCeph says:

            @Alex Godofsky: The takeaway is that you probably could run an economy using math instead of greed.

            Clearly this is an important conclusion.

            Many people believe this to be false, and indeed it is widely assumed that market economies are the only way to do things.

            I’ll admit that I can only guess whether the result would be a better run economy; I think it’s likely that it wouldn’t be much better than the status quo, but that is total guesswork.

            (“There appears to be no remaining useful content to your argument.”)

      • memeticengineer says:

        Market economies are run by humans making < 10,000 or so relevant decisions per person per year.

        How are you measuring decisions? Many “decisions” about what to buy involve a lot more than 1 FLOP. Example: my decision what to have for lunch every day. There’s 600000 restaurants in America. I have to choose one of those ~2^19 places every day. Any algorithm to make that choice other than hardcoding a single value requires significantly more than 1 FLOP, and a lot of information about me in general and about my specific circumstances on any given day. Note: it’s true that on any given day, most of those restaurants are out of practical range, so you can prune the set of choices, but it’s a different subset on different days, so you can’t just hardcode the subset. Probably you could make these decisions for a year by fully simulating a year’s worth of my brain, but a conservative estimate for that would be 10^29 FLOPs, a lot more than your 10^5 per person per year estimate. Simulating an economy’s worth of brains might well be cheaper than 10^70 FLOPs (not obvious because their interactions likely scale super-linearly and you’d probably have to simulate some aspects of the physical environment as well). But it also has no obvious advantage over letting the real brains make their own deicisions.

      • Tracy W says:

        A couple of problems with the comparison:
        -Firstly, that implies 27 decisions a day, or a bit less than 2 per hour, assuming 8 hours of sleep. That seems low, even for people working boring service jobs. A grocery shopping trip could easily blow through a week’s “allowance” of decisions.

        – Those decisions are subject to various constraints (eg money budgets, travelling times, idiosyncrasies like people having deathly allergic reactions to what is otherwise perfectly nutritious food). So one needs various data to make those decisions. Much more than a flop.

        • AlphaCeph says:

          But a lot of them are just repeats: buy toothpaste is a repetitive task that you rarely reassess. And they are also repeated across many people.

          I think the numbers I gave are vast overestimates, especially taking into account the across-person repetition. There are 7 billion people on earth, but we probably fall into a few hundred archetypal person types.

  13. Nornagest says:

    I’m pretty sad that the comment thread for this beautifully even-handed view of a state and ideology that’s usually relegated to whipping-boy status has largely devolved into content-free “hooray for my side”.

    • Multiheaded says:

      I have a fifty-comment plan to move from this Nash equilibrium to a Pareto optimum, but some traitors and Moloch sympathizers here among us wouldn’t let us progress! Please give me mod powers!

      (Also, reference class tennis and poorly sourced numbers are technically content.)

    • lambdaphage says:

      FWIW, I thought this thread had several instances of conspicuously high quality. Although I tend to doubt that the average SSC reader sympathizes much with the economic policies of the Soviet Union, I thought I sensed attempts to at least try to “feel the pull of the argument”.

      It’s also forced me to think a bit more carefully about why I think my team’s parade of horribles was historically contingent, whereas the other team’s is inseparable from their ideas themselves. Which is something I’ve been meaning to think about since I found about this.

      But then again, maybe I’m part of the reason we can’t have nice things, so of course everything looks ok to me.

    • AlphaCeph says:

      For the record, I have long suspected that technology – such as better computers – could change the balance of arguments in relation to centrally planned versus market economies, and I’m glad that SSC has mentioned it.

  14. Shmi Nux says:

    Central planning works pretty decently when it is lubricated by some limited free market (China being the most successful example, but there was some of that in the Soviet Union, as well). Incidentally

    > the killing blow was a paper that pointed out that for him to do everything really correctly would take a hundred million years of computing time.

    is irrelevant if one replaces optimizing with satisficing, and trusts the market to iron out the kinks. Unfortunately (or fortunately, depending on one’s perspective), Gorbachev was no Deng Xiaoping and in his ineptitude started fixing things from the wrong end, slowly giving people more political freedoms instead of economic ones. He probably could not have destroyed the USSR faster and more decisively if he tried.

    • cassander says:

      >China being the most successful example, but there was some of that in the Soviet Union, as well

      I wouldn’t say that central planning is working well in china. China has two economies, the emerging capitalist one in the SEZs and costal regions, and the old socialist one out in the provinces. The old socialist one is as inefficient as ever, what has changed is that the capitalist one is growing fast enough and producing enough money that they can afford to let it keep fucking up as long as it buys social peace and patronage for politicians.

      > Gorbachev was no Deng Xiaoping and in his ineptitude started fixing things from the wrong end, slowly giving people more political freedoms instead of economic ones.

      I agree with the wrong end comment, but not how you mean it. Gorbachev was, by most accounts, a true believer, one of the generation that came up during the kruschev thaw and really thought he could make the USSR work. because he was at the top, and believed in the system, he tried to reform it from the top, which meant upsetting a lot powerful people’s applecarts. Deng took a very different approach. with the exception of agriculture, he didn’t directly attack the problems of the old system, but carved out little areas where he could build new systems, always labling them experiments. when they worked, they grew and, after a couple decades, were bigger than the old system ever had been. top up vs. bottom down, not political vs. economic.

    • Tracy W says:

      Central planning works pretty decently when it is lubricated by some limited free market (China being the most successful example, but there was some of that in the Soviet Union, as well).

      The Soviet Union collapsed.

      • social justice warlock says:

        In that case it was the market “lubrication” that played the critical role, not the central planning. The Soviet leadership in the 80s was dominated by liberals who kept thinking that the proper response to the economy getting more dysfunctional was to liberalize it further. Eventually the whole thing collapsed.

        They themselves did pretty well, though, so I suppose it’s up in the air whether Sturgeon’s law applies.

        • Tracy W says:

          If it was mixing in some market lubrication that was the problem then the Soviet Union would have collapsed under the New Economic Policy in the early 1920s.

    • social justice warlock says:

      I’d say the primary difference was that China had (has) a large surplus of rural labor to move into the cities, whereas the Soviet Union did not. This isn’t an entirely separate issue from economic liberalization (China between the Sino-Soviet split and Deng had an anti-urban bias that was incredibly destructive, and markets were in most important respects a step up from that) of course.

      • Douglas Knight says:

        What is the relevance of the rural to urban migration? China’s economy is growing much faster than its urban population. The current plan is to move 2.5% of the population from rural to urban each year for the next ten years, until they run out of peasants. The average over the last 30 years is 0.8%. So at most 1/3 of China’s growth rate is due to the peasants catching up to the urban population, and the rest is due to the urban population becoming more productive.

        • social justice warlock says:

          There are increasing marginal returns to rural-urban migration, which is why there are returns at all. (Adam Smith discusses this eloquently, as he does most things.) China had comparable rates of growth during the Soviet-style, prourban first 5YP, and while there were other similarities between that and today (probably most importantly technology importation) my reading would primarily credit factor mobilization. IANAEH, though.

          • Douglas Knight says:

            “increasing marginal returns to rural-urban migration” sounds pretty weird to me. Perhaps increasing marginal returns to urban size or density? But China maxes out its cities at 30 million, so I think the current progression is mainly building new cities.

  15. Grant Babcock says:

    In addition to Hayek’s “The Use of Knowledge in Society,” which has already been mentioned you should also probably look at Ludwig von Mises’s “Economic Calculation in the Socialist Commonwealth.”

    The key points are:

    Hayek’s point is that market prices capture local, often inextricably contextual, knowledge held by individual actors that cannot be centrally collated. Not just “it would take too long,” hard can’t. Part of this is because a lot of what you “know” behaviorally isn’t stuff you can articulate verbally. You’re embedded in a set of traditions that constitute the tools you have to navigate life, but which can’t be explained by a rational theory the way physics uses simple rules to explain complex phenomena.

    This means that the “data” are never “given.”

    Second, and this is Mises’ point, even if the “data” *were* given, you would have no basis on which to choose one production setup over another, and it’s not a matter of needing a bigger computer, it’s flat impossible. The chosen production methods can reflect only the arbitrary preferences of the planners. Real life communist countries could sort of dodge this by copying what capitalist economies were doing, but in the absence of that possibility (say communism conquers the world AND eradicates the black market) you’re literally stumbling in the dark.

    TL;DR “real” communism isn’t possible because markets in capital goods solve a problem communism simply can’t.

    • suntzuanime says:

      To be fair, even in a capitalist economy, the chosen production methods can reflect only the arbitrary preferences of the people with money. This is just generalized nihilism brought into an economic context. While it is horrifying, it’s not a basis for choosing one economic system over another.

      • blacktrance says:

        The difference is, those with money are choosing for themselves, while the planners are choosing for everybody. In a free market, if you spend your money unwisely, you suffer the consequence of not having your preferences satisfied, but if you make a bad plan, it’s mostly other people who’ll suffer as a result.

        • suntzuanime says:

          But then we’re back to incentive problems rather than “flat impossible”.

          • blacktrance says:

            It’s not just a matter of incentives, but a matter of knowledge as well, because when you’re choosing for others, you don’t know as much about their preferences. Strictly speaking, it’s not literally impossible, because you could coincidentally create a plan that satisfies people’s preferences as if they had chosen for themselves, but that would be a huge coincidence. Not only does the planner not have the incentive to come up with a good plan, he can’t create an optimal plan even if he wants to, because he doesn’t know what it looks like.

  16. cpopell says:

    I am immensely disquieted at you claiming Glushkov invented cybernetics. He may have re-invented it, or invented it in parallel, but cybernetics was definitively invented by Norbert Wiener (http://en.wikipedia.org/wiki/Norbert_Wiener), though many of the concepts existed elsewhere.

    In addition, I implore you to look into the work of Stafford Beer in trying to apply Cybernetics to Chile, and what advances in control systems have been made since.

  17. oneforward says:

    Scott, if you’re willing to share I’m very curious what else is on your List Of Things To Read.

  18. Joe says:

    I remember being disgusted by communism until I read “Witness” by Whittaker Chambers. There is a very moving scene were the prolitariate help Whittaker get a job building the New York subway. It seems many became communist out of love for the poor. Also monastic life has been called Holy Communism. Makes you wonder what communism could have been if it wasn’t founded on atheism.

    • Mary says:

      One notes that monasticism has traditionally been the practice of the celibate. New members are inducted from the general population, and therefore you don’t have the problem of children departing from the way they were brought up in.

      • Joe says:

        Right I think the monastic analogy is really sort of poetic. It just would have been interesting to see a revolution like the communist one without the brutal mind killing atheism.

    • Harald K says:

      Marx was a Christian before he became an atheist (he was Jewish by ethnicity, but his parents were converts to Christianity). One of the things that turned him off Christianity was British clergy justifying poverty by basically saying the poor are blessed, and we’re doing them a service by keeping all this corrupting money out of their hands. At risk to our souls. Aren’t we making a noble sacrifice?

      He had an old testament prophet’s fury at the economic injustices of his day, and pious hypocrisies like that one. However, he couldn’t justify his anger like a prophet would, by pointing to God’s purpose.

      Instead, he developed this approach: Figure out what the world will look like in the end. That must be the world’s purpose. So he came up with his historicism and declared that in the future, the poor would win (basically). Thus, the poor were the good guys – or their post-ethical counterpart, to borrow Spufford’s expression. Thus, moral outrage against exploitiation of the poor justified.

      • Gilbert says:

        Nope. Marx’ parents converted for entirely careerist reasons but didn’t actually believe it. I don’t know of any sign of Karl ever believing it either.

        At the time he starts commenting on religion he’s already involved with a group of atheists playing status games by accusing other atheists of still having too much Christian memes. (Though that word, of course, wasn’t invented yet.) Then he takes it one more level meta by accusing the accusers of the same thing. (You may have notice this game never quite died out and now is continued, for example, by Moldbuggers. )

        And by the time he moved to Britain his world-view was fully formed and he had already spent years agitating for it.

        • Multiheaded says:

          This comment and the parent… don’t seem incompartible?

        • Harald K says:

          It’s very hard to say what a person really believes, deep down. But there has been preserved an essay the sixteen-year old Marx wrote “The union of believers with Christ according to John 15: 1-14, showing its basis and essence, its absolute necessity, and its effects.” which seems sincere enough.

          • Gilbert says:

            I actually didn’t know that and it clearly does point in that direction.

            That said, there are circumstances making this a lot less decisive than it looks:
            – This is his Abitur essay in religion. In other words, he needed a passing grade from a very religious guy, perhaps even a pastor, in order to be allowed to go on to university.
            – In that situation the title probably was an assignment, not his own choice of topic.
            – The essay is very high on rhetoric and very low on actual content.

            Basically an atheist in that time and place would still have been very likely to produce an essay like this one. So I’ll upgrade to say there is very week evidence rather than none.

  19. The linear-programming approach to allocation wouldn’t have been computationally feasible *at all* then and still isn’t today for provably optimal allocation (but probably we could get by with approximation algorithms that didn’t exist then, on computing power that didn’t exist). Also, think back to algebra word problems and formalizing “let x = how many soccer balls Suzy has …” – hard to capture an economy that way, right? A model that had continuous re-planning and knew about time lags for switching production might let you live through an unexpected design or material quality flaw, a natural disaster or a war …

  20. Mary says:

    “You hear about Stalin, you imagine these guys having total power, but in reality they walked a narrow line,”

    How true.

    Note that their incentives are rather blunt tools, which makes it hard to modulate them.

    Reminds me of a discussion once about Superman’s taking over the world if so inclined. My position was that since he had no mind control powers, his ability to do so would be strictly limited, because you can not actually makes someone do something, only punish him for failure.

    • Samuel Skinner says:

      There was an article about what superman should do with his powers; I believe the best bet was raising satellites into orbit and using the money earned to get people to do what he wanted them to.

      • Illuminati Initiate says:

        I’m not familiar with Superman comics, how strong is he exactly? If he can lift very heavy stuff into space that’s actually pretty important, large spacecraft lifted directly into orbit without the need for liftoff rockets and fuel, he’d basically be a living space elevator…

        • Protagoras says:

          His strength is wildly inconsistent depending on time period, writer, the demands of the current plotline, etc. In the most extreme depictions, his strength is quite absurd, easily capable of lifting far greater loads into orbit than any current spacecraft.

  21. Gilbert says:

    I might regret recommending this book, because Scott is the kind of person to fall for it.

    Anyway, the least unimpressive modern detail-level explanation of how socialism could work is this

    • lambdaphage says:

      Honest question here: did you read Red Plenty as Spufford’s argument for communism?

      If you asked me before I read the book for my fundamental reasons for not being a communist, I probably would have given you a list like:

      1) Socialists lost the socialist calculation debate (Hayek)
      2) Closed societies are bad (Popper)
      3) Dude do you not see the mountains of corpses (five year old with eyes)

      I read the book and really liked it. I thought it was a moving elegy for the Krushchev era that managed to capture the poignant tension between the Soviet hope that the future might be better, and the Russian resignation that it probably won’t. But I didn’t see anything to suggest that points (1-3) weren’t basically correct. Maybe I misread it completely?

      • Gilbert says:

        No, you’re totally right about Spufford.

        Both paragraphs of my comment were meant to refer to the same book, i.e. I might regret recommending Roemer now, not Spufford earlier. (And I actually didn’t recommend Spufford, though I counterfactually might have if some context had made it relevant).

        Basically Scott’s review of Spufford gives me an idea of what kind of specifics he wants socialists to deliver and I’m pointing out that Roemer comes fairly close to doing it.

  22. Anonymous says:

    I dunno. I firmly believe that if the West had started out technologically behind and with less than half the population of the Soviets, we would be talking about all of the foolish Western ways and scoffing merrily about them, like you’re doing here with the Soviets. Every society makes huge, unforgivable mistakes – or at least what can be interpreted as huge unforgivable mistakes if they lose, and get tossed under the rug if they win. You’ll also see this in competitive games – everything the loser did is picked apart, and since nobody can play a perfect game there are always tons of things that are clearly wrong that you can pick apart.

    Here’s the crucial thing – pointing out things that were done wrong means NOTHING, nothing AT ALL, unless you have some sense of the magnitude of those mistakes that goes beyond “well, they lost, so it must have been really bad.” It becomes way harder to get a sense of the real magnitude of mistakes like the ones you’re pointing out here, because they involve humans and anything that involves humans becomes hopelessly emotionally muddled.

    Let’s look at this – here in the West, we have a system where people can be put in jail without a trial for weeks or even months. We have a system where a fair trial is desperately avoided by both prosecution and defense because everyone knows the trials aren’t really anything like fair, and how harsh your sentence is depends on how good a plea deal you’re lucky enough to get. We have a system where you are tried by a jury of your peers, except that any who show intelligence or expertise will be eliminated by the dueling lawyers as too hard to influence through emotional manipulation. We have a system where secret courts can force you to be silent about your prosecution. We have a system where we imprison more people as a percentage of our population than anywhere else on earth.

    Yet somehow this goes nearly unnoticed and is almost never remarked upon by the body politic. Somehow these injustices that are happening here and now aren’t nearly as remarkable or relevant to us as repression in lands far away in space and time.

    If we were to lose a geopolitical struggle or war, and be overrun by an opposing culture, they have all the fodder in the world to cast us as a hopelessly oppressive place, that *of course* lost the war/struggle because of how backwards it was. But all those hopelessly oppressive things about us will, almost certainly, have nothing to do with why we lost the goddamn thing! It’s just the most emotionally resonant thing about us that can be criticized.

    The Soviet Union started out with a much smaller population at a much worse state of technological development. Losing was almost certain unless their system was not just better, but a *lot* better. They came out of the fight a lot closer in technological capability to the West than they went in. Make sure you’re not picking over the bones for what feels the best to criticize instead of doing a bloodless analysis of the game that they played and lost.

    • Tracy W says:

      I firmly believe that if the West had started out technologically behind and with less than half the population of the Soviets

      Didn’t the West do that, with respect to the Chinese Empire?

      (Also, on the courts, I think you meant to be talking about the USA, not the West. I’m fairly sure that, say, Sweden, doesn’t imprison more people as a percentage of its population than China).

      The Soviet Union started out with a much smaller population at a much worse state of technological development.

      Same position as Japan in the 19th century. Or Sweden. Or Italy. Or South Korea in the 20th century.

      They came out of the fight a lot closer in technological capability to the West than they went in.

      After the USA sent a lot of technological capability to them, and the Nazis drove a lot of scientists into escaping to the USA, or killed them, and triggered a war that involved bombing a lot of European factories (including Germany’s).
      What’s more, Imperial Russia had a bunch of famous scientists and mathematicians.

      • Anonymous says:

        >Didn’t the West do that, with respect to the Chinese Empire?

        Not in direct competition until long range power projection was a thing, and the West was the first to get to the point where power projection was a thing. Why? Industrialization. Best explanation I have found is that China and the rest had too many people, labor was too cheap, so it wasn’t as easy to get the great initial returns on labor-saving devices that snowballed into the industrial revolution. Doesn’t much relate to my point, as far as I can tell…

        >Also, on the courts, I think you meant to be talking about the USA, not the West.

        The USA is the biggest deal when it comes to the West. Nowhere else comes close in power, and nowhere else comes remotely close in power projection. It’s what will do the losing if the West loses, and what did the winning when the West won.

        >Same position as Japan in the 19th century. Or Sweden. Or Italy. Or South Korea in the 20th century.

        Sure….? All these places either lost to or never fought the West, and became part of it.

        >After the USA sent a lot of technological capability to them

        USSR was first to space. You don’t become first in something purely by copying what is sent to you.

        >and the Nazis drove a lot of scientists into escaping to the USA

        Relevance?

        >What’s more, Imperial Russia had a bunch of famous scientists and mathematicians.

        Imperial Russia had *potential.* And fell incredibly far short of actualizing it. The USSR seems to have done a far, far better job turning that potential into concrete achievements.

        • Tracy W says:

          I think I may be misreading you. When you started talking about “technologically behind and with less than half the population of the Soviets”, I thought you were talking about the Soviets competing econmomically.

          But now you are saying that Japan or Sweden or Italy or South Korea don’t count, as they lost to the West, or never fought the West, and you talk about “direct competition”.

          And you’re also saying that “The USA is the biggest deal when it comes to the West. Nowhere else comes close in power, and nowhere else comes remotely close in power projection.”

          But the Soviets never fought the USA directly. They fought Germany, which had a population of about 70 million in 1940, while the Soviet Union had a population of around 200 million. So the Soviets were fighting a country with about 1/3 of the Soviet population, who were also fighting a war against Britain at the same time. Plus Germany at the time was being run by the Nazis, who were a bunch of lunatics. So far from fighting the USA, the Russians allied with the USA.

          So, in a military, direct-competition sense, the Soviets never fought the “biggest deal when it comes to the West.”

          Economically, numerous other countries started off with an equally bad or worse starting point, and came out a lot closer in technological capability. What’s more, in the case of the West, with respect to the Chinese Empire, the West did that despite being geographically and culturally much further from China than Russia was from Western Europe (the Russian elite for example spoke French), and without China ever being allied with the West, meaning the West’s ability to learn directly from China in the medieval and Renassiance era was much worse than the Soviets’ ability to learn the West.

          After the USA sent a lot of technological capability to them

          USSR was first to space. You don’t become first in something purely by copying what is sent to you.

          And no one is claiming that the USSR did. But nor was it starting from scratch.

          Imperial Russia had *potential.* And fell incredibly far short of actualizing it

          Third biggest empire by landmass the world has ever seen. Lasted from 1721 to 1917, so much longer than the USSR. Beat Napolean (note France’s populatin in 1820 was 30 million while the Russian Empire’s was only 55 million, so a lot closer in terms of population). Emancipated the serfs. Introduced democracy. Famous writers and composers. Wikipedia has a long list of innovations, in the 19th century, including developing arc welding and the periodic table of elements (Mendeleev). Tsarist Russia had a lot of bad points, and it also had a lot of impressive ones.

          and the Nazis drove a lot of scientists into escaping to the USA

          Relevance?

          The Nazis singlehandedly dealt a vast blow to European science, thus taking the continental Europeans out of the running to get first into space.

          • Multiheaded says:

            Emancipated the serfs. Introduced democracy.

            …lel.

          • Anonymous says:

            The Nazis drove out lots of scientists, but all the best rocket scientists were Nazis. The USSR and USA got to space by harnessing Nazis, not refugees.

        • cassander says:

          Sure….? All these places either lost to or never fought the West, and became part of it.

          >USSR was first to space. You don’t become first in something purely by copying what is sent to you.

