The UN has just announced a big new idea in the war on global poverty, in its just-released Industrial Development Report 2009. In the words of the United Nations Industrial Development Organization (UNIDO) Director General, Dr Kandeh Yumkella, “Our Report represents a major conceptual breakthrough on how to tackle global poverty through sustainable industrial development.”
What was the breakthrough idea? Take government action to reap increasing returns to scale to industrial production, to get out of the free market’s “poverty trap” of low-scale industrial production.
The only problem with this major conceptual breakthrough is that it is 66 years old. It was presented in almost the same words in one of the first and most famous articles of development economics, by Paul Rosenstein-Rodan in 1943:
| Paul Rosenstein-Rodan 1943 | UNIDO report 2009 |
| “It is generally agreed that industrialisation of “international depressed areas ” …is the way of achieving a more equal distribution of income between different areas of the world by raising incomes in depressed areas ….” | “Industrialization is integral to economic development… [For] the bottom billion, manufactured exports are likely to offer more scope for long term productivity growth than either agriculture or natural resources.” |
| “If we create a sufficiently large investment unit by including all the new industries of the region, external economies will become internal profits out of which dividends may be paid easily.” | “Because they still do not have industrial agglomerations, they are unable to be competitive against countries that have…there is a threshold of competitiveness to be surmounted. Once that threshold is crossed, growth is explosive.” |
| “If the industrialization of international depressed areas were to rely entirely on the normal incentive of private entrepreneurs, the process would … be very much slower.” | “There is an important role for
public action, as purely market-driven processes will yield prolonged stagnation.” |
Professor Rosenstein-Rodan in 1943 can be saluted for thinking of a creative new theory. Today’s authors (such as lead author Paul Collier) seem a bit less creative for recycling the exact same theory, especially after 66 years of experience that contradicted every prediction of this theory. Initially very poor countries like South Korea and more recently China and India had no trouble with industrial growth through market forces, and virtually every attempt at state-forced industrialization has failed. UNIDO endorsed many of these state industrialization attempts in the countries that, as a result of many such failures, Collier now calls the Bottom Billion.
Recyling old failed ideas after 66 years might be another small hint that UN accountability is a wee bit deficient.
17 Comments
I’m surprised to see you describe South Korea and China industrial growth as resulting only from market forces. IMO there was quite strong state incentives involved also. Just as in Japan earlier, and in the case of Japan, I know enough to say there was not just incentives, but also strong state involvement.
Now, state involvement will clearly work only in some special case (one could consider that asian countries with strongly entrenched confucianist values might the best known case where it works), but I’d consider that OTOH there are also very few cases where pure market forces were enough to start strong industrial growth.
In the case of South Korea, isn’t it true that export industries have been supported by the state, through subsidies? As for India and China, is it fair to claim that the recent market-led growth had nothing to do with earlier state-led investments? While I agree that it is irresponsible for UNIDO to give a blanket promotion of state intervention, this isn’t reason to dismiss Rosenstein-Rodan in every context.
As empirical economists, we can only say what succeeds or fails on average. On average, state industrial policies, state-owned enterprises, state-planned investment, etc. have all been a miserable failure. If things went better in Korea, India, and China, and they had some such state planning, the first bit of evidence would suggest that they succeed in spite of this state planning instead of because of it. There could also be different effects of state plans in different countries — a plausible but non-falsifiable hypothesis. Responsible policy recommendations can only be based on what works well or badly on average.
“Initially very poor countries like South Korea and more recently China and India had no trouble with industrial growth through market forces”
Er, is this a joke? South Korea (and Japan …) had very interventionist industrial policy, including selectively high tariffs, state aid to industrial conglomerates, etc. The interventions and subsidies were doled out partly on the basis of export performance, so it wasn’t as if market forces played no part – but to suggest that it was all the free market at work is simply wrong. And of course South Korea received a *lot* of aid from the US, as I’m sure you know.
Similarly, China employs a variety of policies that favour domestic firms, including subsidies, significant tariffs until relatively recently, and turning a blind eye to (or possibly even abetting) systematic theft of intellectual property.
None of this means that every poor country can necessarily follow the same route, of course. But it means that the knee-jerk free-marketeer rejection of industrial policy has little basis in the historical record.
I hadn’t seen your follow-up comment when I posted mine, so:
“As empirical economists, we can only say what succeeds or fails on average.”
This seems unnecessarily crude, given the enormous inherent differences between countries. When you include the awful growth record of most sub-Saharan African countries in your sample, pretty much every policy looks terrible, but that may have more to do with reasons unrelated to the policies themselves (this has implications for aid analysis too).
Another way to look at it would be to look at the countries that have successfully industrialised in a short space of time. How many of them used industrial policies, particularly at initial stages (sequencing is very important here, after all), and how many didn’t?
I fully agree that export-led growth is the way to go for development, as opposed to externally-financed Big Plans, as Dr Easterly so cogently criticizes, but I think that there are a couple issues here that seem to be overlooked.
First is that, methodologically speaking, I think we can take history into account and see if, on average, government investment in particular sectors in particular regions has succeeded or failed in facilitating economic development. I dont know enough about East Asia to speak about it authoritatively, but wouldn’t China’s special economic zones and 5 Year Plans be a kind of hybrid between public “planning” and free trade led development? It seems more like planning and less like outright government command of the economy. If this works in China perhaps it can work elsewhere in Asia? Also, perhaps modernization has a particular trajectory or path? Should poor nations in a given region mimic the development pattern of rich nations in the same region?
