Piling it on: Why the Division of Labor is More Important than Comparative Advantage

In a comment thread over at Fly Bottle, Muriego has already inspired two posts by our very own Micha Ghertner. Well, here's a third. Muriego writes:

Ricardo has 3 requirements for comparative advantage to be successful.

1) No flow of capital across borders
2) Full employment and freedom of labor to move from job to job
3) Trade is equal.

None of these requirements is met by our current trade policies. I would argue that I am more “Ricardian” in principle then those claiming our current policies fit the bill.

Micha does a good job of responding to this point in the two posts linked to above, but there is another line of attack worth pursuing. Comparative advantage is only one of two logics of trade that need to be taken into account. The division of labor, which comes from Adam Smith, is probably more important. Incidently, I seem to recall Austrians completely missing the concept of the division of labor, though they do use the term to refer to Ricardian comparative advantage from time to time.

Under comparative advantage, each party to a trade has a fixed set of endowments which affects their ability to produce different types of goods. For example France has it's climate which is conducive to producing wine, China has it's ability to produce silk, etc. For any two goods, there is a ratio between the production of them. This ratio doesn't have to be constant over all levels of production. For example, France might be able to produce 2 gallons of wine for every yard of silk it produces. The benefits of trade from comparative advantage come from the fact that it is extremely improbable that the ratios of production between every set of two goods for one party (whether it be a country, person, or something else), is equal to that of another party. So each party can specialize in producing what they have the most favorable ratio for, relative to everyone else, and trade with everyone else, resulting in increased total production and increased consumption for all.

Note that everything was held constant in the comparative advantage scenario except what was produced. Comparative advantage tells you how to optimize given your constraints. Under the division of labor, things are quite different. By allowing individuals to specialize and trade, the division of labor increases production through very different channels. First, specialization allows everyone to get better at their job, given whatever constraints are in place. Second, specialization saves the time everyone would spend switching from task A to task B to task C. Instead of being wasted, it is spent on producing more, or perhaps more leisure.

Third, and most importantly, specialization directly increases the rate of technological growth. The more familiar someone gets with a production process, the more likely that person can find a way to improve the production process. This can be anything from finding a more efficient way to line up the machines in a factory to a technological breakthrough in the literal sense.

This all comes straight from chapter 1 of Smith's Wealth of Nations. Smith is credited with the principle of absolute advantage when it comes to international trade. For some reason he didn't see the applicability of the division of labor. Well, we do now. Increased international trade means increased specialization which leads to growth for the reasons mentioned above. The borders simply don't matter for the analysis. As long as more people are involved in the specialization and trade process, the benefits accrue. (Note that this is one argument against population doomsayers.) So in essence, if comparative advantage tells you how to optimize given your constraints, the division of labor tells you how to change the constraints in a favorable way. If we take growth theory seriously, it is just this changing of technological constraints which fuels long term growth, which suggests that we should pay more attention to the division of labor than to comparative advantage.

For much more detail, see this econlib podcast.

Share this

I don't think it's entirely

I don't think it's entirely fair or accurate to say that Austrians completely missed the concept of the division of labor. Perhaps they didn't stress the point as much as you would have liked or as much as they should, but they certainly never said anything opposed or in tension with it.

In fact, the whole of Austrian economics can be viewed as an extension of the insights of the benefits that accrue from free trade in the international trade arena--which nearly every educated economist agrees with--to the individual level of single person economic actors--which many educated economists are less likely to accept, for various (unconvincing) reasons. Indeed, as you wrote, Austrians are the ones who truly grasp that "borders simply don't matter for the analysis."

They certainly aren't

They certainly aren't opposed to the policy implications, but I don't recall them ever talking about anything like the consequences of the Smithian division of labor. I was working from memory, so I could be wrong. Here are a few quotes from Rothbard's Man, Economy, and State which suggest that I'm right (gotta love pdfs and the 'find next feature'):

It is clear that conditions for exchange, and therefore increased productivity for the participants, will occur where each party has a superiority in productivity in regard to one of the goods exchanged—a superiority that may be due either to better nature-given factors or to the ability of the producer.

pg. 97, emphasis his. This sounds like an 'if and only if' statement to me, but the language is a bit vague.

