This is a really useful perspective on this debate — and an illustrative survey of the breadth of development literature across a range of angles. But perhaps the breadth is the challenge.
Explain the theory persistence as a measurement problem is very plausible: the tools are weak, history happens once, so we can't decide between competing theories. But there's a question underneath it. What if the theories persist not because we can't measure well enough to decide between them, but because we're asking the wrong question?
The question "Which theory of development is right?" assumes that the theories are competing to explain the same outcome. What if they're not? What if industrial policy works in some institutional environments and fails in others because the surrounding conditions determine what the intervention actually produces?
Nathan Lane's 2025 paper shows Korea's industrial drive worked. The interesting comparison is why broadly similar policies, protection, directed credit, and performance requirements underperformed across much of Latin America in the same era. Korea's bureaucracy could enforce the export discipline. Brazil's, in the 1970s, largely couldn't. Similar policy architecture, different regime, different outcome.
This is actually what you'd expect if economies and polities function as complex adaptive systems — environments where the same input produces different outputs depending on the configuration of the surrounding system. In that kind of system, the search for a universally correct theory of development is probably the wrong research programme. The variety of theories, each working somewhere and failing elsewhere, isn't a measurement problem to be solved. It's the signal. But their divergence also means that there does not appear to be a coherent field of development theory.
Which means the missing thing isn't a better test of which universal theory wins. It's a prior framework for identifying which regime or system a country is in before assessing. And that's a different kind of research than anything currently on the menu. Whether it's tractable is genuinely unclear. But it's at least the right question.
The step both camps are skipping is the upstream one: before asking which intervention works or which theory wins, ask what kind of system you're dealing with. That prior diagnostic question is currently not part of anyone's research programme. And the debate might look quite different if it were.
This is very useful and interesting observation: "What if industrial policy works in some institutional environments and fails in others because the surrounding conditions determine what the intervention actually produces?" and "The step both camps are skipping is the upstream one: before asking which intervention works or which theory wins, ask what kind of system you're dealing with. That prior diagnostic question is currently not part of anyone's research programme. And the debate might look quite different if it were."
It made me wonder if my "Tolstoy" observation might be my naive stumbling in this direction (or I hope I might have been stumbling in this direction)- I think the observation is very useful.
Isn’t this kind of the structural model paradigm, just centering the corollary “it’s really important to be sure you’re using both the correct set of parameters and that you’re correctly assessing what those parameters values are?”
The structural-model paradigm assumes the model's form is given, and then treats parameter values as the open question. The issue I was pointing to sits one level above: i.e., which model actually applies is itself the variable.
Korea-style industrial policy ran on a set of mechanisms (disciplined bureaucracy, export discipline, capital scarcity) that simply weren't present in Bolivia, where any analogous policy would have had to rely on different mechanisms.
A structural model can absorb that only by becoming a different model for each country, which is close to effectively no real model. The harder unsolved problem is meta: a theory of when each model applies.
Thanks. I may be over-extrapolating from AI but this kind of just sounds to me like the model isn't big enough / scale more. Like nothing here sounds like something that you couldn't get with a sufficiently high-dimensional space + gradient descent.
Noah's post hints at the issue - the dataset is not large enough to do this by brute force. If it were, we could go messy data / inconsistent theories -> Complicated Patterns -> AI Rules (hidden logic) -> Heuristics -> (Meta)Theory.
Instead, we need to interpret data and patterns, identify patterns in the anomalies thrown up by existing models, develop new theories, test them against the limited available data, and see if they fit without being overdetermined. This is why I think a complex adaptive systems paradigm might help. It is a parsimonious (meta) theory that can at least accommodate many of the different models. And make sense of the apparently unpredictable nature of events, even if it has limited predictive power on its own.
If the family structure hypothesis is correct, then the different cultures created by different family structures play a foundational (though not completely deterministic) role in defining the developmental potential of different countries/ regions. And this underlying anthropological base may result in some developmental options being largely closed off to some cultures.
