What makes a nation wealthy?

Economists typically explain the wealth of a nation by pointing to good policies and the quality of a country’s institutions.  But why do these differences exist in the first place?

Professor Greg Clark of UC Davis, in his new book-length manuscript, resurrects Malthus, counters Jared Diamond (only recently has the European standard of living surpassed that of hunter-gatherer societies), shows the Industrial Revolution came only slowly, and argues that economists overrate the importance of good policy.  We can separate out the influence of policy by looking at the differential productivity on the factory floor, across regions.  The sheer quality of labor matters more than we used to think.  Quality labor attracts capital, which in turn supports good institutions. 

Here is the conclusion to my column:

Professor Clark’s idea-rich book may just prove to be the next blockbuster in economics.  He offers us a daring story of the economic foundations of good institutions and the climb out of recurring poverty.  We may not have cracked the mystery of human progress, but “A Farewell to Alms” brings us closer than before.

Clark also argues that sub-Saharan Africa is poorer than ever before, and that foreign aid worsens a zero-sum Malthusian trap.  He makes the startling claim that gains in health are the worst thing we can bring to modern Africa.  Here is the full column (by the way, I don’t write the titles or subtitles), which includes a link to Clark’s manuscript. 

The book is not yet out, but it is the best of its kind since Guns, Germs, and Steel


How does the quality of labor argument stand-up given the success of Indian and other immigrants from developing nations in the United States? Admittedly self-selection is involved but isn't the lack of opportunities in the home country due to poor instititutions and other conditions the reason they left? Not having access to the book how did Clark treat the historical immigration patterns in that respect. Wasn't capital availability and good institutions the source of the opportunities available to German, Irish and Italian immigrants in the 1800's?

cute idea.

why does foreign aid does not help Afirca, though?

Let's say, you cross before a starving, poor health beggar at a street corner every morning, and you give him 10 bucks each time out of sympathy. The beggar would not be so unhappy anymore, and if someone says "you need a job",he might angrily respond "hey, don't block my sunlight!"

Basically there are two ways you can help the beggar. the first is to give him a large sum of money so that he need not come out again next morning. the second is to hire him to get you a newspaper at 8:00 every morning for 10 bucks; soon, you will look for rich people who want newspapers to be delivered to their office; some time later, he might have a newspaper stall at the corner. He would be specializing on newspaper delivering then.

The second explains why Japan recoverd so fast after WII during the Korean war.

The aid to Africa are mostly final products in nature; they are soon consumed up. though we give cash, they basically quickly turn that into living needs. so, no accumulation of capital, and no Keynesian multiplier. In a sense, should there had been no aid, they might now have impressive outputs.

The claim about hunter-gatherers having a superior standard of living is not especially far-fetched, although of course there were so many different hunter-gatherer cultures that generalizing can be misleading. Roughly the argument goes that in an industrialized agricultural society you are able to sustain a lot of people because you can produce calories cheaply in the form of grain, but other nutritional complements don't become cheaper as quickly, so a large portion of the population ends up alive but malnourished. In the hunter-gatherer situation the balance of nutrients is likely to be better even if the total carrying capacity is less.
As for height, I seem to recall it did decline in Europe from two thousand years ago, reaching a low sometime during the industrial revolution and then increasing dramatically in modern times.

Malthusian trap argument was very popular in the 60's, but then it was applied to the whole world. When the birth rates dropped in the 70's combined with the discovery that the lower the infant motality rate the lower the brith rate in poor nations, it disappeared. In the 90's some economist were claiming that the very high birth rates in Kenya was good for economic growth. They even had regressions to prove it. We seem to have come a full circle, and it is back to Malthusian traps.

The Malthusian Trap assumes that if there is one person in all of England rather than two, the one person can produce more food per capita than the two people. Diminishing returns and all that.

Malthus can be refuted by observing that there are productivity gains from specialization. Adding more people to a fixed amount of resources (land, say) can increase per capita output rather than necessarily decreasing it - if the greater number of people leads to greater specialization. Rather than the marginal person being less productive than the previous average, the marginal person allows everyone else to abandon their marginally less productive activities.

I like the idea that agricultural societies developed not because they offered better living conditions to the peasants but so that the chiefs could increase the size of the populations they controlled, increasing their wealth and power. Like Kim Jong-Il, lords of agricultural societies couldn't care less whether their peasants were smaller and sicker than their hunting-gathering ancestors. There were more of them and they were easier to control, which made the lords richer and their lives easier.

According to the Malthusian analysis, the subsistence level of income is that at which the death rate equals the birth rate. Assuming birth rates stay constant, the analysis implies that an improvement in technology will result in an increase in population but not an increase in income level - incomes rise at first, but then rising incomes lead to falling death rates which lead to rising population which leads to falling incomes until at equilibrium you're back at the subsistence level of income.