It is clear that, praxeologically, the very fact of exchange and the division of labor implies that it must be more productive for all concerned than isolated, autistic labor. Economic analysis alone, however, does not convey to us knowledge of the enormous increase in productivity that the division of labor brings to society. This is based on a further empirical insight, viz., the enormous variety in human beings and in the world around them. It is a fact that, superimposed on the basic unity of species and objects in nature, there is a great diversity. Particularly is there variety in the aforementioned factors that would give rise to specialization: in the locations and types of natural resources and in the ability, skills, and tastes of human beings.

pg. 96, emphasis his

The latter quote is especially telling. Rothbard thinks that the reason the division of labor brings such an increase in prosperity is because of the differences between individuals and their respective environments. For Smithian specialization, these differences are irrelevant; the prosperity comes with or without them through the three effects listed in my post.

My knowledge of the Austrians is a bit spotty though, so you may be able to find something from Rothbard or someone else which contradicts this.

I don't see how that second

I don't see how that second quote of Rothbard is in anyway at odds with the three effects listed in your post. Read that last sentence of Rothbard: variety may be endogenous or exogenous, it doesn't really matter where the variety comes from. But once you have it, you get incredible benefits from the division (i.e. the variety of separate types) of labor that are mentioned in your three effects.

Rothbard is just positing different ways this variety/division can come about. That doesn't negate the consequentialist and praxeological reasons why benefits accrue once variety/division does come about. And it will come about, because of the potential benefit. Just as in evolutionary theory, in which undirected organisms adapt to their environment, and the ones that do the best at adapting tend to flourish.

Smithian effects need no variety

The point is that Smithian effects need no variety to occur. You and I could be identical twins with the same mental and physical attributes living in the same place with the same possessions and we could still reap the benefits of the division of labor through Smithian effects- we would still get better at our respective jobs, save time due to less switching, and have a better chance of finding a way to do our jobs better. The fact that Rothbard boils down the benefits of the division of labor solely to effects which derive from variety is evidence that he missed the Smithian effects.

Different Meanings of Variety

I think you and I are using different meanings of variety. In my view, Smithian effects even as you defined them need variety to occur. That is, the identical twins with the same capabilities still need to choose different professions or roles for division of labor to occur, even if their choice of which one specializes in what is entirely arbitrary. This is what I meant by exogenous and endogenous variables; wherever the source of variety comes from, even if it is entirely arbitrary, it is still necessary for division of labor to occur. It is the very definition of division of labor - division, i.e. variety.

we could still reap the benefits of the division of labor through Smithian effects- we would still get better at our respective jobs, save time due to less switching, and have a better chance of finding a way to do our jobs better.

Again, I agree with all this, but it is the very definition of variety. Our respective jobs. Meaning each one chooses a different job for whatever reason. Less switching. Meaning each one is focusing on a particular task, not all equally sharing the same tasks. This is variety, however it comes about.

Well then...

I still think Rothbard misses the boat because he is using my definition of variety - variety in endowments. The critical part:

It is a fact that, superimposed on the basic unity of species and objects in nature, there is a great diversity. Particularly is there variety in the aforementioned factors that would give rise to specialization: in the locations and types of natural resources and in the ability, skills, and tastes of human beings.

Location, natural resources, ability, skills, and tastes are all endowments. Also, he is talking about the diversity which is superimposed on the basic constants of human nature. In what sense is our decision to specialize this sort of superimposition? So I take it that he is using my definition.

Endowments are key because comparative advantage is driven by differences in them between two parties which results in differing levels of productivity (or ratios of relative productivity between two goods). Smithian effects apply even when these productivity levels are identical. If I am correctly interpreting the first quote, then Rothbard thinks that the benefits of exchange only come about when we have this difference in productivity, which again suggests that he misses Smithian effects.

