Dani Rodrik Transcript from the End of Growth talk

EMERGING MARKETS: THE END OF GROWTH? 

I have created a transcript of Professor Dani Rodrik “Emerging Markets: The End of Growth?” talk at the Chicago Council of Global Affairs (5th April, 2016). You can find the original talk, recorded by the Chicago Council, by clicking here.

Dani Rodrik is the Ford Foundation Professor of International Political Economy, John F. Kennedy School of Government, Harvard University.

Quick Notes:

  1. This isn’t a professional transcription. So I apologise in advance for any issues. This is an amateur effort.
  2. For readability, I’ve broken the transcript into paragraphs and given them sub headers. These were done by myself alone with no input from Rodrik or anyone else.

[BEGIN TRANSCRIPT]

I’ve been most of my career, professional research career has focused on try to understand, what drives the economies of developing and emerging market economies and I’m not going to make a distinction between developing frontier, emerging, I’m just going to in general lump them together in sort of developing countries and then into my Q and A do a sort of little bit more differentiation.

[Developing countries are typically volatile]

And one thing that we of course all know and you don’t need to be an expert in developing countries’ economies, these are countries that are highly, typically volatile, unstable but I think there’s one thing that’s been even more cyclical and volatile than their economies has been kind of commentary about them in financial press because Al said it wasn’t that long ago that these countries were sort of touted as should have having embarked on a very sort of rapid kind of a growth and catch up process with the advanced countries that seemed to auger a new era.

And I think just to put their gross performance into sort of a bit of a recent historical context, here’s a chart that goes back to 1950 and compares the average growth rates of two parts of the world economy. So the green one, that’s sort of gently trending down, are the advanced countries, the rich economies of the world, and the orange one is the developing countries.

A couple of things you see here are actually quite interesting. One is the until about the 1990s, the rich countries of the world were growing more rapidly than the poor countries of the world and this is not what economics’ logic tells us should have happened because after all you say that developing countries of all kind of this potential of catching up, they can buy, they can access technologies and markets and capitals and finance that’s already out there.

It shouldn’t be that way, the developing countries ought to be growing more rapidly than rich countries. That, by and large, was not the case, of course, there were few exceptions you know such as the East Asian tigers being the main ones that I’ll talk about, but then there was a rather drastic turn around which is that from the 1990s the developing countries began to take off while the developed countries sort of continued their general trend down.

I should say that sort of these charts basically show the average trends and if you want to know what the actual growth rate in any year is, those are the dots. Now the dots, move around a lot. That’s the volatility, there’s a fair amount of volatility in the growth rates of the rich countries as well so I think the trends are represented by these lines. So you can see the pick-up growth in the developing world. A lot of this, as you well know of course is China. And China is having a big impact on this sort of growth take-off in two ways. One is that of course growing so rapidly that it is pulling the average up for the developing world but there’s also a statistical sense which is pulling the orange curve up which is that as growing very rapidly is becoming a bigger and bigger part of the developing world so that is becoming basically even though it’s not the Chinese growth rate, consistently trended up. It’s just that it has become the bigger and bigger part of the developing world that the average developing world growth rate as picked up.

So in case you are wondering why it was steadily increasing even though Chinese growth rate was basically very high but stable. And then you see at the end, so if the tail-end you see a beginning of a downward turn in the growth rate of the developing countries and that’s sort of the on-set of the pessimism that now all the things that suddenly – we previously we didn’t see – we seem to to be seeing corruption, the sort of the hidden liabilities and the various balance sheets, employment problems, political problems and so on, so these have grown.

There’s two ways to interpret this recent experience. One is to say, look this important pick-up in the growth of the developing world from the 1990s on was really fundamentally a real structural change that whether it was because of the globalization because of the more rapid spread of technology whatever or better policy in institutions in the developing world as a whole that there was a breakthrough that developing countries were not able to grow rapidly and that what they were experiencing now was more of a blip and that some say you know go back to their previous growth rate.

