Tuesday, September 06, 2005

How did the East Asian economies take off?

What does it take for an economy to keep on growing fast?

Output comes from combining inputs. Hence:

1. More input, more output.

Combining inputs in new and better ways make for more output. Hence:

2. Better technology, more output.

Since we are talking about economic growth, then "inputs" and "technology" also need to be imagined in very abstract, aggregative terms. So: the labor input aggregates all types of labor inputs in the economy; the capital input aggregates all kinds of capital in the economy. And technology aggregates all kinds of technical know-how in the economy.

There's a little danger in all this aggregation. It may conceal a lot of little goings on that make the aggregate phenomenon possible. For example, "capital accumulation" is a catch-all term, but this requires investment by a large number of firms across many industries; such investments only materialize if investors foresee high future returns from these companies in these industries.

The East Asian High Performing Economies (HPE) are typically listed as: Japan, South Korea, Taiwan, Singapore, and Hong Kong. Other Newly Industrializing Economies in the region are: mainland China, Thailand, Malaysia, and Indonesia.

We limit our list to the East Asian HPEs. Studies have suggested that economic growth was largely the result of accumulation of physical capital, investments in education, and increases in labor force participation (think women!) The importance of rapid capital accumulation is not debated (though there is some controversy about the importance of rapid technological change.)

Clearly, the East Asian HPEs did much to promote investments and education. Physical infrastructure was well provided. Macroeconomic policies kept government deficits trim and inflation well under control. Because of technology investments, the agricultural sector in these economies (except of course Singapore and Hong Kong). Finally, the HPEs started out with relatively low income inequality.

What is less clear is the role played by government industrial policies. While it appeared that the scope of government intervention was massive (a fact pounced upon by the "nationalist economists"), this does not mean that government intervention was partly responsibile for the take-off. In many other developing economies, government intervention was also massive, targeted at industrialization based on substituting domestic for imported manufactures. In such countries there was unfortunately no take-off.

Moreover, measures of government intervention suggest that the quality and severity of intervention differed greatly from the typical import substitution package. The most important difference is that industrial policy aimed at expanding exports, rather than restricting imports. Importation of inputs for export-oriented industries was in fact strongly encouraged. Seen in this light, the East Asian economies were unusually open and market-oriented (World Bank, 1994; Stiglitz and Yusuf, 2001).

So what does it take to take off? Stable macropolicies, an open economy, investments in education, and technical change (especially in agriculture) are all very important. Narrow sectoral promotion, if pursued at all, should be geared towards export competitiveness and not import substitution.

How well is the Philippines doing here? Well consider this news item, which is illustrative of how economic policy is conducted in this country. Open economy? Go figure.


World Bank (1994). "The East Asian Miracle." Oxford University Press, Oxford.

Stiglitz, J., and S. Yusuf (2001). Rethinking the East Asian Miracle. World Bank: Washington, D.C.

Pack, H. (2000). "Industrial Policy: Growth Elixir or Poison?" World Bank Research Observer 15(1)47-67.

[Message edited: econblogger, 1700:5 September 05.]


walker said...

The most important difference is that industrial policy aimed at expanding exports, rather than restricting imports. Importation of inputs for export-oriented industries was in fact strongly encouraged. Seen in this light, the East Asian economies were unusually open and market-oriented.

While it is true that the East Asian developmental state does not follow an "import substitution" strategy to development, this is not quite the same as saying that they do not restrict imports. Protection of the home market in industries considered strategic is a basic strategy of the East Asian developmental state. A locked in home market is an invaluable asset to a good manufacturers, as they can leverage stable profits in the home market to undercut competition in foreign markets.

Econblogger said...


Quite right, they did restrict imports. However it remains to be shown whether this actually contributed to their take-off, or hindered it. Just because the HPEs did something does not quite make it right!

Some empirical studies on the impact of industrial policy suggest their contibution to growth was minor (Pack, 1996). If you have empirical studies that assert otherwise, kindly point me to them.

As for your argument about a "locked-in home market": the work of Anne Kreugger and others suggest that protected home markets breed perverse incentives. Capitalists opt to maintain this protection, rather than undertake efficiency improvements. This is the well-known "rent-seeking" effect. The East Asian HPEs largely avoided this by skewing incentives towards export competitiveness, rather than import substitution. For example, in Korea, duty-free importation of equipment and materials was conditional on growth in export markets - again from Pack (1996).

Anonymous said...

I think to reach at the true cause of economic growth, we need to move beyond anecdotal evidence. The issue I find with a lot of studies is that they take a country, look at its policies, and assign causation. Unfortunately, they ignore other countries who implemented the same policies but ended in failure.