          Both the US and the USSR got to space by copying and scaling up german rockets. the USSR got there a few years earlier because when khrushchev came to power, he shelved stalin’s big bomber program in favor of missile development. In essence, the USSR bought sputnik by not buying b-52s. This decision was made for military reasons. the US had bases near the USSR from which it could operate bombers, which were a known, developed technology with a proven success rate. This made bombers an attractive investment, particularly for an air force dominated by bomber pilots. For the USSR, which lacked bases in the americas and was still having trouble with jet engines, bombers were both less useful and less certain, which made missiles relatively more attractive. khrushchev went whole hog into missiles in space, pouring money made available by canceling other stalin projects. this story says nothing about the success or failure of soviet society, except to point out that once the US began to put the same sort of effort into missiles as the USSR, it quickly excelled them.

          >Imperial Russia had *potential.* And fell incredibly far short of actualizing it. The USSR seems to have done a far, far better job turning that potential into concrete achievements.

          Russia was the fastest growing economy in the world in 1913. it was doing a fantastic job of converting potential into reality. the USSR used 1913 production statistics as the baseline to measure its own efforts, and in general they were not surpassed for at least a decademany not until well after ww2. the USSR did a terrible job of converting potential into concrete.

  23. Vaniver says:

    Scott: I’m an expert in linear programming and numerical optimization. I’m happy to answer questions. (I heard about Red Plenty a few years ago through professional contacts, but haven’t gotten around to reading it because it hasn’t been a high priority.)

    A brief comment, though: the size of problem you can tackle scales with your computational hardware, and most industrially used algorithms aren’t actually LPs. (LP-style algorithms look for two things: a good solution, and a proof that no better solution is possible. Turns out the second isn’t actually industrially useful; people just care about doing better than actual known alternatives, and if you don’t look for the proof, you can save a lot of time.) In the 50s, LPs could tackle a small firm’s allocation problems; today, UPS uses much more sophisticated algorithms to determine which packages get delivered by which driver when, and airlines use them to decide which flight crews to put on which flights when and (I believe) which flights to offer when. Solving the entire production of the USSR centrally would not have been possible, even if the people problem had magically been solved, but numerical optimization is a key part of effective management that is influencing more and more decisions.

    But even if the computational problems were solved, you still need an value function, and socialism is still spectacularly bad at propagating information about values. Prices are the best known mechanism to do this, and I would suggest starting further reading with Hayek’s discussions of the economic calculation problem. (Disclaimer: I am very fond of Hayek, as he thinks basically the way that I think and so suggests all the solutions and criticisms I would suggest.)

    • AlphaCeph says:

      The value problem doesn’t seem insurmountable to me. People could be polled on what they wanted; specialists in predicting latent needs could devise clever new poll options. People could be given fungible but non-transferrable tokens to spend on extant options.

      I recon that the hardest problem would be the incentives of the people runnung the system.

  24. Patrick says:

    Has anyone put serious effort into measuring actual economic growth between the Soviet Union and the US? Given that the Soviet Union seems to have ended WW2 with an economy that looked an awful lot like it was a few centuries out of date, can we really say that this experiment proves that capitalism was better at growth? The end point alone doesn’t answer that.

    And it’s not like capitalism didn’t have massive coordination problems, or didn’t waste human capital brutally slaughtering members of its population as warnings to the others.

    http://en.wikipedia.org/wiki/Mississippi_civil_rights_workers'_murders

    For an idea of the scope of the problem, pay attention to the part where, in dredging for the bodies of three murdered civil rights workers, they find eight other dead african americans they didn’t even know had been murdered.

    • cassander says:

      the problem with measuring soviet economic growth is that so much of what they produced was stuff no one wants. There’s a great bit on econlog about soviet shoes, and how in a country that produced more shoes than any society in history (every one of which was included in soviet GDP figures) people would line up around the block to buy imported shoes because the soviet ones were so terrible. By the end, huge sections of soviet industry were considerably value negative, producing such low quality products that the raw resources were worth more than the final goods.

      http://econlog.econlib.org/archives/2009/09/soviet_shoes.html

      >or didn’t waste human capital brutally slaughtering members of its population as warnings to the others.

      there is no equivalency between the killing of tens of millions to that of dozens.

      • Patrick says:

        Yes there is. If you’re attempting to measure wasted human capital, and the killing was done for the express purpose of economically subjugating millions.

        Also, not dozens. The sad fact is that the US was a terrorist state where you could literally just pick a lake at random, dredge it, and find half a dozen corpses.

        • cassander says:

          Where I come from, killing more people is worse than killing fewer. What insane moral universe do you live in where that is not the case? particularly given the VAST disparity in the level of killing.

      • Multiheaded says:

        there is no equivalency between the killing of tens of millions to that of dozens.

        Now compare what the US has been doing and promoting abroad (SE Asia, South America…) from the late 50s to the 80 to the Eastern bloc’s internal repression in that period.

        • cassander says:

          >Now compare what the US has been doing and promoting abroad (SE Asia, South America…) from the late 50s to the 80 to the Eastern bloc’s internal repression in that period.

          no problem. the only thing even close to the scale of communist killings were the purges in indonesia that followed Suharto taking over. he killed, by most estimates, about a half million people. A vile despicable man. But that makes him more than an entire order of magnitude less vile than Mao and Stalin.

          Nothing else even vaguely capitalist comes even close in scale. After the black book of communism came out, the [left tried to do what you suggest](http://en.wikipedia.org/wiki/Le_Livre_noir_du_capitalisme). Despite ludicrously low standards of guilt, their total was so much lower that they had to add in every single death the second world war to make the numbers balance, an assertion beyond ridiculous given that the war was started when fascists and communists decided to team up to invade poland. The raw facts are utterly unarguable, communists killed more people, in less time, than any other group of people in history. they were orders of magnitude more violent than anyone else.

          • Multiheaded says:

            Adjust for population density, etc. Also, I’m pretty fucking sure that Operation Ranch Hand counts! Also, by your measure none of the famines in China count, because then we get to point out every other 20th century famine under every other regime.

          • cassander says:

            >Adjust for population density, etc.

            that makes communism look worse, not better. the Khmer rouge murdered between 1/4 and 1/3 of their entire country.

            > Also, by your measure none of the famines in China count, because then we get to point out every other 20th century famine under every other regime.

            Again, feel free to do so, it doesn’t help your case. Famines were almost invariably much worse in communist countries than non-communist countries.

          • Illuminati Initiate says:

            I don’t think you can really count the Khmer Rouge because their ideology was very different from anything resembling either Marxism or Soviet style Communism (if you consider those different). They appear to have been Luddite agriculturalists who thought that “primitive communism” was actually a GOOD thing. They were also very nationalistic and the mass killings were often genocides in the racial sense. They seem to have been more like modern North Korea than anything “Communist”.

          • nydwracu says:

            Wasn’t the Khmer Rouge backed by the West against North Vietnam?

          • cassander says:

            @nydwracu

            not exactly, the situation was complicated. in 75, the Khmer, backed by the north vietnamese, ousted the US backed general and took over. Relations almost immediately began to sour as the vietnamese, flush from their victory over the US, began to think they could dominate all of indo-china. The chinese obviously did not like this idea. during the vietnam war, the vietnamese had been very good at playing the soviets and chinese off each other as a way to extort aid from both sides, but post-war, chinese and vietnamese goals were clearly at odds, so vietnam lined up in the soviet camp. the soviets gave them a bunch of equipment and the vietnamese invaded cambodia. this alarmed the chinese who conducted what was, essentially, an extremely large punitive expedition in northern vietnam. the success of that war is highly debated, but the end result was that the vietnamese proved unable to sustain a military sufficient to both protect against the chinese and project power abroad, and began to slowly withdraw. the US at the time was not so much pro-khmer rouge as it was pro-china and anti-vietnam. And quite a few countries condemned the vietnamese attack as naked aggression. I am not sure how much information was available at the time of the true nature of the khmer rouge.

    • Tracy W says:

      Given that the Soviet Union seems to have ended WW2 with an economy that looked an awful lot like it was a few centuries out of date,

      Are you sure – I’m not a military historian but I thought the Soviets did produce a lot of tanks, planes, bullets, etc, to fight the Nazis?
      And there’s all those stories about factories being dismantled in the face of the Nazi approach, loaded onto trains, then being rebuilt a thousand miles away.
      Plus the famous brutal constructions of railways and logging camps and what not by forced labour in Siberia during the 1930s.

      • Patrick says:

        They didn’t have tractors.

        • Fazathra says:

          I’m pretty sure they did. Some quick searching on la wik shows that they lost 137,000 tractors in the war and I highly doubt that was all of them which suggests a pretty well mechanised agricultural sector in general.

          • Anonymous says:

            The tractor factories were all converted to tank factories, with the result that in 1945, they didn’t have any tractors.

          • cassander says:

            @anon

            if the soviets were wanting for tractors during the war, they almost certainly would have gotten some through lend lease. I know for a fact the UK got some.

    • Harald K says:

      Yes. Spufford discusses in the footnotes the attempts to measure actual economic growth; of course the west didn’t blindly believe Soviet numbers. There has also been done “forensic” work in more recent years.

  25. Alex Godofsky says:

    Please describe a mathematical algorithm for figuring out how much money should be spent on people trying to make Snapchat. kthx.

    • Fnord says:

      Lightspeed Ventures and Institutional Venture Partners figured out a number somehow.

      All these decisions ACTUALLY STILL GET MADE in a capitalist economy. They just don’t get made in single centralized location.

      • Alex Godofsky says:

        They don’t get made by a mathematical algorithm. They get made by human decisions.

        And before you accuse me of some kind of mathematically unjustifiable human chauvinism, observe the context: the hypothesis that we could solve the calculation problem with, basically, linear programming writ large. I totally believe that if we invent general AI it can probably do organizing-the-economy stuff, but GAI doesn’t actually exist right now.

        • Fnord says:

          That’s not the hypothesis.

          The hypothesis is that, contra the common belief that Communism was nerds/experts/social scientists trying to run a state, the nerds, experts, and social scientists were, in fact, more or less excluded from running the Communist state, and during the one period when they weren’t, things were actually decent.

          That, contra your apparent point, the state didn’t fail when Kantorovich and those like him went too far in trying to do everything by mathematical formula, but rather when Kantorovich was sidelined and his methods abandoned.

          • Alex Godofsky says:

            Do you not understand how “Kantorovich’s methods cannot work in practice at the scale envisioned” addresses the idea “the state failed when Kantorivich’s methods were abandoned”?

          • Douglas Knight says:

            I don’t understand how it addresses the claim.

          • Alex Godofsky says:

            The claim “the state failed when Kantorivich’s methods were abandoned” carries with it the implication that the author believes the state might not have failed had Kantorovich’s methods not been abandoned. This is further supported by (quoting Scott’s post directly) “What if Kantorovich had been able to make the Soviet leadership base its economic planning around linear programming? How might history have turned out differently?”

            I am not reading tea leaves here or shoving words into Scott’s mouth; he is pretty clearly interested in the question of whether linear-programming-writ-large really could work as a way to centrally plan an economy.

            My response – “no, because there are features of the economy you cannot adequately represent in any of these algorithms” – is on point.

          • Douglas Knight says:

            You’re making an abstract claim to dispute a concrete claim. Do you dispute Scott’s description of history, of when and where Kantorovich’s methods were applied, and when and where the state worked better or worse? “worked” vs “failed” is not a binary.

            Do you merely mean that the USSR did not produce snapchat, therefore it failed?

          • Alex Godofsky says:

            I am not disputing the literal claim “the state failed [chronologically subsequent to the point at which] Kantorivich’s methods were abandoned”. I am disputing the notion that is implicit there, and explicit elsewhere, that had Kantorivich’s methods not been abandoned, the state might not have failed. I am further disputing the more general notion brought up at multiple points by Scott that Kantorivich’s methods might succeed if applied in some future attempt at Communism.

            I am disputing all of these by pointing out one way in which I think his methods, and any similar methods, are completely incapable of replicating an important function of modern capitalism.

            I am suggesting that no “mathematical” algorithm in the lineage of linear programs and similar optimization algorithms is capable of handling this sort of problem, and that these calculations have to be done by people because they are too fuzzy to be formalized.

            Snapchat was a randomly-selected standin for things within its class, roughly “products that did not used to exist and are highly speculative”.

          • Anonymous says:

            And your implicit claim, without evidence that I can see, is that Kantorovich would inevitably try to implement a system where a simple algorithm without human input does everything.

            The reading of that sentence to suggest that Scott thinks linear programming alone, without human input, is all you need to run an economy, rather than using linear programming as useful tool to be used by humans to manage the economy, is uncharitable bordering on a strawman.

            PS: I bet that mathematical algorithms were, at some point, used by potential investors to evaluate Snapchat’s potential future profitability.

          • memeticengineer says:

            I don’t think “without human input” is part of the claim, but it’s not clear how an algorithm along the lines of linear programming would automatically result in collecting more human input than what was actually done. And the problem in all the examples Scott cited was failure to obtain the proper information, and indeed, lack of pathway for anyone with the information to provide it. Thus, anyone proposing to improve the situation with a superior calculation algorithm is likely completely mistaken about the actual problem at hand.

          • Alex Godofsky says:

            And your implicit claim, without evidence that I can see, is that Kantorovich would inevitably try to implement a system where a simple algorithm without human input does everything.

            Kantorovich would not try this because it would be evidently impossible even from the start – simply defining the program would be beyond imagination, let alone solving it.

            However, Scott certainly envisions something like this:

            The book very briefly mentioned a competing theory of resource allocation promoted by Victor Glushkov, a cyberneticist in Ukraine. He thought he could use computers – then a very new technology – to calculate optimal allocation for everyone. He failed to navigate the political seas as adroitly as Kantorovich’s faction, and the killing blow was a paper that pointed out that for him to do everything really correctly would take a hundred million years of computing time.

            That was in 1960. If computing power doubles every two years, we’ve undergone about 25 doubling times since then, suggesting that we ought to be able to perform Glushkov’s calculations in three years – or three days, if we give him a lab of three hundred sixty five computers to work with. There could have been this entire field of centralized economic planning. Maybe it would have continued to underperform prices. Or maybe after decades of trial and error across the entire Soviet Union, it could have caught up. We’ll never know. Glushkov and Kantorovich were marginalized and left to play around with toy problems until their deaths in the 80s, and as far as I know their ideas were never developed further in the context of a national planned economy.

            And:

            Here were people who had a clear view of the problems of human civilization – all the greed, all the waste, all the zero-sum games. Who had the entire population united around a vision of a better future, whose backers could direct the entire state to better serve the goal. All they needed was to solve the engineering challenges, to solve the equations, and there they were, at the golden future.

            Which leads me to answer Scott’s final question:

            The book’s greatest flaw is that it never did get into the details of the math – or even more than a few-sentence summary of the math – and so I was left confused as to whether anything else had been possible, whether Kantorovich and Glushkov really could have saved the vision of prosperity if they’d been allowed to do so.

            No.

          • Fnord says:

            @Memetic Engineer:
            Except, as noted, things actually seemed to be working pretty well for the period when Kantorovich was influential. So something about Kantorovich’s methods (broadly defined) was getting the human input that was needed.

    • Harald K says:

      That’s easy. Let B be the gross national product, let p be the estimated overhead of bureucracy, and let T be the shadow price of potatoes. Then the amount of money that should be spent on making Snapchat is given by this formula:

      F(B,T,p) = 0

      The world would do OK without Snapchat.

      Seriously, though. They came up with one of the best-selling computer games in history, why shouldn’t they be able to come up with other good pieces of software?

      Art and science worked pretty well in the Soviet Union, because artist and scientists tend to be motivated by other things than socioeconomic status anyway. They could do creative work just fine.

      The Soviet Union had excellent computer scientists. What they struggled with was all the boring work of making computers they could play on. For all their focus on heavy industry, heavy industry with all its inputs and output they handled a lot worse than thinking work.

      • Alex Godofsky says:

        I’m not contesting the idea that a centralized, human bureaucracy could allocate resources to innovation efficiently. I’m contesting the idea that an algorithm along the lines of a linear program could do so.

        • dublin says:

          You just have to define innovation as one of the outputs of your economy, that takes hipster glasses and neon hair dye as inputs. Then your algorithm can direct whatever proportion of your economic output you think is best to the production of innovation.

        • Harald K says:

          Well, right. You have to decide how much research you do. But so would you have to under any system.

          This is a fundamentally hard problem. In order to “put the right price” on exploration, you’d have to know what you will discover. And if you knew that completely, you wouldn’t need to explore in the first place.

          Keep in mind, “Use the market to aggregate information” is an algorithm too. Even “let them try whatever and let only the successful survive to the next generation” is an algorithm. These algorithms will fail too: just because a certain rate of exploration was profitable up until now, doesn’t mean it will be in the future.

          • Alex Godofsky says:

            Harald: I know that “use the market to aggregate information” is an algorithm in the broad sense. It is not, however, an algorithm in the narrow sense of something that we have mathematically formalized and implemented on existing desktop computers. It is even less of an algorithm in the sense of “the state of the art in 1950s operations research”.

      • lambdaphage says:

        Art and science worked pretty well in the Soviet Union, because artist and scientists tend to be motivated by other things than socioeconomic status anyway.

        I agree with the second clause, but I wonder if you are forgetting some of the famously poor outcomes obtained whenever art and science ran up against the Party machinery?

        • nydwracu says:

          Socialist realism was a lot better than its American counterpart. You’re mostly right about science — but are there any other cases of Western ideologically-motivated science failure besides the one obvious sham comparable to Lysenkoism in that many people know it’s a sham but are justifiably afraid to speak up?

          This has come up on LW before. The main thing LW came up with is the failure of Geodakian’s theories to get any notice in the West — but LW isn’t the best place for those questions.

          • millericksamuel says:

            I’m fond of Soviet (and all Russian really) art myself (especially the t.v. series Guest From the Future but I’m not sure how you assert that Socialist Realism was so much better than its American counterpart. As far as I can see their was no strict counterpart in America to Socialist realism (sure comic codes and stuff existed but none of these applied to all art). If you mean that you feel that Soviet was simply better in general than American art during the same period I’d be interested to know why you think so.

          • dublin says:

            The other obvious sham comparable to Lysenkoism in that many people know it’s a sham but are justifiably afraid to speak up?

          • lambdaphage says:

            I really tried to do intellectual due diligence before posting that and wrack my brains for the Western analogues of the phenomena I’m gesturing at. But rolling together the Motion Picture Production Code, Tipper Gore and even the CIA’s goofy dabblings in abstract expressionism, I still don’t see how it adds up to anything like official state control over the production of art, with all that entailed. Could you imagine things turning out ok for the Soviet counterparts of Albert Meeropol, Nick Ut or even Norman Rockwell (fer chrissake)?

            That’s without even saying anything about the style of Soviet Realism itself. It strikes me personally as ‘Thomas Kinkade for communists’, a sentiment basically shared by Milan Kundera (who would know), but to each their own 🙂 Obviously there were great artists in the Soviet period, but the examples that come to my mind were mostly working despite the conditions, not as a result of them. The ironic fist-pumping of Shostakovitch’s 5th stands out in particular.

            In re: science, I assume dublin is talking about race and intelligence. So let’s just assume that Phillipe Rushton was right about everything. But consider this: Rushton still had a job. Granted that wouldn’t have happened without tenure, but at least there was a tenure system that was respected the state. Are people dissuaded by informal social pressures from studying intelligence? Sure. Are psychometricians sent to labor camps or shot? No.

            I mean, even if you lay the modern welfare state at the feet of Gould, Lewontin et al., it doesn’t approach Lysenkoism in terms of the damage to the people who opposed it or the civilians who suffered under it.

          • nydwracu says:

            As far as I can see their was no strict counterpart in America to Socialist realism

            There was a style of art that was promoted by the state for political reasons. It was not very good, and yet, to this day, it’s everywhere in government buildings and not hard to find in most cities.

          • Nornagest says:

            There was a style of art that was promoted by the state for political reasons. It was not very good, and yet, to this day, it’s everywhere in government buildings and not hard to find in most cities.

            Don’t be vague.

          • social justice warlock says:

            I suspect he’s referring to brutalism, which is probably the opposite of socialist realism in that the one is aesthetically appealing to cultural elites but not most people, whereas the reverse is true for the other.

          • Cauê says:

            I thought he was referring to the CIA’s use of modern art as an ideological weapon:
            http://www.independent.co.uk/news/world/modern-art-was-cia-weapon-1578808.html

          • nydwracu says:

            I didn’t even think of brutalism, but that too. (Vagueness has its uses.)

        • millericksamuel says:

          There was a style of art that was promoted by the state for political reasons. It was not very good, and yet, to this day, it’s everywhere in government buildings and not hard to find in most cities.

          So a handful of sub par buildings were constructed. Socialist realism applied to every form of art from architecture to children’s books. Every movie, every painting every novel and poem had to keep to its standards (in theory). While it is true that these were relaxed as time passed, and it may well be true that they were superior to whatever architectural style of buildings the US government was fond of, in extent there is no comparison.

        • Harald K says:

          There were areas of research crippled by politics, yes. But even there, there were innovations. Economics was crippled by ideology, but you still had Kantorovich invent linear programming. Biology was crippled by Lysenkoism, but you had people like Raissa Berg (who the character Zoya Vahnsteyn in Red Plenty is based on).

          As for art, it’s often a matter of taste. In “high” art, it was the party who decided what was proper, but I don’t think their choices were so much worse overall than those who decided it in the west – extremely wealthy patrons.

  26. David Barry says:

    They were officially quite certain that any day now Communism was going to prove itself better at economic growth, better at making people rich quickly, than capitalism. Even unofficially, most of their leaders and economists were pretty certain of it. And for a little while, even their capitalist enemies secretly worried they were right.

    There’s an amusing story about Paul Samuelson’s economics textbook, summarised here by Alex Tabarrok:

    In the 1961 edition of his famous textbook of economic principles, Paul Samuelson wrote that GNP in the Soviet Union was about half that in the United States but the Soviet Union was growing faster. As a result, one could comfortably forecast that Soviet GNP would exceed that of the United States by as early as 1984 or perhaps by as late as 1997 and in any event Soviet GNP would greatly catch-up to U.S. GNP. A poor forecast–but it gets worse because in subsequent editions Samuelson presented the same analysis again and again except the overtaking time was always pushed further into the future so by 1980 the dates were 2002 to 2012.

    And:

    When I was in Econ 1, we actually used the infamous 1989 Samuelson text – the one that said, “the Soviet economy is proof that, contrary to what many skeptics had earlier believed, a socialist command economy can function and even thrive.”

