Book Review: How Asia Works by Joe Studwell

How Asia Works has a theory about how development economics works: You must follow The Way. To the extent that you follow The Way, your country will prosper. To the extent that you fail to follow The Way, your country will languish. The Way is divided into three commandments:

  1. Thou shalt enact real land reform.
  2. Thou shalt protect infant industries and enforce upon them export discipline.
  3. Thou shalt repress and direct thine financial system.

The author is supremely confident in his conclusions, considering them “proven” by history. He takes us on tours of Asian countries, to show us the wonders that await the faithful, and the failures of those who strayed from The Way. He does not, however, expect us to take his prescriptions on faith. He explains why there is no other way but The Way, shows how each divergence leads to ruin. At each step of development, he has a model for what is going on, whether or not he quite describes or thinks of himself as having one, and all three models fit into a unified model of how to think about developing countries.

It also implies a lot about the developed world, along with other interesting consequences, but he never mentions this. My guess for why is that they are not conclusions he particularly likes.

From here on in, except where stated otherwise, I’ll be stating the book’s perspective rather than my own, and what seem like its implications.

Let’s take a look at the model’s components, starting with the first one: Agriculture.


The model goes something like this:

In America, farms are big. Really big. No, bigger than that. Farms are big! Labor, on the other hand, is expensive. Thus, we build our agriculture around being able to get a reasonable crop out of a large amount of land, rather than trying to maximize the yield on a farm of a given size.

In most other places, farms are small, and land is often a limiting factor. In places like Japan and South Korea, every inch of available farmland is in use, so it is vital to maximize yield on that land. If you get good farm yields, you can export food to generate precious foreign capital, whereas if you get bad yields, you need to import food, eating up the money you need to spend on industrialization. The question is, what determines yields?

The thesis of this section is that intensive, motivated labor is the most important input, with proper tools and equipment a distant but vital second place.

In order to get the most out of land, you need to know it inside and out, and be willing to work tirelessly day after day. You need to know where you can plant multiple crops at once, and when to weed them and how to scape them so they will both grow, and how to rotate them to never waste a day. You need to work to improve the soil, to rest and renew it, to learn new techniques. The hourly rate on all of this is, shall we say, not great, but it is also non-zero, and there aren’t any good alternative ways to turn your hard work into money.

Traditionally, the land in villages is mostly or entirely owned by the rich. These rich then rent out the land, or hire laborers to work it for them. Either way, this creates at least four problems. The first problem is that the poor workers are not properly motivated to maximize the crop, because even assuming they keep a portion of the crop, they are paying a tax, often 40% or more, to the landlord. What you sell, they’ll give you a terrible price for, since the local rich dude, likely sporting a villain mustache, is a monopsony buyer. The second is that they have even less incentive to maximize next year’s crop by learning about and improving the land. Why work hard to improve someone else’s land? If you make it thrive, they reap the benefits. Third, you won’t have the right equipment, tools or fertilizer. You can’t afford it! Even if you could afford it, you’d be paying to improve a crop you have to pay a huge tax on, so it wasn’t worth it. And if you’re short on funds, the only way to get more is to borrow it, at 100%+ yearly interest rates, from the local rich dude with the mustache.

In this world, the poor are always broke, always on the verge of starvation. Their crops are taxed, their profits cheated, their loans usurious. They have no chance to escape the trap, and will never be the producers of the surplus the country needs, nor the savers, spenders and consumers that are necessary for the next stage in the plan.

The good news is that there is a solution: Land reform!

Land reform is a simple idea. You take the land, and you give it to the farmers.

This tends to be unpopular with the rich dudes with the mustaches. You offer them compensation for the land you took, but it is in the form of money, you pay it out over time at a rate of interest far less than inflation, and it was already much less than land typically is sold for, so mostly you’re just taking the land by force.

They are understandably upset about this, so the landowners do everything they can to stop you. They will try to keep as much land as they can, and the best land, for themselves. Where that does not work, they will do their best to deny the new owners what they need to survive. They’ll buy their crops at bargain basement prices, lend them money for equipment and tools, and money for simple survival when the money runs out, at usurious rates, and cheat in every way you can think of, to get the poor to sell the land back to them.

