Krugman Mangles Smith

September 29, 2010

by Gene Callahan

Here’s Paul Krugman, explaining the meaning of Adam Smith’s pin factory, and why it opposes Smith’s invisible hand metaphor:

“What may not be obvious is the way these two concepts [pin factory and invisible hand] stand in opposition to each other. The parable of the pin factory says that there are increasing returns to scale — the bigger the pin factory, the more specialized its workers can be, and therefore the more pins the factory can produce per worker. But increasing returns create a natural tendency toward monopoly, because a large business can achieve larger scale and hence lower costs than a small business. So in a world of increasing returns, bigger firms tend to drive smaller firms out of business, until each industry is dominated by just a few players.”

And, of course, this monopolistic competition wrecks the operation of the invisible hand, per Krugman.

What is shocking here is that Smith very explicitly denies that an extensive division of labor requires monolithic factories — in fact, he states that it is only in “trifling” instances that this will happen:

“It is commonly supposed to be carried furthest in some very trifling ones; not perhaps that it really is carried further in them than in others of more importance: but in those trifling manufactures which are destined to supply the small wants of but a small number of people, the whole number of workmen must necessarily be small; and those employed in every different branch of the work can often be collected into the same workhouse, and placed at once under the view of the spectator. In those great manufactures, on the contrary, which are destined to supply the great wants of the great body of the people, every different branch of the work employs so great a number of workmen, that it is impossible to collect them all into the same workhouse. We can seldom see more, at one time, than those employed in one single branch. Though in such manufactures, therefore, the work may really be divided into a much greater number of parts, than in those of a more trifling nature, the division is not near so obvious, and has accordingly been much less observed.”

In fact, as Smith’s example of the laborer’s woolen coat shows, he thought of the real division of labor as being carried out across the whole globe, across many, many years, and among a multitude of enterprises. Whereas Smith saw the progress of the division of labor as depending on the “extent of the market,” Krugman translates that to “the size of the factory”!

Now, perhaps Smith was wrong, and an increasing division of labor eventually really does require giant, monopolistic enterprises, as Krugman sees things. But it is not very informed or honest (take your pick) of Krugman to suggest that this is the straightforward meaning of what Smith wrote.

45 Responses to “Krugman Mangles Smith”

  1. Daniel Kuehn Says:

    I think you’re conflating two issues here – the division of labor and increasing returns. Krugman only said the later leads to larger firms which, holding everything else equal, it certainly does.

    To use your phrasing, Krguman saw the progress of monopolization as depending on “increasing returns”, and you translate that to “division of labor”.

  2. Bill Stepp Says:

    PK writes:

    “But increasing returns create a natural tendency toward monopoly, because a large business can achieve larger scale and hence lower costs than a small business. So in a world of increasing returns, bigger firms tend to drive smaller firms out of business, until each industry is dominated by just a few players.

    “But for the invisible hand to work properly, there must be many competitors in each industry, so that nobody is in a position to exert monopoly power. Therefore, the idea that free markets always get it right depends on the assumption that returns to scale are diminishing, not increasing.”

    Rubbish, as a Brit economist would say.
    Even if bigger firms drive smaller firms out of business, this doesn’t prevent new competitors from popping up. Wal-Mart drove out a lot of mom-and-pop competitors, but a lot of new retailers have started since Sam Walton founded his firm in 1962. Ditto for McDonalds. I see new burger places popping up from time to time in New York. Apparently these retail and burger entrepreneurs never got the memo.
    This is the economics equivalent of historical determinism.

    And what’s the exact definition/number of “many competitors in each industry”?
    It’s not free markets that get it right or wrong, but individual entrepreneurs, who compete in a free market (except those who belly up to the public trough, and get goodies paid for by taxes forked over by the likes of Wal-Mart and McDonalds).
    His last sentence is wrong.

    Krugman is a pseudo-economist. I still can’t figure out what he did that was Nobel Prize worthy, or even valid.

  3. gcallah Says:

    “I think you’re conflating two issues here – the division of labor and increasing returns. Krugman only said the later leads to larger firms which, holding everything else equal, it certainly does.”

