Friday, January 14, 2011

Adam Smith on poverty, progress, wages, and "universal opulence" — A theoretical conundrum

In Fall 2010 I taught an undergraduate seminar on modern social & political theory—or, to view it from another angle, the social & political theory of modernity—which included reading & discussing substantial portions of Adam Smith's Wealth of Nations. Two students in the seminar, Daniel Albornoz and David Kanter, raised a question about one point in Smith's argument that has perplexed interpreters, disciples, successors, and critics of Smith since at least the early 19th century. The issue they identified is important and contentious, both theoretically and substantively, and the way they framed the problem was perceptive. So after we had talked about it a bit, I wound up sending them a moderately long response via e-mail.

It occurs to me that this discussion might be of more general interest, and other Adam Smith aficionados might have opinions on this subject different from mine. So (with the agreement of Dan & David) I'm putting it out in the public sphere. Any thoughts out there?

(Page numbers refer to the Liberty Classics edition of WN.)

Yours for theory,
Jeff Weintraub

=> UPDATE (1/16/2011): Gavin Kennedy of the University of Edinburgh posted a two-part response to this discussion in his blog, Adam Smith's Lost Legacy ...
... which was partly re-posted, with a bit of further commentary, by Mark Thoma in his Economist's View blog. My partial & preliminary counter-response is HERE.

Sunday, October 10, 2010

Hi Dan & David,

This follows up a loose end from our conversation during office hours on Friday. You raised an important question regarding Adam Smith's theory (a point that had already been touched on in your discussion group's Smith I memo). I didn't want to say too much in response, partly in order to let you wrestle with the issues yourselves ... but having thought about it further, perhaps I ought to say a bit more than I did.

=> As we all know, Smith believes that the long-term process of socio-economic development, driven by the dynamics of the market and the division of labor, will steadily increase the overall productivity, prosperity, and "opulence" of societies over time. He also believes that the benefits of this increasing prosperity will be spread throughout the society (though, of course, unequally). In particular, he expects that, given the "natural progress of opulence," the standard of living of the poor—by which he means the great majority of people in society—will increase steadily and significantly (as long as the market is allowed to operate unhindered). From Smith's point of view, that consequence is one of the strongest justifications for the whole market system and its long-term developmental logic (e.g., WN pp. 22-24, 95-96, etc.). After all, "No society can surely be flourishing and happy, of which the far greater part of the members are poor and miserable" (WN p. 96).

Smith further believes that, broadly speaking, he has plenty of historical evidence for expecting an ever-increasing "universal opulence that extends itself to the lowest ranks of the people" (p. 22), and he appears confident that this conclusion follows plausibly from the overall logic of his theory. However, it's fair to raise the question of whether or not Smith has provided an adequate and convincing theoretical account to explain how and why this happy outcome should occur.

In your discussion group's Smith I memo, the four of you expressed your skepticism this way:
On what does Smith base this claim? Is it that prices will continue to drop as production becomes more efficient? Is it that wages will rise as there is more and more competition among capitalists? In our discussion we struggled to flesh out Smith’s reasoning here. In light of how simple Smith envisions most tasks to be when the division of labor has reached an advanced state, what sort of incentive would capitalists have to pay labor more than survival wages? Smith seems to think that most anyone can do most any super-specialized job (hence his worry about “pinheadism”). If that is the case, any laborer could be replaced by any other laborer. And if that is the case, wouldn’t the equilibrium wage just settle at the wage necessary to keep the laborer and a few of his offspring alive? That sort of existence does not sound all that appealing.
Good questions. As I've said before, your discussion here points to some genuinely interesting and significant issues, and the way you frame the problem is perceptive. Smith does try to address these issues and related ones—he does so primarily in Book I, ch. VIII of WN—but that doesn't necessarily mean he's done so in a fully successful way. So let me reframe and elaborate your question a bit, and fill in some of the logical steps more explicitly.

=> The first key point is that, according to Smith, one feature of the long-term process of socio-historical development is that the great majority of the population will "naturally" and necessarily become wage-laborers (or the wives and children of wage-laborers). That is, they will gain their livelihood, not by selling the products of their labor, but by selling their labor itself as a commodity. Labor, like every other commodity, has an exchange-value determined (more or less, and in the long run) by market processes and expressed in its price. So the material well-being of the vast majority of the population depends on the price of labor, which is to say the level of wages.