By default, the rich will succeed. Even without most of the land, they’ll still have a lot more wealth and power. With no margin for error, every year is an invitation to disaster. One by one, the new owners will fall on hard times and have to borrow, or sell the land back. Once they start falling behind, it gets steadily worse. Because of this, the government needs to do more than just land reform. It also needs to make sure the new owners have a whole support infrastructure in place. They need buyers to ensure a decent price for their crops, irrigation and fertilizer for their land, a tractor they can share, new higher quality seeds, and other neat stuff like that. Otherwise, you fail – for example, Meiji Japan tried land reform without enough secondary help, and it worked for a while, but by World War II the old owners had back most of the land.

With enough help, the result is self-sustaining and rapidly pays for itself. With their own land to work and improve, and the tools they need, yields double and triple. Even in places where land reform mostly didn’t work, there are pockets – he looks at one in The Philippines – where it succeeds because an NGO or other source steps in and provides that help. Soon everyone is not only not starving but has a surplus, a surplus that creates a domestic market for your new industries, and savings and exports that will be your sources of hard currency.

Land reform also has the side benefit of killing support for a communist revolution. Since communist revolution is the actual worst thing that can happen, that is a pretty great side benefit.

This already has lots of interesting implications. Huge, if true!

Once you have this under control, your country is ready for step two.


In the How Asia Works model, industrialization is fragile and requires careful guidance. By default, you will do all the wrong things, because the most short-term profitable things to do don’t lead to long term success. Long term success depends on developing the skills and knowledge to compete on the international stage, and not rely on foreigners for the highest value parts of the production chain. The problem is that this is a long-term project, and it will lose money for a while.

The first step of The Way is to protect your infant industries. The book says everyone has always done this: England, America, Germany are the prime examples outside Asia, and the book claims there are no exceptions. Without protection, your first attempt at a product will suck and no one will buy it when they could import something instead, so you have no market, no profits, not even reasonable feedback loops for learning, and you never get off the ground on your own. Instead, all you do is make your labor available to foreigners who will run the factories, keep all the vital knowledge to themselves, and also keep most of the profits.

Now that you have protected yourself with tariffs and used land reform to create a rural surplus, you now have a captive domestic market for your goods. That starts with textiles because they are low skill and can actually be competitive, so everyone always starts with textiles, but the quicker you move beyond that, the better. They will need your cars, your steel, your widgets of all shapes and sizes. This provides the incentive necessary to build a factory that produces crappy cars and another that produces crappy steel, so you can learn how to make cars and steel that aren’t as crappy.

By granting an oligopoly the right to produce for the domestic market, and giving them access to low-interest loans, you make sure that production happens, and soon you have captains of industry. The problem is, captains of industry do not want to be captains of industry. That sounds like a lot of work, and not that much profit. Sure, they will happily use foreign expertise to sell their crappy cars and crappy steel to a protected domestic market. That part sounds great! But learn how to compete for real? Invest in the future? Why do that when you can make a quick buck and use it to build casinos and luxury condos?

In cases where these seekers of rent are allowed to do that, that’s exactly what they do. Don’t get me wrong, I like a good poker game and I love me a good luxury apartment, but they are not the path to industrialization and a modern economy. On a fundamental level, they don’t produce anything. They are consumption goods, without the positive externalities you are looking to compound over decades.

A smart government, therefore, will make sure that doesn’t happen. In exchange for the right to sell to this captive domestic market, the captains need to invest in the future. The government will decide what industries and technologies are important, and guide them, using the levers of finance and regulation to force compliance. The captains must take on the projects of Vital National Interest, even if the returns are not as good. Year by year, they must use more domestic inputs, even if they need to develop those inputs themselves. They must use every deal as a way to get the foreign devils to give you access to their technology, and you reverse engineer everything you need and learn to operate it on your own, even if that slows you down. If you don’t understand it, you don’t do it, period. Remember, you’re getting massive subsidies from the government, so doing things the hard way is survivable.

Most importantly, they must be subject to export discipline.

Export discipline is the magic that makes all of this work. If you don’t have to export, you never get the feedback on what the consumer market wants, and you never stop producing crap. The Soviets learned this the hard way, and never learned how to produce consumer goods. But if you make the exporting a condition of the massive (and oh boy are they massive) subsidies you are throwing at these captains, they will sell their products overseas no matter what. They will take a loss. They will find no one wants to buy, even then, and they will ask why, and they will learn. They will iterate. And over time, they will figure out how to make something that can compete.