On a micro-scale, let's take the example of the much-vaunted MITI. People have always pointed to MITI as justification for government invervention in strategic industries. However, while MITI has had some successes, they also have a number of high-profile failures. One website that talks about it is: http://www.rieti.go.jp/en/special/policy-debate/11.html
The interesting thing is that the MITI programs that have been successful included companies that competed globally, while the failed programs involved companies that "competed" domestically. Makes you wonder if it was really MITI that contributed to success, or if it was competition.

I wonder if there has been a study that looks at MITI's overall success rate to determine whether it is a statistically significant contributor to economic growth.

As a more macro-level example, for protection of the home market, there are more examples of countries that have tried this and failed abysmally, hence my wariness in believing that this contributes to economic growth.


Econblogger said...

Thanks AT for the link. Sorry, the Pack paper is referenced for 2000. Complete set of references are posted in the edited entry.

Anonymous said...

Just read the Pack paper. It's pretty good. Actually answered a lot of questions I had in my comment.


walker said...

Thanks for the Pack article, it was an interesting read, but there are some issues article that make it less than convincing.

One, correct me if I am wrong but the reliance on the total factor productivity metric would underestimate the gains to economic growth made during periods of substantial capital investment. Since targeting took place in capital-intensive industries, I am not sure that TFP is appropriate (even ignoring the theoretical and conceptual problems with TFP).

Two, the "spillover" effect of benefits to "neglected" industries is estimated by using data on purchases from "favored" industries. Such a measure doesn't even begin to capture the positive externalities born through technological advancements, workforce development, eased entry into new technologically sophisticated industries, etc.

Three, the terms of trade gains are not acknowledged. The targeted industries are typically in tradable goods, and success in such sectors will tend to increase the value of the home currency, and thus increase overall purchasing power and national economic power.

Fourth, there is a deeper issue with the methodology. Dissecting the effect of any one policy measure and coming up with a quantitative factor of contribution is very difficult. A combination of policies can have an effect that is greater (or less) than the sum of the individual policies.

Finally, the paper takes a very narrow view of industrial policy -- equating it with redirecting capital from one sector to another. Policies that increase the overall domestic savings and investment rates or keep the currency undervalued, encourage technology acquisition from foreign industry, etc. also constitute industrial policy but are not taken into account.

A useful analysis requires a bit more subtlety. I am not aware of an empirical account that covers all industries in depth but "Computers Inc: Japan's challenge to IBM" is a very good play-by-play of Japanese industrial policy at work in electronics.

Econblogger said...


ON TFP: If you follow the simple growth-accounting decomposition, then growth unaccounted for by factor growth must be growth in TFP. Since the idea of protection is to obtain "dynamic comparative advantage", clearly TFP growth is relevant, rather than factor growth.

SPILLOVERS: hard to quantify anyway. Any suggestions?

TERMS OF TRADE (value of currency): the changing value of the currency merely reflects the fact that the currency by which to buy a country's rising exports, obviously becomes more valuable to foreigners. That is all. "National economic power" almost religious!

COMBINATION OF POLICIES: Good point. But unless you come up with the measures, however crude, this is speculative, no?

INDUSTRIAL POLICY: protection precisely favors some industries over others. (Nationalists favor sexy stuff like steel, machinery, computers,etc.) Increasing the domestic savings rate is not typically called industrial policy. And I am a bit confused about your view of the exchange rate - keeping the currency value low is good for industry, but a high currency value implies "national economic power"?

Is it possible for me to access that material? From its title, I can only suppose the greater subtlety you are referring to involves a more detailed type of anecdotal analysis.

walker said...

"National economic power" religious? Perhaps, but it is a common one -- the ability to project economic power (make acquisitions, buy influence, etc.) is something that policy-makers generally care about, especially in East Asia.

On terms of trade, the point is that imports become cheaper as the currency rises but the author does not attempt to quantify these gains when accounting the share of GDP growth due to "industrial policy". This is a glaring omission given that favors are generally provided to tradable goods industries.

And I am a bit confused about your view of the exchange rate - keeping the currency value low is good for industry, but a high currency value implies "national economic power"?

Let me explain. By maintaining an undervalued currency domestic manufacturers are able to undercut their competition in world markets, and thus increase their market share. Over a period of time the competition will weaken as domestic manufacturers meet more of world demand. A trade surplus is the result putting upward pressure on the currency, but having in the interim built up industrial capacity and competitiveness (through economy of scale benefits, infrastructure upgrades, R&D, etc.) the currency can then be allowed to rise without hurting business to the same degree as before.

In this way the currency is kept undervalued, but will steadily rise in accordance with the increased competitiveness of a nation's firms.

Regarding the book, it is certainly available (google it) but the text is not online if that is what you mean. It is a detailed historical account, not an attempt at quantitative analysis like the Pack article. I say it is more subtle because it indeed more nuanced in regarding the effects of industrial policy as well as less abstract. It would be nice if a quantitative analysis were available that provides similar insight while at the same time making precise claims, but on such a complex question such an analysis is difficult.