    Sorry, Daniel, I think it’s Krugman who is conflating two issues here. Smith quite explicitly offers the pin factory as an example of the division of labor. It is Krugman who slides from one two the other: “the bigger the pin factory, the more specialized its workers can be, and therefore the more pins the factory can produce per worker.”

    Which was my very point.

  4. Daniel Kuehn Says:

    Bill Stepp –
    I wasn’t aware of any point where he said that competitors couldn’t emerge – only that the extent of monopoly power is related to scale economies.

    Markets “getting it right” or “getting it wrong” is somewhat vague – but it’s a NY Times article, not a journal article. I don’t really see why you feel the need to read it the way you do. Doesn’t he just mean that the more monopoly power there is, the more we’re going to depart from competitive equilibrium. I never read “monopolies are bad”. I never read “you need millions of small firms for competitive solutions” (a straw man that you often hear people complain about but you rarely hear anyone assert!). I just read a layman’s version of “the more monopoly power, the lower quantity and higher prices you’ll get relative to competitive equilibrium”.

    It’s amazing to me this FOUR YEAR OLD article is so offensive to people.

    And if you read the article – he’s making the case that economists should pay more attention to Smith (and Schumpeter too, for that matter)!

  5. Daniel Kuehn Says:

    Sorry, Daniel, I think it’s Krugman who is conflating two issues here. Smith quite explicitly offers the pin factory as an example of the division of labor. It is Krugman who slides from one two the other: “the bigger the pin factory, the more specialized its workers can be, and therefore the more pins the factory can produce per worker.”

    Smith notes that increasing returns to scale and the existence of a wide market allow for the division of labor, which is more efficient. The pin factory is an example of the division of labor, but it’s able to come about because of increasing returns.

    We’re really dealing with three economic phenomena operating on Smith’s example:
    1. Market demand
    2. Productive technology (ie – increasing returns), and
    3. Division of labor (ie – the institutional response to the first two in a market economy).

    Krugman cites the example in toto, and then focuses on the increasing returns element of it. What could possibly be wrong with that? How could this possibly incite controversy???

  6. Bill Stepp Says:

    Daniel Kuehn,

    The term “monopoly power” makes no sense in descibing a firm operating without a government grant of monopoly power, which always means explicitly restricting competition. Think of the USPS. Now name one free market monopoly, that is, a firm that is a sole supplier of some good or service.
    To quote an enonomics professor of mine, “we can wait all afternoon.”

    In fact, as Murray Rothbard pointed out, the term “power” should be confined to whatever the State does. As Mao put it, power grows out of the barrel of a gun.
    An example occured about a month ago here in the Soviet People’s Republic of New York. An ex-“offender” had been released from jail and was supporting himself and his pregnant wife by selling some sort of iced desert product out of a cart in Tompkins Square Park. Along came two cops, who demanded to see his purveyor’s license, which he didn’t have. They shut him down and stole his cart. I wonder how he’s making a living now?
    The cops can do this, because they work for the criminal-monopoly entity known as the State, which employs a monopoly police farce to enforce its monopoly statute law book, using Maoist “barrel-of-a-gun” tactics.
    You won’t see Krugman ever leap to the defense of someone who has the temerity not to comply with the Maoist State’s diktats.
    No, Krugman is an advocate of the State, the bigger the better, and an enemy of the free market.

  7. Daniel Kuehn Says:

    Are we talking about “monopoly power” or are we talking about a “monopoly”. I’d agree that a strictly defined “monopoly” is probably only a product of government intervention (or a temporary phenomenon). It might not even be a real monopoly then.

    But none of that semantics makes discussion of monopoly power or pricing irrelevant.

    As for “Maoist state diktats” – I don’t even want to touch that one. Feel free to wait all afternoon if you want.

  8. Bill Stepp Says:

    What’s the definition of “monopoly power”?
    Can a firm that isn’t a monopolist exercise it? I don’t see how it could.
    Sounds like Orwellian double-speak to me.

  9. Daniel Kuehn Says:

    Ever heard of “monopolistic competition”? It’s not Orwellian double-speak. It’s jargon at the very worst, but it shouldn’t be especially insurmountable jargon for commenters on a page run by an economics professor.