Other countries will of course try to tell you that this isn’t fair, and you can’t be doing this, but you just don’t listen to them and you do it anyway. They won’t be happy, but they benefit enough from trading with you that they’ll put up with it.

The companies that can’t or won’t export, you cut off from the free flow of cheap money. They die.

Thanks to export discipline, together with massive subsidies, Japan, South Korea and Taiwan learned what the global market wanted, and figured out how to provide it, reaching the technological frontier in the places their governments led them. At this point, you have a Sony or a Samsung, and you can (mostly) let them find their own way.

Think of the model as saying that some investments have massive positive externalities for the country. The captains of industry won’t capture those gains, so they will make choices that are right for them but wrong for you, the same way that the mustache-twirling lords of the countryside were doing things that were good for them but horrible for the country. Once again, you must force them to do what you want. Using massive bribes everywhere, and outright coercion when necessary (e.g. General Park of South Korea showing he was willing to throw his capitalists in jail if they didn’t do what he wanted) you can correct the incentives, point them towards your country’s long term needs, avoid the easy paths that don’t have a future, and be on the road to success.

There is just one more problem you need to solve, and that’s finance.


Finance has the same nasty habit as the captains of industry do, which is the desire to make as much money as possible. Like the captains, the bankers get to use their position to make tons of money. They have a captive source of savings that they can pay negative real interest rates. Financial repression makes sure of that. On top of that you have even more cheap funding from the government because the government needs you to make even more loans than that so it can keep subsidizing all those captains of industry it wants to build shiny new factories that take a decade to make any money.

The government even has to bail the bankers out if they fail. So being a banker is a pretty nice work if you can get it.

The problem, from a development perspective, is that the way to make as much money as possible is to do one of two things.

Option one is to loan money out at the highest possible rate of interest. They then tend to go out and create things like casinos and luxury condos, often not even in your country. That’s the fair and honest play.

Option two is to loan money to your friends and engage in massive corruption. That’s the natural outcome if you’re not careful.

You want neither of these things to happen. Instead, you want your bankers making loans to people who will produce things, in your country, that lead to industrialization via the learning process in part II, and further improve the fiscal situation. You especially don’t want the money leaving the country for projects elsewhere.

The way you do this is that you directly reward banks that make loans to the industries and companies you like, by rewarding them with more cheap funds and renewing their charters. The banks that try to make loans to people who want to borrow money for “non-productive” purposes, or who loan to cronies or other projects that don’t prove themselves, find themselves on the outside looking in, without additional funds. You take the money from taxes, and the money from financial repression, and funnel it into development.

Where you do this, the financial system is an asset and works for you. With so much subsidy, the fact that these loans are at moderate interest rates, often negative real interest rates, is fine; the bank is de facto acting as a branch of the government much of the time. If you do not do this, you get banks using the people’s money to build casinos and condos, or worse, outright stealing the money via corruption.

Capital controls are also vital. If you allow foreign money to flow freely in and out, three very bad things happen. First, they will use that money to produce the dreaded consumption goods, which will then get consumed and will consume people’s money, hurting your development. Second, to the extent the money is productive at all, it will go towards exactly the short-term kind of production that doesn’t teach your country how to do things for themselves, letting the foreigners to reap the benefits. Third, because the money is constantly seeking the best possible returns and moves in giant flows of greed and fear, when trouble comes, all the money will run away and turn a manageable situation into a crisis like in 1998.

Now you know The Way.


The natural enemy of The Way is The Washington Consensus. Largely because of the ideological commitment to capitalism and struggle against communism, people like the IMF are fanatically pushing “free markets” which means not protecting your industries, not controlling the flows of money, and counting on everyone’s self-interest to fix all your problems. The model says this is disastrous, because the externalities involve steering resources in all the wrong directions.

Alas, the IMF is one of the only sources of badly needed money, so once South Asian countries fell behind, they had no choice but to kowtow.

The other enemy is lack of understanding of what must be done. The South Asian states like Indonesia and Malaysia understood that Japan and South Korea had a good thing going, but they did not realize what caused their success. Export discipline was dropped. Land reform was incomplete. If you do a half-baked job of The Way, The Way will do a half-baked job of getting you there, and you will fall further and further behind. This also forms the fully general excuse: If something went wrong, something wasn’t done correctly.

He also makes some comments about China, but I will skip those as they did not seem especially insightful.