    From the sounds of things (my knowledge of this topic runs as far as those two blog posts), this was hardly a universal view amongst American economists during the Cold War years. But Samuelson’s is the biggest-selling econ textbook ever, so the notion that industry could be commanded to great heights of efficient investment was not a fringe idea.

  27. Jess Riedel says:

    The arguments are easy to understand. Under capitalism, plutocrats use the profits of industry to buy giant yachts for themselves. Under communism, the profits can be reinvested back into the industry to build more factories or to make production more efficient, increasing growth rate.

    Under capitalism, everyone is competing with each other, and much of your budget is spent on zero-sum games like advertising and marketing and sales to give you a leg up over your competition. Under communism, there is no need to play these zero-sum games and that part of the budget can be reinvested to grow the industry more quickly.

    Under capitalism, everyone is working against everyone else. If Ford discovers a clever new car-manufacturing technique, their first impulse is to patent it so GM can’t use it, and GM’s first impulse is to hire thousands of lawyers to try to thwart that attempt. Under communism, everyone is working together, so if one car-manufacturing collective discovers a new technique they send their blueprints to all the other car-manufacturing collectives in order to help them out. So in capitalism, each companies will possess a few individual advances, but under communism every collective will have every advance, and so be more productive.

    These arguments make a lot of sense to me, and they definitely made sense to the Communists of the first half of the 20th century.

    It seems like Scott (along with many other folks in this thread) is under the impression that the question of whether Capitalism or Communism has greater economic power is completely contingent on which of their respective disadvantages are greater: Capitalism suffers from negative-sum competition and from selfish consumption rather than investment of excess resources for the greater good; Communism suffers from computational, coordination, and incentive problems. This sort of thinking would lead one to believe that Communism could outperform Capitalism if technological or societal developments shifted the relative weights of the respective drawbacks. For instance, maybe the creation of better computers and communication equipment in the past few decades (while human nature remains relatively unchanged) means the balance has actually tilted in favor of Communism, and if only some country would try it as an experiment then we could see Communism succeed before our eyes. (Obviously, this argument places aside the highly important issue of whether the human freedom in Capitalism has intrinsic rather than merely instrumental importance.)

    The problem with this sort of thinking is that Capitalism already has the ability to simulate patches of Communism inside the economy, at least insofar as many of the above advantages are concerned. These are called “corporations”, and they indeed provide several coordination advantages to their constituents compared to the simple exchange of good between individuals. They remove a negative sum advertising competition between the piston manufacturer, the engine designer, and the safety engineers, and allow them to all work together. If a someone in the R&D department discovers a new cupholder design, he doesn’t use his patent as a weapon against the guy making the audio system; the whole company gets to use the invention.

    Now, most corporations are for-profit, but of course it’s entirely legal to set-up non-profits directed by (a) an altruistically minded board, who simulate benevolent single- party rule, or (b) an employee-elected board, who simulate a fully democratic version of communism (=”extreme socialism”?). (The second one is what many communism apologists think would have all the advantages of communism without the authoritarianism.) These other leadership structures are only occasionally realized, but they do exists and in any case that doesn’t detract much from the main argument that corporations simulate communism (with most of the advantages and disadvantages described by Scott) in pursuit of whatever the goals of the corporation’s owners are.

    And of course, what we find is that some industries naturally benefit from the advantages provided by corporations, leading to consolidation, whereas other industries do not. Now consider this argument: we should almost certainly apply our trust-busting laws to any company that becomes a monopoly in its industry, even if that company were a non-profit simulating democratic communism. The reason is that (a) such a corporation could become permanently entrenched using monopoly powers even as it became highly inefficient and evil and (b) this risk isn’t much mitigated by having a non-profit, nominally altruistic board. Communism is, in a sense, the ultimate monopoly. And if you’ve ever worked inside a massive corporations (like, say, IBM…), then you know — even if that large corporation is a good employer for you at the moment — how terrifying is the thought that such a corporation could take over the world leaving no other employment options. Every time a start-up company displaces an industry giant by inventing something the giant wouldn’t have thought of in a million years, you’re seeing why monopolies (and, hence, Communism) are so dangerous.

    (This sort of thinking immediately prompts the question of whether a Communist society could invert this and create “uncorporations”, which would be economic zones where capitalist markets operated within a broader communist framework. These regions could even contain each other in a hierarchy down to lower scales, like the nested Indo-Bangladesh enclaves that exist down to third order. And indeed, there are at least a few cases where corporations have tried to set up internal exchanges.)

    One natural way to steer this train of thought is towards the sort of libertarian idea that Scott has been sympathetic towards in the past: that the most important thing is voting with your feet, i.e. the freedom to leave one country and join another. Similarly, I’m a massively more comfortable with the existence of IBMs so long as people have the realistic ability to leave and go work for someone else.

    Obviously, this is a crude model and there are several reasons why corporations can’t be taken as complete Communism simulations. But I think you’re a lot better off looking at them for experimental data than by making theoretical arguments about how communism might be really good now that we have smart phones. And who knows, maybe in the future we’ll find that every industry is undergoing rapid consolidation, which would be a good indicator that the technological balance had shifted in favor of Communism. But even in this case, it seems we’d ultimately want a Capitalism “back end” and would maintain trust-busting abilities.

  28. Evan Gaensbauer says:

    I agree wholeheartedly. I may yet purchase ‘Red Plenty’, and if I do, it will be through whatever link is on Slate Star Codex for Scott as an Amazon ambassador, or whatever.

    • 27chaos says:

      I don’t really understand Amazon Smile, but does Scott’s link go through it? (Is that what the upper left hand logo means?) Because that would be a good idea if it is possible. IIRC GiveDirectly is an available option.

  29. Anatoly says:

    As someone who grew up in the Soviet Union you’d think I might be able to say something germane to the book, but it seems there’s little overlap: I wasn’t around when Khruschev thawed and Brezhnev froze again. If anyone’s curious what life in the 80ies was like for a Soviet kid, on the other hand, I could fill in lots of details.

  30. Shenpen says:

    Dear Scott,

    How could you ignore one of the most frequent counter-arguments, namely that motivation to hard work does not simply mean to bang a hamme real fast, but also to motivate engineers and managers to think smart, to pay attention to details, to be efficient at deciding daily things and so on? Which means, you largely need to extend linear programming into allocating rewards: the manager who makes better decisions and does a better job at managing details than others should 1) get rewarded in person, with some luxury 2) should have more resources at his command. The market does this – windfall profits to smart capitalists can be used to buy a yacht or to expand the business, buy more equipment and hire more people.

    This is possible – the shadow money could be used for both, just like real money. But no amount of refinery is useful if it does not meet the customers needs, and thus customers should have a vote in shadow money allocation. And this vote should _cost_ customers, so that they really have skin in the game, they really need to think hard to decide if they want to reward some manager or not.

    So, to sum it up, we need a way of customers making voluntary sacrifices which should be given to managers who, according to customers, do a really good job running their businesses, and this thing that was sacrificed, given, by customers, should be equally usable by the manager to buy a nice luxury as a reward or to expand the business.

    Now, that is for all practical purposes a market.

    Of course, we could also do it so: if customers really like the product or services provided by a business, they give themselves a few slaps in the face and record it and upload it. (Yes, it should be something hurtful, that is the point: that is how we know their praise is serious and honest. Forking out $50K for a new car is a honest praise, slapping your own face swollen is a honest praise, if it does not hurt it is not reliable at all.) Managers who earned the most self-slaps get resource points allocated by government which can be spent on personal luxuries or on growing the businesses.

    This could also work, but isn’t a monetary market sort of easier?

    • Viliam Búr says:

      the manager who makes better decisions and does a better job at managing details than others should 1) get rewarded in person, with some luxury 2) should have more resources at his command.

      Should the “luxury as a personal reward” and “more resources at command” be necessarily of the same type? In Capitalism, if you make a good decision that brings you $ 1 000 000, you can use a part of it for personal luxury, and part of it for further production, and split it as you want. But it may seem that $ 1 000 000 for personal luxury is too much, while $ 1 000 000 for production is not enough (okay, depends on what exactly you are going to produce). Maybe we could have a system where you get both the reward and the resources to command, but they are not of the same type. For example: rewards = chocolate; commanding power = your votes in the resources allocation committee get extra weight. Or would this inevitably create a black market, where the chocolate is traded for the votes?

      In other words, is it possible to give people power over more resources in such way that they are able to use it for production, but unable to spend it all on personal consumption? Simply said, to give someone $ 1 000 000 that they can spend on buying a new machine for the factory, or employing more workers, or investing in research… but not on their yacht? In a way that couldn’t be hacked by trades like “I will invest in research that actually helps a different company (let’s say, a travel agency) instead of my company, and in exchange they will let me use their yacht (which, as a travel agency, they were allowed to purchase) for my personal enjoyment”.

      • Nita says:

        Maybe we could have a system where you get both the reward and the resources to command, but they are not of the same type.

        Isn’t that exactly what the Soviet Union did? Rewards = extra pay, nice summer vacations for you and your kids, praise and medals; commanding power = a job higher up the chain. Scott’s quips about factory workers being motivated by blatant death threats aren’t based in reality, as far as I know.

        Unfortunately, various shady economies did thrive in this setup.

        • Mary says:

          If you were late to work you could be criminally convicted of being a saboteur.

        • Harald K says:

          > Scott’s quips about factory workers being motivated by blatant death threats aren’t based in reality, as far as I know.

          They’re from Red Plenty. The book says that getting late to work three times would at one point get you branded a saboteur and shot. I don’t find it implausible, by all accounts Stalin became paranoid and imagined traitors and saboteurs everywhere.

          I can’t find the footnote right now, but Spufford is a conscientous researcher, so I assume it’s correct.

          • Anonymous says:

            Soviet joke: three men enroute to the gulags lament their fates.

            One was always ten minutes late to work and was denounced as a saboteur.

            One was always ten minutes early, and was denounced as a spy.

            One was always on time, and was denounced as having an American watch.

      • Christopher says:

        Are food stamps in the US a form of this red money / blue money distinction, in that they’re good as cash in return for groceries and whatnot, but can’t be used for certain other things? Is there a black market for food stamps in the US? I’m tempted to say the analogy is ineffective because they don’t buy anything that money can’t, but I don’t know the details.

        • gattsuru says:

          Are food stamps in the US a form of this red money / blue money distinction, in that they’re good as cash in return for groceries and whatnot, but can’t be used for certain other things? Is there a black market for food stamps in the US?

          TANF and other supplemental income programs are pretty unlimited : some states block a few purchases, but federally the money’s supposed to be usable for nearly anything. In most states, SNAP and other food stamp money can only be used for allowable (non-alcoholic, not-pre-cooked) food items, or material used to grow foods. WIC funds are further restricted, theoretically to healthy food and in practice to ‘healthy’-sounding food.

          The extent and degree that people bypass these restrictions is controversial, but it’s obviously possible: buying items and reselling them isn’t difficult

        • nydwracu says:

          National Review says there’s something going on.

          “Well, you try paying that much for a case of pop,” says the irritated proprietor of a nearby café, who is curt with whoever is on the other end of the telephone but greets customers with the perfect manners that small-town restaurateurs reliably develop. I don’t think much of that overheard remark at the time, but it turns out that the local economy runs on black-market soda the way Baghdad ran on contraband crude during the days of sanctions.

          It works like this: Once a month, the debit-card accounts of those receiving what we still call food stamps are credited with a few hundred dollars — about $500 for a family of four, on average — which are immediately converted into a unit of exchange, in this case cases of soda. On the day when accounts are credited, local establishments accepting EBT cards — and all across the Big White Ghetto, “We Accept Food Stamps” is the new E pluribus unum – are swamped with locals using their public benefits to buy cases and cases — reports put the number at 30 to 40 cases for some buyers — of soda. Those cases of soda then either go on to another retailer, who buys them at 50 cents on the dollar, in effect laundering those $500 in monthly benefits into $250 in cash — a considerably worse rate than your typical organized-crime money launderer offers — or else they go into the local black-market economy, where they can be used as currency in such ventures as the dealing of unauthorized prescription painkillers — by “pillbillies,” as they are known at the sympathetic establishments in Florida that do so much business with Kentucky and West Virginia that the relevant interstate bus service is nicknamed the “OxyContin Express.”

        • Mary says:

          Yes. There was an article about how — I believe it was soda — to launder the money. You bought the soda and exchanged it elsewhere.

      • drethelin says:

        In a sense this is done in America by having different tax rates on capital gains as on income, as well as with things like property and luxury taxes. You are in a sense getting money that can only be used on investing in the form of lower taxes on investment earnings.

      • Jaskologist says:

        Define “personal consumption.”

        I’m going to plant some persimmon trees on my property, because I like fresh fruit. Is this personal consumption or production? The food produced is real enough, and will likely offset food I would otherwise purchase. Under US case law, this would probably even qualify as “interstate commerce.” But really I’m just planting them because they’re tasty; I could subsist just as well on other food.

        What if I install new lights and a better paint job in the factory because it was getting kind of gloomy in there? Or nice, new, ergonomic chairs for the cube-dwellers? These luxuries might well improve worker productivity.

  31. Lalartu says:

    czarist Russia had been dirt poor and almost without an industrial base

    That is a huge overstatement. Russian Empire had industry in general comparable to German or French, but lacked hi-tech such as motor building.

    In certain cases, Russians were very well-incentivized by things like “We will kill you unless you meet the production target”. Later, when the state became less murder-happy, the threat of death faded to threats of demotions, ruined careers, and transfer to backwater provinces.

    That is essentially the same sort of motivation that slavery used. It does not work very well, it’s a problem known at least since Roman times.

    What if Kantorovich had been able to make the Soviet leadership base its economic planning around linear programming? How might history have turned out differently?

    No way. “Linear programming” is just a method of solving already formalized optimization problems. It has nothing to do with choosing optimization criterias. The only result of it’s successful implementation in USSR would be more junk that nobody needs.

  32. Salem says:

    I worry that you are giving Communism far too generous a reading. Yes, the lack of this Hayekian dispersed information was a problem, but it was by no means the biggest issue. They had shortages of bread, which is one of the simplest production chains imaginable. Their problems went much deeper.

    Allow me to illustrate by way of a joke. My father grew up in the Soviet sphere of influence, and his favourite joke was that a citizen saved up his coupons for a new Lada, and was told by the apparatchik that it would be delivered on 22nd February – in 20 years time. “Is that the morning or the afternoon?” queries the citizen. “What the hell do you care?” wonders the apparatchik. “Well, I have the plumber coming in the morning.”

    The key Soviet problem was lack of accountability of the top to the bottom. Prices, and profits and losses, are a form of accountability, but they aren’t the only ones – voting, social pressure, exit rights, etc. all provide it in different forms. But the Soviet system was designed so that accountability flowed in one direction only, from bottom to the top. Yes, you might be executed if you didn’t please your boss. But if your boss didn’t please you, you were SOL. And the government was everyone’s boss, meaning that there was no sense in which government plans were accountable to the people, or even reality. And if you aren’t accountable, why bother? Hence the joke – the apparatchik is amazed that the citizen is planning his life as if the official pronouncements had any reality behind them. It’s a deeply corrosive system that engenders great cynicism in all directions.

    But the system was necessarily designed like that, it’s what puts the “command” in a command economy. Besides, if the government is answerable to the people, you’ll end up with the “Moloch” of democratic tattonement between parties competing for popular support, which will lead to the same old bourgeois liberalism, not the glorious shining future of the proletariat. Why, the people may not even support Communism at all, afflicted as they are with false consciousness and whatever other epicycles we may need to make up to explain why the party of the people isn’t popular. But even if this stuff was originally meant in earnest (which I do not for one moment concede), it’s impossible to maintain seriously in an actual Communist country, which leads to Spufford’s great line about gangsters just pretending.

    And I think it was necessarily designed like that on another level. It’s not hard to notice the contempt for ordinary people as they really exist (as opposed to some mythic ideal of the proletarian Everyman) that lies at the heart of Communist ideology. It’s not, to my mind, an accident that this same ideology contains so many different ideas that denigrate most people – “false consciousness,” “wage slavery,” “Vanguardism,” “consumerism,” “the new Soviet Man,” etc. I suspect that for many, the sneering contempt for their fellow man is psychologically prior to the Communism that tries to justify it. When Orwell wrote that a millionaire may enjoy breakfasting off orange juice and Ryvita, but if you’ve had a hard day you just want something tasty, he was already in the process of breaking with mainline Communism, and I think that human insights like that are a large part of it.

    • Shenpen says:

      I am from the same parts (Hungarian) and basically our Eastern European culture has always been very hierarchical and contemptuous of the common man, and Commies may have inherited it. Or their own contempt, from Marx, mixed with our traditional gentlemanly contempt of the stinking peasants.

      Daniel Berzsenyi in 1833:

      “A feeling of honor is central to motivation, it makes hard work feel like play, death in battle feel easy, and it makes people not be content with the necessary, but always strive to achieve more. But this instinct can blossom if people are treated like humans. The people understand that not everybody can be a lord, but they also understand that nobody has a reason to be contemptuous and rude to the poor. Even the smallest officers or lords give no respect to the serfs, even their local leaders will be called ‘thou’ or get pulled around, and this supresses a sense of honor in the people. To be despised suffocates the sense of honor, and in weaker people like Slovaks that leads to self-hatred and depression, and in stronger people like Hungarians that leads to reversed sense of honor, where people seek their honor in committing evil acts, acts of crime. It is all right to give commands, but not to despise and not to make feel inferior: because contempt makes the small smaller and the honorable hostile. People either in the saddle or at the plow [lords and serfs] feel their own dignity, and are able to help or harm here or there if they want to: and if they like you or hate you, they will want to. And it would be very stupid to expect more from being hated than from being liked.”

      • memeticengineer says:

        I can’t speak for Hungary — the current political climate there does not seem so great. But Czech, formerly-East Germany, Poland, Slovakia, Romania and the Baltics all seem to be doing ok on respect for the common man in their post-Communist eras. At least no worse than Western Europe.

        • Viliam Búr says:

          If we measure “respect for the common man” on a scale where Communist era was 0, and Western Europe is 10, I would rate Slovakia these days probably as 8 or 9.

          So yes, it’s similar to Western Europe, but “no worse” sounds wrong. The differences are there. You can’t have communism for decades and expect it to stop having power over minds of people, just because it is formally over.

          • memeticengineer says:

            OK, I’d agree that probably all the examples I cited are around the 8/9 level. But I think the primary factor affecting respect for the common man is systems of governance and lingering effects thereof, not something intrinsic to Eastern European culture, since changing the system has a large effect.

      • Salem says:

        I do not think the Communists inherited these thoughts from Eastern-European culture specifically, as this style of thinking is so common in Western Communists.

    • 27chaos says:

      This is the best explanation of the incentive structure I’ve come across by someone making that argument. Could you explain the importance of this incentive structure? Why would it be so completely devastating to growth if bosses could be bad?

    • lockswriter says:

      The key Soviet problem was lack of accountability of the top to the bottom. Prices, and profits and losses, are a form of accountability, but they aren’t the only ones – voting, social pressure, exit rights, etc. all provide it in different forms. But the Soviet system was designed so that accountability flowed in one direction only, from bottom to the top. Yes, you might be executed if you didn’t please your boss. But if your boss didn’t please you, you were SOL.

      This. All of the stories in this review have something in common — people afraid to talk to those in charge, because they were accountable and those in charge were not. The pig farmer afraid to say “I need more wood.” The tire factory employees afraid to say “This quota is too high.” The tire-making machine factory employees afraid to say “This machine may be lighter, but it’s better.” Accountability at the bottom, but not at the top, didn’t make people afraid to screw up or do less than their best, it made them afraid to tell the truth.

      If you can’t tell your boss the truth, you have a problem, but so does your boss. This essay describes how a fictional empire, usually thought of as having been destroyed by its enemies, was in fact brought down by similar organizational defects.

      • Viliam Búr says:

        Soviet Communism sounds a bit like Feudalism, doesn’t it? Even with the small details, like people not allowed to move to another city without a permission from their masters.

    • Sniffnoy says:

      But the system was necessarily designed like that, it’s what puts the “command” in a command economy.

      Well, one can distinguish between central control and lack of feedback; one does not imply the other. I guess you addressed this with your next point, but I do think it is worth pointing out that this point is not really accurate.

  33. Tracy W says:

    The contrast is Tanzania. Julius Nyerere seems to have been about as good, morally, as politicians get. He tried central planning in Tanzania, got economic disaster.

  34. Viliam Búr says:

    Seems to me that many comparisons between Capitalism are Communism are essentially about this:

    Capitalism: Moloch. We have to stop him.

    Communism: Moloch will magically go away, because… uhm… because I believe so.

    Of course, accepting this premises, we have a strong argument for Communism. The problem is the credibility of the second part.

    Or:

    Premise 1: Choosing the best point in the option-space is by definition better than anything we could achieve by economy and democracy.

    Premise 2: The Communist Party will choose the best point in the option-space, because… uhm… because we trust them they will.

    Conclusion: We need to give unlimited power to the Communist Party.

    Reality check: The Communist Party chooses torture chambers for their opponents, because the leaders who put their competitors in the torture chambers outcompete those who don’t.

    • Nita says:

      Capitalism: Moloch. We have to stop him.

      Eh, more like: Yay, Moloch! Every man for himself, and let the Invisible Hand sort it out!

      • Viliam Búr says:

        I meant “We have to stop him” as a reaction that many opponents of Capitalism have (not the proponents). The problem is, they think Moloch is equivalent to Capitalism, while in reality, Moloch can also find its way into alternatives to Capitalism. So getting rid of Capitalism does not necessarily mean getting rid of Moloch… but some people believe it does.

        • MugaSofer says:

          Specifically, the USSR transferred purchasing power from money to status, so everyone began competing for status instead of money.

          • nydwracu says:

            …and in America, where purchasing power has not been transferred from money to status, there’s an entire caste of people who compete for status anyway.

          • Anonymous says:

            Replace “Status” with “Friendships/Connections with the right people”, i.e. sometimes in order to obtain something that was hard to get (known as “deficit”) you had to personally know someone who works in a relevant industry.

  35. R. says:

    They realized that they’d started with a handicap – czarist Russia had been dirt poor and almost without an industrial base – and that they’d faced a further handicap in having the Nazis burn half their country during World War II – but they figured as soon as they overcame these handicaps their natural advantages would let them leap ahead of the West in only a couple of decades. The great Russian advances of the 50s – Sputnik, Gagarin, etc – were seen as evidence that this was already starting to come true in certain fields.

    Couple of things.