All of these can and should be guided empirically, but I suppose my core point here is that a people can use statistics and history to guide their search for what works. Whereas previous searches were literally fumbling around in the dark, trying any- and everything, because we have science we can make educated guesses at what will be successful in our search for development and what will not.
Also, there seems to be good evidence in economics that institutions and policies matter. Keep in mind that I am categorically not advocating making aid conditional on a government adopting these policies for reasons mentioned in this blog and elsewhere, but it does seem to be the case that democratic institutions, government-enforced laws against monopoly and other forms of collusion, having a police force, lighthouses, courts to enforce contracts, yada yada all have a positive impact on development, just as does liberty and freedom from external interference. Indeed the lack of oversight – or outdated oversight – by the government seems to be what got our own country, and hence the world, into the mess we’re in now.
So government and institutions do have an important role to play in the economy, even if it is clear that they should not command it.
Just some musings.
“As empirical economists, we can only say what succeeds or fails on average.”
With this statement, you demonstrate that you have the same cookie cutter approach to development that you criticize in others. As someone has already said, this is a rather crude way of evaluating the success of development strategy. And of course, as has already been mentioned, the state was heavily involved in the development of the economy in China, Korea, Japan etc.
For more on this, see Ha-Joon Chang book, Bad Samaritans.
The “what succeeds or fails on average”-approach is, from my point of view, not appropriate in complex economic and political context, that is the context of most developing countries.. I’m very surprised to see that that Easterly embraces that approach.
I agree with the elusive novelty and creativity of the UN report, however, Bill, you do miss out an important point (which I learned from you, btw):
Incentives!!!
Governments with incentives to pursue good development policies will most likely be capable of doing so, governments with an incentive to do something else, will most likely do something else.
The Kuomintang government of Taiwan was, in 1940’s, notorious for being one of the most corrupt and least capable governments in the world. Calling them incompetent would be an understatement. But they knew that the only hope they had to stay alive was through successful development of what little they had left – Taiwan – let alone their aspirations of redeployment on the mainland.
They HAD to delevop a strong industrial base and a sound economy. They HAD to outperform Mao. And they certainly did.
South Korea faced a similar threat from North Korea, that turned another incompetent government into a competent one because they knew they had to.
The incentives to pursue good policies can come in various forms, but if any such incentive is absent, we know what we will get: Your average government run development. Most developing country givernment do not have any strong incentive to develop their countries – they have strong incentives to implement corruption and predate on natural resources and aid. That do tell us something about the charachter of the challenge – developing country governments need checks and balances in oder to shape their incentives – but that doesn’t tell us anything more revolutionary than the fact that incentives matters.
And ranting on about government driven or market driven development in this context is irrelevant. If markets works and gives market agents the proper incentives – fine! If governments are facing the proper incentives – then that is fine too!
It is widely recognized that South Korea and Chinese industrial growth did not result only from market forces but often involved significant state intervention into the economy. Where Collier’s theory may share some characteristics to that of Rosenstein-Rodan, at least it is not a bold faced lie. This blog post by William Easterly is completely irresponsible, and reeks of ideology. We should expect more from an academic of this stature.
so allow me to summarise your position Bill:
you should offer policy recommendation based on your prior (ideologically rather than analytically based) beliefs on what works regardless of the place in question. The job of policy recommendations is explicitly not to try and understand why certain policies worked in particular countries at particular times, and certainly not to tailor policy recommendations to a particular context.
Understood. Good point actually.
Could you post a link to the 1943 Rosenstein-Rodan article? JSTOR and Google are failing me for some reason.
It’s worth considering the history of another idea from the international community.
It is widely recognized that South Korea and Chinese industrial growth did not result only from market forces but often involved significant state intervention into the economy. Where Collier’s theory may share some characteristics to that of Rosenstein-Rodan, at least it is not a bold faced lie. This blog post by William Easterly is completely irresponsible, and reeks of ideology. We should expect more from an academic of this stature.
The best answer to this question about government interventions would in the famous words of one of our presidents in Venezuela be phrased as “neither this nor that but just the opposite”.
I myself and signing up on much of what was said above by Tord Steiro with respect to incentives, always try to use a matrix that include all the combinations between the market and the government having the right incentives or not, and possessing the resources or not, all in order to produce a second best list, that goes from worse to best.
As you can understand, as a citizen from an oil-cursed country, my best of the best is always having the market with incentives and resources and the government with incentives but NO resources. There is nothing as bad as an independently wealthy government.
“As empirical economists, we can only say what succeeds or fails on average” – on average, the person with her head in the oven and the feet in the fridge is doing just fine! the challenge is not to swing from one extreme (the state is everything) to the other extreme (the market is everything) every few decades or so, but rather to find a way to incorporate local context into general theories.
A physicist, a carpenter and an economist go hunting. Seeing a deer on the hill, they all take aim. The physicist misses by a yard to the left, the carpenter by a yard to the right. The economist happily shouts: “We got it!”
The troubling beauty of the ‘East Asian miracle’ is that there’s something in it for all ’schools’. I’d celebrate the day more researchers took the complexity approach.
Reading Suggestion: ‘The End of Certainty: Time, Chaos, and the New Laws of Nature’, Ilya Prigogine