I have been thinking about this a lot since reading Growth and the Case Against Randomista Development in 2020... there's almost no one that I know that is earnestly trying to eliminate extreme poverty other than Luke Eure of https://www.noidlesitting.com/profile/posts
Having been thinking about this question since 1969, I have two comments: 1. These theories are not mutually exclusive, of course; and 2. If we could re-run world history since 1945 a few thousand times with random changes in policy choices, we would then have some chance of knowing the answer. But I do have a personal comment. I worked on South Korea in 1972 and Kenya in 1973, when I was an economist at the World Bank. The different subsequent trajectories of these countries have not, to put it mildly, been a surprise.
While one can guess, still if I may this is perhaps just a wee bit overly cagey "I worked on South Korea in 1972 and Kenya in 1973, when I was an economist at the World Bank. The different subsequent trajectories of these countries have not, to put it mildly, been a surprise." (but understandable)
Though not fitting precisely, and some outliers, doesn’t development just mainly match the IQ of the population? East Asians super smart - get Taiwan, Japan, Singapore, South Korea, Hong Kong, all rich as hell. Only thing holding China back is communism. Low IQs all over Africa so doesn’t matter what geography, institutions, education system etc etc each country has - all are poor. Communism and corruption held back Eastern Europe for a long time but those who have ditched it getting rapidly richer, like Poland - as per underlying high IQ population. Australia a barren desert with no close neighbours but dump some Europeans there and in only 100 years they are one of the world’s rich countries.
I don't really buy the IQ argument in part because of the Flynn effect. What kinds of IQ scores would you have gotten if you tested South Koreans in 1920? (Or 1850?)
I can't consider the research here to be anything resembling science until IQ and other behavioral differences between groups is considered. But I don't think that IQ alone is as powerful of an explainer as you give it credit for. You say there are outliers, but I think that understates it. East Asians have higher IQs than Europeans, but industrialization was behind by a couple centuries. And the Europeans who landed in Australia brought their institutions with them as much as their IQs.
But again you run up against endogenicity. (ugh, that's a hard word to spell) At the biological level, malnutrition and parasitic infections are known to be hard on brains. At a more cultural level, consider the Flynn Effect -- it's well-documented that over the past 150 years or so the average IQ of the highest developed Euro-American countries has increased by about 30 points, which is one s.d. It's clear that wasn't due to any sort of evolutionary selection so the cause must be environmental. The best guess I've seen is that the kids have been raised with a far richer information diet as "the media" have become more and more elaborate. But of course, parallel effects would be expected from formal schooling, and the amount of schooling the average person has seen has increased greatly.
If you look at a place like Congo, they've got plenty of malnutrition and chronic parasitic infection, weak schooling and underdeveloped media. So to make the comparison with the US fair, you'd first have to fix *all of those*. Then measure Congo's average IQ and economic growth.
But for that matter, several African countries have seen very good growth rates at times. IIRC Ethiopia managed 5% per year for a decade or two. But even those successes are invisible to an outsider -- It's hard to notice that doubling the GDP/capita from $600 to $1200 is a fantastic achievement.
Yeah I was thinking the same thing. Fernández-Villaverde asks why so few people are studying the problem and I thought "they think they already know the answer but can't say it." Even if the hereditarian explanation turns out to be wrong, the taboo deters people from studying the question in the first place.
I suspect we will not come up with a good model here because there are variables that are off limits. If I want to research why South Korea got rich while Bolivia didn't I would have to consider that South Korea has an average IQ of 107 vs 96 for Bolivia. But the researchers will never do this because that is heresy and a career ender. It certainly isn't the only factor here. Presumably North Korea and South Korea had a similar IQ average. But to leave it out as a variable is simple unscientific.
Same thing with culture "a culture of progress, innovation, and openness to technology" as you describe it is not going to cut it. It's vague to the point of uselessness. What needs to be done here is an actual quantification of the behavioral differences between a Bolivian and a South Korean, but that is also getting into heretical territory.
If there's a theory to be found here it's not going to be found by going over the same factors over and over again. It will be found by going where everyone else has been afraid to go.
I think you're right about the inherent difficulties, but I'd say that economists need to learn more history. I'm not against the math-stats revolution that has swept the profession, but 'the economy' is an in inherently multi-diciplinary field. A good analogy is medicine: if you study something as complicated as bodies, you need physics, chemistry, cell biology, and a bunch of other subfields. If you study 'the economy,' you need math and statistics, but also political science and history.