Now that we have Studwell’s perspective, it is time to turn this into a model of a how one might develop a country. At the start, there is a giant pool of unused and under-utilized labor. By giving those workers land and other complements to their labor, you allow them to work extra hard. By default, you earn huge returns on your investments, because rural interest rates are effectively over 100%, even if you can’t capture those gains directly. Labor stops being wasted and starts producing more output, which creates a surplus of goods and labor. Using financial repression and protected markets, you then capture much of that surplus, which you can use to hire the labor, and to industrialize.

Industrialization and finance create natural oligopolies, which means those who get to participate will make economic profits. You increase those profits by feeding the system the rural surplus. By controlling who gets to participate in these inherently profitable activities, you have immense leverage over their activities. You use that leverage to steer them towards actions that drive production rather than consumption, and that have positive externalities in the form of technological know-how to produce more. You also steer them towards creating jobs, because that is also a strong positive externality; you ensure labor isn’t standing idle.

All of these activities then create productive capacity and technological catch-up. You avoid buying things that do not do this, so your balance of payments is good. The longer you can steer everyone away from consumption, or from taking their surplus elsewhere, the longer you can enjoy this compounding interest, and your country rapidly gets richer. Eventually, you’ve caught up, at which point you can engage in free trade and free capital markets and let your people have nice things.

Export discipline is necessary for two reasons. One, it forces industry to learn what people actually want and how to produce it, in real competition. They get the benefits of a competitive marketplace. Second, it allows you to judge whether industry is doing real things, and cull those who are not performing. In theory, this is a way around the socialist calculation problem as well as give people the incentive to work.

That is the heart of the trick. The system aims to simultaneously give everyone a surplus, and also get the incentive and discovery benefits of robust markets. That is when markets work best. Everyone has slack, and externalities and market failures are mostly being fixed.

Rural markets break down because there isn’t enough density to create competition, allowing the local rich to use monopoly power to capture all or more than all of the surplus. Rather than destroying free markets, smart redistribution creates them.

Local markets provide a place to develop and test early stage manufactured goods without being wiped out by foreign competition. Export discipline then allows you to get the price signal and other information that full competition gives you, without getting wiped out by competitors that have a huge head start, or allowing outsiders to extract the rural surplus.

Financial markets under this system are more heavy handed, with banks serving mostly to execute industrial policy, but if that policy is staying sensitive to price signals and can remain free of corruption, maybe that is all right. The system guards against finance capturing the surplus, or creating a financial crisis every few years.


Should one buy any of this?

Good question. Studwell believes something. Clearly, he is massively overconfident. He considers his thesis “proven” by the historical record, as if that was even a thing. He is also being selective and presenting facts that support his thesis, taking us on tours of carefully selected portions of carefully selected regions, and not even trying to be even handed. His theory has a central concept that reduces his complexity penalty, but it’s still a pretty big penalty. At each step, there is a correct set of things to do, and the extent to which you comply perfectly determines success. You can rank countries from most compliant (Japan and South Korea, then Taiwan) down to least compliant. No doubt the details are overfit and given oversize importance.

That is true even without checking his facts, or looking for things he so conveniently overlooked to strengthen his case. Two colleagues saw me reading the book. Both commented that the book seemed small, given its title.

My answer, despite all of that, is a tentative ‘partially.’ I think there’s true and important things being pointed at, especially with regards to agricultural policy and its role in success. I extracted a useful model, another perspective on how things can work. Given what I normally get out of similar books, that has to be considered a roaring success.


What does this then imply about potential interventions now? What does it suggest about first world policies?

There is a clear intervention being suggested heavily, which is land reform and rural aid. The book mentions that there are NGOs that do this. One can go in, buy land, and then distribute it to local farmers, and give them the tools and equipment, and market access, they need to succeed. If we model the rural poor as existing in a world where interest rates are over 100% per year, then investing in them should have very good returns. If they can stop borrowing money at those rates, or selling their crops at harvest time when they have to accept steep price discounts, what was a failure to make ends meet suddenly is a surplus that can be used to improve the farm or send kids to school. If it works as advertised, this becomes self-sustaining.