The second perspective and the one that I think is the one that is more consistent with the facts – and I’m going to try to explain why I think so – is in fact that very rapid growth in the 1990s and the 2000s sort of the last two decades before the growth started to dip down. That rapid growth was fundamentally unsustainable, that it was driven by highly cyclical factors in the world economy, having to do with commodity cycle, having to do with cheap capital, having to do with recovery from various internal and external shocks that these countries went through in 70s and 80s and so forth. And in some sense what the developing countries today are now experiencing is returning back to their longer run, more sustainable growth rates.

So it’s a return back to normal rather than sort of a deviation from their new trend. Having said that, there’s one other thing that you should know from this chart is that the developing world has a long way to come down before they actually start to reach the growth rates of the rich world or in fact below. So I’m going to argue that the developing countries are about to come down but I’m not going to argue the fact that they all are going to come down to growth rates of the advanced countries.

So the good news is in some sense convergence is going to continue, the developing countries will continue to grow more rapidly than rich countries but not nearly by the same margin as they did earlier. So that’s a bit of a sort of a where I’m going. Now when you study economic growth in terms of trying to understand “what are the fundamental drivers of these kinds of trends?”, the one thing that you notice from economic history is that the kind of very rapid growth rate that developing countries experienced very recently is extremely, extremely rare.

[Three regions where we see growth post-1950s]

If you look at the period before 1950, virtually no country in the world experienced growth rates that were 4%- 4.5% on per capita basis like these countries were experiencing in the period before. After the 1950s there had been periods of sustained growth of such high rates. And I imagine the script there is could be too small for you to tell each one of these countries but I’m going to tell you the story.

[First Region: Post-WW2 Western Europe]

Essentially there had been 3 groups of countries that experienced sustained and very rapid growth in the periods since 1950. One, the first group on the column on the right hand were countries that were in immediate periphery of Western Europe that experienced very rapid growth in the aftermath of the Second World War. That’s the 50s and 60s and the early 70s. Those are countries like Italy, Spain, Portugal and Greece. Yes, Greece, Greece actually grew very rapidly in the period between 1950 and 1980.

[Second Region: South East Asia]

The second group and now I’ll jump to the very bottom, those were basically a cluster of countries in Eastern and Southeast Asia. Those are the ones that we immediately associate with very rapid growth. These are the East Asia tigers, South Korea, Singapore, Taiwan and a few others like Malaysia, Indonesia and, of course, more recently now China.

[Post-WW2 Western Europe & Southeast Asia: Rapid Industrializations]

Now what actually was more common, two clusters of countries in different parts of the world, those in the European periphery and those in East Asia at somewhat different periods of time but there was one thing that was common in the experience of these two countries which I’m going to argue has been critical in generating high growth was both of those were based in very rapid industrialization. Manufacturing was key driver of growth in these countries.

[Third Group: Small Resource Rich Countries]

The third group, that sort of the middle group, that you’d immediately recognize, is a relatively small number of resource rich countries. There are a lot of countries in the world that are resource rich but it turns out only a few of them, maybe half a dozen or so have managed to grow very rapidly on very sustainable basis because of a variety of reasons. There’s something called Resource Curse which means having a lot of resources is not good for your long term development comes with a lot of volatility, comes with corruption, comes with a lot of back rate and economic instability and so forth. And therefore just a few countries have managed to grow at those rates for a period of two or three decades. Now I’m going to leave the resource rich countries because those are very much the exception and just sort of go back to this point: that if you are looking for a common factor in countries that experience their rapid growth, that it is the countries that have experienced their rapid industrialization.

[Industrialization has run out of steam]

I’m going to take a little time trying to explain why I think manufacturing is so critical to generate economic growth. And I will do that because when I come to the story why I don’t think this growth, the future holds the possibility is of similar kinds of growths, I’m going to say that’s because manufacturing has run out of steam. That the possibilities of industrialization are significantly weaker as we move forward as compared to the recent historical past. Okay.

So why is it that manufacturing is critical to economic growth in particular, why is it critical in driving rapid economic growth of the kind that these two cluster of countries, East Asia and the European periphery have produced?