    I. Tzarist Russia had an industrial base. It’s economy was growing rapidly until WWI. It was a big exporter of foodstuffs and raw materials.

    Between 1905 and 1914 it’s economy grew at an annual rate of 6%. Fortunes were made too. It also managed to produce interesting stuff, like say, the world’s first practical four-engine plane in 1913.

    Need I mention Russia went from the bread-basket of Europe to an agricultural basket case because the regime killed or ousted all people who could farm? Russians are IIRC again producing grain surpluses by now. Something they almost never managed under socialism.

    It’s pretty naive to think Russia would not have kept industrializing rapidly under a capitalist regime and that it would not be a much more pleasant place to live by now.

    Maybe it would not have had most of the world’s tanks and combat planes by 1941, but I’m pretty sure that without commies there would have been no Hitler or Operation Barbarossa.

    II. Soviet Industrialization would have been impossible without capitalist engineers and their know how. Soviets drove out or killed most of their industrialists (duh) and people who knew how to run an industry. Well known fact: Koch sr. earned a lot of money and got a first-hand knowledge of communism by building up Soviet petrochemical industry.

    Lesser known facts: Stalingrad Tank Plant: built by.. GM? Anyway, US engineeres provided the expertise, plans, directed construction at top and mid-level. Same for lots of other big factories around the USSR.

    There’s a funny book about a Jewish communist US architect and structural engineer who volunteers for a couple of years to go help Russians industrialize.

    ^regarding their industrialization and industrial-military complex, Anthony C.Sutton wrote a whole book about it, and as far as know, it’s not factually incorrect. Link goes to the subchapter about Soviet tank factories (which by 1941 had produced more tanks than the rest of the world combined)

    He later got into conspiracy thinking (Rockefellers, CFR) and so on. Probably occupational hazard for a historian. Maybe it’s easier to imagine there is evil at work than to accept people are just that bloody fucking stupid.

    III. WWII destroyed a big part of the country, but Soviets managed to move entire industries in just a couple of months, and due to their focus on economies of scale were soon outproducing the Germans.
    Perhaps way more amazing achievement than Dunkerque evacuation. Moving people, easy. Dismantling whole industries and re-assembling it a couple of thousand kms east.. at a time when your country is being over-run by a rapidly advancing enemy.

    IV. Red Plenty seems like a thoroughly uninteresting book to anyone who has first or second hand knowledge of Soviet-style economics.
    An economy where factories have to build everything from scratch because they can’t be assured of being able to source components from elsewhere is just never gonna compete. They could copy stuff. Czechoslovakia (and Bulgaria iirc) made it’s own PC clones in the 1980’s. I remember having a programming lesson on one in the nineties.

    My grandpa worked as a ‘buyer’ for a state construction company. Basically he travelled all around the country and tried to get other state companies to sell his company construction supplies so they could get their job.

    My father has related how almost no one in comBloc thought the regime would ever end: people were hopeful that ‘some sort of reform’ would fix the economic dysfunction and things would get at least materially better. Kept hoping so for decades.
    And there were many attempts at fixing state industries.
    Socialist industries were sometimes truly bizarre..

    For example, one of the manufacturers of 8bit computers in Czechoslovakia was an agricultural collective farm.

    • cassander says:

      the impact of american lend-lease supplies on the communist war effort should not be understated. the communists built a lot of planes, but 2/3s of the fuel they flew on was provided by the US, since they had trouble making the high octane fuel the planes needed to fly. moving those factories thousands of miles east does you no good without railroads to move what they make west, and virtually all of the new locomotives and rolling stock were provided by the US. the commies were decent at making tanks, but mechanized armies need trucks to carry the fuel and ammunition for those tanks, and all the trucks came from the US.

      • Tracy W says:

        I read a book once about the economics of Nazi Germany and the Soviet Union. The author wound up commenting with some amazement on how much production the two countries managed, under brutal human circumstances (it was very clear from the writing that he thought badly of the Nazis overall).

        I was mildly impressed, but I thought that the ability of the Brits and Americans to produce like they did without squeezing their populations to the bone was much more impressive.

        • cassander says:

          I’ve always been amazed at the degree to which german society did not simply collapse. invaded from both sides, bombed from above, they simply kept on functioning until virtually every square mile of ground was occupied.

          As for the US and brits making more, you have to remember that the brits and US were much larger, much richer societies, the US in particular. pre-war, the US had about as much industrial capacity as everyone else in it put together.

          • Tracy W says:

            Yes, the wealth of the UK and the USA during WWII is what impresses me, even more than the ability of the Germans and Soviets to keep functioning while going through hell.

    • Anonymous says:

      You left out your link to Sutton, and even the name of the book – Sutton wrote a lot on this subject.

  36. pwyll says:

    Felicitously, Radish has just published an exploration of the cladistics of Atheism that touches on, not Red Plenty, but the Red Terror:

    Also, we found a chair like a dentist’s chair which still had attached to it straps for the binding of its tortured victims. And the whole of the concrete floor around the chair was smeared with blood, and the chair itself studded with clots of blood, and fragments of human skin, and bits of hairy scalp. And the same with the premises of the district Che-Ka, where, similarly, the floor was caked with blood and fragments of bone and brain. There, too, a conspicuous object was the wooden block upon which the victims had had to lay their heads for the purpose of being brained with a crowbar, with, in the floor beside it, a traphole filled to the brim with human brain-matter from the shattering of the skulls.

    Or:

    For a beginning she was raped and outraged generally — the raping taking place in order of seniority of torturers, with a man called Friedmann raping her first, and the others in regular sequence. And, that done, she was questioned further as to the whereabouts of the jewellery, and further tortured by having incisions made into her body, and her finger tips nipped with pliers and pincers. Until at last, in her agony, with the blood pouring from her wounds, she confessed …

    More here, if you have the stomach for it: http://radishmag.wordpress.com/2014/09/22/reign-of-reason/#red-terror

    Scott writes:

    And then it all went wrong.

    No. It was evil from the very beginning.

    Also, a you can find a handy meme-ified version of “true communism has never been tried” here: http://cheezburger.com/8311814656

    • suntzuanime says:

      Those sorts of horrors are pretty much orthogonal to whether or not you have a working economic system. I don’t think most people are sympathetic to death squads and torture chambers, but that’s not a feature of economic Communism, it’s a feature of social Totalitarianism, which is not the subject of this blog post.

      • Viliam Búr says:

        Torture chambers are the outside view of Communism.

        Some people believe you can magically get Communism without torture chambers, but where is the evidence?

        • suntzuanime says:

          America is a Communist country, don’t you reactionary folks always say that? Where are our torture chambers?

          (Uh, other than Iraq. And Cuba. But committing atrocities on foreign soil hardly counts, and anyway Guantanamo Bay is not the Killing Fields of Cambodia by any stretch.)

          • nydwracu says:

            How many thousands of words does Moldbug spend on the difference between capital-C “Communist” and lowercase-c “communist”?

          • memeticengineer says:

            Maybe steelman Viliam’s position? It’s possible to believe that there is no evidence you can get communism without torture chambers, without also adopting an idiosyncratic definition of communism that would include America. In fact, that’s probably true of most who hold that belief, so your rebuttal does not really address the argument.

          • suntzuanime says:

            To be a little more serious, welfare socialism has become the accepted order in the West, and there has been substantial flirting with wage and price controls and centrally planning large sections of the economy, most saliently to me the attempts by FDR to fight the Great Depression. Despite those flirtations with Communism, the West has not really seen a proportionate rise in death squads.

            It seems to me that the problem is actually Communist Revolution, not Communism per se. Revolutions tend to select for the aggressive chickens, and then when they succeed those chickens are the ones that end up ruling the roost. Note that in the Communist states that lasted a while, they tended to ease off the death squad/torture chamber stuff as they grew more distant in time from the initial Revolution.

            I guess at this point I’m just echoing Scott’s “Empire/Forest Fire” theory of mass murder: https://slatestarcodex.com/2013/12/01/empireforest-fire/

          • memeticengineer says:

            It’s true that most Communist states that lasted a while decreased use of torture chambers in favor of secret police and arbitrary imprisonment. Turns out you can produce lots of fear without having to kill quite so many people. But they also became significantly less Communist over time. In some cases to the point that the Communist Party ceased to be in power, in other cases still maintaining nominal Communism. Notable exception: North Korea, which still has high levels of both Communism and torture chambers.

            The revolution theory could have some explanatory power, but Communist revolutions seem to produce higher levels of ongoing torture/death than, say, merely anti-monarchist ones.

            As for Western flirtations with elements of central planning, and widespread social welfare programs: I don’t think it’s reasonable to count these as flirtations with Communism, as that was neither the intent nor the effect. FDR explicitly and deliberately cartelized industries rather than nationalizing them for example.

            It does seem like pretty high levels of social democracy style socialism are compatible with low levels of death squads though.

          • pwyll says:

            It seems to me that the problem is actually Communist Revolution, not Communism per se.

            Interesting. That would fit in with Moldbug’s assertion that governments are more repressive when they are weak, and once they’ve consolidated their power they tend to relax a bit.

            As for the assertion that “reactionaries believe America is a communist country, and communist countries invariably have torture chambers, so where are the American torture chambers?”…

            Even if we agree with the reactionary that America is a communist country, we observe that the implementation of it has differed dramatically from the Russian flavor. For one thing, it seems much less centralized and more distributed. Might the torture chambers take on a more distributed form as well? It’s an apples to oranges comparison, but I’m reminded of the violence (and threats of violence) in the ’60s committed by leftists (e.g. Eldrige Cleaver) that was treated relatively leniently by the elites.

          • suntzuanime says:

            Man, what anti-monarchist revolutions turned out well? I can think of some that were actually secessions that left the local power structure in place, some that slow-boiled for a while before turning into fascism, and some that were stomped by concerned neighbors or counterrevolution before they got out of control.

            Iran hasn’t turned out that badly I guess? Relatively speaking? You could argue that was in essence a secession that threw off the yoke of the western puppet. You could say that of a lot of decolonized states I guess.

            Was Turkey a revolution? My sense is that it was more of a bloodless coup following a loss of the Mandate of Heaven.

          • memeticengineer says:

            There are very few violent revolutions I would endorse overall. Peaceful transitions are almost always a better option. I’m just comparing relative levels of post-revolution death/torture. It seems like Communist revolutions are outliers in this regard.

          • suntzuanime says:

            Well the French Revolution was really pretty awful (although I guess we can credit them with the invention of the guillotine to give their victims a humane death?). How did the Spanish Revolution turn out? I know they went fascist but maybe that’s not so bad, relatively speaking? At least on the metric of mass murder?

            EDIT: Welp. Doesn’t really sound like a friendly environment. http://en.wikipedia.org/wiki/White_Terror_(Spain)#Post-war

          • Tracy W says:

            Turns out you can produce lots of fear without having to kill quite so many people.

            I once read a discussion of the French Revolution by Thomas Macaulay saying that the revolutionaries killed so many people because they had no experience in ruling and couldn’t think of a less bloody way of maintaining control and collecting taxes and so forth.
            (He attributes their inexperience partly to the past French monarchies’ depotism and partly to that any revolutionary with experience served on the first revolutionary government, who then nobly but stupidly ruled themselves out from standing again.)

          • Multiheaded says:

            The French revolution shortly resulted in, at least, the Napoleonic code and Jewish emancipation in most of Europe. I don’t know much about the economic side of things, but here’s an article by some Trot (wait, which of the confusingly-named British commies are Trots again?)

          • Princess_Stargirl says:

            The USa has 1 about 1% of its population currently in Jail. This is a huge scale human tradgedy (albeit not the killing field).

            Of course parts of Europe have market socialism and no horrors on the elvel of the US prison sysytem. But the USA is a bad example of a country that implemented socialist policies without any horrors.

          • Multiheaded says:

            “Elvel” is a pretty good neologism/spoonerism!

          • pwyll says:

            Man, what anti-monarchist revolutions turned out well?

            Yes, exactly! The reactionary would put the American, French, Russian, and Chinese revolutions in the same category; all are populist, anti-monarchist revolutions. In that analysis, the American revolution seems to come off as by far the least bad. (except of course that it seems to have inspired the French.)

          • Anonymous says:

            pwyll, sun tzu excluded the american: “actually secessions that left the local power structure in place”

          • pwyll says:

            Yes, one big difference between America and the rest is that it happened in a distant colony and didn’t directly challenge the monarchy at home in Britain. I’d bet that that’s one reason why it was less bloody than the rest.

          • Anonymous says:

            Sun Tzu said two things: “secession” and “leaving local power structures in place.” Whereas you say “distant colony.”

            Distant colony => secession => preserving local power structures.

            But there are lots of examples that aren’t secession, let alone distant colonies. It’s just that they are usually called “coups” rather than “revolutions.” eg, the Glorious Revolution.

            Though maybe I’ve lost track of the point of this thread. What does “anti-monarchical” mean? Quick and painless palace coups happen all the time. But if it means changing the structure of government to get rid of the king, that’s pretty likely to have repercussions all the way down the line.

            How about the Second Republic?

          • coffeespoons says:

            @Multiheaded – the SPGBs seem to be an odd one – they are “anti-leninist” according to their wikipedia article. They’ve been around since 1904 so presumably they didn’t support the bolshevik revolution. I am surprised – I had thought that all the political parties in the UK with socialist in their title were Trot groups.

          • coffeespoons says:

            Hmmm, I’m curious now. I’m going to have to make a list of all the still ongoing British Marxist groups along with their allegiances.

          • BenSix says:

            The Socialist Party is Trotskyite but the Socialist Party of Great Britain isn’t.

            Socialist Party: Pro-Trotskyism/anti-capitalism

            SPGB: Pro-Impossibilism/anti-Leninism

            CPGB-ML: Pro-Stalin/anti-Orwell

            SWP: Pro-self promotion/anti-SWP defectors

            AWL: Pro-AWL/anti-everyone else

            WRP: Pro-nostalgia/anti-extinction

      • Tracy W says:

        But, as both Marx and Hayek would tell us, social structures can’t be separated from economic structures.

        Hayek’s explanation in the Road to Serform is that central planning can’t cope with a plurality of views about the good life. It doesn’t have prices to encourage people to cooperate even if they don’t want to for ideological reasons (eg Jehovah Witnesses). Steve-Jobs or Peter-Jackson visionaries require either re-jigging of production to fulfil their crazy perfectionist visions (I was living in Wellington while the LOTR movie was being made and frankly I’m surprised no employee throttled Peter Jackson in the process) or suppression somehow.

        Thus, I think, the proliferation of associated political repression (about the only Communist country that escaped it was Tanzania, where Julius Nyerere resigned, and even he had his Preventative Detention Act used against political opponents, and burned people’s homes to force collectivism in agriculture.

      • cassander says:

        >but that’s not a feature of economic Communism, i

        then why has every single attempt to implement economic Communism produced such death squads?

        >it’s a feature of social Totalitarianism,

        communism explicitly calls for totalitarianism. Marx himself called for revolutionary terror.

    • Harald K says:

      You should read Red Plenty. It’s by no means unacknowledged.

      Thing is, Kruschev was an utilitarian (of sorts, probably not of the SSC type). He knew about the atrocities, he had survived Stalin’s terror himself and been complicit in many – and he commits more even in the book – the Novocherckassk massacre.

      The book also points out that the reason for the protests in Novocherckassk was that they for once tried to do as the economists advised, and even tried to let the public in on it. It by no means paints an “if only they’d listened to Kantorovich” picture.

      The only thing that could make up for the atrocities (to K), was if it was really true, that there was really an utopia coming. The only thing that could justify those blood-clotted drains was a thousand-year kingdom of justice, peace and plenty. Unlike later communists, he was sincere about it, and unlike earlier ones, he had doubts he needed to fight down.

    • MugaSofer says:

      Meep. I must admit, I find it hard to argue with that. Gah.

      *shudders*

      Regardless.

      >It was evil from the very beginning.

      That doesn’t explain anything. You can’t actually just say “the enemy are evil” and leave it at that.

      • pwyll says:

        I don’t think the Soviets were evil, just that they adopted an seductively evil memeplex that appeals to the worst in people.

    • Multiheaded says:

      Look at every damn regime that the likes of Henry Kissinger have ever promoted. (Like a million murdered in Indonesia alone, hundreds of thousands in South America…) Look at what happens when we try to avoid repression altogether. (Ok, another try to do everything the nice slow and democratic way went a little better). Or we can go in circles over and over again.

      • pwyll says:

        If you’re arguing that American foreign meddling was as toxic as Soviet foreign meddling, I don’t have much of a disagreement. Heck, without Woodrow Wilson’s getting us into WWI possibly both the Russian revolution and the rise of the Nazis could have been avoided.

        • Anonymous says:

          Wasn’t American entry rather late to affect Russia? Or do you mean something WW did prior to formal entry?

          Early 1917 in quick succession: Germany escalated submarine warfare, the February Revolution, USA declares war, Lenin allowed passage. Half a year later the October Revolution.

          • Protagoras says:

            Presumably Pwyll is imagining that Germany would have won in this counterfactual (not completely uncontroversial, but let’s grant it). If so, the Soviet Union would have ended up with considerably less territory, and it doesn’t seem unlikely that the hypothetical victorious Germans would have offered a lot of support to the White Movement in Russia, possibly further weakening the communists. Still, it seems a stretch to imagine this getting rid of the Soviet Union completely.

          • Anonymous says:

            Russia more or less lost the war. It signed a treaty ceding land to Germany, negotiations starting immediately after the Revolution. Why would it have lost more if America hadn’t entered?

            In the real world, the Allies invaded Russia to support the Whites, so why would you expect Germany to do so in the hypothetical? Germany was allied with the Reds – it made a deal with Lenin that he could go to Russia and try to conquer it in return for promising to take it out of the war.

            Anyhow, pwyll specifically said the Russian Revolution, not just the USSR or the USSR as we knew it.

          • RCF says:

            Didn’t the USSR renege on much of the treaty ceding territory to Germany in return for peace, once Germany was defeated?

          • Anonymous says:

            No, I don’t think the USSR reneged on its peace treaty with Germany. The actual victors stripped the land from Germany, but they did not give it to the USSR, instead creating new countries, like Poland.

          • Anonymous says:

            I don’t deny these points, but I think the section on Wilson here is very interesting:

            http://radishmag.wordpress.com/2014/07/31/arab-spring/#democracy-promotion

            as is this:

            http://foseti.wordpress.com/2011/07/20/tom-lamont/

      • BenSix says:

        Look at what happens when we try to avoid repression altogether.

        I grant that it would be foolish to expect records of unwavering benevolence from anyone governing large societies but if one is to maintain that sadism on the scale of that practiced by Dzerzhinsky and his comrades was justified by the existence of enemies of the regime one has no principled grounds on which to complain about the toppling of Allende or the killings in Chile. (It amuses me, indeed, that there are people – perhaps not yourself – who lament the evils of Tail Gunner Joe, who deprived people of their jobs, while offering sympathy to those who deprived others of their innards.)

        • Multiheaded says:

          Justified? Never. Inevitably expected in environments of such hostility? That’s more like the claim we’re making. We also claim that the other side was the aggressor, and could have de-escalated at its leisure, but chose not too, because it was too damn afraid all the time. Afraid of our ends, not just our means. We reacted disgracefully and disastrously (and had a few high-profile psychopaths; who hadn’t?) – but we never started this.

          • BenSix says:

            …we never started this.

            There was no one “side”. The enemies the Terror targeted ranged from monarchists to anarchists, and was inspired by a would-be assassin from the Socialist Revolutionary Party. While the civil war exacerbated Bolshevik agitation (and I am not fit to judge who was responsible for that) it did not define their attempts to impose order on a population that was unenthusiastic about their plans. Perhaps it’s that that is inevitable, and one’s perspective of its significance doubtless relies on one’s enthusiasm for communist ends. I’ll give Eric Hobsbawm credit for being honest.

          • RCF says:

            “We” “our” “other side”. Sure are a lot of unclear terms in your post. This is identity politics at its extreme: your politics aren’t even intelligible without knowing what “we” you identify with.

      • cassander says:

        We will ignore the fact that Suharto did his killing years before Kissinger came to power. the indonesian kills, which are typically clocked at more like half a million, were the absolute worst white terror, by a wide margin. That makes them at least an order of magnitude less bad than the worst communist terror, so if we are going to judge the systems based on the single worst thing they did, communist is at least 10 times worse. If you take a more holistic view, the total gets even more skewed against communist. American meddling was inept, it often had bad consequences. but to say “it was just as bad” is a gross insult to the tens of millions murdered by communists.

        • Multiheaded says:

          so if we are going to judge the systems based on the single worst thing they did, communist is at least 10 times worse

          …but, ahem, we shouldn’t? Maybe how things look in aggregate is a better metric?

          Again, please understand that every time a teenaged factory worker is raped by a boss, gets pregnant, loses her living and dies in the gutter, we want to count that as a consequence of capitalism. And yes, I’m going to resort to this kind of shock porn if you keep it up with the absurd reference class tennis.

          P.S. Ironic how our avatars match, no? (…)

          • Anonymous says:

            Do you have a good measure of whether capitalist or communist bosses are more likely to rape factory workers?

          • cassander says:

            >Maybe how things look in aggregate is a better metric?

            Patience much? I advocate doing that in literally the next sentence. It doesn’t make communism look any better.

            >Again, please understand that every time a teenaged factory worker is raped by a boss, gets pregnant, loses her living and dies in the gutter, we want to count that as a consequence of capitalis

            Alright, get me some accurate figures for how many people were raped by communist party members in the, along with how many of those victims died in gutters, or maybe gulags. Then we’ll compare.

            >P.S. Ironic how our avatars match, no?

            lol. I hadn’t noticed.

          • Multiheaded says:

            Do you have a good measure of whether capitalist or communist bosses are more likely to rape factory workers?

            Yes; the likelihood that the hypothetical girl would instead be in high school under a hypothetical Communist regime. And so on.

          • memeticengineer says:

            Multi, that’s a sketchy theoretical mechanism, not a metric. As far as I can tell, there aren’t good crime statistics from the Soviet Union so we can’t tell if rates of rape were higher. But if we extrapolate “starving in the gutter” to all forms of starvation, USSR does not look so good compared to USA or Western Europe. Having a system with a theoretical goal of no one starving in the gutter counts for nothing when the actual results involve lots of starvation. At least if you are even slightly consequentialist.

          • Multiheaded says:

            Absent a miracle (that as Scott explains almost looked like it could happen), Russia wasn’t going to look good compared to the US and Western Europe, period. We just compare the Second World with the Third World and conclude that by the dismal standards of actually existing humanity it did better.