Say technology improves so that food output doubles. Incomes rise, death rates fall, population rises, marginal food output declines, incomes fall, death rates rise, equilibrium again. Fine. But that assumes that death rates are tied to a given level of food production per person (income here is measured in food) regardless of technology. If the only technological development is in food production, then sure, that makes sense. But technology improvements happen in every aspect of life - and unlike the Malthusian assumption about food production, they scale with population size rather than with land mass.

Say sanitation improves. That means a given level of per-capita food production will mean fewer deaths. That means at the Malthusian equilibrium the "subsistence income level" (as measured in food production) will be lower. Lower per capita food production means that a smaller percentage of the population can be engaged in activities other than food production, which is why we assume that mortality is higher with lower per capita food output - there's fewer doctors available to treat the sick, for example.

But at equilibrium, there's just enough excess food to pay for just enough doctors to keep the death rate the same as it had been before sanitation. Sanitation gives us better productivity per doctor, thus the same health care output with fewer doctors. But there's no reason to think that sanitation is the only improvement that's happened. Just as there will be fewer doctors per capita, there will also be fewer stonewrights and architects and shipbuilders... but if masonry and architecture and shipbuilding technologies have all improved along with sanitation, then real income can rise even if per-capita agricultural output has fallen.

Clark conflates "level of food production" with "standard of living" and concludes that the average Joe in 1800 was no less miserable than cave dwellers.

"How does the quality of labor argument stand-up given the success of Indian and other immigrants from developing nations in the United States? Admittedly self-selection is involved but isn't the lack of opportunities in the home country due to poor instititutions and other conditions the reason they left?"

In the "old country", there's a hell of a lot of idiots and a few smart people that can't make full use of their abilities because all the idiots are saddling them with poor institutions. The smart people come to the United States, leave the idiots behind, and make better institutions to attract even more smart people. There's not that many of them in any given country, but when the US draws them from all over the world, they add up pretty quickly.

It's consistent with the "quality labor attract capital, which makes better institutions" theory, anyway.

"Also, the agricultural society will leave behind less healthy skeletons because those who had slightly less than excellent health tended to die quickly"

That should be "... those in hunter-gatherer societies with slightly less than excellent health tended to die quickly"

One of the more compelling arguments in the 'institutions' case is that, when a migrant moves from a poorer country to a richer one, his real income increases. It's the same person, so there is no difference in human capital or work ethic, or whatever, to talk about. In the richer country, the same laborer gets hooked up with more capital and is more productive. Something was preventing him from being appropriately capitalized in his home country, but many kinds of capital are mobile, so what was it?

But even taken on its own terms, the case for reducing health aid puzzles me. If you need quality labor to develop, then you need to have people living long enough to develop the appropriate skills and recoup the costs of acquiring those skills. Premature death and disability is a barrier to labor quality.

Ken: There's some evidence for your (implicit) model in my new paper here:


It looks like high-IQ groups are better at playing repeated prisoner's dilemmas than low-IQ groups. A study published in Evolution and Human Behavior, looking only at twins, found the same thing:


Since so much of successful society-building is (we're told) the process of avoiding prisoner's dilemmas--i.e., choosing cooperative outcomes when it's tempting to cheat in the short-run--it would seem that high-IQ societies would build better, more trustworthy societies all-around. Then a few low-IQ folks could pop in now and then and reap the fruits of the society built by the cooperation of others.....A great gift from the IQ-haves to the IQ-have-nots.....

Oh, and if you're looking for an easily-verifiable difference in "labor quality" that can generate Manuelli and Seshadri's 27-percent difference in productivity, it appears that you need look no further than IQ. I point that out on page 22 of this paper:



Hunter-gatherer societies have a natural form of
birth control in the fecundity-fertility difference.
Their low carb diets tended to make women infertile
while they were breast feeding, thus naturally reducing
the birth rate.

Garrett Jones,

Interestingly, in the original repeated PD game
experiments put on by Dresher and Flood at RAND
that annoyed Nash, it was the economist Alchian
who kept "cheating" and a mathematician (forget
his name) who complained that Alchian was an
"idiot" who could not see the advantages of

The idea that had occurred to me a while back regarding aid to Africa was this: If African countries are in the Malthusian trap, then giving them aid without increasing productivity will result in population growth beyond (or closer to for those countries close to climbing out) subsitance level. Suffering and famine ensue. Is this anything like Clarks argument?

I was disappointed to see that in the entire 453 page manuscript of Clark's book, there is no mention of Lynn & Vanhanen's 2002 book "IQ and the Wealth of Nations," which goes a long way toward quantifying a big part of what Clark apparently more or less means by "quality of labor."