[Manufacturing is powerful because it’s quick to adapt & scale]

One is, there is something special about the abilities of the countries to require manufacturing technologies off-the-shelf in a way that doesn’t exist in other parts of the economy whether it’s agriculture or services. I don’t want to get too much into it but basically that, part of it is that any of you who have visited say a Toyota factory in different parts of the world, will know they don’t differ a whole lot. Toyota factory in United States, Japan, South Africa is pretty much the same.

It doesn’t mean that you can just put these things in place, automatically you can produce with the same kind of productivity but you can get a huge sort of boost in productivity, you can almost sort of catch up to the frontier, global levels of productivity in manufacturing in ways that are simply not possible in say public services or education or you know in traditional agriculture. where a lot of people are. So there’s something that manufacturing enables you to access frontier technology in a much more quickly and much more directly than the other parts of the economy. But that’s sort of one aspect of it.

[Manufacturing is also powerful because you can hire unskilled labor at scale]

The second thing that it does so while employing and it has been traditionally the case, while employing the resource that is in greatest abundance in poor countries which is relatively unskilled labor. So the traditional model of industrialization has been that you basically get a lot of people from the country side with very little skills and they become factory workers, they become production workers. This was the story of industrial revolution in Britain, it was the story of Japanese industrialization, it was the story of South Korean, Taiwan industrialization, it was the story of Chinese industrialization.

[Taken together, manufacturing provides a quick 3x-4x leap in productivity]

The key thing here is that the basically these people you bring in from the country side are workers, don’t need to be highly skilled or educated to experience a significant jump in productivity, a three-fold, four-fold jump in productivity from the kinds of things they were doing in the countryside growing the crops, the crops that they are doing. So this is wonderful, unlike so many other parts of the economy which could be very productive. Those may require high skills. So for example one problem with Indian development today is that it is really based on highly skilled intensive kind of specialization: services, BPO and IT services. And, the reason that those kinds of services cannot drive Indian economic growth in a significant way is because they rely on highly skilled workers. That’s not what, India has a lot of in absolute numbers but 95% of the Indian labor force is actually not very skilled workers. And you will get much more growth if you take those numbers, 95% of the people and put them in higher productivity jobs as manufacturing traditionally has been able to do. So this labor absorption capacity of manufacturing again is the second special feature.

 [Unlike most services, you can scale via exporting manufacturing, no domestic demand constraints]

Now the third is one you have this factory, you are not constrained by your domestic demand. In other words, all you need to do is get a few manufacturing industries going and they face essentially boundless markets in world globally. And, they are not constrained by domestic incomes and productivity so they can grow and expand very rapidly without the constraints of ex domestic market stocks. Again this is very different than most services.

What that means you can actually focus on part of the economy as very critically Japan and South Korea and Taiwan and China did. And, develop capabilities and expand manufacturing industries, while a lot of the rest of the economy is lagging behind but the major productivity boost that is happening is because you are shifting your labor your workers from those lower productivity activities to your more productive ones. And, then this can go on for a long time because there is nothing that stops your manufacturing from expanding because as long as you are a small part of world market.

[Industrialization has run of out of steam, we need new models of development]

It’s important, I want to spend a little bit of the time to understand why manufacturing is important because I’m going to argue that once manufacturing you know the forces of industrialization run out of steam, models of growth based on the sectors of economy like services are not going to have this kind of property and this is I think the fundamental reason we are going to run into problems. So now of course you needed to have the right kind of policies to generate the industrialization so obviously not all countries could do industrialization, it didn’t happen naturally you had to have a right kind of a background.

So this is my one slide version of what I think what creates successful industrialization. You’ll see it’s sort of a very modestly stated but I think there are 3 critical features of what all successful countries did and I think all 3 of them were equally important. One is the sort of basic macro-economic environment, the basic policy environment was conducive to private environment. What I mean by that is not a free market economy. I don’t mean inflation at 5% or lower. I don’t mean a basic, a corruption free economies. So you know it’s the reasonably stable macro-fiscal policy, reasonably business friendly policy regimes. So it’s not a very demanding requirement.