            Institutions that were merely screwey and dangerous rather than the barebones flavour of Chaotic Evil that you get in “capitalist” third world countries; rapid and significant gains for women; education and infrastructure; a social safety net. Things that can really make or break an ordinary life – and when they break one, it’s barely one billionth of a footnote in history books, it’s so common.

            And yes, I’m ok with using a hypothetical girl as a salient example; I mentioned the systemic factors that would’ve got her and that Communist regimes could and did reduce.

            As far as I can tell, there aren’t good crime statistics from the Soviet Union so we can’t tell if rates of rape were higher.

            The kind of abuse and violence that matters for this example is the kind that rarely ends up in crime statistics, because “crime” statistics are a construct of the system too.

            P.S.: Senpai, give me a hand over here!

          • memeticengineer says:

            I don’t think it’s reference class tennis to say you can’t fairly compare a real body count to a purely imagined probability of a hypothetical scenario. Also, that you can’t compare an imaginary “it could have turned out better” scenario that has never been observed to real systems that do better than what actually happened, but not as well as our imaginary best-case scenario.

            Compare real to real, or ideal to ideal. I don’t think it is logically valid or particularly persuasive to mix and match.

            Jason Brennan’s Why Not Capitalism? makes a much more elaborated version of the argument about comparing real to ideal.

            Also, I think the argument that the ideal was even possible was pretty weak. Many commenters on this thread (including me) explain why it seems unlikely that a solution based on linear programming could address the failure scenarios of central planning even in principle.

          • RCF says:

            If you’re going to insist that a panhandler on whom I pour gasoline on and set on fire, and a panhandler whom I ignore and dies for lack of being able to afford food, belong in the same reference class, then the inferential distance between our two worldviews is so massive that any productive dialog is unlikely.

          • The Anonymouse says:

            we want to count that as a consequence of capitalism

            Of course you want to. The biggest stumbling block you have in these discussions is that it is terribly obvious–and gets more obvious every time a hitherto-closed archive is opened–that the most prominent examples of communism being put into practice have created seven- and eight-digit body counts. I wouldn’t want to argue against those examples either! It’s hard to spend all your time justifying and/or handwaving away the dude do you not see the mountains of corpses factor before you can get to your ideological payload.

            And that’s how you end up trying to equate the Holodomor, Great Leap Forward, and Killing Fields with a hypothetical teenager, hypothetical boss, and hypothetical starvation.

            I have abstract sympathy for the obstacles to argument you have, but since the argument you are presenting has consistently led to villains so vile that their only real competitors in real human terms are the anopheles mosquito and H1N1 influenza, my sympathy only carries so far.

    • John Hartman says:

      Saying that Communism is *necessarily* plagued by the horrors inflicted by CHEKA types is absurd – mainly because from the mid 1950’s onwards things were actually pretty chill. The police sure as hell weren’t raiding random people’s houses and raping the inhabitants, even after Khrushchev left. When Scott says that ‘Things went wrong’, he’s referring to economic stagnation, not mass terror.

  37. John Hartman says:

    The USSR probably wasn’t as great as it could be, but I don’t think any amount of mathematics could compensate for the informational asymmetries in a country as huge as the Soviet Union – no computer will be able to tell the government how much wood each pig farmer needs. Similarly, it excelled in technology that was too monolithically expensive and too external for any market to provide, such as space exploration – but lagged behind in almost every other field, which were fueled by the curiosity of eccentric entrepreneurs rather than the galactic ambitions of an empire.

    It’s nice to think of alternate histories where energetic reformers like Kruschev stayed in and the sterile kleptocrat Brezhnev never existed – and sure, things *probably* would have gone a lot better. But in terms of industry the USSR only caught up by copying what the capitalists had done 100 years earlier, and once they did they had nothing new to do. Programming and mathematics can rearrange what exists, but until we hit an intelligence explosion innovation can only be done by innovators rather than the government.

    • Multiheaded says:

      innovation can only be done by innovators rather than the government

      How can you say that given the last 70 years? An enormous lot of innovation has been paid for and organized by the government. WW2 advances in tech, the space race, DARPA…

      • nydwracu says:

        …and how much innovation not done by the government was done by a monopoly like Bell? #timepreference

        • david says:

          did Bell behave in a way that is consistent with a hypothesis of self-interested long-term time preference, or did it generally fail to monetize its discoveries?

          • MugaSofer says:

            The point is that monopolies kill competition. That’s why we have laws against them, to promote competition.

          • Auroch says:

            It monetized some of the discoveries, and generally made large sums from those. As is usual with funding basic research, most of the ideas didn’t pan out economically, though many probably led to later ideas, some of which did. Whether that’s strictly a matter of time preference, who knows, but it’s certainly true that monopolists seem to spend proportionally more R&D money on forward-looking projects that aren’t necessarily expected to pan out quickly.

            See also: modern-day Google Labs, Microsoft in the years around the dotcom bubble popping.

          • Nornagest says:

            To be fair, Microsoft’s notorious for funding lots of really interesting R&D and then not using any of it.

            Google seems to be better at this.

          • Anonymous says:

            Can you list some discoveries that Bell Labs monetized? The really famous ones seem to have been great for the world, but do not seem to have been used by AT&T.

            For example, Shockley left AT&T to commercialize the transistor. It was certainly worth all the money spent on Bell Labs that AT&T was able to buy transistors, but they failed to pursue them.

            There may be other inventions that they did pursue for internal use, but it’s hard to tell because they didn’t sell them. For example, they probably did make great internal use of Shannon’s information theory.

      • John Hartman says:

        Again – all of these are massively external things that only a government can throw money into. The USSR had massive technological gains after WWII too, and there was nothing stopping them from making a DARPA equivalent. But it took eccentric capitalists to turn DARPA into the modern internet – I can guarantee you that if the Soviets had built the world wide web, in 2014 the only be able to access an official government website for updated quota information. Almost all of the groundbreaking applications of the internet were built upon the wild ideas of risk-seeking young people – and there’s no funding for probably insane visonaries in a command economy. Entrepreneurs don’t fill production quotas (at least in part because their product doesn’t exist yet).

        So yeah. Governments can fund enormously expensive and external projects that leave plenty of technological breakthroughs in their wake – but turning those breakthroughs into products that people need or want is something only entrepreneurs do well.

    • cassander says:

      > Similarly, it excelled in technology that was too monolithically expensive and too external for any market to provide, such as space exploration

      I wouldn’t say excelled. Despite having the absolute highest priority in the economy, the aerospace sector was still plagued by the same problems that faced the rest of the economy. post-ww2, despite having german jets to reverse engineer, the russians couldn’t get the mig-15’s engines to work, so they had to import a british model. their moon rocket ran into trouble because of the unreliability of its 30 engine design, but that design was necessary because of their inability to produce larger engines. the USSR’s early successes in the space race came almost entirely from them throwing money at their german rockets and scientists while the US was building B-52s. Once the US decided to throw money at its germans, it quickly excelled them.

  38. Said Achmiz says:

    Here were people who had a clear view of the problems of human civilization – all the greed, all the waste, all the zero-sum games. Who had the entire population united around a vision of a better future, whose backers could direct the entire state to better serve the goal. All they needed was to solve the engineering challenges, to solve the equations, and there they were, at the golden future. And they were smart enough to be worthy of the problem – Glushkov invented cybernetics, Kantorovich won a Nobel Prize in Economics.

    Indeed. There’s a reason that utopia as imagined even by two of the smartest and most aware-of-the-system’s-flaws writers in the USSR — the Strugatsky brothers — is basically: “Communism has been successfully implemented.” Revulsion at capitalism is in large part what drives it; and that vision of utopia remained the utopia of choice for many Russian intellectuals even to this day.

    • Ilya Shpitser says:

      I am a huge fan of the Strugatsky brothers. However, this is not entirely fair.

      One of their most famous stories is about an artifact smuggler (in a successful communist society?) who operates in a very bleak, very grey, recognizably Russia-very-far-from-communism kind of setting. Have you seen the Tarkovsky adaptation?

      That aside, their Strangers/Wanderers universe books writes about the scientific elite, and they don’t spend much time on worldbuilding, as they are heavily into character instead.

      I don’t actually know their political views at all, but it is hard to tell what they have done apart from: “what is the path of least political resistance for our book setting?”

      • Multiheaded says:

        One of their most famous stories is about an artifact smuggler (in a successful communist society?) who operates in a very bleak, very grey, recognizably Russia-very-far-from-communism kind of setting. Have you seen the Tarkovsky adaptation?

        Uh, no, the novel is set in a fictional Western capitalist nation, and one of the minor characters is a Soviet scientist working there.

        • Said Achmiz says:

          (Said Soviet scientist is also pretty much the only unambiguously Good character in the book: intelligent, decent and kind, intellectually curious, professionally accomplished and competent. Everyone else, including the main character, is flawed in various ways. Make of that what you will.)

      • Said Achmiz says:

        Yes, what Multiheaded said. Heck, Roadside Picnic (edit: the novel Ilya was referring to) is, in large part, about the sort of environment that capitalism creates. It is pretty much the exact opposite of a critique of Soviet communism.

        The Noon Universe books have plenty of worldbuilding. Read Noon: 22nd Century; it’s pretty much nothing but worldbuilding. (Of course it’s one of their first, and weakest, novels; and later stories that take place in the Noon Universe do a lot of deconstruction of the very world they built — see, for instance, Beetle in the Anthill. Nonetheless the worldbuilding in Noon is never canceled or retconned, and many of the other novels elaborate on it.)

      • Said Achmiz says:

        As an aside: Tarkovsky’s Stalker is a great work of filmmaking, but a faithful adaptation of the novel’s themes and message it is certainly not. (Arkady and Boris Strugatsky are on record as not being terribly fond of it.)

  39. declan says:

    If you have lots of libertarian friends, you have read Hayek’s “Use of Knowledge in Society”, right?

    • david says:

      I think Grossman and Stiglitz (1980) trump Hayek’s UoKiS in the abstract, and Myerson and Satterwaithe (1983) in the particular, but you’d never hear a Hayekian concede that…

      • declan says:

        Well, no, because they both measure “efficiency” relative to the very high baseline of a perfect information equilibrium. I guess it is interesting (although not particularly surprising) that a world with imperfect information is less efficient than a world with perfect information, but this has very little relationship to what Hayek was talking about.

        Anyway, here is a link for anyone who hasn’t read it – it is really very short, and brilliant:

        http://www.econlib.org/library/Essays/hykKnw1.html

        “The peculiar character of the problem of a rational economic order is determined precisely by the fact that the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess. The economic problem of society is thus not merely a problem of how to allocate “given” resources—if “given” is taken to mean given to a single mind which deliberately solves the problem set by these “data.” It is rather a problem of how to secure the best use of resources known to any of the members of society, for ends whose relative importance only these individuals know. Or, to put it briefly, it is a problem of the utilization of knowledge which is not given to anyone in its totality.”

        • david says:

          No, you’ve missed the point. Hayek never bothers to prove that trade in pursuit of a profit motive actually aggregates dispersed information into a meaningful form. This is actually okay because UoKiS is a 1945 essay and the Samuelson revolution hasn’t happened yet; Hayek handwaves because handwaving is how economics proceeded back then. Nonetheless, now that the conceptual framework for talking about dispersed information exists, there’s no reason to uphold UoKiS as revelatory, especially since it is not entirely accurate.

          That aside – Hayek, too, is making an argument from the baseline of imperfect information; his whole argument is that trade reveals asymmetrically-held information. And so it does! But it turns out that trade only reveals some of this imperfect information, not all of it – and remember that the alternative is not central planning but central regulation based on the already-common-knowledge information.

          If the argument is weakened to the triviality that trade reveals only some private information, there is still plenty of scope for a regulator to either make assumptions about the remaining private information, or to structure markets so that more information is revealed – externality regulation via subsidies or taxes, for instance. Indeed, Hayek’s own business cycle theory depends crucially on the argument of prices-as-knowledge operating as a unique unifying signal for time preference for the whole economy, so Grossman/Stiglitz is very much worth worrying about here. If the signal turns out to be shabby, central bank intervention based on compiled aggregate data might well be better.

          • declan says:

            Are we reading the same thread? Red Plenty is about the Soviet Union in the 1950s, and Hayek’s targets in UoKiS are that style of central planning, Lange-style “market socialism”, and the (perfect information) Samuelsonian revolution you mention. So I can’t see how “the alternative is not central planning but central regulation.”

            Yes, there are a whole lot of interesting questions about optimal regulation under imperfect information (though I don’t think the specific papers you mention say anything about it, being quite early in the development of the field). I’m not particularly “Hayekian” or “Austrian” on those, although I’m not quite on the same page as Stiglitz either – I think he tends to be optimistic about both the level of knowledge and benevolence of the regulators.

            All I am trying to do is give UoKiS a shoutout as a short, readable and highly regarded paper about the same issues raised in the post (i.e. the problem with socialism is not simply that people won’t work hard, but that their work is sabotaged or misdirected because the planners who tell them what to do don’t have all the necessary information).

          • Tracy W says:

            Actually, experimental tests do find that markets reveal asymmetrically held information, when repeated trade is permitted.

            See for example http://teaching.ust.hk/~bee/papers/040918/1982-Forsythe_etal-asset_valuation.pdf, or http://teaching.ust.hk/~bee/papers/040918/1982-Plott-Sunder-insider_info.pdf

      • Blake Riley says:

        Myerson-Satterthwaite is a universal result that applies to all bargaining with independent, private information. Since perfect efficiency is impossible under any system, it can’t trump the price system in particular. It just shifts the question to whether markets can achieve the second-best outcome. Rustichini, Satterthwaite and Williams (1994) show inefficiency due to strategic behavior disappears really quickly as more people participate.

        I’m not as intimately familiar with Grossman and Stiglitz, but I’m guessing the same thing applies. Market prices can be imperfect and still be the best.

  40. Crimson Wool says:

    I’m not familiar with this book; my primary reading on the subject of the Soviet economy was Farm to Factory: A Reinterpretation of the Soviet Industrial Revolution. It has a very different conclusion about why the Soviet Union started to falter after Khrushchev, which was (basically) this:

    The Soviet/Russian economy had too many farmers. They had too many farmers due to the commune system. If you had land that five farmers could manage, but ten farmers lived in the commune, the ten farmers split the land between them, rather than having five of the farmers work the land while the other five headed to the city to become industrial workers. Collectivization and other developments (e.g. new fertilization methods) saw large numbers of farmers “freed” from the land by removing this inefficiency; these farmers moved to the cities to become industrial workers; these industrial workers provided the warm bodies to staff more and bigger factories.

    Over time, of course, this surplus of farmers was exhausted, and when it was, economic growth slowed. By that time, the clowns (Brezhnev, et al) were in charge, and the rest, as they say, is history.

    • david says:

      this doesn’t seem like a unique analysis – it follows straightforwardly from Solow growth models disaggregating growth into increases in inputs (labour, capital) versus efficiency advancements.

      it’s true that the Soviet Union advanced enormously through mobilizing inputs rather than becoming more efficient per se, but this was also true of the Western postwar boom. It’s also true of the Far Eastern postwar boom, a couple of authors (most prominently Krugman) have pointed out. Nonetheless, the West and the Far East have both managed to run out of farmers to educate and still keep growing, deindustrialization or not, albeit more slowly.

  41. david says:

    I would recommend Galbraith’s The Affluent Society and The New Industrial State as a partner to Red Plenty – Galbraith heavily influenced Western conceptions of their own mixed economies on the eve of the “new left” social changes, the War on Poverty, etc.

    The Soviets in the 1960s were very much aware of Western criticisms of the inability of the Soviet economic planning mechanism to meet diverse consumer desires (see the ‘kitchen debate’ between Khrushchev and Nixon). Conversely, the West felt obliged to be able to make concrete statements of the ability of average working-men to afford particular material qualities of living. State of the Union addresses treating aggregate demand, aggregate production, inflation, etc. as general indicators was (at the point) novel to the West. The parallels are quite stark, although of course the West survived stagflation whereas the Soviet bloc did not.

  42. calef says:

    The more I read about this, the more I think that Moloch is actually just fundamental–that large enough organizations of people spontaneously summon a god of corruption and entropy, in accordance with some Law of Thermodynamics which governs all sufficiently large groups of humans who hope to coordinate with each other.

    Call it “The Inefficient Government Hypothesis”. I’d like to know if economics theory has a more formal way of stating this.

  43. Ken Arromdee says:

    But second, a sense of what could have been. What if Stalin hadn’t murdered most of the competent people? What if entire fields of science hadn’t been banned for silly reasons? What if Kantorovich had been able to make the Soviet leadership base its economic planning around linear programming? How might history have turned out differently?

    A question like this treats Stalin and company as pure accidents that are unrelated to Communism except they happened to be there at the same time and used it. Perhaps that’s wrong. Perhaps history couldn’t have turned out differently, because Communism creates incentives that lead to a Stalin, to bans on knowledge, and all the other problems, and the only way it could have not had those incentives was to not have been Communism in the first place.

    • Multiheaded says:

      Perhaps history couldn’t have turned out differently, because Communism creates incentives that lead to a Stalin, to bans on knowledge, and all the other problems, and the only way it could have not had those incentives was to not have been Communism in the first place.

      An enormous part of this is the hostile environment, IMO. In various ways. Mao was the only realistic chance China had, and he really did need some kind of unilateral power to fix the horrible and disastrous state of things, and did a lot of objectively great stuff before the 60s; Allende upheld democracy and died a martyr in the face of ruthless American anti-socialist intervention, Tito… didn’t do so bad from what I’ve heard.

      What caused Stalinism as we know it is pretty complicated, and it cannot be reduced to “evil sociopaths rising to the top”; did J. Edgar Hoover really have a better moral character than Lavrentij Beria?

      • Ken Arromdee says:

        That sounds a lot like the neoreactionaries, except with Mao, Allende, and Tito instead of kings.

        • Viliam Búr says:

          Sometimes it seems to me there are more similarities between neoreactionaries and communists than there are differences. Maybe the most important difference is: “Our king, or your king?”

        • Jaskologist says:

          NRX does often sound extremely Marxist to me, especially when they say “give us a king to lead us.”

          I guess one difference they would/could claim is that they want a hereditary monarchy, which supposedly gives the king an interest in his country doing well after his death.

      • The Anonymouse says:

        An enormous part of this is the hostile environment, IMO.

        I’m not sure where this goes. If you’re arguing that communism necessarily creates murderous leaders and systemic oppression when it must be implemented in the presence of competing socioeconomic systems, I’ll probably agree. But the necessary correlate–that hey, we should try it some more until the murderous leaders and systemic oppression destroy all competitors, every society is subject to murderous leaders and systemic oppression, and /then/ we get to have all the noble leaders and everyone gets to be free–does not seem to be a project that needs to be tried any more than it already has.

      • Tracy W says:

        It is fairly clear from history that taking a country that has been through a political upheaval and disintegration of the level China underwent from the mid-1850s to the 1940s is incredibly hard and I’m struggling to think of any examples where this was managed while the country was democratic.

        And arguably democracies might have particular problems, eg the failure of Reconstruction in the southern USA, although certainly dictorships and monarchies have also had bad failures in recovering after civil wars.

        On the other hand, the Great Famine in China seems like an unforced error.

      • Jake says:

        Hoover harassed a bunch of people. Beria kidnapped, tortured, and killed a whole lot more people. Whether or not their inherent moral characters were different (although comparing Hoover’s shameful sexual secret (transvestism) with Beria’s (serial rape) may shed some light on that issue), Hoover operated in a system with checks and balances that prevented him from causing all that much damage. Beria operated in a system that actively encouraged him to cause as much damage as possible.

        • Multiheaded says:

          Well yes, that’s what I’m getting at. In case you haven’t noticed, I’m not a Stalinist myself, I just think that the standard dismissal of Stalinists is piss-poor. And Robert McNamara and Henry Kissinger also operated in a system that encouraged them to do a lot of damage. (trigger warning!)

          • cassander says:

            >And Robert McNamara and Henry Kissinger also operated in a system that encouraged them to do a lot of damage.

            and yet, when the south fell, millions of people fled the country, with up to half of them losing their lives in the process. that did not happen when McNamara and Kissinger took over vietnam, which should tell you all you need to know about the preferences of the people involved.

        • Sniffnoy says:

          This is unlikely to be relevant to your point, but it’s worth pointing out that (as far as I’m aware) there’s very little evidence that Hoover ever engaged in cross-dressing and it’s probably made up. Going by Wikipedia, there seems to be better evidence for him been gay than for him cross-dressing. For whatever reason, though, that idea, whether true or false, doesn’t seem to be repeated nearly as often.

    • Matthew says:

      While it was, finally, modernizing, the Romanov Empire was really brutal. I think some credit for the way the early Bolshevik leaders turned out has to do with the fact that they came out of the Okhranka’s prisons, not simply the fact that they were communists. They took the methods of a pre-revolutionary Russian state that was highly despotic but extremely low in penetration and then applied them on an industrializing scale.

  44. Ken Arromdee says:

    Communism is good at caring for the poor and promoting equality.

    Considering how terrible Communism was at those things, if this is part of a settled modern explanation, I’d hate to see an unsettled explanation.

      • suntzuanime says:

        Wow, that unlabeled quantity went from 41 to 71 in only 45 what-I-assume-to-be-years!

        Something should be added to the comments policy about meaningless or deceptive graphs.

        • Protagoras says:

          I believe it is life expectancy (at least, that seems to match some numbers mentioned in the other link).

          • Viliam Búr says:

            So, is the lesson that when you stop mass-killing people, the life expectancy increases?

          • BenSix says:

            So, is the lesson that when you stop mass-killing people, the life expectancy increases?

            As if the Maoists were still butchering people in the 50s, 60s and 70s. To give the devil his due, the mobilisation of public against health risks yielded real achievements.

            But such campaigns also yielded dramatic failures, as when the Four Pests Campaign led to the eradication of sparrows, booming populations of crop-eating insects, and a fall in rice production just when food was most needed. Mao’s Mao’s war against nature was an especially vivid example of the failure of communist states to incorporate self-doubt (which, in fairness, is not a vice alien to their capitalist rivals).

        • Multiheaded says:

          I can’t find any good non-paywalled articles on life expectancy in China, but all the sources do say that it grew like this after the Chinese revolution.

          Also, part of the explaination:
          http://en.wikipedia.org/wiki/Health_in_China#Post-1949_history

          • Tracy W says:

            Before the Chinese Communist Revolution, there was occupation/war with the Japanese from 1931 onwards. The Japanese were terrible in their treatment of non-Japanese (I remember visiting the JEATH railway museum in Thailand, which is mostly about the experiences of the Western POWs who funded it, but a plaque on the wall mentions that the death rate in the Chinese labour camps building it was 50%, compared to 20% of the Westerners. The Japanese also did medical experiments in Manchuria, the one that stands out in my mind is removing people’s blood and replacing it with animal blood, apparently one victim lived for 8 hours in agony.)