Rates of wages, like the prices of all other commodities, are "naturally regulated" (WN p. 72). How does that work? Like every other commodity, labor has both a "natural" price and an actual market price, and in the long run market forces (supply & demand, etc.) will push the actual price of a commodity toward its natural price. At any given moment, the actual price of a commodity may fluctuate above or below its natural price, but the natural price is "the central price to which the prices of all commodities are continually gravitating" (WN p. 73) as long as the market is allowed to operate without serious interference or distortion. Basically, the natural price corresponds to the normal cost of producing the commodity (relative to the costs of producing all other commodities in the system). If more efficient production techniques, technological improvements, and/or other factors allow that commodity to be produced at lower cost, then its natural price will go down—and vice-versa.

If we follow that logic, the "natural" price of labor in a particular society at a particular moment would be determined by the overall cost necessary to produce & reproduce laborers (including children who will become future laborers) and to maintain their capacity to go on working—i.e., a subsistence wage. So why should we expect wages (i.e., the price of labor) to keep increasing? Why wouldn't the "equilibrium wage," as you put it, "settle at" a subsistence wage? (And why shouldn't cheaper prices for food and other necessities simply lower the "natural" price of labor, and thus the level of a subsistence wage?)

As you also point out, this problem should be further sharpened if in fact, as Smith predicts, the "progress of the division of labor" means that the work performed by most workers, "the great body of the people," becomes increasingly narrow, specialized, simple, and mindless (WN pp. 781-782). Aside from making workers in the labor market increasingly interchangeable (which you mention), wouldn't this tend to reduce the cost of producing and reproducing labor (less training and expertise needed, for example), thus further lowering its natural price?

So why aren't workers stuck at a subsistence wage, even as the productivity and prosperity of the society around them continues to increase?

=> OK, here are some of the ways that Smith tries to deal with these issues.

Smith does acknowledge (e.g., on pp. 85-86 of WN) that wages can sometimes sink to this subsistence level (though, for obvious reasons, they can't stay below that level for an extended period). However, he adds (p. 86), there are "certain circumstances" that allow the normal rate of wages to go "considerably above that rate". Clearly, he expects (and hopes) that these "circumstances" will be common, not exceptional. What are they?

Basically, he argues that labor will command a higher price, and therefore workers will have a better standard of living, whenever the (effective) demand for labor outstrips the supply of labor. Are there any "circumstances" in which this might be a steady rather than merely transitory condition? In general, that is most likely to happen in periods of continuous economic growth (the more rapid the better). And one of the factors driving this process, by the way, is that as long as the economy is humming along pretty well, every increase in the accumulation of capital will produce an increase in the demand for labor (since the capitalist who has accumulated those resources will want to put them to profitable use, for which he needs workers). Under these circumstances, the actual price of labor will remain above its natural price—not in a temporary or fluctuating way, but systematically.

(For these arguments, see especially WN pp. 86-89 & 99-104.)

Under these circumstances, let us remember, a capitalist will not pay higher wages because he wants to (generally speaking), but rather because he can (since there are more resources at his disposal), because he has to (since he's in competition with other capitalists who also want to hire workers, and labor is scarce relative to capital), and because it's worth his while (since he can put the additional labor to more productive and ultimately profitable use). If he and other capitalists could keep the going wage closer to a subsistence level, they would (and, Smith adds, they try).

=> In bare outline, those are the main elements of Smith's theoretical response to the question you raised.

(At least, it's an important part of his response. Can you think of any other factors or mechanisms, compatible with the overall logic of Smith's theory of the market, that might lead wages to increase systematically as overall productivity and "opulence" increase in a society?)

Is that a strong and entirely satisfactory response? Well, that's for you to ponder.

I will say, though, that I think there are plausible grounds for wondering whether Smith's solution to this particular conundrum is entirely successful. It certainly does look like a peculiarly ad-hoc, equivocal, and in some respects paradoxical solution to such a central theoretical problem.

At the heart of Smith's whole vision of society and socio-historical development, after all, is a theory of the market as a self-regulating impersonal system which, if allowed to operate unhindered, "naturally" or spontaneously achieves a (more or less) optimal allocation of economic resources and activities and promotes ever-increasing growth in efficiency, productivity, and overall social wealth. And the market system is coordinated, above all, by the mechanism of prices. But the market can be expected to deliver optimal overall results only if, and to the extent that, the actual prices of commodities conform (roughly, over time) to their "natural" prices. That's why interfering with "natural" market processes to fix or distort prices so that they diverge from "natural" prices will tend to generate sub-optimal or even pathological results. However, it turns out that the main benefit, and justification, of the market system depends on the price of one particular commodity, labor, systematically diverging from its "natural" price. That may or may not strictly contradict the overall logic of the theory, but at the very least there seems to be a theoretical tension there.