  10. Troy Camplin Says:

    Apparently Smith thought his readers were going to be smart enough to understand that “increasing returns” didn’t mean into “infinity and beyond.” In the real world — to which Smith was referring, and not to the Krugman fantasy world, where anything can happen according to his mood — increasing returns flatten out. While one company is in the flat portion, another is on the exponential portion, while yet another is engaging in entrepreneurial innovation to allow it to gain even more increasing returns. The monopoly is undermined everywhere (except the government monopoly, which Krugman loves so very much). Krugman is an enemy of truth. Or he’s not smart enough to understand Smith. Take your pick.

  11. Jerry O'Driscoll Says:

    Krugman and Kuehn got it wrong and Callahan has it right. Smith was talking about the division of labor. That is an entirely separate issue from returns to scale.

    There are diminishing returns in the short run. In Smith, supply and demand curves shift dynamically over time. In manufacturing, that might result in observed prices being constant or even decreasing over time. The latter would not properly speaking be increasing returns to scale, and it would not imply monopoly. Monopoly comes from state intervention.

  12. Glen Says:

    On reading Krugman’s essay, I’m with Gene here. Krugman is misunderstanding Smith. Krugman says there is a contradiction between division of labor (and he does use that term explicitly) and the invisible hand (meaning the beneficial effects of competition). He perceives a contradiction because he sees the firm as the only means of achieving a greater division of labor. But as Gene and Smith make clear, the division of labor can be and often is achieved by decentralized processes. By looking at a single firm, you will only observe a small fraction of the total division of labor in the market.

    So what about increasing returns? Division of labor does, indeed, allow for increasing returns. Smith doesn’t use that term, as far as I know. But his logic still applies. We can achieve increasing returns through greater division of labor, and that division of labor does not necessarily require centralization.

  13. Daniel Kuehn Says:

    Jerry and Gene – what do you all think of George Stigler’s 1951 paper on this, then (JPE I think?)?

    I’m really surprised this was that controversial. This is essentailly what Stigler said too, isn’t it. Firm size and division of labor increase when there are increasing returns to be exploited. Holding all else equal, that means monoply power is going to increase. Since when has that been controversial?

    Troy is right that you eventually exhaust scale economies (Troy is wrong to say that Krugman says they go on ad infinitum, of course – and it’s telling he can’t cite Krugman saying that), but that doesn’t mean that firm size and the division of labor don’t increase as the point at which scale economies are exhausted increases.

    Jerry – I’m not sure what observed long run price behavior is supposed to have to do with the question. Presumably production get’s more efficient over time. That’s going to be a function of technological development and a host of other factors. It doesn’t change the fact that the returns to scale are going to affect (1.) firm size, and (2.) the division of labor. And Smith’s whole point is that a pin manufacturer supplying lots of pin users more efficiently than single pin users can supply themselves. How in the world is that “separate form” increasing returns. That is increasing returns to scale in action, and driving the division of labor – just as Stigler pointed out.

  14. Daniel Kuehn Says:

    *I noticed in my third paragraph I wrote “scale economies” accidentally. Obviously that’s closely related to returns to scale, and seems like it shouldn’t make a difference at all on the issues that are being discussed here – but clearly a little different from the returns to scale that Krugman mentions. Sorry about that.

  15. Troy Camplin Says:

    It is implied in his belief that “increasing returns” necessarily leads to monopoly. Krugman is using an enthymeme, meaning he is leaving out the minor premise, but that doesn’t mean it’s not implied by the argument. Of course, one of the benefits of such a structure is that people can then say, “Well he (I) didn’t actually say that.” Well, of course he did — it’s implied in the logic of the argument itself.

  16. Daniel Kuehn Says:

    Troy –
    That sort of case is used to say Krugman thinks war is good for the economy too, and yet he’s recently had to walk those people back to the beginning and clarify the obvious – that he thinks no such thing. In other words – I acknowledge the rhetorical strategy you point out, but it seems to be so frequently misapplied to Krugman that I think the burden is ultimately on the accuser, at least in my mind.