            Along this was warlordism and open fighting between different political forces in China itself.

            So comparing Communist China to pre-Communist China is comparing relative peace to not merely war, but a simultaneous combination of external and civil war, and the external war against an enemy with no moral compunctions about how they treated those under their rule.

          • Multiheaded says:

            So comparing Communist China to pre-Communist China is comparing relative peace to not merely war, but a simultaneous combination of external and civil war, and the external war against an enemy with no moral compunctions about how they treated those under their rule.

            If Maoist brutality was no accident, the preceding brutality and chaos were no accident either!

          • Nornagest says:

            You have to go back pretty far before the Communist era to find a stable China. The Chinese Civil War period was unusually messy, but the Qing Dynasty was essentially being propped up by European powers (who’d done their part in destabilizing it, to be fair) from the Treaty of Nanjing onward, and they weren’t doing an especially good job: in particular, the Taiping Rebellion (1850-1864) led to more deaths than any European wars before WWI. WWII, by some estimates.

          • Multiheaded says:

            The Taiping Rebellion, you say? Sure, Moloch did that!

          • Anonymous says:

            WWIII by yet others.

          • Nornagest says:

            Moloch does a lot of things.

            (I’m not going to follow that link, though; I tend to find the SCP Foundation mildly obnoxious at best.)

        • John Hartman says:

          I think there’s pretty good evidence that the communist revolution dramatically increased literacy, life expectancy and general well being quite quickly, especially during the Stalinist period. At least that’s what wikipedia says. http://en.wikipedia.org/wiki/Economy_of_the_Soviet_Union)

          Stalin was a stone cold psychopathic murderous bastard, but he got shit done.

          • nydwracu says:

            Yes, industrializing is kind of useful. Can socialism be distinguished here from industrializing? (Or is socialism necessary in order to industrialize? Or is there a confounding effect, where most leaders studied in the West and picked up the ideas of the Western intelligentsia?)

          • Harald K says:

            Or, you could say that there was so much low-hanging fruit (it seems, for instance, that any government remotely caring about literacy will succeed in increasing it to near 100% levels) that even a person as self-destructively murderous as Stalin couldn’t fail to reach some of them.

          • Multiheaded says:

            And yet somehow many right-wing or centrist regimes failed to.

          • BenSix says:

            Or is socialism necessary in order to industrialize?

            Japan was undergoing industrialisation at the same time and achieved even more dramatic results.

          • Multiheaded says:

            Japan was undergoing industrialisation at the same time and achieved even more dramatic results.

            Please, what about the massive, enormous, vital American supply/basing contracts during the Korean war, and later on during Vietnam, and all the while with American protection?

          • BenSix says:

            I was referring to its pre-war industrialisation – and not with unqualified praise* but as an example of independent industrialisation in a non-socialist state.

            * Because, of course, it also had the unfortunate consequence of strengthening the drive towards empire. The next step in this argument is for someone to maintain that industrialisation is doomed to end in tears regardless of what system governments end up adopting. I sometimes fear that it is so.

          • John Hartman says:

            I don’t think so. Stalin’s rabid industrialism wasn’t even close to unanimous – it was something that he specifically fought for while others (such as Trotsky) had different ideas. It ultimately paid off, too – it’s hard to imagine the USSR existing past 1941 if they hadn’t advanced past a 1917 military.

          • Hanfeizi says:

            The most effective programs of industrialization have generally consisted of state planning within a market driven system.

            Japan. South Korea. Taiwan. Post 1980 China. Imperial Germany. America under the system of Henry Clay.

            Listean Economics = It Gets S–t Done

      • memeticengineer says:

        Having experienced being poor under communism and poor under capitalism, the latter was much more physically comfortable and less scary. The difference was so huge that it was hard to believe at first. No lines for bread? You can buy as much meat or chocolate as you want *without ration cards*? We might be able to have them every day! Our TV is *how* big? And in *color*? You don’t have to have a party member for a parent to buy Legos? The kids get actual beds, not ancient dirty fold-out couch cushions? We have *how* much space in our apartment? Amazeballs!

        And that’s while we were still on social assistance. It’s not even counting the massive option value of the opportunity to become less-poor or even not-poor, which my family had and exercised under capitalism, but not under communism. Ancestral lack of both social climbing skills and ability to feign ideological conformity among our peasant and worker stock, so we were never destined to be nomenklatura. Yes, capitalism also rewards those things to some extent, but there are a lot more opportunities to become not-poor without being good as social climbing.

        Also not yet counted: the value of not fearing imprisonment by the secret police at any time.

        Honestly, I might be too triggered by this topic to discuss it rationally. I find it galling that communism can still be discussed in polite society as a valid and possibly desirable option in a way that fascism or feudalism cannot, when its reality is at least as grim and brutal. I can accept it from Scott only because I know he’ll give fascism and feudalism dispassionate reviews as well. But the fact that being a communist is any less shameful than being a neoreactionary in broader society is just crazy. (For clarity, I am neither.)

        • Multiheaded says:

          If a poor person were to move from Indonesia or South Korea to the GDR or Yugoslavia in like the 60s-70s, what would they say?

          • memeticengineer says:

            Not entirely sure of the point of your question, but a few responses:

            (1) Communist countries spent more effort on keeping people in than keeping people out. Revealed preference shows there were not that many people who would have considered it an upgrade. So the best evidence we have seems to show they would not be impressed.

            (2) Alternately, given the expense of leaving a very poor country in the first place, anyone who was able to do so would pick a better option. Whatever such a person thought of the GDR, they almost certainly would have preferred the BRD. And evidence seems to show that, in fact, poor people who managed to leave third-world conditions predominantly chose capitalist countries. And the delta would only have grown from the 60s to the 80s, if your point is that we should compare growth curves rather than conditions at a static point in time.

            (3) Speaking of growth curves, by the 80s, the GDR would probably not compare very favorably to South Korea even in base material terms, even for a poor person.

            If you merely want to make the point that a single snapshot comparison is not a full and fair comparison of two different systems, then point granted. But I think your alternate comparison is not very favorable to communism either.

          • I’ve wondered why, if poverty in capitalist countries is considered to be a moral failure by the larger society, the much more pervasive poverty in communist countries isn’t considered to be at least as bad a failure.

          • Viliam Búr says:

            why … the much more pervasive poverty in communist countries isn’t considered to be at least as bad a failure

            Because allegedly they are all starving equally. And equality is so nice it trumps everything.

            Which is a complete crap, of course. The powerful Communists never starve. Well, except when they are victims of a political purge by even more powerful Communists. On the other hand, people who are guilty by having their grandparents entrepreneurs or priests, or having an emigrant in their family… or women who refused to have sex with those powerful Communists (using the salient example proposed by Multiheaded)… those sometimes do.

        • MugaSofer says:

          Firstly … yup. Communism was shitty. I’m sorry you experienced that. It’s easy for us, when theorizing about what might work better, to forget that this is basically the best any society has ever gotten.

          ***

          I feel like it’s important to note that this “capitalism” included social assistance. Which is an add-on to regular capitalism, (produced by, well, socialists.)

          There are places and people out there without social safety nets.

          Most of them are in the third world, because capitalism can survive without a stable government, and socialism can’t. But ask a homeless person sometime what it’s like not to be caught by a social safety net.

          • memeticengineer says:

            I tentatively think the best feasible option (barring availability of superintelligence) is laissez-faire capitalism + guaranteed basic income, AKA Bleeding Heart Libertarianism. I recall that Scott called it “correct” once though I’m not sure if he still agrees with that. It seems like BHL would avoid the failures of central planning, avoid the failures of paternalistic control over the poor, and let people live with needs met and in dignity.

            By the way, I am not going to give socialism credit for social assistance. It bears no relationship to who owns the means of production, and was often implemented with the explicit aim of holding off socialism. OK, so maybe indirect credit.

            I’m not sure if third-world economic systems are best described as “capitalist” in all cases. Most score rather badly on the Economic Freedom Index. Like if a country scores worse than China it seems dubious to call it capitalist. Conversely, countries that sustain high economic freedom for a few decades seem to stop being third-world.

            I agree, being homeless is very bad. There are worse things, though, than slipping through the social safety net. For example, if the safety net that catches you is a Communist-era mental institution, or forced labor camp. I would prefer to be homeless in a Western country to either of those.

          • Creutzer says:

            There is another subthread where this was brought up, but I feel it’s a valid point and relevant here: it’s highly questionable to what extent the totalitarian elements of the USSR have anything to do with communism as an economic system.

          • Ken Arromdee says:

            “I feel like it’s important to note that this “capitalism” included social assistance. Which is an add-on to regular capitalism, (produced by, well, socialists.)”

            Is that the counterpart of “Communism has never worked because it’s never been really tried”, as “Capitalism has never really failed because it’s never been really tried”?

            I think it’s clear that even with social assistance, the society he was in was a lot more capitalist and a lot less Communist and socialist than the one he was in before. Just because there’s still some degree of socialism doesn’t change the fact that he’s comparing a system with a lot of it and a system with not so much of it.

      • cassander says:

        That graph does not prove what you think it does. in 1949, Mao ended a massive civil war that had been going on for, conservatively, 4 decades. Longer, if you include the imperial decline that preceded it. It was also just a few years after a foreign army invaded and shot up the most populous parts of the country. Mao does not get credit for the rise in life expectancy that came about from the end of a war he was one of the participants in.

        • Multiheaded says:

          Mao does not get credit for the rise in life expectancy that came about from the end of a war he was one of the participants in.

          When Mao was in desperate straits and embarking on the Long March, Chiang still couldn’t crush the warlords and keep out the Japanese. Because he was a corrupt shithead. Max Hastings, hardly a commie sympathizer, makes it quite clear in Retribution.

          • cassander says:

            I don’t dispute chiang’s general ineptitude (though he got badly fucked by marshal), but the fact is that mao was half of a massive civil war. he doesn’t get credit for LE increases that come from winning said war.

          • Multiheaded says:

            He does, because: 1) he did win it decisively, while the war would’ve gone on at a slow boil without him, because Chiang couldn’t even crush the warlords, and 2) do you really think that this oh-so-maoist healthcare program of his did nothing? look at India.

            Hastings describes how the US bent over backwards to prop Chiang up, but the pathetic ingrateful shit wasted all of it. Of course people like Stillwell grew to hate him. Everyone would!

          • cassander says:

            >while the war would’ve gone on at a slow boil without him,

            no, it wouldn’t have. without mao, The war would have ended years, maybe decades, earlier, Chiang was nominally in charge of everyone but mao by the 30s at the latest, and if he hadn’t had mao nipping at his heels, he would have been much more able to suppress the warlords. Certainly there wouldn’t have been a war after in 1945.

            >how the US bent over backwards to prop Chiang up

            It did and it didn’t. the US aided the nationalists, but George Marshal actively prevented chiang from crushing Mao when he had the chance. he threatened to withdraw US aid if chiang did not try to form a unified government.

            http://en.wikipedia.org/wiki/George_Marshall#Post_War:_China

          • Multiheaded says:

            What gave Mao the continued ability to nip at Chiang’s heels during and after the Long March was Chiang’s inability to get shit under control. Like, the commies had struck upon the brilliant idea of not robbing the peasants to win their trust; “government” troops and officials, not so much.

          • cassander says:

            Mao was, at the very least, actively making the problem harder to solve, not easier. This was particularly the case after the japanese invasion, when american aid began to make both mao and chiang much stronger vis a vis the others. I have no interest in quibbling over exact percentages, but at any point mao could have subordinated himself to chaing and made the situation better. Instead, he held out for his own power and made things worse. he does not get credit for recovery from a situation he created. this is particularly the case when life expectancy was growing fairly quickly in the 1930s, up until the japanese invasion. most of the serious losses took place between 37 and 47.

          • Emile says:

            I’ve talked at (mainland) Chinese people who hated Mao and blamed him for the sucky economy (and killing people and all that), considering that Chiang would have managed it much better.

            The general impression I get from the history of the war (and talking about it with Chinese people…) is that Chang’s government did most of the work fighting the Japanese, but that when the Japanese left the Chinese got most of their weapons and had the upper hand (apparently due to Soviet help).

            (afterwards, the Chinese Communist Party made a lot of propaganda movies about the Brave Struggle Of The Communists Against The Japanese to put their side in a better light, and as a way of reminding people of their legitimacy; that’s still going on now, hence sustained anti-Japan hate etc.)

      • alexp says:

        Congratulations Maoists! You did better than a brutal occupying regime that liked to bayonet babies for fun!

        Snark aside, 1890-1950 was a period when China was ruled by competing warlords, chaos, a brutal foreign power/occupying regime and civil war. From such a low base, it would have been quite an achievement not to improve.

        • Multiheaded says:

          Snark aside, 1890-1950 was a period when China was ruled by competing warlords, chaos, a brutal foreign power/occupying regime and civil war. From such a low base, it would have been quite an achievement not to improve.

          Again, do you think China ended up like that by historical accident? Or was this the work of Moloch, whom the communists set out to challenge?

          • alexp says:

            I don’t get the point of this question.

            How and why China disintegrated is an interesting and massive question, beyond the scope of this discussion.

    • tom says:

      This. That the myth of the communist system being good at caring for the poor has spread so much in US is plain scary if you had the “opportunity” to experience it first-hand. You couldn’t buy bathroom tissue sometimes. Meat was rare. Public transport was much worse than NYC’s MTA and that’s quite an achievement. And that’s all in the economy which both before USSR and after its fall has been functioning very well.

      • memeticengineer says:

        OMG I’d forgotten about the huge disparity in quantity and quality of available toilet paper. I think anyone who grew up in the West or especially the USA cannot imagine the vileness of communist toilet paper. Having to resort to newspaper pages a lot of the time was almost a relief. And it’s not just this one thing, it’s thousands like it.

  45. Anonymous says:

    Check out this too:
    https://en.wikipedia.org/wiki/Project_Cybersyn

    Interesting that so many projects like this were aborted. Just to put my batshit crackpot hat on, I wonder How Many Cybernetic-Communist-System Failures Is Too Many

    People have often remarked on the failures of communism to be implemented as evidence it just doesn’t work, but I’ve never seen anyone explore the possibility that it works so well that it unleashes some kind of cybernetic-economic lovecraftian horror that devours the universe.

    • Multiheaded says:

      I’ve never seen anyone explore the possibility that it works so well that it unleashes some kind of cybernetic-economic lovecraftian horror that devours the universe

      What if the cybernetic planning mechanism inevitably comes to the conclusion that Human Nature indeed means communism would fail, and so breaks down the boundaries between souls and turns everything into a swirly glowing collective consciousness? And then two Objectivist teenagers within it find the will in their heart to state that A=A, and galt off out of the Commstrumentality to recreate capitalist civilization?

  46. Matthew says:

    Crooked Timber had a symposium on Red Plenty a few months ago, including a guest post by Spufford, IIRC.

      • suntzuanime says:

        ” It makes no sense for the goal of the economy, as a whole, to be to maximize its claims on itself.” Somebody tell the capitalists, who keep prattling on about GDP as an indicator of economic health.

        • Multiheaded says:

          Yup. Not even the individual companies are strictly and rationally profit-driven; the 50+ hour week is widely suggested to be suboptimal even in the short term and seriously hampers the reproduction of labour power in the longer term – but apparently it’s hard, being an American capitalist, to shake the perception that you aren’t exploiting your workers as much as you could be and, worse, giving in to their demands.

          • suntzuanime says:

            You’re certainly optimistic about the extent to which social science “knows” things. Say not “it is known”, but rather “some studies suggest”. Even better if you can link to the studies so your interlocutor can assess the methodology for themselves.

          • suntzuanime says:

            I said “link to the studies”, not “link to other people asserting the same thing is known”.

            For all that I don’t trust peer-reviewed social science journals, I surely trust “alternet.org” even less.

          • Multiheaded says:

            http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2149325

            (You sure do seem to trust reactionary blogs, right? At least I’ve never seen you being this demanding about research and sources when talking to them.)

          • suntzuanime says:

            I’m not being “demanding” – if you had said “some studies suggest”, rather than “it is known”, I would have let it pass without comment. I suggested linking to the studies as something you could do that would make you even more intellectually virtuous, but by no means was I demanding that you do so.

            However, when I said “link to the studies” and you provided a link to a popular science article on a politically-activist news site, I started to worry that you did not know the difference between the two.

          • Quixote says:

            Germany’s labor productivity is very high, a lot higher than most of the rest of Europe, or Korea or Japan. By some measures it’s higher than the US. It does this with social coordination on shorter work weeks and longer vacations and laws and social norms working to prevent defection where people work longer hours to grab status at the expense of actual productivity.

            It’s easy to argue with studies, it’s harder to argue with results, growth and tangible wealth. Those are harder to fake.

          • Multiheaded, your source doesn’t support your position, particularly if you account for labor market rigidities.

            One of the numbers cited is in manufacturing – a 10% increase in overtime results in a 2.4% decrease in hourly productivity. So a 10% increase in hours results in a 7.36% increase in production. This # comes from the bottom of page 5.

            I’ll assume similarly that a 10% decrease in hours results in a 2.4% increase in hourly productivity.

            Now suppose you have a fixed per employee cost – for example, health insurance, a 1 desk per employee policy, etc. If you have employees working 30 hours/week, you’ve reduced your per-hour costs by 2.4% but you’ve increased your per-worker costs by 30% (since you need to hire more employees). If you have employees working 50 hours/week, you’ve increased your per-hour costs by 2.4%, but you’ve reduced your per-worker costs by 22%. (All % numbers reference a baseline of 40%.)

          • Tracy W says:

            There are serious problems with even defining labour productivity in the services. Eg if a health clinic saves one person’s life with $50 of antibiotics and another’s with $100, 000 of chemotherapy how productive are they?
            What’s the value of having people writing books in specialised genres like steampunk-detective novels versus everyone reading Cecilia.
            How does having an ipod compare to attending a live orchestra performance?

            How much value does in-house IT support add to a business?

        • suntzuanime says:

          (EDIT: Misplaced response to Quixote upthread; the deletion system is finicky so I’ll just leave it here.)

          It’s totally easy to argue with results! Results are an uncontrolled experiment with a sample size of 1! Who’s to say Germany didn’t just get lucky, or that the phenomenon doesn’t stem from, say, Teutonic racial superiority or political domination of the European Union or something?

          Even if you mapped working week vs. labor productivity across various countries, there’s still too many confounders and too few experiments to really conclude much.

          It’s not even obvious that labor productivity is a good metric to measure social welfare. For example, one way of looking at the result is to say that the more you induce/require people to forgo working hours of labor, the more valuable the marginal hour of forgone labor is. That makes the result semi-tautological (you’d forgo the least valuable ones first, right?) and it makes it clear that in some sense you are leaving labor hours to “rot on the vine”.

      • Kaj Sotala says:

        Basically came to the comments to post a link to Shalizi’s, in the unlikely event that nobody would have done it before. Want to particularly highlight it, since it goes into detail on e.g. Scott’s question about whether successful central planning would actually have been mathematically feasible.

      • Anonymous says:

        I disagreed with the section which begins, “The most plausible candidate is to look for problems which are ‘separable’”. My intuition was that, even though the planning problem for a real economy would contain many couplings, the important global couplings seemed likely to have a low dimension, scaling as some fractional power of the problem size. Likewise for couplings important in the equivalents of “local subproblems”. (Exposing this structure in a problem might require adding synthetic variables. For example, “labor everywhere in 1960″, constrained to be equal to the sum of labor in all individual categories in 1960, or “labor everywhere in Georgia in 1960″, etc., or ones automatically generated during the computation.)

        Shalizi seems to give up too early on formalisms whose naive application he can associate with, e.g., “the kind of thinking which leads to ‘where’s my jet-pack?’ ranting on the part of geeks of a certain age”. I think this is a flaw in his rationality.

        • suntzuanime says:

          I agree. A key insight he seems to miss is that it is manifestly possible to organize the economy on a level competitive with capitalism; capitalism does it. I’m not sure how you’d estimate how much distributed brainpower actually goes into running the global economy, or how much computing time that equates to in a computation-theoretic sense, but clearly it puts an upper bound on how hard the problem can be in practice.

    • Harald K says:

      Yup, Spufford is a regular poster over there and a really interesting fellow.

      And then there’s our old friend Cosma Shalizi, the guy who knows all the things I suspect Scott Alexander wished he did about heavy statistics, math and causal inference 😉 (I know I wish I knew. He posts lecture notes and stuff. Maybe we can convince him to do a MOOC? I’d sign up!) In his post on the Red Plenty symposium, he gets into the limits of what their approaches could do, and the limits of markets, which are pretty bleak too.

  47. Paul says:

    “Then the makers and the things made turned alike into commodities, and the motion of society turned into a kind of zombie dance… Living money and dying humans, metal as tender as skin and skin as hard as metal, taking hands, and dancing round, and round, and round, with no way ever of stopping; the quickened and the deadened, whirling on.”

    Other than periodic injection of the gratuitous “zombie,” “deadened,” “dying,” etc., this reads to me like a charming description of what capitalism aspires to be.

    “And what would be the alternative? … A dance to the music of use, where every step fulfilled some real need, did some tangible good, and no matter how fast the dancers spun, they moved easily, because they moved to a human measure, intelligible to all, chosen by all.”

    And this paragraph seems to make clearer than almost any other I have read how difficult the project of communism must be. (It also sounds as though it gravely mischaracterizes the problems people actually face.) That said, the fact that these paragraphs strike me as so accurate does make me more inclined to like the book.

    Linear programming is a good idea, but it’s a very long way from that to the technology to actually plan an economy. When I think about how to engineer things by hand, I am consistently amazed by how beautifully capitalism works (e.g. market prices are the same as shadow prices, it’s not just a superficial similarity, and the market computes them much more robustly). I do think the problem of how to engineer an economy receives much less attention than it ought to, but I strongly suspect we should be focusing on incremental improvements.

    • Anonymous says:

      >Other than periodic injection of the gratuitous “zombie,” “deadened,” “dying,” etc., this reads to me like a charming description of what capitalism aspires to be.

      Can you explain this? As Scott notes, it seems like a restatement of the idea of Moloch. Not a good thing.

      • memeticengineer says:

        If you take out the fnords, it reads as:

        “Then the makers and the things made turned alike, and the motion of society turned into a kind of dance… Living money and humans, metal as tender as skin and skin as hard as metal, taking hands, and dancing round, and round, and round; the quickened, whirling on.”

