Bryan Caplan  

A Farewell to Alms: Overview of My Critique

PRINT
Happiness Research at a High S... Ph. D Curriculum in Economics...

I've finally finished Gregory Clark's A Farewell to Alms. Like everyone else, I agree that it's well-written and addresses important topics. And I think lunch with Clark would be fun. But frankly, I can't find much else good to say about this book. Every major argument that I can evaluate without getting knee-deep in the data is over-stated at best, and usually deeply misguided. His data-intensive arguments might be fine, but since Clark didn't build up much credibility with me elsewhere in the book, I'm skeptical.

I'm going to start with a quick overview of my position, which I'll flesh out in future posts.

1. Malthusianism. Clark adamantly argues that the Malthusian model explains human history until 1800. A key assumption of the model is that, for humans as for hyenas, more population means lower per-capita income. Other than a brief dismissal of Kremer's endogenous growth theory, Clark literally seems unaware that anyone disagrees with this assumption.

Even worse, Clark repeatedly misstates the implications of his own model. He delights in counterintuitive claims like "[I]n 1776, when the Malthusian economy still governed human welfare in England, the calls of Adam Smith for restraint in government taxation and unproductive expenditure were largely pointless." To the contrary, even in a Malthusian model, more production and less waste is unambiguously good for living standards. Population growth eventually returns living standards to their original level, but that may take generations.

2. Efficient Institutions. Clark repeatedly scoffs at the importance of institutions/policies for growth. He argues both that (a) Good institutions/policies failed to cause growth in important cases, and (b) Institutions naturally evolve toward efficiency anyway. His "proof" consists in some examples where good institutions/policies failed to produce rapid growth, and some examples where extremely inefficient institutions disappeared after a couple centuries.

Clark makes no effort to explain why other economists might have come to the opposite conclusion. He ignores the large differences in economic freedom, corruption, etc. between rich and poor countries. He ignores the compelling work of e.g. Sachs and Warner showing that, at least in the post-war era, avoiding a short list of bad policies has been virtually a sufficient condition for growth. He also ignores massive, long-lasting policy disasters like Communism; comparisons between West and East Germany, and North and South Korea; and can't even acknowledge post-Mao Chinese growth as a mind-blowing demonstration of the power of policy.

In the face of all this evidence, Clark throws up his hands and says that economists don't know how to create growth. Give me a break. If voters and politicians around the world since 1800 had just done what Adam Smith told them to do in The Wealth of Nations, poverty would already be a thing of the past. Economists have known how to create growth for centuries. The problem is that, all too often, non-economists choose not to listen.

3. How Important Is Labor Quality? Clark repeatedly dismisses competing theories by pointing out the imperfect predictions of their most simple-minded formulations. But he goes a lot easier on his own story. Why are rich countries so much richer than poor countries, according to Clark? Because they have lower-quality workers.

Obvious objection: If that's the problem, why does moving these low-quality workers to the West quickly raise their wages from $1/day to $40/day? Yes, that's below average for the West, but it's in the same ballpark. If that isn't iron-clad proof that institutions/policy matter a lot, what is?


Comments and Sharing





COMMENTS (12 to date)
eric writes:

Per the value of unskilled labor in different countries. Is the wage of unskilled labor in developed countries a function of the legal structure (property rights, free trade), or is the legal structure of a country a function of the distribution of human capital (ie, a critical mass of human capital is necessary--but insufficient, see communism--to create these legal structures)?

John S Bolton writes:

I'm glad we have economists who can be just sent with plenipotentiary powers to Haiti, Liberia or Honduras and ordain per capita economic growth out of a book, and an 18th century one, at that. Of course we must select them so carefully that their anti-discrimination qualities are high as can be, so that they will be incapable of falling into that defeatism which assigns great importance to the supposedly high and lasting differences in quality of population. They must also fully believe that institutions create more than they reflect, lasting differences between populations, and that you can never have too fast a labor force growth. To think of the terrible suffering in the poor countries, caused by our hoarding of ideas which could make them rich so quickly, and our hiding from them the secret that certain economists with total power, could eliminate age-old problems at a stroke. If only they knew, that quality of population is the same everywhere, that belief in equality and brotherhood of all mankind is the answer, institutions could grow like mushrooms, and imperialists could shape the benighted natives at will, with the right incentives and conditioning to alleviate their anti-foreign bias.