[Industrialization was also possible in the past, because nations could create trade rules illegal today]

Second what I think is common in all of these countries that achieves very rapid industrialization, you had a particular type of a meddling gift because this industrialization doesn’t take on its own. You see time and again going all the way back to Japanese experience, Korean experience, Taiwanese experience, you see the government meddling in ways that’s going to actually increase the animal spirits of private sector, the entrepreneurs, basically stimulate private investment not in the old industries but in the new ones.

Sometimes it could be quite a heavy handed way. When president Park took over in Korea in 1961, he basically threw all the businessmen, the big businessmen in jail and said I’ll let you out only if you build factories for export and they said okay we will and so that’s how the export industry… Not a strategy to recommend everybody.

[A lot of good economic policy in the past consisted of highly heterodox and unorthodox policies]

This is just a point in fact that there was a lot of sort of things that is not consistent with our normal understanding of what good economic policy is. A lot of this was highly heterodox and unorthodox. And,  I’ve given you some of the examples there you know tariff protection, various subsidies, managing currencies, a lot of things that you can’t anymore do under the WTO rules and so forth.

And then finally of course you know this was a global environment, that was relatively development friendly – which is not to say that you had complete free trade, end of sixties and seventies or even eighties we’re talking about the pre-WTO days. Tariff rates were coming down but certainly not zero. You needed reasonable openness of the world economy and that reasonable benign neglect on the part of the major economies, vis-à-vis the kind of unorthodox meddling interventionist policies of these countries.

So nobody gave South Korea and Taiwan really hard time in the sixties and seventies because they were using extensive industrial policies. Even China didn’t get much of a hard time throughout much of its take off. And, I think this enabled these countries to engage in this kind of creative policy making.

[Countries are premature deindustrializing at lower levels of incomes]

Now, so all of that was really backward-looking to set the stage as to how I understand growth miracles and what generates very rapid growth and why is it that as we go forward we are not going to be experiencing the same kind of path. And, the main reason is that I’ve become, sort of looking at the data on trends on industrialization, I become very pessimistic about the prospects of industrialization as we go forward. And, in fact I think the right term to use which I’ve used in a number of writings is “premature de-industrialization”: That is that countries are beginning to de-industrialize today at levels of income which are a fraction of income levels than countries richer than them began to de-industrialize. And, then they never reach the levels of de-industrialization that countries before them did.

And, this has been true over the course of history that the countries like the UK and Japan for example UK and Germany put 30%-35% percent of their labor force in manufacturing at the height of their industrialization. For Korea, it was more like 25%-27% percent. China never approached this and India’s now hovering around 10%-11% percent and beginning to come down.

[Challenges: Manufacturing requires more and more advanced education and complex skills]

So the possibilities of increasing your manufacturing industries seem to have really come down. Why is that? I think there are two background factors at work here which are really working against generalized industrialization. One is that the technology of manufacturing globally has become in fact much more skill intensive and capital intensive than it used to be.

Remember one of the arguments that I made as to why manufacturing was as important as a driver of growth was that you could take a worker from countryside and put him or her in a factory and basically nothing else was required for that worker’s productivity to increase by a factor of 3 or 4. So to tell the story for example from the perspective of China, when the Chinese manufacturing boom, export oriented industrialization was happening, if you’re hiring a worker from the countryside to let’s say produce in a shoe factory in China, basically the only test that the owners would do to see if this worker was suitable for employment was that whether the worker could in fact do this. The most basic kind of manual co-ordination and if you could do this, you could produce shoes.

Now the same factory, the same company that’s investing in Africa, in Ethiopia, which in some sense should be the next frontier in low cost industrialization because they have lots of workers, and it could in principle replicate the kind of pattern the East Asian countries and then the European periphery before them did. The same company is now hiring in Ethiopia, basically university graduates, so it’s a very different kind of employment situation. So in that way manufacturing is becoming more like you know your IT sector or your high skill services and it no longer has the capacity to absorb a lot of unskilled labor and therefore constraints on rapid industrialization is that much stronger.