        That sounds like a cheer for the dance of production. Or it could be a description of a post-singularity transhumanist utopia. So it seems like it’s the fnords that make it sound Molochian, not the actual semantic content.

        Here’s the fnord-only version for comparison:

        “commodities zombie no way of ever stopping deadened”

        • MugaSofer says:

          [sorry, that was me, forgot to log in.]

          “No way of ever stopping” is content, not fnord.

          The point is that the human dancers have lost control, ceded it to their “quickened” possessions.

          Money doesn’t share human values. That is not a transhuman utopia I’d want to live in.

          • memeticengineer says:

            I take it as a fnord because it’s an intentionally negative-connotation way of describing ongoing continuity, by implying that one would very much want to stop if only one could. One wouldn’t describe Christian heaven as “unfathomable joy, with no way of ever stopping” but one might describe Christian hell as “horrible torture, with no way of ever stopping”. So even if the content is important, the wording is smuggling in a connotation in a major way.

            Compare to “no need or desire to ever stop” or a more neutral “continuing indefinitely” or “eternally”. I would accept subbing for one of those as a valid defnording.

            Money is just a unit of caring. It doesn’t specify what to care about. It doesn’t share human values but it can serve human values. But certainly money has been culturally coded as evil in many ways so I can see how one would have a negative reaction to it.

          • MugaSofer says:

            >implying that one would very much want to stop if only one could. One wouldn’t describe Christian heaven as “unfathomable joy, with no way of ever stopping”. One wouldn’t describe Christian heaven as “unfathomable joy, with no way of ever stopping”

            One wouldn’t describe Christian heaven that way because you can, in fact, opt out of it. (Indeed, it’s usually considered difficult not to accidentally opt out.)

            Regarding whether it’s somehow deceptive to point out that we can’t opt out of capitalism, because no-one could ever possibly want to …

            Consider Tragedies of the Commons and coordination problems. Consider poverty, and war, and famine.

            Consider Moloch.

            Capitalism does not optimise for our ends, and we can’t stop it when our goals do not align.

          • Mary says:

            Of course, at what point have humans ever really been in control? Prior to capitalism, it was usually Mother Nature calling the shots.

          • nydwracu says:

            And what have we learned today?

            That something is content doesn’t mean it’s not a fnord; that something is a fnord doesn’t mean it’s not content. “No way of ever stopping” is both.

            If I say I just delivered a righteous curbstomping to a racist cracker, you can’t put “I delivered a curbstomping to a…” in the content bucket and “… righteous … racist cracker” in the fnord bucket and just read the content bucket; “racist cracker” is both fnord and content.

          • RCF says:

            “Consider Tragedies of the Commons”

            Why does Tragedy of the Commons keep showing up in criticisms of capitalism? “Everyone owns the pasture in common” sounds like socialism to me, not capitalism.

            And this “can’t stop it” is just another way of saying that it’s a stable attractor. If we create a FAI, we won’t be able to stop it. If we create a self-sustaining egalitarian society, by definition people won’t be able to overthrow it.

            Every system can be stopped if everyone coordinates opposition to it. So the only non-trivial way in something is “stoppable” is if it’s stoppable by a minority Is that a positive attribute? Being able to be overthrown by a small disaffected group?

          • Nornagest says:

            Why does Tragedy of the Commons keep showing up in criticisms of capitalism? “Everyone owns the pasture in common” sounds like socialism to me, not capitalism.

            That’s just an artifact of the thought experiment. Analogous situations come up all over the place that aren’t as easily privatized: for “commonly held meadow”, for example, you could read “fisheries” or “air that isn’t polluted” or something else you can gain a commercial advantage by exploiting disproportionately or unsustainably (this latter in the technical, not the buzzword, sense).

            That said, I don’t think tragedies of the commons are actually the biggest stumbling blocks to laissez-faire economics. Most commons can be enclosed, or can be managed according to schemes that maintain a sustainable equilibrium but don’t interfere in markets beyond the expectations of all but the most doctrinaire libertarians. It’s much harder to engineer away asymmetrical information, or natural monopolies, or short-term/long-term tradeoffs that don’t have to do with resource consumption.

            (Lest I give the wrong impression here, I’m actually pretty sympathetic to laissez-faire — as a default, at any rate.)

      • MugaSofer says:

        Here’s what the grandparent specified, for the record; all the – “gratuitous”, apparently – references to humans losing some sort of metaphorical life-force removed from the quote:

        Then the makers and the things made turned alike into commodities, and the motion of society turned into a kind of dance… Living money and humans taking hands, and dancing round, and round, and round, with no way ever of stopping; the quick and the quickened, whirling on.

        • Auroch says:

          There are a couple words that change referent when you removed the negative fnords but not the positive ones. Pulling out “living”, “quickened”, and “ever” (which is a fairly fnordy word generally), and tweaking a bit of punctuation to preserve the grammatical structure, here’s the fully-defnorded version.

          Then the makers and the things made turned alike into commodities, and the motion of society turned into a kind of dance… money and humans taking hands, and dancing round, and round, and round, with no way of stopping, whirling on.

    • gattsuru says:

      I am consistently amazed by how beautifully capitalism works (e.g. market prices are the same as shadow prices, it’s not just a superficial similarity, and the market computes them much more robustly).

      As much as I cheerlead for capitalism, I’m not sure this is actually the case. Not only in the presence of black- or gray- markets, but in the sheer amount of signaling that goes on. I’d give at least a 75% confidence that several sectors of modern western economies have an order-of-magnitude mismatch between shadow prices and market prices, and a 25% confidence that a slim majority of the world’s GDP is based on on such mismatches.

      We’re /less/ vulnerable to Goodhart’s Law, because money is necessarily expended when used and not recouped, but one of the more optimistic responses to Sowol’s productivity paradox is that we’ve become dominated by it still.

    • Hanfeizi says:

      Extraordinarily, what the Soviet Union failed to make use of, capitalism has taken as a fundamental tool. Linear programming is on the syllabus of virtually every MBA program as a fundamental tool of quantitative finance, marketing and operations.

  48. Multiheaded says:

    Also: HELL YES! Comradefists all around! We did it, senpai! I always knew Scott would come around!

    ~♥♥♥☭☭☭♥♥♥~

  49. Protagoras says:

    When I read a biography of Otto Neurath which talked about his role in the government of the brief socialist republic of Bavaria and in Vienna in the Red Vienna period in the 20s, it made me wish there had been some larger scale, longer term experiments with having people like Neurath running things. Sadly, people like Neurath do not seem to be an abundant resource. Still, while it is certainly an issue worth thinking about when something has repeatedly defeated efforts at implementation, it is also true that sometimes people find solutions to problems. It is hard to know when something continues to deserve further investigation and when it’s time to just give up.

  50. Multiheaded says:

    “No one gave us this beautiful plane. We built it ourselves, we pulled it out of nothing by our determination and our strength. They tried to crush us over and over again, but we wouldn’t be crushed. We drove off the Whites. We winkled out the priests, out of the churches and more important out of people’s minds. We got rid of the shopkeeppers, thieving bastards, getting their dirty fingers in every deal, making every straight thing crooked. We dragged the farmers into the twentieth century, and that was hard, that was a cruel business and there were some hungry years there, but it had to be done, we had to get the muck off our boots. We realised there were saboteurs and enemies among us, and we caught them, but it drove us mad for a while, and for a while we were seeing enemies and saboteurs everywhere, and hurting people who were brothers, sisters, good friends, honest comrades. Then the Fascists came, and stamping on them was bloody, nobody could call what we did then sweetness and light, wreckage everywhere, but what are you going to do when a gang of murderers breaks into the house?

    And the Boss didn’t help much. Wonderful clear mind, but by that time he was frankly screwy, moving whole nations round the map like chess pieces, making us sit up all night with him and drink that filthy vodka till we couldn’t see straight, and always watching us: no, I don’t deny we went wrong, in fact if you recall it was me that said so. But all the while we were building. All the while we were building factories and mines, railroads and roads, towns and cities, and all without any help, all without getting the say-so from any millionaire or bigshot. We did that. We taught people to read, we taught them to love culture. We sent tens of millions of them to school and millions of them to college, so they could have the advantages we never had. We created the boys and girls who’re young now. We did the dirty work so they could inherit a clean world.”

    (Adam Kotsko’s thoughts and review. Take a look, people.)

    • nydwracu says:

      Speaking of Kotsko, and all the Kotskos of this country…

      But the Soviet experiment had run into exactly the difficulty that Plato’s admirers encountered, back in the fifth century BC, when they attempted to mould philosophical monarchies for Syracuse and Macedonia. The recipe called for rule by heavily-armed virtue—or in the Leninist case, not exactly virtue, but a sort of intentionally post-ethical counterpart to it, self-righteously brutal. Wisdom was to be set where it could be ruthless. Once such a system existed, though, the qualities required to rise in it had much more to do with ruthlessness than wisdom. Lenin’s core of Bolsheviks, and the socialists like Trotsky who joined them, were many of them highly educated people, literate in multiple European languages, learned in the scholastic traditions of Marxism; and they preserved these attributes even as they murdered and lied and tortured and terrorized. They were social scientists who thought principle required them to behave like gangsters. But their successors – the vydvizhentsy who refilled the Central Committee in the thirties – were not the most selfless people in Soviet society, or the most principled, or the most scrupulous. They were the most ambitious, the most domineering, the most manipulative, the most greedy, the most sycophantic: people whose adherence to Bolshevik ideas was inseparable from the power that came with them. Gradually their loyalty to the ideas became more and more instrumental, more and more a matter of what the ideas would let them grip in their two hands…

      Sounds familiar, no? (Hint: replace the concept ‘material goods’ with the concept ‘status’.)

      • Multiheaded says:

        Hmm, sure, knew you’d say that…

        Improved communications technlolgy gives individual philosopher-apparatchik activists the illusion of having a say in the generational flow and the selection pressures… but they have poor/lacking incentives to actually do policing by the most inconvenient and nuanced principles in the book, and it’s much easier to motte-and-bailey one’s way around the worst, and then the offenders stay part of your support base…

        Some cool people do keep trying, but it increasingly does feel that we might even need to renounce the words “SJ” at some point. There’s a slight problem with people already claiming to be against The Toxicity and for “real justice” not being leftists at all and just squatting on the meme to parasitize on the parasites.

        We need a robust internal instituation, not these clowns. (To be fair, the people yelling at them are often very much part of the original problem.)

        • nydwracu says:

          Discovery of social technology to solve the question of Kotskos and Leiters (and Zoe Quinns and worse) is definitely important, especially given that y’all fucked up so hard on SA that… well, it wouldn’t have made a difference in the end, since their particular manifestation of the crocodilian tendency has been here for a hundred years and was bound to be fully unleashed someday — but there aren’t as many chains left unbroken on it as there would have been otherwise.

          (I figure anyone like me ought to own guns and know how to use them, and always keep enough in the bank for a short-notice one-way ticket out of the country. I doubt Leiter’s wish for concentration camps will be fulfilled anytime soon, but Sweden-style acts of terrorism enabled by the media [after the Expressen hack, one of the outed commenters’ houses was bombed] and ignored by the government don’t seem unlikely — nor do crushing economic sanctions, which have already started to form.)

          • Mark says:

            I really don’t think that poem was seriously advocating internment camps for conservatives. I think it was instead trying to lampoon the (perceived) rhetoric of their (perceived) post-9/11 hysteria by flipping it around on them. If that’s the case, then it’s very much coming out against such camps (but also still against conservatives, of course).

            Not sure if you really believe the literal interpretation of the poem was intended, but I’ve seen you remark on it a few times before and that’s the impression I got.

          • Multiheaded says:

            I’ll leave it to senpai to comment on the “crushing economic sanctions”.

          • gattsuru says:

            Sweden-style acts of terrorism enabled by the media [after the Expressen hack, one of the outed commenters’ houses was bombed]

            I thought you were joking. Nope — and apparently not just a random commentor, but a commentor that was a politician, which I’d have expected to be part of the Protected Class.

          • Anonymous says:

            A politician from SD is the opposite of the protected class. Is there any reason to believe that the blog comments caused the bombing, and not just being an SD politician? (Is something being lost in translation? Is this really a politician, and not just a member of the party?)

          • nydwracu says:

            Is there any reason to believe that the blog comments caused the bombing, and not just being an SD politician?

            Here.

            The victim was recently outed by the tabloid Expressen, which published the names of several Sweden Democrat party representatives who had left controversial comments on sites such as Avpixlat.

            I like this part:

            “At the moment, this has been classified as vandalism that endangers the public (allmänfarlig ödeläggelse),” Michael Carlzon, Malmö police spokesman, told Sveriges Television (SVT). “Usually when it comes to explosives, it is classified as aggravated assault, but that could change.”

            I’m assuming that vandalism that endangers the public is a lesser charge than aggravated assault.

          • Multiheaded says:

            Nydwracu: y’know… so that’s apparently all true, but on the other hand, the kind of shit in Stieg Larsson’s trilogy was apparently based in reality too. And on the third hand, Stieg Larsson himself was the kind of guy to go train rebels in Africa.

            So maybe the conclusion is that Sweden is a grim and frostbitten place where people generally act in a grim and frostbitten manner?

          • Anonymous says:

            So, yes, something was lost in translation. He’s not a “politician” but a “party representative.” At least I know that I don’t know the meaning of that phrase.

    • cassander says:

      > but what are you going to do when a gang of murderers breaks into the house?

      The chutzpah of a communist, particularly a communist who drank with stalin, calling any other group of people on the planet “a gang of murderers” beggars disbelief.

      • suntzuanime says:

        Well, I mean, the Nazis? You might try to argue that the Soviets were worse overall than the Nazis, but surely not so much worse that it beggars belief that the Soviets would call the Nazis names.

        • cassander says:

          It’s like mugger calling someone who short changed him a thief. He’s not technically wrong, but you just want to hit him over the head and say “pot, this is kettle, we need to talk….”

          • suntzuanime says:

            I don’t think that the difference between the Soviets and the Nazis is on a scale similar to that between a mugger and a short-changer. I think that it’s absurd if you do.

          • Multiheaded says:

            I think the difference between the Stalinists and the Nazis is the difference between a mean-spirited gang boss and a serial killer who occasionally also robs his victims.

          • Mary says:

            If that were so, Hitler would not be third on the list of mass murderers.

            Also note that the death toll from the USSR includes both before and after Stalin.

        • Weirdly, the Soviets did more damage than the Nazis because the Soviet ideas weren’t quite as bad.

          Nazism was like a disease that kills its host quickly. They believed the best thing was that a smallish ethnicity should conquer the world. Even though the ethnicity was rather capable, the world is a good bit bigger.

          Communism was at least somewhat internationalist, and its ideal sort of person was a worker rather than a soldier. These are much more practical ideas.

          • cassander says:

            >They believed the best thing was that a smallish ethnicity should conquer the world

            not really. the idea that hitler wanted to conquer the world is pure propaganda. According to him (and he is surprisingly a reliable source, upon gaining power he did exactly what he said he was going to do in Mein Kampf), the plan was to make germany a great empire by doing to the slavs what the US did to the indians. This would ensure the german people not world domination, but their place in the sun, to borrow a phrase from an earlier era. Now, of course, you could argue that such an entity would inevitably be a danger to other countries, but if the US fought world war two to prevent one country from dominating all of eastern europe, we did a piss poor job of it.

    • Tracy W says:

      “We got rid of the shopkeeppers, thieving bastards, getting their dirty fingers in every deal, making every straight thing crooked. ”

      Yeah, those people who got supplies from farms and factories to people who needed them, what a terrible lot. Now people march out to the fields to buy their food directly from the farmers, and then march to the shoe factory to get their shoes, and so forth. Okay, they spend all day walking back and forth, and are absolutely exhausted, but hey, we got rid of those shopkeepers.

      “We dragged the farmers into the twentieth century, and that was hard, that was a cruel business and there were some hungry years there, but it had to be done, we had to get the muck off our boots.”

      Because the only reason farmers would go into the twentieth century was if they were dragged? NZ agricultural history strikes me as an obvious counter-example.

      Adam Smith had a good point (if I’m recalling my source correctly), about a lot of city dwellers regarding farmers as ignorant and stupid because farmers aren’t used to being in cities talking amongst large crowds and strangers, overlooking the massive complexity in what successful farmers do.

  51. Bugmaster says:

    That was in 1960. If computing power doubles every two years, we’ve undergone about 25 doubling times since then, suggesting that we ought to be able to perform Glushkov’s calculations in three years – or three days, if we give him a lab of three hundred sixty five computers to work with.

    This is only true if the complexity of the problem remains constant. However, due to the same factors that drive Moore’s Law, as well as general population growth, this complexity also increases exponentially, thus negating some (if not all) of the advantages of Moore’s Law.

    In fact, this is what makes capitalism so attractive (at least, on paper): the calculations required to optimize it are distributed. This means that as the economy (and society in general) grows, so does the computational capacity required to optimize it — because this computation runs on the economy itself, and not on some external device.

    • Paul says:

      Note that you can distribute the computation of communism easily, and that the world doubles once every 15 years, not every 2 years, so if the dependence was anything less than cubic in the size of the economy you are still doing smashingly (the dependence of linear programming is much less than this)

    • Jess Riedel says:

      The computation isn’t done “by the economy”, it’s done by the brains of the people making buying and selling decisions within the economy. Formally, you can always mimic the computation aspect of capitalism within communism by just asking people to make the same hypothetical decisions they would make under capitalism; the raw computational power is the same. However, the answers you’ll actually get will be much worse because those people don’t actually face the consequences of those decisions.

      Capitalism out-performs communism because of incentive reasons, not computational ones.

      • 27chaos says:

        We don’t have computers nearly that good.

        I think that most people are insensitive to incentives. I think people will innovate for very little gain, sometimes just out of pride in their work. Not as much, but sometimes people try to explain the rise and fall of empires by appealing to innovation, which has always thrown me. I can understand that the more innovative empire should outgrow the other, but why should the less innovative empire decline just for not being *as* productive.

        I have in mind Acemoglu and Robinson’s Why Nations Fail when I’m making this complaint, btw. Good book, but oversimplified I think. And it has too few predictive claims, and too many chaos theory inspired ones, for my taste.

        • Tracy W says:

          “I think people will innovate for very little gain, sometimes just out of pride in their work.”

          But innovation by itself doesn’t do much good. You also need production and distribution. Eg, my life was saved when I was 11 years old by a course of antibiotics. That didn’t require just Flemning and Florey and Chain, it also required some factory churning out the antibiotic, and packaging, and a truck driving it to the pharmacy, and the pharmacist handing out the antibiotics, and someone providing the fuel to the truck, and someone making the truck in the first place, etc.

          If that chain, or an equivalent one, hadn’t happened, I’d’ve died painfully, despite the original innovation.

          “why should the less innovative empire decline just for not being *as* productive.”
          People move to the more productive one, for the better standard of living?

          • 27chaos says:

            I agree that production and distribution networks are important. I think that the hard part about them is knowing what should go where and when. I think we’re on the same page here.

            I am doubtful that either social mobility or wars with neighbors would cause getting out competed to death. They played a role certainly, but I think it would be overestimating it to say they cause it. I have no evidence I can think of here, though.

          • Tracy W says:

            I am doubtful that either social mobility or wars with neighbors would cause getting out competed to death.

            How often do empires get out-competed to death? The Spanish are still there. The British are still there. The Romans are still there. The Chinese are still there. The Japanese are still there. The Russians are still there. The Mongols are still there. The Persians are still there (albeit renamed).

            The Aztecs and the Incas could be fairly described as disappeared, but disease had a lot to do with that, didn’t it?

            There probably are some empires I’m not thinking of, as they got out-competed to death, but I don’t think death is an inevitable result of losing an empire.

        • MugaSofer says:

          >We don’t have computers nearly that good.

          We have people. In theory, you could pay them to make these calculations.

          Of course, in practice it’s not quite so easy.

        • Circumstances (climate, neighboring states…) change, so you may need to innovate just to not be worse off.

      • Kzickas says:

        I think you could create a communist system that uses prices for computation. If every business is state owned (that is to say it’s managers don’t have any right to remove money from it, by profit or setting their own wages) you can allow their managers to freely trade with each other and the public while the system is still communistic (is that a word?)

        • Tracy W says:

          But if you can’t remove money from a business, what happens if a new and possibly more efficient option appears? Eg you’re running a buggy whip factory and the motor vehicle comes along.

          Or take the pig raiser wanting the excess wood: how could they trade if money can’t be taken out of either business?

          • suntzuanime says:

            Presumably the state closes down the buggy whip factory and builds a motor vehicle factory. It’s probably easier to notice when motor vehicles are invented than when every little pig farmer needs wood.

            The pig farmer takes the pig farm money he manages and uses it to buy wood from the factory. This is totally legal, because the pig farmer freely trades with the factory manager. The pig farmer and the factory manager don’t keep the profits or eat the loss, but presumably they have some career incentive to get the deal done.

          • Kzickas says:

            “But if you can’t remove money from a business, what happens if a new and possibly more efficient option appears? Eg you’re running a buggy whip factory and the motor vehicle comes along.”

            I said the people running the business couldn’t. The owner (the state under communism) definitely could. Since both the imputs and the outputs of both would be publicly traded it should be reasonably able to tell when it ought to transfer capital from one to the other.

            “Or take the pig raiser wanting the excess wood: how could they trade if money can’t be taken out of either business?”

            The pig farming business would buy the wood from the factory? I’m sorry I don’t understand this objection, businesses being allowed to trade goods freely was part of the assumption.

          • Tracy W says:

            If only the state can take money out of a business, then you’re back at the knowledge problem of the state trying to identify new profitable businesses.

            Note I specified “possibly more efficient”. There’s been a bunch of possible new technologies that have failed for one reason or another (the Sedgeway, at least in mass, the Concorde).

            And if the pig farmer can freely trade for the wood, presumably the management can freely trade for, say, a hot tub in the executive suite?

          • Kzickas says:

            “If only the state can take money out of a business, then you’re back at the knowledge problem of the state trying to identify new profitable businesses.”

            No more so than the lack of knowledge of possible new profitable businesses in capitalism, and since you can accurately gauge the efficiency of a factory if you have real prices then you can solve it in the same way: trial and error.

            “And if the pig farmer can freely trade for the wood, presumably the management can freely trade for, say, a hot tub in the executive suite?”

            That would probably come under the heading of compansation rather than business decision, so it would probably happen, but probably be illegal.

          • Tracy W says:

            No more so than the lack of knowledge of possible new profitable businesses in capitalism

            On the contrary, rather more so. A million people can know far more things than 1 person can.