jaim klein writes:

Paragraph 3 has been written in a haste and has to be rewritten. The main point of Clark's thesis is that rich countries have improved workers. Moving individual workers from one country to other proves not that workers are freely interchangeable a la Schweik. Well lead and structured organization like an army are able to absorb and assimilate a large percentage of unsuitable individuals, but over a certain bar they break down and becomes what in Argentina we call a "quilombo", an unruly mess. In my opinion, only high quality individuals are able to create and maintain effective institutions, so Clark's and your arguments are, in fact, identical.

Jason Malloy writes:

He also ignores massive, long-lasting policy disasters like Communism; comparisons between West and East Germany, and North and South Korea


No he doesn't. You may disagree with his conclusions, but he doesn't 'ignore' Communism, which he argues dissolved in an eye-blink of history, consistent with a view that destructive institutions are in an unstable balance with the evolved dispositions of the population.

... This "efficient institutions" view can accept, especially in dealing with long-run history, that there may be periodic ideological pushes to adopt inefficient institutions as a result of religious fervor or social turmoil (p 213)... Thus it is no real surprise that China, despite nearly a generation of extreme forms of Communism between 1949 and 1978, emerged unchanged as a society individualist and capitalist to its core. The effects of the thousands of years of operation of a society under the selective pressures of the Malthusian regime could not bed uprooted by utopian dreamers. (p 188)

"The large differences in economic freedom, corruption, etc. between rich and poor countries" is a chicken or egg problem. Sachs and Warner are "growth epidemiology" (plausibly suffering from "healthy user bias" writ large). Clark at least makes some headway in demonstrating that there were times that had all the right institutions but no growth to show for it. Institutions may be necessary but not sufficient for growth. And institutions themselves may require further necessary conditions to work in the constitutional qualities of the population.

Matt writes:

The endogeneous growth model suggests aggregate productivity increases more than enough to offset each new additional worker, at least in the West under current conditions.

The problem with the theory is that is does not explain how an economy is productive if it cannot fill its own population needs.

Rather the endogeneous theory simply proves what we, and Brian, already know, that under some circumstances, we can rearange global population to net out better efficiencies.

Brian needs to compute the other half of his worker productivity argument. What is the cost of bringing a worker here, for high efficiency, rather than moving the tecnology there. Brian's argument is that the cost of establishing the proper institutions in the less developed nations is currently too high.

nordsieck writes:

[Comment deleted for supplying false email address. Please email the webmaster@econlib.org to request restoring this comment. A valid email address is required to post comments to EconLog. We do not use email addresses to send spam.--Econlib Ed.]

Absurdism writes:

Arnold, you write:

"If that's the problem, why does moving these low-quality workers to the West quickly raise their wages from $1/day to $40/day? Yes, that's below average for the West, but it's in the same ballpark. If that isn't iron-clad proof that institutions/policy matter a lot, what is?"

But what about complimentarities with other workers/consumers/etc? An unskilled worker would surely benefit from the higher stock of human capital present in the West, among other things, which is not necessarily the same as better institutions.

Gavin Kennedy writes:

The Malthusian Trap is compelling but limited in what it explains. As production rises, per capita incomes rise above subsistence (itself slowly changing in content, as changing notions of subsistence take effect over long periods), which has the effect, over generational time, that more children survive to adulthood and reproduce and continue the Malthusian cycle. Eventually, population returns per capita incomes to subsistence, child mortality increases and growth in population ceases.

Gregory Clark documents the Malthusian Trap with data, showing that over millennia long periods, per capita incomes remained at subsistence (the calorific count) to 1800, and concludes that no economic growth was strong enough to overcome the recurring constancy of the Trap.

His central question is why until 1800? What happened to locate the ‘IR’ in England and not elsewhere in Europe, and as an extension of that conundrum, why didn’t it happen in other countries and at other times in history (including some surely marginal candidates)?

