Diego Comin, William Easterly, and Erick Gong write,
The finding of this paper is a simple one: centuries-old technological history still matters today. The most surprising part of the finding is just how old the history can be and still matter. Our most robust finding is that technology in 1500 AD matters for development outcomes today, itself remarkably old when we consider that most history discussions of developing countries start with European contact and colonization. Even more surprising is that technology in 1000 BC and 0 AD has a significant effect in many specifications.
This finding is at least suggestive that technology is a strong candidate for being a principal determinant of development; this paper increased our prior weights on technology vis-à-vis competing explanations such as institutions and factor accumulation (not to go to the extreme that those latter things don’t matter)…
What does seem inescapable from this finding (if it is taken at face value despite the caveats) is that development is a very long run process. The tendency of policymakers and international institutions to overemphasize the instruments under their control may have contributed to an excessive weight being placed on the behavior of modern-day governments and development strategies as a determinant of development outcomes.
Pointer from Tyler Cowen, who perhaps can me sort out its relationship to the findings he so breathlessly cited in my preceding post. What they have in common, I believe is that both the labor-quality and the technology-adoption-history view of development imply that it is difficult for outsiders to shape economic development using instruments like aid, capital, or even legal institutions.
READER COMMENTS
Tim Lundeen
Nov 2 2006 at 12:58pm
Perhaps differences in IQ are a parsimonious explanation for all of these observations. I’ll have to read the book 🙂
dearieme
Nov 2 2006 at 1:24pm
“it is difficult for outsiders to shape economic development using instruments like aid, capital, or even legal institutions.” Or armies?
spencer
Nov 2 2006 at 5:08pm
It still goes back to the cost of labor.
If labor is cheap there is no incentive to change production methods and living standards stagnate.
If you want living standards to rise make labor expensive and this leads to greater investment and a rising capital/labor ratio and rising standards of living.
show me where this is wrong.
Standard conservative/liberaterian economics is just the opposite, keep labor cheap. But if labor is cheap why increase capital per labor
or standards of living?
James
Nov 2 2006 at 9:58pm
Spencer,
The standard libertarian view is that things work best if all factors, including labor, receive their marginal product whether that marginal product is cheap or expensive. Can you cite where you read that the standard libertarian view is to keep labor cheap?
The reason your argument is wrong is that it disregards the theory of the firm, and the relationship of changes in output to changes in living standards. Prior to whatever mechanism made labor more costly, the firm was operating on an isoquant tangent to the initial buget constraint. If labor becomes more costly, the immediate result will be a tighter budget constraint for the firm. The firm may well reorganize to have a higher capital labor ratio as a result, but total output will be lower because the firm’s new budget constraint is below the initial budget constraint, and therefore, also below the prior isoquant.
The stats tell a similar story. One variable that will make labor more costly is the federal income tax. (Recall that tax incidence is not the same as tax burden.) If your theory is correct, when the federal income tax is higher, we should see more fixed investment in subsequent periods. When I run a rank correlation between the federal personal income tax as a share of GDP against percent changes in fixed private investment, I get a z-score of about -2.158. The comments form here is not so good for displaying regression output, but after goofing around with a whole mess of different functional forms, I’m unable to find a specification, even with lags, controlling for other variables, and so on, where fixed private investment appears to rise faster subsequent to an increase in labor costs. So the capital/labor ratio may be increasing, but more of this increase is coming from the decrease in labor purchased than from an increase in investment. But download the data yourself and see what you get.
Bruce G Charlton
Nov 3 2006 at 1:42am
Very interesting paper – thanks for the link.
spencer
Nov 3 2006 at 4:03pm
James — if I run regressions between marginal tax rates and real gdp growth going all the way back to 1913 I get solid results using current year, lagged time frames and long run moving averages that raising taxes leads to stronger real gdp growth.
But I do not know if it proves any more then your comments.
But you might find your understanding improves a bit if you cconsidered some dynamic feedback loops in your analysis rather them putting everything in a static purely competitive model.
James
Nov 4 2006 at 3:42am
spencer,
Could you cite that source I asked for?
Marginal tax rates reflect how the incidence of taxation is distributed, but they don’t show how much more costly a tax policy makes labor. Also, you ought to test the steps of the causal mechanism that you conjecture. Does a higher tax on labor lead to increased capital formation? I doubt it.
You’re right to point out the value of a more dynamic model. Did you have one in mind that you could specify? As a starting point, I assume that both capital owners and workers are trying to maximize some objective function subject to constraints. I’d be impressed to see a dynamic model (outside of monopsony) where imposing an additional constraint such as a higher labor cost leads to a greater maximum for the objective function of either workers or laborers.
Comments are closed.