I read 'How Asia Works' after already having a Masters in Economics for a long time, and doing a lot of outside reading--and I was absolutely dumbstruck by the chapter on land reform. My first thought was "how the HELL had I never even HEARD of this?" How had I had five years of formal education in economics, and another decade of self-education, and I'd never even heard of the idea that land reform accelerated growth? I'd even taken a graduate level development economics course! And the idea that land reform increased agricultural production never came up, even as a hypothesis.
Reading that book, it struck me that economists need to learn more history. It's shocking that you can get a Phd in Economics and basically be a statistician who's learned very little about the most basic economic facts. Yes, history has a lot of limitations, but it's crazy that economics education barely even requires it any more.
Useful reflection, I wonder however if perhaps one should profitably invert the Tolstoyian phrase All "happy families are alike; each unhappy family is unhappy in its own way" for econ dev success on national geography (to eximine if the unhappy, i.e. the unsuccessful are more alike than not in combinations of policy and other error or other handicapping (as like national state incoherence [Nigeria])
Although it's not totally on topic, I'm really tickled to see your regular mention of medical research targeting the biology of aging, since I've had an intense interest in it for several years and the techno-optimist in me is still fascinated by the idea of indefinite healthy lifespan. I'd be curious to see whole posts where you explore economic and social implications of compressed age-related ill health and increased healthy lifespan.
Here's some reading, although you might already be familiar with it:
Noah, this cuts to the heart of something I've been writing about. The development industry is obsessed with the seen—schools, clinics, kilometres of road—while systematically ignoring the unseen infrastructure that makes any of it yield lasting returns. Contract enforcement is the textbook example. Everyone in the profession knows it's foundational, but what do development institutions actually do about it? A bunch of useless seminars, training programmes for judges who will never rule independently, and "capacity building" reports that gather dust.
I've tried to move beyond critique to design. Satyapur is a proposal for a privately-chartered commercial court system within Bangladesh—a new jurisdiction modelled on Delaware, Singapore's judicial credibility, and Dubai's DIFC. It requires zero public funding, operates on an entirely opt-in basis, and funds itself through registration and filing fees. The idea is to create a space where contracts are actually enforceable, so the economic activity that currently flees to Singapore or Delaware can stay home.
The two manifestos lay out the full architecture and political strategy:
If development economics is serious about moving from the seen to the unseen, charter jurisdictions like this are one concrete place to start. Would be curious to hear your thoughts.
Though of course a private, opt-in system isn't like what happened in developed countries, where a well-functioning legal system was definitely *not* opt-in. Also, the ugly part is that in the early stages of development, good *land rights* are critical, and the adjudication of land rights is always the most slow-changing part of any legal system.
The maddening thing looking at underdeveloped countries is it sure seems like all they need is a bunch of people with the correct know-how to come in and start industrializing everything. And human capital is cheap! All you need to do is take your bountiful supply of cheap human capital, put them through some education, and some of them will prove quite skilled! Why is a single generation of universal literacy and internet connectivity not enough to pull a poor country up to near parity?
Really good write up. I always for some reason get surprised when really smart people that I respect and who seem to be open minded and pragmatic fall into the trap of trying to look for simple answers to complex problems, or refuse to think that multiple things can be true at once.
The Washington Consensus aimed directly at those big questions, but that wasn't "Proressive." The World Bank turned away from pusing policy reform. USAID in the GWB Administrtion had a few.