Excluding institutional (widely defined) influences and qualitative growth in knowledge, this represents a challenge to current conventional thinking among economists and historians.

What does the Malthusian Trap argument leave out? In my view, it is too a narrow view of history. That the majority of a population remained on subsistence for 10,000 years, despite the agricultural ‘revolution’ – an ‘event’ that took several millennia – is not decisive.

The agricultural mode of subsistence (Adam Smith’s third ‘age of man’) did not do much better in subsistence terms, and perhaps did worse, pace J. Diamond, than hunter-gathering, did bring about definite changes that made it possible for ‘IR’ to occur millennia later, which persisting with hunter-gathering only, would not have done and didn’t do, namely it created a mode of subsistence that did not require the entire unit of the hunting-gathering band to produce the per capita subsistence of the entire population.

This change in economic roles created the inequality in per capita subsistence, power, and non-work roles for a minority and new work roles for what became ‘armed retainers’, servants and religious strata, and later the manufacturing labourers. It also created, or activated, the human ‘propensity to truck, barter, and exchange’, which led directly to divisions of labour in production and differential levels of per capita consumption (Adam Smith).

In the course of history, the ‘surplus’ of subsistence above that required for the producers of subsistence, net of the amount required for the per capita consumption of increases in population. It is not the constancy of the per capita consumption (an pure arithmetical result that hides the differential rates of consumption), and, tellingly, the subsistence of those ‘elites’, including producers of knowledge and innovators of technology that is decisive for answering Gregory Clark’s conundrum. It is what the elite that controlled the modes of subsistence did with their net surplus over long periods of time.

That is why I have argued that the history of what made modern commercial society possible is not a history of the labouring majority of the population; it is the history of the minority of elites. This not an ethically acceptable statement to some people who confuse their social preferences, admirable as they are, (and which I share) for what drove history through all its diversions in waste, cruelty, rapine and vileness, memorialised in the detritus of past-failed civilisations and the bloody lives of many elites.

So, examine what the elite, their retainers, executioners, innovators, philosophers and preservers of knowledge, artists, architects and infra-structure builders, and so on, did during these millennia, and, crucially, what those who traded were doing in creating the foundations of the commercial society (Smith’s 4th age of man), from which in due course, given the concatenation of evolutionary processes and circumstances, brought about the possibilities for what became ‘1800’ and thereafter.

What those circumstances amounted to is something I shall consider in more detail when I return to Scotland from France and consult my library and other sources.

William Newman writes:

I agree with the broad point that Clark needs to seriously address the ways that workers from poor countries do well in advanced countries. But the particular way that you point out is not a particularly strong challenge to the quality-of-workers idea, I think. It's quite possible in theory that the side effects of high quality workers could be the sole cause of $40/day niches for low-quality workers. As an example of such a niche (a tiny niche, but one that happens to be easy to see), I'd guess it mightn't take great ability to scavenge $40/day in resale value of consumer goods from dumpsters here in North Dallas.

A much bigger problem for the theory, I think, is any case where people come from a country with screwed-up institutions and not only do better than natives in their old country, but do better than the natives in their new country. And of course the theory is also challenged by the problem that you mentioned, of institutional boundaries being imposed on reasonably homogeneous populations and being reliably correlated with large differences in outcomes.

Isegoria writes:
In the face of all this evidence, Clark throws up his hands and says that economists don't know how to create growth. Give me a break. If voters and politicians around the world since 1800 had just done what Adam Smith told them to do in The Wealth of Nations, poverty would already be a thing of the past. Economists have known how to create growth for centuries. The problem is that, all too often, non-economists choose not to listen.
Perhaps we should attempt to solve the larger problem of actually creating growth, rather than the fairly narrow problem of coming up with a "solution" that won't get implemented.
Teddy writes:

Industrial revolution? People became too productive for the model. Compared to other responses that seems too simple.

Aaron Swartz writes:
If voters and politicians around the world since 1800 had just done what Adam Smith told them to do in The Wealth of Nations, poverty would already be a thing of the past.

Pray tell, what is this magical solution to poverty?

Comments for this entry have been closed
Return to top