The other intervention is less personally actionable and more well-known: allowing free trade with developing nations, even if they are not giving us symmetrical terms. If export discipline is key to success in development, and development is the best way to lift people out of poverty, then shutting people out of our markets, or demanding that they engage in ‘fair’ practices that effectively shut them out, is the worst thing you can do. These developing nations are effectively buying an education from us, making little or no (or less than no) profits to get the benefits of real market tests. That’s win-win, so denying them this isn’t just cruel, it’s downright evil.

For the first world’s internal house, there are a number of parallels that imply policies.

A big implication is that government should work to shape the incentives of companies and individuals in a way that better accounts for the externalities they have on the economy (also externalities on other things, of course) while retaining strong market mechanisms. For example, if you believe that creating jobs creates a large positive externality, those that create jobs should get paid for doing that, the same way that a company should be taxed or punished for polluting a river. We should tax consumption rather than production, and in particular we should seek to heavily tax consumption of positional and signaling goods. This also happens to have lots of other nice properties for other reasons.

The market mechanisms are even more important. Everywhere in the book that a government tried an intervention without forcing the participants to face a continuous market test, the results went nowhere. Everywhere around us today we see areas of developed economies that no longer face good market tests, and thus we get the recent discussions of cost disease in health care, in education, in housing costs, in subway construction. Export discipline is a great trick because it means that not only the participants must pass a market test. The rules and regulations you impose must also pass a market test. If outsiders have to choose to buy our products, without any compulsion to do so, that forces us to get our house in order, no matter whose ‘fault’ it is.

The key insight here might be that the market test can be distinct from the bulk of how you distribute the good. So long as some purchasing is done in a free and open market, you can use those market prices and standards as a baseline. Maybe instead of throwing the entire health care market open, we can create some sort of smaller test world (any state want to volunteer?) where we can ‘export’ our care into a free market (foreigners who come here for health care definitely help, but their willingness to pay is through the roof, so they’re not going to solve our cost problem for us), and pay only a fraction of the costs while getting much of the benefit of market signals. Then use the same trick in other places. Maybe.

This entry was posted in Reviews and tagged , , , , . Bookmark the permalink.

6 Responses to Book Review: How Asia Works by Joe Studwell

  1. Anonymous says:

    Interesting – particularly the idea of how export markets can be used as a source of honest price signals without full exposure to market competition.

  2. Glen Raphael says:

    It sounds like the author answered the question “when it worked, what policy was being followed?” rather than “when countries followed this policy, how often did it work?” One trouble with industrial policy that tries to steer the market in particular directions is that often the people doing the steering don’t know where they are going. Does he discuss things like Japan’s Fifth Generation project, where they spent an ungodly amount of money trying to invent AI and next-generation parallel computing and it all just kind of fizzled out and didn’t get anywhere? (The old conventional wisdom – based in part on things like that – was that successful Asian countries were fast followers but that once you “catch up” you can’t use the same policies that got you there to *keep* innovating.)

    • TheZvi says:

      He did not mention Fifth Generation; facts that did not fit the narrative seemed like they were unwelcome. He did mention some things that he felt Japan did wrong after it caught up, such as its massive continuing farm subsidies. The claim that countries should not use industrial policy after being caught up implies he understands the problem, though – that if you are copying someone else, you can have a reasonable idea what to do, but if you are at the technological frontier, as he calls it, things are a lot trickier.

  3. Quixote says:

    Does the land reform success case suggest capital reform in the West?
    Oligarch with mustache:: 100 families with inherited wealth ~= half US nonland assets
    Farmer who doesn’t work hard because doesn’t own land:: Office worker who doesn’t own company
    Greedy barons who recapture farmer’s land:: student loan companies that prevent worker capital accumulation
    confiscate capital, give everyone a million dollars to start a business. Watch productivity go up. Profit.

    • TheZvi says:

      I was consciously laying it out with that as an implication, one which the book seems to deliberately not think about other than to say that things after development are different.

      I would say that land reform in developing countries seems to have a strictly better case than capital reform in developed countries such as the United States (the breach of trust matters more, the difference in returns not only being smaller but not having an obvious directional sign, etc etc) – if you support the second you should support the first, and if you support the first, that should increase your willingness to do things like the second. It in no way commits you to anything.

      The student loan part comes closest to being an obvious parallel, and is the place where the current system seems most obviously both unfair and inefficient. I am very much against tulip subsidies (

  4. Quixote says:

    Note, not saying I fully endorse the above. Just saying that how you feel about one claim should relate to how you feel about the other.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s