[Challenges: Global division of labor makes it more difficult to build diverse industries]

The second thing of course is which is related is the whole global division of labor today is that much more marked. So, before the global economy was so integrated every country could produce their own simple manufacturers. They could sort of develop capabilities in producing for the whole market. Every country had a toy industry. Every country had a bicycle industry because those stuffs wasn’t really imported. But, today you really can’t have a toy industry or a bicycle industry or even sort of basic furniture industry in a country like Ethiopia because despite the distance it is still so much cheaper to import it from China or from Vietnam or from Cambodia or countries that are sort of heard of.

So what that means: the global division of labor is good news for those countries that have achieved an early mastery of manufacturing, that means that manufacturing is going global. Production of manufacturing and employment in manufacturing is going to be concentrated in a much smaller set of countries, those with a head start. And, the ones that you want to follow up, the African countries, the lower income countries are really being hampered.

[Today’s Growth is a U Pattern of Development]

We can see this in the kind of very striking patterns. I’ve said about “premature de-industrialization,” this is very much in the data, which is if you look at the relationship between income levels and how much manufacturing employees is a share of the total economy that typically is a kind of inverted U shape.

First as the economy gets richer, manufacturing becomes bigger and bigger but at some point the service economy starts to take over and manufacturing share of employment starts to come down. After the 1990s this curve essentially has shifted down and towards the origin which means that basically you’re starting to deindustrialize at much lower levels of income and your peak level of industrialization looks nothing like what countries before you managed to achieve.

This is probably too much to look at but the most striking thing here is really what’s happening, as I already mentioned in Africa because Africa in some sense would be the frontier for low cost industrialization and this is just not happening.

So if you saw for example that the very rapid growth in Africa since the 1990s countries like Ethiopia have been growing 10% average over a decade or more that you thought that this was the same kind of growth that was happening in East Asia, the same traditional pattern of growth based on rapid industrialization, that would be wrong because there was basically the manufacturing industry in countries like Ethiopia barely kept pace with the rest of the economy. This was by no means industrialization-based growth. In fact, you can see that if anything, since the 1990s there is some indication that on average even that the share of employment in manufacturing has been trending down in Africa.

So if you look at it very schematically the kind of transformation and the structural changes that happens over the long course of development, think of dividing up the economy into the top row of three kinds of economic activities. Agriculture, manufacturing and services.

And then, you could organize each one of these activities either in an informal manner where its family farms or family or self-employment with relatively primitive technology outside the networks of the public, of the institutions of the state so forth, or in the organized formal parts where you have some chance of catching up with productive technologies and so forth.

[Informal Agricultural > Organized Manufacturing > Organized Services]

The traditional mode of economic development has been that basically you have this movement from informal agricultural production to organized manufacturing and then from there to organized services. This is what every country that got rich essentially this is the kind of transition that they have experienced.

[What’s happening today?]

What’s happening today is something that’s actually very different. What’s happening is something like this: that it’s not as if agriculture isn’t shrinking, it’s not as if people aren’t leaving the country side but they’re still flooding to the cities except that what they do in the cities is pick up jobs in the informal sector – petty services and services – where essentially their productivity isn’t that much high than what it was in the country side. And, the movement into manufacturing is actually relatively thin and in fact a lot of that manufacturing is also informal manufacturing where your possibilities for plugging yourself into global networks and global value chains and global technology are significantly less as well.

So the paradox in some sense is the paradox only if you thought that the growth of that we’ve experienced in the developing world is the kind of growth that would’ve been sustainable, the paradox is that basically all the countries that experience very rapid growth over the last twenty years before growth began to decline, this is the kind of growth they were experiencing and this I think is even in the best of circumstances I think the kind of a future that awaits many of these countries.

[Can Services create jobs? The way Manufacturing did?]