Furthermore, even if we accept that Smith has demonstrated that a situation of continuous economic growth and capital accumulation prevents the actual price of labor from sinking down to its "natural" price and getting stuck there, would that analysis provide us with any reason to feel confident that economic growth will lead to a steady increase in wages over time (rather than simply enriching landlords and capitalists, while average wage rates remain more or less constant)? That's less clear.

In short, I suppose I share your skepticism about whether Smith can offer a solid theoretical justification, or even explanation, for his optimistic conclusion about the tendency of the market system and the division of labor to produce that ever-increasing and "universal opulence that extends itself to the lowest ranks of the people" (WN, p. 22)—at least, within the theoretical framework he's constructed. If one wanted to 'save' that optimistic conclusion about the overall social benefits of the market system (as Smith certainly would), then doing so might well require some modifications in the core theoretical apparatus. (Or maybe not?)

=> I suppose it's worth adding that, for almost two centuries following Smith's work, a wide range of people argued, from different ideological and theoretical directions, that his optimistic conclusion was not convincing, plausible, or correct. That is, they argued that the "natural" operation of the market system does not automatically, spontaneously, or necessarily spread the material benefits around to the workers and steadily increase their standard of living. Some even argued that it couldn't do this, except perhaps in temporary bursts.

I say "from different directions" because the people who reached this conclusion included some of Smith's own followers and disciples within the tradition of classic political economy, e.g., Malthus, and critics of the market system and of Smith's theory, including, but definitely not limited to, Marx and his followers. (There is a widespread impression, going back to the 19th century, that Ricardo shared Malthus's unequivocally gloomy outlook on this question; but, for reasons indicated in my Ricardian P.S. below, that's not quite right—Ricardo's position is a bit more internally conflicted.)

Actually, this remains a live issue, though nowadays economists who address it generally do so using a somewhat different conceptual apparatus. Of course, it is now widely accepted that, at least in certain circumstances, a capitalist market economy can generate steady long-term economic growth in a way that spreads the benefits through most of the population, producing broad-based and ever-increasing overall material affluence (perhaps with occasional hiccups). But how and why that happens, as well as the factors that might explain whether and to what extent it happens (since those patterns happen to vary considerably in practice)—all those remain open questions, theoretically and empirically.

=> Meanwhile, I hope I've given you some more food for thought to chew on.

Yours for theory,
Jeff Weintraub

A Ricardian P.S. Incidentally, one indication that the argument I outlined above really is Smith's response to this problem, and not just my own interpretation of Smith's reasoning, comes from way it was restated by one of Smith's most influential successors and disciples in the next generation of British political economists, David Ricardo. But Ricardo's overall treatment of these issues may also help to bring out the shaky and uncertain features of Smith's solution.

In the opening pages of Chapter V of The Principles of Political Economy and Taxation (1817), "On Wages," Ricardo reproduces the central logic of the Smithian analysis I outlined above and makes it even more explicit. Here is how Ricardo sums things up in the first few pages of Chapter V (pp. 52-53):
  • "However much the market price of labour may deviate from its natural price, it has, like [other] commodities, a tendency to conform to it."
  • But "It is [only] when the market price of labour exceeds its natural price that the condition of the labourer is flourishing and happy [....]"
  • So is it unrealistic to expect workers' standard of living to improve over time? Not necessarily. "Notwithstanding the tendency of wages to conform to their natural rate, their market rate may, in an improving society, for an indefinite period, be constantly above it; for no sooner may the impulse which an increased capital gives to new demand for labour be obeyed, than another increase of capital may produce the same effect [....]"
Of course, Ricardo's positions are not always identical to Smith's, to say the least. And even in this chapter on wages, Ricardo elaborates his analysis in some ways with which Smith might not entirely agree. But when it comes to the theoretical core of Smith's attempt to explain how and why wage-laborers might share in the increasing wealth of a society—at least in the sense of having a standard of living above the minimum subsistence level—I think it is hard to deny that Ricardo has closely reproduced the central thrust of Smith's analysis.

So far, so good—from the perspective of workers and humanitarians. However, leaving it at that would be misleading. If we go beyond the first two pages of Chapter V and consider Ricardo's overall treatment of these issues, the tone and substance of his analysis looks a good deal more ambivalent and less optimistic than those formulations, taken in isolation, might seem to suggest. In fact, as the discussion proceeds, the apparently reassuring implications of those statements get increasingly qualified, and at times even seem close to being repudiated. What Ricardo gave with one hand, he gradually takes back with the other.