    Could you explain why “it’s implied in the logic of the argument itself”? I don’t see how it is. I, for example, made the same point about increasing returns while explicitly noting that increasing returns will be exhausted. So clearly the logic of the argument doesn’t require what you claim it requires. The question then becomes “why does Troy Camplin consistently resort to the most damning and the most ridiculous sounding version of what is left implied?”.

    If the logic of the argument unambiguously implied something, that would be a different story – but it doesn’t here.

  17. Jerry O'Driscoll Says:

    We have Smith. Why do we need to read Stigler on Smith? And let’s not “read backwards” into Smith.

    Smith was interested in innovation and technological progress. As I already noted, they shift cost curves (supply) out. That doesn’t represent economies of scale.

    Is the division of labor separate, or part of the innovation process?

    On cost and output, I would turn to Alchian and his article of that title (or close to it) — not Stigler.

  18. Current Says:

    Daniel. In your description increasing returns to scale is the starting point and that leads to larger firms and more division of labour. Why do you think of it in that order?


  19. Would it be tacky to ask for text to support these interpretations?

    Here is text (Liberty Fund edition): “The division of labour, however, so far as it can be introduced, occasions, in every art, a proportionable increase of the productive powers of labour” (15).

    Agriculture is part of the division of labor, but agriculture is subject to diminishing returns (16).

    The use of machinery is associated with the progress of the division of labor (19-21). That is an increase in labor productivity.

    The division of labor is linked to “improvements in machinery” (21). That is technical progress, not increasing returns to scale (the shifting supply curves again).

  20. Eric Hosemann Says:

    Okay–so the division of labor leads to “monopoly,” whatever that means. Would Daniel and Krugman argue that technological advance does too? Or maybe international trade? Since according to Smith the division of labor is limited by the extent of the market, and if dividing labor plants the seeds of monopoly, then extending a market across borders causes monopoly too.

  21. Eric Hosemann Says:

    …and–pardon the officiousness of two subsequent entries–isn’t comparative advantage a form of the division of labor at the global scale? So doesn’t that, in some sense, create a “natural tendency towards monopoly?”

    Krugman’s best work, so I’ve been told, involves shining light on the benefits of international trade and specialization, and apparently, circa ’06, shilling FOR monopoly!

  22. Bill Stepp Says:

    Presumably Wal-Mart has increasing returns to scale, but the retail business has low margins (on a good day). Wal-Mart also gives its customers “everyday low prices.”
    The retail business is certainly not characterized by a few big firms. On the contrary, it exhibits lots of creative destruction and is extremely competitive.

    Could you (or PK) give an example of an industry that is like what he writes about in the article? (Industries with state-granted monopolies don’t count.)


  23. In retail, distribution is the source of economies of scale. Everything else has diseconomies. Otherwise we’d all be dealing exclusively with Walmart.

  24. Daniel Kuehn Says:

    Current –
    I don’t think of it in that order. You’re not going to have division of labor unless there is a prospect of increasing returns – but usually it’s the division of labor itself that makes the increasing returns possible, right? Don’t you agree?

    Jerry –
    Why do we need Stigler? Cause he was a smart guy. Just because a smart guy wrote some smart stuff in the 18th century doesn’t mean I’m going to jettison others in the 20th that thought about what that guy in the 18th century said. You seem to think the same thing with Alchian (and I would agree on him too), so I’m not sure why you would ask “why do we need Stigler”. The division of labor is absolutely part of the innovation process. It’s probably the most important part. Eureka-style innovations have their place of course, but a lot of innovation occurs in the midst of the market process and as a result of the specialization inherent in the division of labor.

    Why do you separate technological progress from increasing returns? “Increasing returns” is just a way of talking about how we exploit the technology that we have at our disposal. I’m not sure why you’re acting like these are separate.

  25. Daniel Kuehn Says:

    Eric –
    There are counter-vailing forces as well. You are doing what Troy has done when you speak as if these processes go on ad infinitum. Increasing returns get exploited through market processes over time. International trade, insofar as it increases demand, clearly motivates the division of labor. John Mclaren at UVA has written some about that.

    You ask “does technological development too?”. Doesn’t this depend on what kind of technological development we’re talking about? If it’s the sort of technological development that allows more efficient production at larger scales, you’re going to see fewer, larger firms. If it’s another sort of technological development you’re not. This is practically tautological.