High quality narratives like How Asia Works are not only attentive to a lot of details but also do a soft, verbal version of cross-country regressions. For example Studwell himself observed that the basic ingredients of the policy mix he recommends (land reform, export discipline - i.e. support conditional on market performance, especially on export market performance because the latter is independent from local political connections - and control of financial institutions) were all present in the countries that were growth successes (Korea, Taiwan, Japan, later China), while these good policies were not followed in the countiries whose catch up was unsuccessful (Indonesia, Malaysia, Thailand and Philippines). This is quite convincing even if it is not a mathematic and scientific method. If this is coupled with microeconmic work on the effect of policies on certain industries, as was done by Nathan Lane on the Korean heavy and chemical insdustry drive, which verifies the narrratives, we can be as certain about the results as one can get in these fields where counterfactuals and experiments at large scale cannot be run. I am certainly convinced and I don't think any greater certainty is needed for policy. If one is more demanding in terms of evidence to drive policy, a state would be paralyzed as no policy option can be supported by mathematical, unquestionable proof. Imposing such requirements would effectively condemn a state to doing nothing, which is simply laissez-faire policy. There is no neutral policy in any event. Further, I don't mind if the focus of microeconomic work and narratives is policy because policy is exactly what countries can change - they cannot change their resource endowments and any attempt to fickle with things like culture is bound to fail. Countries can change the quality of their human capital by education but Studwell is also very instructive here as he lists a number of countries with good eductional systems and high quality human capital, which failed economically (Cuba, Soviet Union, Philippines etc) and argues that spending more on education alone is not a game changer. In effect, usually policy driven growth creates demand for educated people and this leads to the increase in the quality of eduction rather than the other way around. So focusing on policy and not getting stuck with unmutable things like resources, ethnicity, culture etc. is the right way to go (with the necessary humility of course).
This is a really useful perspective on this debate — and an illustrative survey of the breadth of development literature across a range of angles. But perhaps the breadth is the challenge.
Explain the theory persistence as a measurement problem is very plausible: the tools are weak, history happens once, so we can't decide between competing theories. But there's a question underneath it. What if the theories persist not because we can't measure well enough to decide between them, but because we're asking the wrong question?
The question "Which theory of development is right?" assumes that the theories are competing to explain the same outcome. What if they're not? What if industrial policy works in some institutional environments and fails in others because the surrounding conditions determine what the intervention actually produces?
Nathan Lane's 2025 paper shows Korea's industrial drive worked. The interesting comparison is why broadly similar policies, protection, directed credit, and performance requirements underperformed across much of Latin America in the same era. Korea's bureaucracy could enforce the export discipline. Brazil's, in the 1970s, largely couldn't. Similar policy architecture, different regime, different outcome.
This is actually what you'd expect if economies and polities function as complex adaptive systems — environments where the same input produces different outputs depending on the configuration of the surrounding system. In that kind of system, the search for a universally correct theory of development is probably the wrong research programme. The variety of theories, each working somewhere and failing elsewhere, isn't a measurement problem to be solved. It's the signal. But their divergence also means that there does not appear to be a coherent field of development theory.
Which means the missing thing isn't a better test of which universal theory wins. It's a prior framework for identifying which regime or system a country is in before assessing. And that's a different kind of research than anything currently on the menu. Whether it's tractable is genuinely unclear. But it's at least the right question.
The step both camps are skipping is the upstream one: before asking which intervention works or which theory wins, ask what kind of system you're dealing with. That prior diagnostic question is currently not part of anyone's research programme. And the debate might look quite different if it were.
This is very useful and interesting observation: "What if industrial policy works in some institutional environments and fails in others because the surrounding conditions determine what the intervention actually produces?" and "The step both camps are skipping is the upstream one: before asking which intervention works or which theory wins, ask what kind of system you're dealing with. That prior diagnostic question is currently not part of anyone's research programme. And the debate might look quite different if it were."
It made me wonder if my "Tolstoy" observation might be my naive stumbling in this direction (or I hope I might have been stumbling in this direction)- I think the observation is very useful.
It’s early Spring in Michigan, and you really do need to check the weather before going outside.
Isn’t this kind of the structural model paradigm, just centering the corollary “it’s really important to be sure you’re using both the correct set of parameters and that you’re correctly assessing what those parameters values are?”
I would say not quite.
The structural-model paradigm assumes the model's form is given, and then treats parameter values as the open question. The issue I was pointing to sits one level above: i.e., which model actually applies is itself the variable.
Korea-style industrial policy ran on a set of mechanisms (disciplined bureaucracy, export discipline, capital scarcity) that simply weren't present in Bolivia, where any analogous policy would have had to rely on different mechanisms.
A structural model can absorb that only by becoming a different model for each country, which is close to effectively no real model. The harder unsolved problem is meta: a theory of when each model applies.
Thanks. I may be over-extrapolating from AI but this kind of just sounds to me like the model isn't big enough / scale more. Like nothing here sounds like something that you couldn't get with a sufficiently high-dimensional space + gradient descent.