            And in particular, recognizing new profitable businesses can depend on the skills and knowledge you already have, eg Bill Gates knew a chunk about coding already, and thus was better-placed to guess what might be a successful software business than someone who’d spent their teenage years working on cars, or studying economics, or what not.

        • Hanfeizi says:

          ” If every business is state owned (that is to say it’s managers don’t have any right to remove money from it, by profit or setting their own wages) you can allow their managers to freely trade with each other and the public while the system is still communistic (is that a word?)”

          Congratulations, you just reinvented Deng Xiaoping’s conception of “Socialism with Chinese Characteristics”. His vision was of a combination of huge state-owned enterprises and small village co-ops and small private businesses all operating in an essentially market-driven economy. It’s worked fairly well, which is why superficial comparisons between Soviet economics and contemporary China really don’t wash; but at the end of the day, it’s hard to distinguish it from Western European postwar Dirigisme or the Chaebol/Keiretsu capitalism of the two other Northeast Asian economic powers.

  52. social justice warlock says:

    The book’s greatest flaw is that it never did get into the details of the math – or even more than a few-sentence summary of the math – and so I was left confused as to whether anything else had been possible, whether Kantorovich and Glushkov really could have saved the vision of prosperity if they’d been allowed to do so.

    You might be interested in the work of Paul Cockshott. Here‘s a book outlining his sketch of how to plan an economy, you can also find his more mathy/computer-sciencey stuff around the web. And in fact he had a friendly debate with Spufford after Red Plenty was published.

    • Scott Alexander says:

      Thanks!!!

      • ElTighre says:

        It’s to be remembered that this is no doubt an intensely political subject. I am given to understand that Friedrich von Hayek’s (yes, the economist beloved by Thatcher and Glenn Beck) best work was on the price system, and why centrally planned economies had no hope of successfully aggregating information in a timely fashion to continually and correctly choose both what to produce and how to produce it.

        Writing that, it occurs to me that I would really want this disadvantage to be quantified when compared to the positive attributes of the centrally planned system you listed. However, my original point was that as you research this, you will find a huge contingent of people who are motivated to believe Hayek was correct, and not all of them even on the right.

        I haven’t read Hayek, but I suspect it would be tricky even if I was sure I was capable of A. correctly understanding him and any critics and B. locating flaws in their arguments myself.

        Afterthoughts: the liberal blogosphere generally holds Hayek’s Road to Serfdom to be a work of political economy too reliant on slippery slope arguments to make an effective argument. I’m not sure where Hayek’s price system theory work would be, but that would be all you would be interested in.

        • kaninchen says:

          A very good place to start would be The Use of Knowledge in Society, which is a single paper (perhaps a tad longer than Scott’s median blog post) arguing that central price management cannot match the market for spreading information which may only be known by a few people. American Economic Review selected this paper as one of the twenty best papers they published in their first century of operation.

        • Viliam Búr says:

          Maybe with the help of computers and internet we no longer have the dilemma between “centralized” and “decentralized” decision making.

          To explain what I mean, as a starting point imagine a completely free market, except that all money is virtual, and all payments are done using a single website. Someone puts there “selling apples $0.30 a piece”, someone else clicks “buy 5 apples”, and the deal is done. Assume that everyone has a smartphone with internet access, and this is how you pay literally for everything.

          Now should we call this “centralized”, because all the economy is done on a single website? Or should we call it “decentralized”, because each transaction is decided individually? I think it has some important properties of both. Everyone is free to act according to their personal preferences, and yet the government experts can have precise aggregate data about everything, and can make some global decisions according to it.

          One problem with the Soviet-style planned economy was that people and organizations without political power were unable to express their preferences in a way that would make the system care. Some “shadow money” is necessary for that. But a computer could easily handle many different forms of “shadow money” without problems.

          • 27chaos says:

            http://mises.org/journals/qjae/pdf/qjae16_2_5.pdf

            tl;dr nope

            That said, while pure central planning would definitely fail, the possible idea of hybrid systems is something I’ve been interested in for a while. We shouldn’t assume individual consumer decisions will aggregate towards strong efficiency, if even supercomputers can’t figure it out. Obviously the market does something right, but whose to say how close to perfect it is? Also, the paper doesn’t really address optimization algorithms; though they are nowhere near sufficient they would still be somewhat useful.

            This is one of my pet ideas, in case it’s not obvious.

            One idea I like is that free markets are very good at climbing slopes but very bad at escaping local maxima. Moloch eats anyone who tries to escape.

          • Viliam Búr says:

            From the PDF:

            consider a small community of just 1,000 people with
            1,000 different consumer goods that they are trying to distribute
            among them. When compared to a real economy, this one is
            quite small. However, the system of equations required to solve
            the problem involves 1,001,000 equations

            This fails a simple sanity check. I refuse to believe that humans are able to calculate those equations even with 1/1000 of the speed of the computer — or, equivalently — I refuse to believe that 1000 humans are faster at calculating equations than 1 computer, just because they are distributed.

            What really happens is that those humans don’t really calculate all those equations, but use some simplified version. But then, why does the computer need to use the 1,001,000 equations instead of using similar simplifications, or approximations, or better algorithms, or maybe caching the results of the previous calculations and recomputing only the necessary changes, or…

            This argument simply proves too much. It probably disproves Google.

          • 27chaos says:

            I largely agree with your comment in both senses, perhaps you never saw after I finished my edits. However, finding optimizations for problem solving is extremely hard, so I think it’s worth glancing at the giant numbers.

            I would be interested to see what Google did if it tried to make a centrally planned market.

            Also, your comparison is slightly false in that individual humans don’t have to make decisions about thousands of other people.

            The fact that the humans are distributed also has more importance than you’re perhaps assigning, since their specific location and resources all contain information relevant to the overall market. I’m reminded of Lewontin’s Fallacy, if that makes any sense.

            Another interesting factor to take into account would be the influence of size on the efficiency of free vs. planned markets. No idea what would go on here, but I bet it would be cool. I know some work has been done about city sizes and economic productivity that seems vaguely similar, might be a place to look for inspiration.

          • Tracy W says:

            In other words, with a centralised website, you could nicely replicate a market economy.

            This is quite different from the original hope of central planning which was that it would operate much more efficiently than a market economy.

            And, if you start using “shadow money” to communicate people’s preferences through the system, particularly if you use it to not merely communicate consumer preferences, but also work preferences (eg how much do people like or dislike working outdoors, where you get fresh air but bad weather), then what you’ve got is basically money, not just “shadow money”.

          • nydwracu says:

            What really happens is that those humans don’t really calculate all those equations, but use some simplified version.

            Capitalism computes with input from those humans, sure — but it’s not apparatchiks sitting in a room. The incentive and feedback mechanisms can’t be ignored, and how do you build those into a central computer?

          • haishan says:

            I’m not sure one should take to heart a study in the Quarterly Journal of Austrian Economics about how bad central planning is. Seems like the equivalent of going to the Creation Museum to learn about unsolved questions in evolutionary biology.

          • social justice warlock says:

            The paper in question adds the additional burdensome detail of goods being assigned to consumers by paternalistic computers, not deriving them from choices in a monopsonistic consumer market, as is the case in almost all historical and theoretical planned economies.

        • Quixote says:

          It seems like the difficulty of the task of aggregating and centrally using information is related to the cost / speed of storing and processing information.

          If information needs to be manually written down, mailed and processed by hand by people, the it may be literally impossible to succeed at central planning.

          The more information is automatically cpatured (say by barcode at point of sale) the less hard it should be. The cheaper computation is te less hard it should be. The better your algorithmic sophistication the less hard it should be. The faster information can be transmitted the less hard it should be. All those areas have trends in the same direction. If you project those trends out (allways perilous) then you might eventually see a switch where a decentralized market is no longer the best system.

          For transportation we may allready be at that point, uber is probably better than lots of decentralized private car services.

        • social justice warlock says:

          The technical argument is pretty much settled here: the Austrians were wrong, if you have consumer goods markets everything is quite solveable.

          Road to Serfdom is less original or interesting, but at least retains baseline plausibility, since no representative-democratic government has yet attempted comprehensive planning.

          • ckp says:

            >The technical argument is pretty much settled here: the Austrians were wrong, if you have consumer goods markets everything is quite solveable.

            Got a source on that?

          • Tracy W says:

            How do you distinguish consumer goods from business goods? Is a light bulb a consumer good? Or a PC? Or a plot of land?

            And without markets how do you decide between particular ways of producing resources? If you have a choice between a more labour-intensive but less-steel intensive option and the opposite which one do you pick? If an aluminium plant burns down, who can do without aluminium the best? For a good for which consumer demand is very variable in the short-term how much resources do you have ready for the high demand days?

    • memeticengineer says:

      It seems to me that advanced math or computer programming would not solve all of the failures of central planning, because they often are problems of information and modeling, not of suboptimal calculation.

      This applies even to some of the examples cited.

      The pig farmer: The allocation model had failed to account for the possibility that a pig farmer might require wood as an input and that this may depend on climate. Applying linear algebra or cybernetics to the economy would not automatically put that in the model, nor would it automatically create an avenue for the pig farmer to provide that information. It doesn’t even seem like constructing a formalism would incidentally make you gather information about edge case production requirements like that. So it doesn’t seem like it would help.

      The tire factory: It seems their quota was set without regard to the actual production capacity of the plant. Maybe the formalism would force you to at least consider this. But doing it requires study of the specific facilities the factory has available, a fact that may change over time, which requires a lot more information gathering and thus more work than just arbitrarily decreeing a quota. So a fancier algorithm may make you realize you need to model this, but it would not automatically give you the means or the incentive to gather the information to do so accurately.

      The tire-making machine factory: It’s a non-obvious problem how to measure the productivity of production facility for capital goods or raw materials. Dumb metrics like quantity or weight were used because it’s hard to come up with something more sophisticated, and thus they produce perverse incentives. But just number of machines times productivity/year of each machine is also not a sufficient metric. You have to consider quality of output, cost of building and maintaining fewer more productive machines vs more less productive ones, cost of labor to operate different machines, etc. It is really hard to imagine an exercise in linear programming that would fully balance all the different factors here.

      On the other hand, it seems like decentralization of planning (even partial, piecemeal decentralization) could have helped each of these problems in obvious ways.

      If instead of specific inputs, the pig farmer was given some tokens of exchange (lets cal them “money”) to operate his farm, he could have spent some of them on wood without having to make sure a central model knew he would need it. If instead of a quota the tire factory was given an incentive based on some function of production and cost and freedom to decide how to implement that, they wouldn’t need to sabotage their own equipment to avoid failure. And if the tire-making machine factory was not given a quota but rather had to compete to get factories to use their stuff, they’d make machines that factories consider more desirable if they could get more money that way.

      Now, it might be that these “shadow prices” are an attempt to implement decentralization of this sort, with incentives and localized decision-making authority about resource allocation. But if so, the charge that it’s a way of smuggling in capitalism is true.

      • social justice warlock says:

        But if so, the charge that it’s a way of smuggling in capitalism is true.

        Depends on what you mean by “capitalism.” The normatively relevant part of capitalism that most socialists want to replace is private ownership of the means of production, not markets or prices, hence why “market socialism” is a thing. (In fact, Cockshott’s model includes explicit markets and price mechanisms, although he is not a market socialist and has argued against those who are.)

      • Glen Raphael says:

        The key bit of missing information is not merely THAT the edge-case pig farmer needs wood, but HOW MUCH wood the pig farmer needs. So far as I can tell, no algorithm could determine this. If you just ask this pig farmer, he might say he needs a lot of wood and use the excess to build a bigger house or sell the excess to buy a better car. If every farmer in the nation miraculously starts out perfectly trustworthy and buys into the program, you might initially get honest estimates. But over time some people will tend to overestimate what they need and those people will get richer; it will be discovered that honesty doesn’t pay.

        I don’t see how even a perfectly efficient computer program gets us around this. The use of “money” gives farmers an incentive to economize on resource use; the use of “I think I need this much stuff” claims plugged into a computer program does not do that.

        • Daniel Speyer says:

          The farmer didn’t request wood through official channels because it would mean bugging his supervisor and that seemed politically unwise. Let us assume he was right that it would be unwise. The supervisor wasn’t set up to deal with such requests.

          Imagine the modern variant…

          The farmer opens the red Amazon web page and goes to wood. He selects what he needs. The website reminds him that he might need nails and to consider if corrugated sheet metal might serve him better (these suggestions cone from automated analysis of previous requests). He adds a one sentence explanation of why he needs it and hits Request.

          The computer looks at the quantity, the available woods in the system, the farmer’s karma, the historical tendency of wood to be abused for personal luxury, the longtailnesss of wood usage, and a random number generator. It either fulfills the order or puts it on a supervisor’s ticket queue. The queue is optimized for fast evaluation. The supervisor has buttons for Approve, Deny, Request Clarification, and Flag as Dishonest, but the last one requires more work on his part. The farmer doesn’t have a specific supervisor — instead the tickets are spread evenly over a pool.

          Instead of annoying someone who had other plans, the farmer has a chance of adding a ticket to the workload if whoever was already stuck on tickets that week.

          All fulfilled tickets get aggregated and fed into the regular central planning algorithm for the next cycle.

      • Tracy W says:

        If instead of a quota the tire factory was given an incentive based on some function of production and cost and freedom to decide how to implement that

        If the tire factory doesn’t have a quota of tires, how do you determine the cost functions for the factories that use tires – such as the car factory or the wheelbarrow factory or the factory that makes the trucks that transport the rubber to the tire-making factory?

    • Samuel Skinner says:

      There are many flaws in the pdf. I’ll try to cover the most salient.

      Page 12- “Suppose a worker is employed for 40 hours a week. If he is not to be exploited, then the wages he gets for that work should allow him to buy goods and services that took 40 hours of work to produce.”

      This is completely unworkable for goods that require different skill levels, machinery or value of raw material to produce.

      Page 12- they measure “value added by labor” by the difference between sales and cost of non-labor inputs. However the cost of non-labor inputs is not the same as the value they add; otherwise value added by labor would decrease as electricity became more expensive. They are assuming marginal benefit= marginal cost for inputs, which is not true for the average.

      This means they overestimate the value of labor to production.

      Page 13- It is unclear how they measure profit. Does management count as wealth producers under their measure or not?

      Page 15- “The proceeds of exploitation are devoted to two main purposes. They are either distributed as dividends or interest payments, or they are used to finance capital accumulation on the part of the company.”

      If capital accumulation is based on exploitation, it is unclear how socialism would eliminate this and, if this is an inherent feature of economics, why it should be called exploitation.

      Page 15- asserts that wealth is retained intergenerationally

      Page 16- assumes that the reason England doesn’t return to policies employed from 1945-1970 is because the rich wish to exploit the poor. Fails to notice that the golden age occurred in multiple countries and broke down; it was caused by a high growth rate and specific domestic conditions due to WW1 and 2. It is not repeatable.

      Page 30- The solution to getting people to be willing to get an education when pay is equal is to count student as a form of productive employment.

      Page 31- And since that will result in shortages for the harder fields we will use “directed employment”. Nothing like being ordered to volunteer!

      If you want a full overview, ask, but as it stands the authors were tone deaf to economics and irony. They advocate giving everyone the same income, except some people should have higher incomes to encourage individuals to work those jobs.

      • social justice warlock says:

        Page 12- “Suppose a worker is employed for 40 hours a week. If he is not to be exploited, then the wages he gets for that work should allow him to buy goods and services that took 40 hours of work to produce.”

        This is completely unworkable for goods that require different skill levels, machinery or value of raw material to produce.

        The Marxist concept of socially necessary labor time already includes those – you count both the direct hours worked plus the wear and tear on the machinery (which took a certain amount of social labor to produce, &c.)

        Page 12- they measure “value added by labor” by the difference between sales and cost of non-labor inputs. However the cost of non-labor inputs is not the same as the value they add; otherwise value added by labor would decrease as electricity became more expensive. They are assuming marginal benefit= marginal cost for inputs, which is not true for the average.

        This means they overestimate the value of labor to production.

        This is a very well-phrased objection, and while I disagree with you, it merits a longer answer and I don’t have the time or energy to clear the inferential distance at the present moment. Feel free to pester me about this/regard it as an unusually whiny and pompous admission of defeat until I do, &c.

        Page 15- “The proceeds of exploitation are devoted to two main purposes. They are either distributed as dividends or interest payments, or they are used to finance capital accumulation on the part of the company.”

        If capital accumulation is based on exploitation, it is unclear how socialism would eliminate this and, if this is an inherent feature of economics, why it should be called exploitation.

        You’re affirming the consequent. If rabbits become mainly pets or stew, it doesn’t follow that stew must be rabbits.

        Page 15- asserts that wealth is retained intergenerationally

        Yes?

        Page 16- assumes that the reason England doesn’t return to policies employed from 1945-1970 is because the rich wish to exploit the poor. Fails to notice that the golden age occurred in multiple countries and broke down; it was caused by a high growth rate and specific domestic conditions due to WW1 and 2. It is not repeatable.

        I don’t see the contradiction.

        Page 30- The solution to getting people to be willing to get an education when pay is equal is to count student as a form of productive employment.

        Page 31- And since that will result in shortages for the harder fields we will use “directed employment”. Nothing like being ordered to volunteer!

        He discusses a variety of incentives and rejects “being ordered to volunteer.”

  53. suntzuanime says:

    So what you’re saying is, it’s not that Communism failed, it’s that Communism has never really been tried.

    • Scott Alexander says:

      I know you’re not serious, but…

      This reminds me of something Kenzi said this weekend in California, which is that her least favorite kind of CFAR applicant is the one who says “I have come up with the optimal plan for how to improve my life and the world, but instead of doing it I just sit around and play video games. I want to go to CFAR to gain the ability to carry out my optimal plan.”

      Her commentary was that if the plan is impossible to follow, then it’s not optimal. “Ability to be carried out” is one of the criteria to optimize for.

      I think there’s a lot of philosophical complexity contained in this territory but that there’s at least one sense in which she’s right, and it’s the same sense in which you’re not allowed to say “Communism works fine, except for the part where every time you try it you get something other than communism.”

      • 27chaos says:

        How many times have we tried it? Do we count the USSR satellite states as one or many or somewhere in between?

        IIRC the historical objection to democracy was that practically every time nations attempted it they collapsed into anarchy. But it turns out that democracy can actually work fairly well, if you go about it right.

        • Geirr says:

          “IIRC the historical objection to democracy was that practically every time nations attempted it they collapsed into anarchy. But it turns out that democracy can actually work fairly well, if you go about it right.”

          Going about it right demands a mass literate population and a state ruled by, or at least highly influenced by the bourgeoisie (but I repeat myself). I’m also not sure many people want real rule by the people, Jacksonian democracy was brutal and corrupt but reflected the will of the people quite well.

          Venice and Switzerland were long running republics but calling Switzerland a democracy before ~1900 is stretching the term very far.

          To me it looks like representative democracy showed up more or less as soon as it was socioeconomically viable and the socio is doing very little work there. It was mostly a matter of technology. Democracy arose at the same time as self-conscious nationalism and for the same reasons.

        • Nornagest says:

          There were an awful lot of communist revolutions. A number of these produced, or eventually led to, Soviet client states, but not all; even in Europe you had Yugoslavia under Tito, and outside of Europe the Soviets tended to let Communist-aligned states do their own thing as long as they supported Soviet foreign policy goals and didn’t look like they were getting too chummy with the West (or, to a lesser extent, China).

          On the other hand, most of those revolutions, even when they developed largely independently of Soviet or Chinese influence, still tended to profess Leninist or Maoist interpretations of Communism. Yugoslavia is again a notable exception. Cambodia under Pol Pot was another.

          • Auroch says:

            Where does Cuba fall? I know that up to the late 80s American leftists still considered it a shining example of Communism Done Right, and that today it’s been gradually drifting towards capitalism. But I don’t know much about whether those American leftists were justified or how popular its government was.

          • Nornagest says:

            I don’t have a particularly deep knowledge of Cuban history, but my understanding is that its post-revolutionary government started out relatively independent, grew closer to the Soviet model starting in the early Sixties (following the American embargo and attempted invasion), and started going its own way again after the fall of the Soviet Union — though somewhat less so than most of the other nations that still profess Communism.

            A more detailed analysis would take information I don’t have. But it does seem to have turned out better than most states in similar situations.

    • Matt C says:

      Indeed. Not only not tried in Soviet Russia under various leaders, but also not tried in many other places around the world to similar effect.

      • JB says:

        Communism had another failure mode that deserves consideration, because it strongly weighs on the observed evidence of the performance of communist countries, but rarely is taken into account.

        Throughout the cold war, any time a country attempted to democratically elect a socialist government, they had a high likelihood of being subverted by an American-backed military coup, assassination, or other form of covert regime change. The US worked very hard and successfully to prevent communism from being experimented with.

        http://en.wikipedia.org/wiki/Covert_United_States_foreign_regime_change_actions

        As a result, many communist states failed before even starting, and the types that made it through that filter may have tended more strongly toward the madman dictatorship end of the spectrum than the constitutional democracy end.

        • Multiheaded says:

          Exactly what I’ve been repeating ad nauseam, but nobody’s listening. For a more extreme example, see the Spanish revolution.

          • The Anonymouse says:

            I’m listening! 🙂

            And the claim may or may not be true. But if your starting premise is “Assume a world with no competing state actors…” we’re not really talking about anything relevant, are we?

          • cassander says:

            because it’s complete and utter nonsense. the spanish revolution, for example, started mass murdering almost immediately, tens of thousands of priests and nuns murdered for the crime of being catholic. and there were a great many social democratic states that did just fine, like most of europe.

    • Sniffnoy says:

      It’s interesting that here Scott is going on about the possibility of Communism in the sense of a centrally-planned economy, when in the last Communism thread, Anish was going on about how doing Communism properly would be all about decentralized coordination mechanisms. Of course, we already have one of those: It’s called the free market! No Communism necessary. (I mean, except for the part where we also restrain it somewhat with a government, but A. it’s not a Communist government, and B. it’s certainly not what Anish wants because government is centralized.) But if there are better decentralized coordination mechanisms out there, it would be interesting to see.

      • social justice warlock says:

        Surely we can agree that there are multiple kinds of decentralized coordination mechanisms (as well as multiple forms of property and distribution thereof a free market might trade in.)

        • Sniffnoy says:

          Certainly! I don’t mean to imply that the free market is optimal among decentralized coordination mechanisms. Just that pointing out decentralization as a distinctive feature seemed strange, when historically the case has been closer to the opposite.

  54. Pingback: Gunlord500