Now, you can see all kinds of problems that this might, this is going to raise. It’s not just a problem of growth it’s a problem of exclusion, it’s a problem of inequality, it’s a problem of being able to generate employment for a lot of people who are leaving the country-side and coming to the urban areas and that’s essentially not happening and wasn’t happening even when growth was high.

[There are highly productive service industries, but they require high skill]

Now, couldn’t just services substitute for manufacturing?

Now, when you look at services broadly, services really come in two types. One is basically the very high productivity segment: banking, business servicing, insurance, IT, right? So productivity in those services in developing countries if anything is probably even higher than manufacturing. So, it’s not as if there aren’t high productivity services. But, these service sectors are bound to remain enclaves in developing countries that simply do not have enough skilled well-trained people so the high productivity services are not going to be able to absorb.

[India needs two more generations to get the right type of workers]

Again India is a highly main example here because they’re not going to absorb the workers India has today and India needs another two generations before it can transform under the best of circumstances its workforce today to a workforce that can be absorbed into high productivity services.

[Hong Kong quickly transitioned from manufacturing to services industry]

The only country that I know has managed this transition extremely well, in the sense of, being able to move very quickly its manufacturing labor force into high productivity services is Hong Kong because Hong Kong went from a country that’s hyper-industrialized in the 70s-80s. Remember Hong Kong was the famous “Gang of Four” export-oriented industrializer, builder sort of garment exports and so forth.

But when China began to open up essentially Hong Kong became the service provider for China’s global trade economy. So basically, Hong Kong experienced a very, very rapid de-industrialization but ended up with a stronger economy because everybody was displaced from manufacturing-ended up in these high productivity services in finance, logistics, shipping and all of that and being a relatively small country with strong education that kind of transition was feasible in Hong Kong within sort of a generation or two.

So this transition is a race between how quickly you’re skilling up, you’re educating and increasing the skills of your labor force against how quickly is de-industrialization happening. The bulk of the services in a typical developing country or like in Hong Kong are basically these very low productivity. And, these cannot act as growth pulls, as growth escalators for reasons that I’ve already talked about.

 [Ethiopia is experiencing high growth, not based on export-oriented industrialization]

So, let me just end very quickly with a somewhat of a brighter spot.

I kept mentioning Ethiopia because in many ways Ethiopia is the country that exemplifies all the different aspects of the story I’m telling. The very rapid growth over the last two decades but one that actually was not based on export oriented industrialization.

And, therefore the question is “what was it based on? Can we learn something from them?” It turns out, in the case of Ethiopia, it was basically based on a very rapid ramping up of public investment, very rapid increase in public investment. The public sector invested in extension services, in their culture productivity, huge electric dam, transport network and so far managed to raise the resources by a combination of again orthodox and heterodox policies.

The orthodox policies was the relatively conservative fiscal policies, the heterodox part was significant dose of financial restrictions and financial repression which transferred resources from private sector for these public investments. So far it has worked but it’s a very different kind of model and typically we know that public investment booms end up sort of in tears also that these are not, typically create their own problems down the road.

[Future development will be slower, but driven by more diverse domestic factors]

Alright so let me just end with just a couple of quick conclusions. So I don’t think we are likely to see the kind of very rapid growth experience East Asia type or the kind of very rapid growth over the last two decades that that we’ve seen. I think because that kind of growth which we saw for the most part quite ephemeral and driven by external and temporary drivers.

I think convergence will continue as I said partly because the developed countries are not going to be doing that well, but I think even though growth in the developing world will come down it’s not going to come down nearly as much as that in the rich countries.

And, I think what we’re going to see is that sort of the growth experience in the developing world is going to be much more diverse, much more heterogeneous. Now, because once it is domestic factors that is driving the growth everything that you happen to do domestically, that’s going to be the key thing so I think we’re going to be seeing much greater heterogeneity. Some countries doing relatively well, other countries doing actually quite poorly and the main thing is going to be this heterogeneity in performance rather than the common.

So let me just stop this and I’m happy to take some questions.

[End Transcript]