Noah's post hints at the issue - the dataset is not large enough to do this by brute force. If it were, we could go messy data / inconsistent theories -> Complicated Patterns -> AI Rules (hidden logic) -> Heuristics -> (Meta)Theory.
Instead, we need to interpret data and patterns, identify patterns in the anomalies thrown up by existing models, develop new theories, test them against the limited available data, and see if they fit without being overdetermined. This is why I think a complex adaptive systems paradigm might help. It is a parsimonious (meta) theory that can at least accommodate many of the different models. And make sense of the apparently unpredictable nature of events, even if it has limited predictive power on its own.
If the family structure hypothesis is correct, then the different cultures created by different family structures play a foundational (though not completely deterministic) role in defining the developmental potential of different countries/ regions. And this underlying anthropological base may result in some developmental options being largely closed off to some cultures.
I have been thinking about this a lot since reading Growth and the Case Against Randomista Development in 2020... there's almost no one that I know that is earnestly trying to eliminate extreme poverty other than Luke Eure of https://www.noidlesitting.com/profile/posts
I did really enjoy reading How Africa Works recently though! (Oliver Kim's treatment of it is pretty good too if you're short on time https://www.global-developments.org/p/how-africa-works )
(https://forum.effectivealtruism.org/posts/bsE5t6qhGC65fEpzN/growth-and-the-case-against-randomista-development)
I liked How Africa Works too! I didn't like Oliver Kim's treatment so much though. Read the book!
Ooh I would love to know where you disagreed!
Having been thinking about this question since 1969, I have two comments: 1. These theories are not mutually exclusive, of course; and 2. If we could re-run world history since 1945 a few thousand times with random changes in policy choices, we would then have some chance of knowing the answer. But I do have a personal comment. I worked on South Korea in 1972 and Kenya in 1973, when I was an economist at the World Bank. The different subsequent trajectories of these countries have not, to put it mildly, been a surprise.
While one can guess, still if I may this is perhaps just a wee bit overly cagey "I worked on South Korea in 1972 and Kenya in 1973, when I was an economist at the World Bank. The different subsequent trajectories of these countries have not, to put it mildly, been a surprise." (but understandable)
When writing a short comment I expect people to use their imaginations.
Though not fitting precisely, and some outliers, doesn’t development just mainly match the IQ of the population? East Asians super smart - get Taiwan, Japan, Singapore, South Korea, Hong Kong, all rich as hell. Only thing holding China back is communism. Low IQs all over Africa so doesn’t matter what geography, institutions, education system etc etc each country has - all are poor. Communism and corruption held back Eastern Europe for a long time but those who have ditched it getting rapidly richer, like Poland - as per underlying high IQ population. Australia a barren desert with no close neighbours but dump some Europeans there and in only 100 years they are one of the world’s rich countries.
I don't really buy the IQ argument in part because of the Flynn effect. What kinds of IQ scores would you have gotten if you tested South Koreans in 1920? (Or 1850?)
I can't consider the research here to be anything resembling science until IQ and other behavioral differences between groups is considered. But I don't think that IQ alone is as powerful of an explainer as you give it credit for. You say there are outliers, but I think that understates it. East Asians have higher IQs than Europeans, but industrialization was behind by a couple centuries. And the Europeans who landed in Australia brought their institutions with them as much as their IQs.
Sure and that goes back to there being other factors like institutions.
But again you run up against endogenicity. (ugh, that's a hard word to spell) At the biological level, malnutrition and parasitic infections are known to be hard on brains. At a more cultural level, consider the Flynn Effect -- it's well-documented that over the past 150 years or so the average IQ of the highest developed Euro-American countries has increased by about 30 points, which is one s.d. It's clear that wasn't due to any sort of evolutionary selection so the cause must be environmental. The best guess I've seen is that the kids have been raised with a far richer information diet as "the media" have become more and more elaborate. But of course, parallel effects would be expected from formal schooling, and the amount of schooling the average person has seen has increased greatly.
If you look at a place like Congo, they've got plenty of malnutrition and chronic parasitic infection, weak schooling and underdeveloped media. So to make the comparison with the US fair, you'd first have to fix *all of those*. Then measure Congo's average IQ and economic growth.