For example, the biggest economic phenomenon in the world at present -- the rapid growth of China -- can be explained as largely an arbitrage exercise hammering out the disparity between the high average IQ of the Chinese (apparently a little higher than that of Americans) and the low wages they earned due to the legacies of various historical - political catastrophes.

Not relevant to this discussion at all, but I thought it was entertaining that "A Farewell to Alms" was also a book by Trudeau (the guy that wrote Doonesbury").

Given all the publicity its receiving on the blogs, I hope people aren't buying the wrong book.

I think the whole crux of the matter is "aid without increasing productivity." In the economies of the industrial revolution, people consumed only resources equivalent in value to what they produced. Therefore, no matter the birth rates of these societies, people will continue to become more productive as they strive to increase their standard of living.

However, in the case of Africa today, productive developed societies give resources to unproductive developing societies without expecting any resources in return. The developing countries then quickly use up these resources for sustenance, increasing the number of unproductive persons as the aid meets health needs. Unlike the economies of the industrial revolution, these new persons will not produce the resources they consume but will instead partake in free resources from developed countries. The amount of these resources typically does not keep up with populations and thus the average person of the developing country will see a lower standard of living.

With regard to IQs, a person's intelligence is only one of the variable in a person's production. Between countries, institutions play a far more vital role in raising the standard of living. Take, for example, North Korea and South Korea. These two countries are like two identical twins separated at birth, with an identical race with an identical gene pool and an identical language. Despite these similarities, South Korea has far, far more prosperity than North Korea.

I've heard of a recent theory explaining the turn from hunter/gathering (short life but a merry one) to settled farming (labour in the sweat of your brow) by climate change, i.e. worse weather - perhaps more drought? - that would have reduced the number of animals. Farming as a way out. This is of course not at all contrary to the theory that increased numbers of hunter/gatherers (in aboslute tersm or in carrying capacity terms) would have made their lifestyle unsustainable, i.e. the Malthusian trap. Is this the explanation of paradise, the golden age etc?

I also think that it is posisble that they lived well, relatively speaking, in the areas they then controlled. It probably leads the presetn day observer wrong to look at the few sorry remnants of the hunter/gatherers extant now - they have been forced out of their best hunting ground.

I believe that writ large people get the kind of societies they want. So it seems to me that the transformation from hunter-gatherer to farmer must have been voluntary -- people saw advantages. They perceived a higher standard of living; they made a choice or series of choices.

So how can we assume that they were wrong? And that their "standard of living" went down?

Greg Clark's writing may be good economic history. But if one is looking for good contemporary writers on the possible influences of future trends on economic growth, one really should read Ray Kurzweil.

These are my estimates, based significantly on his work:




I feel slightly slimy for asking this, but:

I downloaded the paper last week, but my computer crashed and ate it. When I went back to download it again this morning, it's no longer vailable. Anyone willing to email it to me?

Prof Cowen,
I read your article in the NYT about Mr Davis' book, and the example you gave (from his book?) about India.

Perhaps the following small facts could explain the difference in productivity:
Average Life expectancy:
year U.S Indian Subcontinent
1900 47.3 ~20 years (1911-1920)
1950 68.2 ~32 years (1947-1950)

I would like to know which is more important for productivity: whether the average worker is sufficiently well nourished to function, or whether his/her culture is more responsible?

Or is it your belief that the average worker in the mills in India was as well fed and had the same state of well being as the average American worker? I find this example curious, given how easy it is to find statistics to provide simpler and less nebulous explanations of this phenomenon (differences in productivity). As other writers have pointed out, Indian productivity was higher before the British takeover and corresponding collapse of the economy, than its European counterparts. Which is a greater factor ? Health and malnutrition, or the nebulous thing known as "culture"?

I hope this example is not representative of the book and if it is, there is a stronger argument for the role of culture; for else, it is a poor and shoddy one.

This is all very well, assuming that Clark's fundamental hypothesis - that there was no fundamental divergence of European living standards from world norms before 1800 - is correct. In fact there's good reason to believe that this wasn't the case. Greg's full of valuable insights, but on this one point I don't think he's brought us any closer to a conclusive answer: I think we need to be looking far earlier. Resolving present-day poverty is going to remain a monumental task requiring every mechanism available if it's to be more than a faddish slogan.

1. Age of Empires II and the conquerers expansion
2. Worms 3D
3. Starcraft and the Brood War expansion
4. Battleship
5. Star Wars Battlefront
6. any of the warcraft trilogy
7. Star Wars Republic Commando
those are just a few, comment back if you need more

Comments for this post are closed