    As for your “shilling for monopoly” point – you act as if “monopoly” is a bad thing. Do you think it is? The economy simply behaves differently when increasing returns are pervasive – that’s the thread that runs through Krugman’s work. I think you’re inferring value judgements that aren’t there, and I’m not sure why you’re doing that.

  26. Daniel Kuehn Says:

    Could you (or PK) give an example of an industry that is like what he writes about in the article? (Industries with state-granted monopolies don’t count.)

    I was talking with a friend of mine the other day who does electronic discovery and database work for law firms. It’s a relatively new field – he deals with all those millions of emails that companies have to account for in court, etc. Over the course of a decade or so, as you might expect, demand for these sorts of services increased dramatically. The kind of expertise required for the work and the substantial demand for it by all law firms allowed his company to grow very rapidly, and compete with only a few competitors. There were increasing returns to this sort of database work – more efficient for one firm to do the work for a bunch of law firms that were short on computer experts than for them to do a sloppy or ad hoc job themselves. So his company was, for a time, one of the few in the business – and it was one of the biggest providers of this service.

    Now, he was telling me, electronic discovery has become so important for lawsuits that their clients are now actually needing these services at sufficient scale to make it affordable to bring it back in house. Not all of them, of course, but many of the big ones. Pricewaterhouse Coopers is one in this area, for example, that is starting to do all its own electronic discovery work.

    So increasing returns to this sort of business initially afforded the scope for monopolization of the industry and the division of labor – and the increase in demand sufficient for individual firms to supply their own needs at an efficient scale is now scaling back the division of labor.

    How’s that for an example? All “increasing returns” means is “it’s more efficient to produce at greater scale”. All Krugman and I are saying is “if it’s more efficient to produce at greater scale, firms will produce at greater scale”.

  27. Troy Camplin Says:

    Daniel,

    Good job of accusing me of saying literally the opposite of what I said. I said that these processes *couldn’t* go on ad infinitum, while Krugman clearly assumes they could. If you’re going to criticize me, what say you get me right, hmm? It is Krugman who assumes they go on ad infinitum, until monopoly is created; I argued that he is wrong, that they cannot do so, and that therefore monopoly is not going to result. Or do you not even remember your argument against me, above?

  28. Daniel Kuehn Says:

    One of the things that always complicates discussions of specialization is that for Smith the division of labor really operated on two levels: the division between pin-factories and pin-users (ie – division of labor as outsourcing), and the division of labor within the pin factory itself (ie – more of a “scientific management”/assembly line version of the division of labor). As Eric points out above – comparative advantage is just an elaboration of the division of labor. The two kinds of division of labor are related and go hand in hand, of course.

    This is, again, why I would advocate reading Stigler to understand Smith. And I’d advocate reading Coase and Williamson to understand Smith as well. Stigler looks at “division of labor as outsourcing”, Coase looks at “division of labor within the firm”, and I suppose you can think of Williamson as putting both of those back together. You get the same pin-factory story, more rigorously. Alright – I’ve said more than enough at this point.

  29. Troy Camplin Says:

    Daniel,

    You don’t even know what the term “division of labor” means. Just because it is being done inhouse, that doesn’t mean that the labor isn’t being divided. Division of labor is not the same thing as “different corporations doing it.”

  30. Bill Stepp Says:

    I know a bit about the industry Daniel refers to, and there are several companies in it. A monopoly, to repeat what I said earlier, means a SINGLE supplier. Not two or three or four, etc. but ONE.
    Think of the USPS, formerly known as the US Postal Service, before the term “going postal” acquired a new meaning. If you try to deliver something to someone’s (supposedly government-owned) mail box, you can be charged with a felony. That’s a libertarian-certified monopoly.
    So the industry he talks about is not a monopoly.

    Krugman talked about one with a few competitors. That is not a monopoly.

  31. Daniel Kuehn Says:

    Troy –
    You might want to read the post where I agree with exactly what you just said. Division of labor itself doesn’t require contracting for that labor, although it can. But if we want to think about the relationship between the division of labor and firm size (which is the subject of the discussion, after all), we have to talk about both the division of labor and when it is and isn’t contracted for. In other words – we have to put Smith, Stigler, Coase, Williamson, and others together.