But for that matter, several African countries have seen very good growth rates at times. IIRC Ethiopia managed 5% per year for a decade or two. But even those successes are invisible to an outsider -- It's hard to notice that doubling the GDP/capita from $600 to $1200 is a fantastic achievement.
But everywhere was like that before industrialization, and yet they industrialized anyway.
"For every complex problem there is an answer that is clear, simple, and wrong. - HL Mencken
Yeah I was thinking the same thing. Fernández-Villaverde asks why so few people are studying the problem and I thought "they think they already know the answer but can't say it." Even if the hereditarian explanation turns out to be wrong, the taboo deters people from studying the question in the first place.
Overall good article.
But I think the analogy to aging research is misguided
We should be spending a lot more time and money on aging research.
Moreover. We will be able to definitively answer aging eventually. Because it's something that you can test in the real world
And in fact, the first tests are occurring now in FDA trials in humans. Life Bioscience is testing age reversal for human eyes.
I suspect we will not come up with a good model here because there are variables that are off limits. If I want to research why South Korea got rich while Bolivia didn't I would have to consider that South Korea has an average IQ of 107 vs 96 for Bolivia. But the researchers will never do this because that is heresy and a career ender. It certainly isn't the only factor here. Presumably North Korea and South Korea had a similar IQ average. But to leave it out as a variable is simple unscientific.
Same thing with culture "a culture of progress, innovation, and openness to technology" as you describe it is not going to cut it. It's vague to the point of uselessness. What needs to be done here is an actual quantification of the behavioral differences between a Bolivian and a South Korean, but that is also getting into heretical territory.
If there's a theory to be found here it's not going to be found by going over the same factors over and over again. It will be found by going where everyone else has been afraid to go.
“Where it is a duty to worship the sun, it is pretty sure to be a crime to examine the laws of heat.” - John Morley (1st Viscount Morley of Blackburn)
Examining IQ and culture is a crime when worshipping at the altar of "diversity" is de rigeur in academia.
Great summary of the state of the field.
I think you're right about the inherent difficulties, but I'd say that economists need to learn more history. I'm not against the math-stats revolution that has swept the profession, but 'the economy' is an in inherently multi-diciplinary field. A good analogy is medicine: if you study something as complicated as bodies, you need physics, chemistry, cell biology, and a bunch of other subfields. If you study 'the economy,' you need math and statistics, but also political science and history.
I read 'How Asia Works' after already having a Masters in Economics for a long time, and doing a lot of outside reading--and I was absolutely dumbstruck by the chapter on land reform. My first thought was "how the HELL had I never even HEARD of this?" How had I had five years of formal education in economics, and another decade of self-education, and I'd never even heard of the idea that land reform accelerated growth? I'd even taken a graduate level development economics course! And the idea that land reform increased agricultural production never came up, even as a hypothesis.
Reading that book, it struck me that economists need to learn more history. It's shocking that you can get a Phd in Economics and basically be a statistician who's learned very little about the most basic economic facts. Yes, history has a lot of limitations, but it's crazy that economics education barely even requires it any more.
Useful reflection, I wonder however if perhaps one should profitably invert the Tolstoyian phrase All "happy families are alike; each unhappy family is unhappy in its own way" for econ dev success on national geography (to eximine if the unhappy, i.e. the unsuccessful are more alike than not in combinations of policy and other error or other handicapping (as like national state incoherence [Nigeria])
Although it's not totally on topic, I'm really tickled to see your regular mention of medical research targeting the biology of aging, since I've had an intense interest in it for several years and the techno-optimist in me is still fascinated by the idea of indefinite healthy lifespan. I'd be curious to see whole posts where you explore economic and social implications of compressed age-related ill health and increased healthy lifespan.
Here's some reading, although you might already be familiar with it:
https://www.nature.com/articles/s43587-021-00080-0
https://onlinelibrary.wiley.com/doi/10.1002/hast.5007
https://www.afar.org/hallmarksofaging
Noah, this cuts to the heart of something I've been writing about. The development industry is obsessed with the seen—schools, clinics, kilometres of road—while systematically ignoring the unseen infrastructure that makes any of it yield lasting returns. Contract enforcement is the textbook example. Everyone in the profession knows it's foundational, but what do development institutions actually do about it? A bunch of useless seminars, training programmes for judges who will never rule independently, and "capacity building" reports that gather dust.