Yes, Ricardo does appear to accept that it is sometimes possible for continuous economic growth to keep workers' average standard of living above the subsistence level "for an indefinite period". (I see no hint of ever-increasing "universal opulence" in Ricardo.) But is that outcome likely?

Smith's response to that last question would be positive (at least, for a "civilized" and "well-governed society" [WN p. 22]). That positive response is crucial to his overall picture of the nature, dynamics, and prospects of modern "commercial" society.

Ricardo's response turns out to be, at best, uncertain and highly conditional. The conditions that make it possible for wages to remain above the subsistence labor depend on maintaining a (beneficial) imbalance between the supply and demand for labor. But even though it is sometimes possible to maintain that imbalance "for an indefinite period," it remains an anomalous deviation from the central logic of the market system, which should eventually bring the actual price of a commodity into line with its natural price. For Ricardo, it seems highly unlikely that this kind of socially beneficial anomaly could be maintained permanently. In practice, the mechanisms that can temporarily keep the actual price of labor above the natural price are most likely to be effective in economically underdeveloped, socially backward, and badly governed societies, especially those with a lot of unused or under-used land (p. 56). In such societies, reforms that unleash the magic of the market can indeed lead to rising wages for a while. But in "rich countries, where all the fertile land is already cultivated," those exceptional mechanisms will cease to operate (p. 56). The ordinary laws of the market will then reassert themselves, pushing average wage rates down to the natural price. Thus:
In the natural advance of society, the wages of labour will have a tendency to fall, as far as they are regulated by supply and demand; for the supply of labourers will continue to increase at the same rate, whilst the demand for them will increase at a slower rate. [p. 57]
In so far as this tendency can be counteracted in economically developed commercial societies, the remedy would have to lie in "a reduction in people" (p. 56) through lower birth rates (pp. 61-62). (Others who shared this perspective also recommended emigration as a way of helping relieve population pressure, but I don't think Ricardo himself emphasized that "remedy".)

=> The pessimistic side of Ricardo's analysis of wages is due in part to his acceptance of the grimly discouraging implications of Malthus's theory of population. But it's worth emphasizing that Ricardo's departure from Smith's optimistic expectations about ever-increasing improvements in most people's standards of living is also rooted in more fundamental theoretical reservations. Ricardo is keenly aware that expecting the actual price of one particular commodity, labor, to diverge permanently from its natural price, and to do so on the basis of a permanent (and permanently beneficial) imbalance between supply and demand, is in tension with the central logic of the theory of the market as he and other (broadly) Smithian political economists understand it. And this theoretical anomaly clearly bothers him.

Later in the book, this theoretical tension and its implications burst out sharply in the first two paragraphs of Chapter XXX (p. 260). Ricardo begins by restating and emphasizing, in general terms, a key element of Smithian price theory:
It is the cost of production that must ultimately regulate the price of commodities, and not, as has been often said, the proportion between the supply and demand: the proportion between supply and demand may, indeed, for a time, affect the market value of a commodity, until it is supplied in greater or less abundance, according as the demand may have increased or diminished; but this effect will be only of temporary duration. [my emphasis]
In the long run the interplay of supply and demand merely serves to bring the actual price of a commodity into line with its natural price. Increased demand for a commodity, by bumping up its market price, will simply call forth increased supply, thus restoring the "natural" balance. Ricardo then uses a concrete example to spell out the implications:
Diminish the cost of production of hats, and their price will ultimately fall to their new natural price, although the demand may be doubled, trebled, or quadrupled.
It would thus appear that, in the long run, the market price of any commodity must "ultimately fall" to its natural price. OK, but would Ricardo make an exception for one particular commodity, labor? The next sentence disposes of that possibility.
Diminish the cost of subsistence of men, by diminishing the natural price of the food and clothing by which life is sustained, and wages will ultimately fall, notwithstanding that the demand for labourers may very greatly increase.
Thus, if increasing productivity generates cheaper and more abundant food and other necessities, the result must "ultimately" be, not an improved standard of living for wage-laborers who depend on these commodities, but simply a corresponding fall in wages. So much for "universal opulence". I think it's clear that the implications of this discussion run sharply counter to Smith's own hopes and expectations, and that he would not like Ricardo's conclusion one bit. But I also think Ricardo could plausibly argue, at least, that his more pessimistic conclusion accords with the central theoretical logic of Smith's own theory. If Smith wanted to argue against that claim, how would he do so?

—Jeff Weintraub