But again, Rothbard doesn't

But again, Rothbard doesn't seem to be saying that variety in endowments are absolutely necessary. He isn't ruling out the possibility that mere arbitrary concerns (i.e. flip a coin, you do task X and I'll do task Y) could lead to specialization. But as it happens, this is generally not the way things work. We don't generally need to rely on purely arbitrary concerns because variety in natural endowments occurs. Rothbard is simply describing the world as it is, not as it logically must be in all possible cases. Now, I can see why that might be confusing, and perhaps he would have an even stronger argument if he dropped all talk of endowments and focussed solely on even the exceptional case of arbitrary variety. But I don't think you can fault him for making the argument in a less abstract, easier-to-grasp way.

I read Rothbard as saying that variety in endowments is a sufficient condition, but not a necessary one. And I understand the risk of confusion by him saying so. But the fault lies not (solely or primarily) with Rothbard but with the people confusing sufficient conditions with necessary ones.

Mises on differentiation (Human Action)

5. The Effects of the Division of Labor
The division of labor is the outcome of man’s conscious reaction to the
multiplicity of natural conditions. On the other hand it is itself a factor
bringing about differentiation
. It assigns to the various geographic areas
specific functions in the complex of the processes of production. It makes
some areas urban, others rural; it locates the various branches of manufac-
turing, mining, and agriculture in different places. Still more important, how-
ever, is the fact that it intensifies the innate inequality of men. Exercise and
practice of specific tasks adjust individuals better to the requirements of their
performance; men develop some of their inborn faculties and stunt the devel-
opment of others
. Vocational types emerge, people become specialists.

Good enough?

And that's an important

And that's an important point Mises is making: human endowments are neither solely endogenous or exogeneous. They feed into each other, minor differences in genetic and circumstantial endowments creating incentives for great differences in exactly which endowments each of us personally chooses to develop. The great benefits of specialization itself provide huge incentives to enlarge our differences with each other by specializing.

The focus should not be on the source of variety in endowments (since there is no single source; it is an ongoing process), only that we are lead as if by an invisible hand to vary our endowments to benefit from the division of labor.

The concept is probably

The concept is probably there. The first sentence is the only thing that generates a bit of uncertainty - multiplicity of natural conditions is irrelevant for Smith. However, whether human beings consciously recognize this is another question. At the very least, Mises has seems to have had a confused notion of Smithian effects and he may well have completely understood them. It is sufficiently vague that readers unacquainted with the concept might not catch it resulting in, perhaps, a blind spot among many of his followers.

For what it's worth, the other relevant part of my memory is the absence of any mention of the Smithian effects from Hulsman's talk on the division of labor and exchange from my Mises U tenure. However, that very well could be due to a defect of memory rather than a defect in the talk. I wasn't aware of the difference until a bit later when I listened to the econlib podcast I linked to in the original post, and it is, after all, only an anecdote.

For what it's worth, the

For what it's worth, the Mises Institute flavor of Austrians, taking after Rothbard as they inevitably do, do tend to undeservingly denigrate the ideas of Smith and Ricardo:

Rothbard contends that Adam Smith, far from being the founder of economics, "contributed nothing of value to economic thought … he introduced numerous fallacies, including the labor theory of value, and thereby caused a significant deterioration of economic thought from previous French and British economists of the eighteenth century" (Economic Thought, p. 463). In the second volume of his work, Rothbard assails David Ricardo; and, though he sadly did not live to finish a volume on twentieth-century economic thought, it is safe to say that Alfred Marshall would not have fared well in it.

First sentence should remove uncertainty

The first sentence is the only thing that generates a bit of uncertainty

Actually I think it should remove uncertainty because it is inserted as a point of contrast. The words "on the other hand" contrast what comes after from what comes before.

Div of labour

"Incidently, I seem to recall Austrians completely missing the concept of the Incidently, I seem to recall Austrians completely missing the concept of the division of labor

If you had noted the odd phrasing here, you might have avoided the mistake you made -- I think the right thing to say was 'I fail to remember Austrians discussing the division of labor."

Also, Ricardo did not contend that comparative advantage didn't apply in the absence of those three conditions, but that it must apply in their presence.