What is the lesson from the East Asian miracle?

By Scott Sumner.

"Tyler Cowen recently linked to an interesting essay by Byrne Hobart on East Asian development. Here’s the opening paragraph:

I’ve always had highly libertarian instincts, for both pragmatic and ideological reasons. You say civilians should be able to own rocket launchers, I demand that these rocket launchers not face a sales tax. But for me and people like me, the East Asian economic miracle poses a serious challenge: the greatest anti-poverty program in history involved not just a lot of capitalism, but a ton of state intervention as well. The history of East Asian economic growth in the last half of the twentieth century is a history of academics and the World Bank insisting that their policies couldn’t possibly work, followed by decades and decades of torrid growth.

The last half of the 20th century is a long time.  When most of these growth miracles were occurring during the 1950s-80s, academics and World Bank types tended to favor the statist model used in Latin America, rather than the more free market model used in East Asia.  Reading the final two sentences of this opening paragraph might give readers almost the opposite impression, that the academics favored free markets and the East Asian model was seen as violating that consensus.
In fairness, things changed during the 1980s and 1990s.  Expert opinion began moving toward the neoliberal model, aka the “Washington Consensus”.  But when I was studying economics in the 1970s, the debate was between the relatively statist Latin American model and the relatively market-oriented East Asian model.  Latin America did fairly well in the early post-WWII decades, but by the end of the 20th century the East Asian model was clearly more successful.
Here’s another questionable claim:

No American company or investor has improved as many lives by as wide a margin as MITI, Park Chung-Hee, and Deng Xiaoping. But the methods are, at heart, startlingly similar: identify a critical inflection point, make a bold bet on an unproven market, “blitzscale” as quickly as possible, deftly react to crises, and carefully expose budding monopolists to hormetic doses of competition until they’re strong enough to monopolize on their own merits.

Sorry, but I don’t see any relationship at all between the model described in the final sentence and the policy views of Deng Xiaoping.  There was state involvement in the economy under Deng, as there was under Mao.  But Deng’s key reforms were all about moving away from industrial planning.  Empowering individual farmers, allowing township enterprises in the countryside, allowing foreign investment to boost exports, etc.  It was all about the state pulling back from any attempt to control the economy.

This paragraph may be more accurate as a description of Japanese and Korean policies.  But that doesn’t mean those specific policies were successful.  Here’s a question for readers better versed than I am on these issues.  Are there rigorous, empirical, scientific studies that show MITI was successful in boosting Japanese growth?  I ask because there are plenty of development experts who don’t think it was effective, and plenty of anecdotal evidence that it was counterproductive.  Here’s one example from the auto industry:

MITI Proposes a “People’s Car”

In May 1955, the Ministry of International Trade and Industry announced the “People’s Car” Plan, which gave Japanese car manufacturers an excellent opportunity to develop original models of their own. This plan called for a car weighing less than 400kg with an engine displacement of 350-500cc, fuel efficiency of 30km/liter, 2- to 4-person occupancy and a cost of not more than \150,000 (this price was later revised to \250,000).
The response of manufacturers to this proposal was that such a vehicle would be “impossible to manufacture with the performance and sales price requested”. After much back-and-forth discussion between the industry and government, the “People’s Car” Plan was eventually scrapped.

Another example is Honda. MITI bureaucrats told Honda not to produce cars, to stick with motorcycles.  Honda’s CEO later said the company would have been more successful without being “bullied” by MITI.

There are many more such anecdotes, and there are also lots of anecdotes about how MITI did help Japan to succeed.  That’s why I ask if there is any rigorous scientific study of this issue.  I’m not saying it wasn’t effective—perhaps MITI helped overcome institutional barriers to needed industrial mergers.  I’m asking if there is non-anecdotal evidence for that claim.
Bryan Caplan has a post discussing Bayesian reasoning:

Suppose someone sends you a new article claiming X.  Intuitively, we think, “This will either make you more likely to believe X, or have no effect.”  Once you understand Bayesian reasoning, however, this makes no sense.  When someone sends you an article claiming X, you should ask yourself, “Is this evidence stronger or weaker than I would have expected?” If the answer is “stronger, ” then you should become more likely to believe X.  However, if the answer is “weaker,” then you should become less likely to believe X.

Hobart cites a number of books on East Asian development, some of which were written by proponents of interventionist policy models.  Readers of these books should not ask themselves whether the arguments are persuasive.  If you are not an expert in this area, then of course they are persuasive!  Readers should ask if the arguments are more persuasive than expected.

If the author talks about the merits of planning, does he or she explain why Hong Kong grew extremely fast without planning?  If they discuss the merits of selective protectionism, do they explain why Singapore grew very rapidly with free trade?  If they discuss the merits of trade surpluses, do they explain why South Korea’s period of fastest growth coincided with current account deficits?

And do they make misleading claims, such as this item from Hobart’s article:

As a consequence of all this, Japan’s GDP per capita in 1950 was under a fifth of the US level. Today, Japan’s GDP per capita is about two thirds of the US level.

This claim is true, but misleadingly suggests that Japan has recently been doing something right.  In fact, Japan’s per capita income was more than 2/3rds US levels in 1990, and over the past three decades Japan has grown slower than the US; it’s falling further behind. This despite the fact that you’d expect countries engaged in “catch-up growth” to grow faster than the US.  Not to mention that the Japanese work longer hours than Americans and are better educated.  From this perspective, the Japanese model looks like a relative failure.

Readers of these books on Asian development should also ask what policies the author’s focus on.  Let’s say the author is a left-leaning skeptic of free market capitalism.  Do they talk about the fact that most of the fast growing East Asian economies have much lower taxes than Western economies, or do they focus on policies that might be more appealing to skeptics of “market fundamentalism”?  Conversely, if a conservative discusses the successful neoliberal Nordic model, do they also mention the high taxes?

Just to be clear, Hobart wrote a very thoughtful piece, based on extensive reading.  I’m no expert on East Asian development, and don’t know enough about institutions like MITI to have a well-founded opinion on the subject.  Rather I’m encouraging people to be skeptical of any group of sources that do have an agenda."