I've tried to move beyond critique to design. Satyapur is a proposal for a privately-chartered commercial court system within Bangladesh—a new jurisdiction modelled on Delaware, Singapore's judicial credibility, and Dubai's DIFC. It requires zero public funding, operates on an entirely opt-in basis, and funds itself through registration and filing fees. The idea is to create a space where contracts are actually enforceable, so the economic activity that currently flees to Singapore or Delaware can stay home.
The two manifestos lay out the full architecture and political strategy:
Satyapur: The Delaware of Bangladesh (https://mdnadimahmed888222.substack.com/p/satyapur-the-delaware-of-bangladesh)
Satyapur II: Strategic Implementation and Political Sustainability (https://mdnadimahmed888222.substack.com/p/satyapur-ii)
If development economics is serious about moving from the seen to the unseen, charter jurisdictions like this are one concrete place to start. Would be curious to hear your thoughts.
Though of course a private, opt-in system isn't like what happened in developed countries, where a well-functioning legal system was definitely *not* opt-in. Also, the ugly part is that in the early stages of development, good *land rights* are critical, and the adjudication of land rights is always the most slow-changing part of any legal system.
That's true. A good legal system for the corporate sector can at least be self funded and also contributes disproportionately to productivity growth.
My proposal uses the Singapore appeals process to overcome the reputation problem.
I know that I am a bit late to the party, but I published a response to this article:
https://frompovertytoprogress.substack.com/p/progress-studies-needs-synthesis
I agree with many of your points, but I arrive at a very different conclusion.
The maddening thing looking at underdeveloped countries is it sure seems like all they need is a bunch of people with the correct know-how to come in and start industrializing everything. And human capital is cheap! All you need to do is take your bountiful supply of cheap human capital, put them through some education, and some of them will prove quite skilled! Why is a single generation of universal literacy and internet connectivity not enough to pull a poor country up to near parity?
RCT - what is it? AER? Not the easiest piece to read, not as interesting as I had hoped.
Really good write up. I always for some reason get surprised when really smart people that I respect and who seem to be open minded and pragmatic fall into the trap of trying to look for simple answers to complex problems, or refuse to think that multiple things can be true at once.
The Washington Consensus aimed directly at those big questions, but that wasn't "Proressive." The World Bank turned away from pusing policy reform. USAID in the GWB Administrtion had a few.
High quality narratives like How Asia Works are not only attentive to a lot of details but also do a soft, verbal version of cross-country regressions. For example Studwell himself observed that the basic ingredients of the policy mix he recommends (land reform, export discipline - i.e. support conditional on market performance, especially on export market performance because the latter is independent from local political connections - and control of financial institutions) were all present in the countries that were growth successes (Korea, Taiwan, Japan, later China), while these good policies were not followed in the countiries whose catch up was unsuccessful (Indonesia, Malaysia, Thailand and Philippines). This is quite convincing even if it is not a mathematic and scientific method. If this is coupled with microeconmic work on the effect of policies on certain industries, as was done by Nathan Lane on the Korean heavy and chemical insdustry drive, which verifies the narrratives, we can be as certain about the results as one can get in these fields where counterfactuals and experiments at large scale cannot be run. I am certainly convinced and I don't think any greater certainty is needed for policy. If one is more demanding in terms of evidence to drive policy, a state would be paralyzed as no policy option can be supported by mathematical, unquestionable proof. Imposing such requirements would effectively condemn a state to doing nothing, which is simply laissez-faire policy. There is no neutral policy in any event. Further, I don't mind if the focus of microeconomic work and narratives is policy because policy is exactly what countries can change - they cannot change their resource endowments and any attempt to fickle with things like culture is bound to fail. Countries can change the quality of their human capital by education but Studwell is also very instructive here as he lists a number of countries with good eductional systems and high quality human capital, which failed economically (Cuba, Soviet Union, Philippines etc) and argues that spending more on education alone is not a game changer. In effect, usually policy driven growth creates demand for educated people and this leads to the increase in the quality of eduction rather than the other way around. So focusing on policy and not getting stuck with unmutable things like resources, ethnicity, culture etc. is the right way to go (with the necessary humility of course).