    Bill –
    All we’ve been talking about is (to use Krugman’s words) a “tendency towards monopoly”, not a strict, one-firm monopoly. I’m not sure why you’re so hung up on that. Nobody said there would just be one firm – the point is, monopoly power will increase. Monopoly power isn’t contingent on a single firm. Monopoly power is a factor in oligopolistic situations and monopolistic competition situations. Why do you think this “is it just one firm or not?” thing is so important? Analytically, it seems entirely trivial to me.

  32. Troy Camplin Says:

    Go back and re-read what you just said, which contradicts what you say now. I think, if you look at precisely what you said, without knowing what you intended to say, that you can see why I argued that you said what you said. You appear to contradict yourself quite often, precisely because you do this sort of thing.

  33. Daniel Kuehn Says:

    Troy –
    You seem to be reading my earlier comment with the understanding that the only division of labor Smith ever talked about was the division of labor between individuals. That’s the pin factory, of course. But Smith also talked about different establishments – the brewer, the baker and the butcher. Outsourcing is part and parcel of the division of labor, just like more specialized jobs within a firm are. See Eric’s comment above, too, about how comparative advantage is a sort of division of labor.


  34. Gene’s post concerned what Adam Smith said. The only text for that question is Adam Smith. Daniel has once again sent the discussion off track.

  35. Daniel Kuehn Says:

    Jerry –
    It’s not getting things off track when considering the question “what did Smith say?” to look to what other people said on the question “what did Smith say?”.

    If that was “getting off track”, there’d be no point in reading what Gene has to say about what Smith had to say in the first place! We’d all just sit on our own reading Smith. Smith is a good read, granted, but that doesn’t sound very interesting.

  36. paul e. Says:

    A charitable interpretation of Krugman might be that he was trying to get across the idea of increasing returns being incompatible with perfect competition. In the neoclassical model, p.c. is the epitome of the “invisible hand” argument. But he couldn’t spell out the theoretical nuances in a general interest article. Adam Smith’s pin factory was just a neat way to get across the core idea – increasing returns is not compatible with perfect competition – to lay audience who don’t care about all the underlying theory.
    If Krugman was addressing the History of Economic Thought conference, he might have said things another way.

  37. serge d'agostino Says:

    Sorry but I don’t read in Krugman’s paper what you read (if I well understand your conclusion). In his paper, Krugman says that he sees an opposition between the model of the pin factory and the one of invisible hand. I don’t see a sentence… where he asserts that Smith says that division of labor will tend to monopoly. The Krugman’s criticism concerns the internal logic of Smith’s theses, that’s all. For Krugman Smith was wrong because he forgot the incidence of increasing returns in scale. Every author has the right to make that kind of criticism; it’s not dishonest to do so.
    PS : You’re right when you say that, for Smith, it’s the size of the market that leads or not to division of labor. But it’s obvuious that Krugman agrees such an idea. His paper doen’t deny that, it deals with something else

  38. Current Says:

    Daniel,

    I think that we all agree that there are two different types of division of labour. One between individuals within firms, which Smith discussed with his example of the pin factory. The other being between firms, which Smith discussed with his example of the wollen coat.

    The problem here is that Krugman is claiming that division of labour itself negates the invisible hand. He is claiming that it tends towards monopoly. That isn’t true. If the only sort of division of labour that is possible is that within firms then that will increase the returns to scale of firms, and that will reduce competition. However, division of labour isn’t necessarily within the firm, it can happen generally too and be exploited between firms rather than within them. In that case it doesn’t increase returns to scale and may just as easily increase competition as reduce it.

  39. Allan Walstad Says:

    Krugman quoted by Callahan: “But increasing returns create a natural tendency toward monopoly, because a large business can achieve larger scale and hence lower costs than a small business. So in a world of increasing returns, bigger firms tend to drive smaller firms out of business, until each industry is dominated by just a few players.”

    Actually, this is correct as far as it goes. If increasing returns were the ONLY factor — if there were no innovations in technology or organization, no changes in taste, no discoveries of new resources or depletion of old, no tendency for monopolies to get fat and lazy and thereby vulnerable to lean, mean upstarts — then sure, monopoly might be the end result. Indeed, in that case, even the government, i.e., the ultimate monopoly, might eventually “get it right” and socialism would work.

    But that’s not the real world. In the real world, there is technological and organizational innovation and there are changes in tastes and resource availability, etc.; then the relevant question is, who can adapt most quickly? It ain’t generally the dinosaurs. The obstacle to clear thinking here is the notion — very nearly an oxymoron — of “competitive equilibrium.”

  40. Daniel Kuehn Says:

    serge –
    I think you’re on the right track, but I think this is even a little strong: “For Krugman Smith was wrong because he forgot the incidence of increasing returns in scale.”

    I don’t see where he says that “Smith was wrong”. All he’s saying is that there are countervailing forces at work. Increasing returns increases firm size which increase monopoly power in markets, holding all else equal. But division of labor (and increasing returns) fosters growth too.

    There are a number of forces under the surface of the pin factory which work in different directions simultaneously. That’s not invalidating Smith – that’s just showing the richness of Smith’s observations – it’s a richness that Krugman argues has been largely abandoned for a lot of the history of economic thought.

  41. serge d'agostino Says:

    @ Daniel
    I badly expressed myself (sorry my english is not so good!). I wanted to say that the invisible hand could not function well when economies of scale are increasing. When I wrote “Smith Was Wrong”, it was to express my doubt face the supposed virtues of the invisible hand.
    Of course that Smith’s analysis is rich: for example his thesis about free trade that does not prevent him to show that protectionism may be desirable in some important circumstances. Same thing for the role of State in a market economy. Etc.

  42. Eric Hosemann Says:

    Let’s apply thinking about the division of labor and increasing returns to Krugman himself. Oh, and there’s only one division of labor–mostly because there’s only one labor–work, the world’s most unappreciated disutility.

    Let’s also pretend for a moment that Krugman’s recent abuse of economics and Adam Smith is actually a good.

    Krugman the Younger, as he works diligently by lamplight, tabulating facts and figures and crunching numbers, is at first merely one of many. He produces but a small portion of total production because he lacks experience and has little capital. He has few ways to multiply his efforts.

    As he matures into Krugman the Elder, he gains experience, accrues capital, and gathers about him a small army of acolytes willing to work tirelessly for the few shekels he tosses their way without a thought.

    What do these acolytes do for him? Some type. Some research. Some mow his lawn. Others cook for him and launder his clothes. Those who’ve drawn the shortest straws watch O’Reilly and Hannity and produce detailed summaries for him to digest over a snifter of brandy.

    Krugman’s maturation process culminates in his ability to outsource–i.e. divide his labor–to all of those acolytes, including the New York Times, whom Krugman hired to print and widely circulate his opinions. He writes and thinks and produces more than he ever did, and more people benefit from his willingness to exchange his capital for their effort. But let us not forget the most important fact of all: there is only one Krugman. The individual we know as Krugman has a monopoly on the production of economic pronouncements laced with pious self-regard and condescension and garnished with sneering “Um, no’s.” And the only way he could have become a monopolist: the division of labor.

    In other words: so what if dividing labor leads to monopoly? What doesn’t? If monopoly theory is taken seriously, then everyone is a monopolist, including Krugman. If Krugman is so concerned about the apparent errors of Smith’s reasoning–that is, if he’s interested in moving the discipline forward–he should air that concern in scholarly work. Instead, he’s worried about the dangerous laissez faire brothel to which such reasoning might lead, and so his thoughts on Smith become just another whispered pop-intellectual emission: “Gather close, and I will reveal all: the free market actually depends on decreasing returns! So subtle! So sublime!”

    Only someone who completely takes for granted the division of labor and the increasing returns derived thereby would believe production is actually served by making it more difficult.

  43. liberty Says:

    Wow – “The parable of the pin factory says that there are increasing returns to scale”?????

    Krugman needs to go back and read a little more carefully! It’s hard to believe that an economist could come away with such a lesson.

    The pin factory was about division of labor, not increasing returns! It’s spelled out across the whole tale.


  44. @liberty,

    You are correct.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: