1. Based on the book "How Asia works" isn't it clear that most of Africa should focus a lot more on increasing agricultural productivity until they are more-or-less self sufficient? That would allow them to expand their foreign currency earning portfolio besides mining ores or fossil fuels (if the country has them), increase national savings from farmers, and stop using up foreign exchange to artificially prop up their currency to import food? It appears having a base level of food and fuel self-sufficiency is a baseline for a country to pursue export led growth. You can't expect a poor country that depends on cheap food and energy imports, to depreciate its currency for manufacturing exports when that policy would make food imports dramatically more expensive.
Most of Africa is in the 1.75 tons of food per hectare. Kenya a relative "star" of Sub-Saharan Africa is even lower at 1.49, Ethiopia is one of the few Sub-Saharan African countries that massively boosted food fields from less than 1 to 2.79 which has contributed a lot to their growth. Bangladesh and Vietnam have had a massive productivity boom to 5 and 6 tons respectively.
To me it doesn't appear that export-led growth has failed. It literally hasn't even been tried yet in many poor countries in Africa since agricultural productivity has been neglected.
2. Would you agree that part of the reason why Asia sucked up manufacturing was because they were near Japan and Japan was basically the first foreign direct investor in Taiwan, SK, Thailand, and then later China when it opened up? Meanwhile, at the same time, large swarths of Africa were nationalizing industries and the would be big investor - South Africa - was an apartheid state focused on destabilizing its neighbors rather than invest factories in them.
As a bangladeshi citizen Noah being optimistic on the future of Bangladesh makes very optimistic as well. I have my own theory on the future of my country. Below I'm sharing a reddit post I wrote a few days ago.
Bangladesh economic prospects in 2041 (3 Possible Scenarios)
Scenario 1: Industrialisation stalls because of poor infrastructure investments. Unlikely since the recent government's actions and also because we're an easy country to build infrastructure for. We have huge population and a large network of waterways. Unlike Philippines which is a bunch of islands. Hence our infrastructure costs per capita is low and return on investment is also higher.
Per capita income stuck at 3.3k (modern Philippines)
Scenario 2: Unable to create globally competitive contract manufacturing because of weak electricity and gas supply. I think technology will solve some of this problem because of solar panels, batteries and industrial heat pumps. By the 2030s we should be privatising electricity generation and retail (with transmission in public hands). Gas pipelines will be largely irrelevant since both households and businesses will move to electric heating systems.
Per capita income stuck at 5.5k (modern Indonesia)
Scenario 3: We manage to become effective contract manufacturers but we can't innovate. Basically we can create Foxconn but not a Apple. Contract manufacturing is a low margins business, majority of the money is engineering design, marketing and after sales services. There is a reason Walton has their R&D labs in Korea and not in Bangladesh. This is because of weak legal institutions, political instability, and culture. A open society that enables innovation is an underrated thing. No one is entirely sure how something like this can be created.
Per capita income stuck at 7.7k (modern Thailand)
A falling fertility rate is also a big pain point. An aging society will place additional burdens on the next generation. Additionally there is less roam for innovation if the country is full of a bunch of old farts.
It’s not an accident Mexico has become the largest trading partner for the U.S. Maquiladoras established northern Mexico and industrial/manufacturing belt. Mexico also assembles/test high-end medical equipment, etc. ENPH (micro-inverters) was ahead of the curve, four years ago moving its manufacturing operation from China to Mexico. This moved it closer to its largest market (California), simplifying logistics and reaping significant savings in re shipping costs.
Why can something similar be tried/incentivized in Central America? A number of vendor business radiate from large manufacturing operations. Freight destined for the North American market creates truck driving and/or shipping port jobs and upgrades. This could help stem some of the migration from Central American countries (light from gang violence and lack of employment), a benefit to Mexico and the U.S.
There’s nothing wrong with working with India and Southeast Asian countries to develop economic ties, but that does little to ease migration. It seems more cost effective and politically beneficial to do more of this with our southern neighbors.
Lack of state capacity has been the primary bottleneck in Africa's development. Although technology is changing the need for governments to have an activity role in industrialisation. For example, for manufacturing you need at least three things, electricity, gas lines, railways or robust road network. Not having this stuff was a huge barrier to even basic labour intensive industrialisation. But in the coming decades, small scale manufacturers in Africa can benefit from solar, batteries and industrial heat pumps for their energy needs. Starlink for their internet needs. More speculatively airships as a replacement for a bad road/rail network.
I think both stories omit an even more important factor: state capacity and overall quality of institutions. You need a pretty effective government to be able to implement the kind of development path that Studwell, for example, advocates. First of all, you need to be able to reallocate land efficiently, rather than as a source of patronage. Then you need to allocate capital to where you might have a comparative advantage, and not to cronies. Finally you need to actually allow firms of wealthy and well-connected people to fail. I don't see how most developing nations (e.g. Chad, South Sudan) have anywhere near this level of state capacity.
This is my general problem with Studwell. Sure the success could be due to industrial policy, but it could also be due to an effective government that would have been successful regardless of the development path that it took. And even if industrial policy is optimal in the presence of high state capacity, it could be terrible for low state capacity countries because it provides so many opportunities for elite capture. I don't understand how you can even begin to talk about economic development without addressing this fact, and yet Studwell makes no mention at all.
Noah, I'm wondering if you've ever read Global Trade and Conflicting National Interests by Baumol. It's an older book and you may already know everything in it, but years ago it was my first exposure to serious, academic, non-neo-liberal, non-Ricardo-based trade theory. I still recommend it, and it has a lot to say about industrial policy.
I am reminded of something Jane Jacobs wrote in one of her books on economics. I can't remember the exact details, but she talks about "appropriate technology". She said that economists from Western, developed countries expected the economies in poor, undeveloped countries to behave like their home ones, forgetting the process of industrialization that occurred. Therefore they were recommending developing countries embrace high levels of automation because in the West, the idea that it could cause mass unemployment is Luddism, but in the global south, their economies simply weren't growing fast enough and so they were actually reducing the opportunities in agriculture and manufacturing as the population grew.
As usual, she didn't cite her sources or any statistics, but the similarities are striking.
Interesting article.... though I find that the discussion is done in a vacuum. Industrialization in general or even in a focused vertical market form is not a complete enough discussion. Much like individuals or companies, one has to get very specific about the competitive differentiation which can be projected toward SPECIFIC customers/markets. Poorer countries have natural advantages in price, but that may not be enough for particular market categories.
Also, one idea which is not discussed very often is protectionism to allow for the formation of a local industry until the domestic industry is competitive worldwide. One of the successes for China, South Korea, and to a smaller degree India has been some policies which explicitly have these ideas in mind. Economically this is very inefficient. But has been famously said. we are all communists at home and capitalists outside the home. You might wonder where these countries learned about these sorts of policies....perhaps it was from the US in the 1800s as the US was trying to catchup with the technology juggernauts in Europe. We can thank most of the mills in New England based on some IP theft from England. it is best not to be too pious about these matters.
I'm sympathetic to Rodrik's thesis. Manufacturing used to be 32+% of global gdp before the 90s and now it's around 22%. So it is definitely a shrinking pie. I'm with Richard Baldwin who thinks the future of trade is services. Although no country today has gotten rich from services exports primarily, the theory does hold some potential. Since the late 2008s, trade as a share of global gdp has stagnated around 50%. However, trade in goods has dropped (as percentage of global gdp) but tradable services was making up an increasing share of trade. So tele-education, telemedicine, and telerobotics might unlock another wave of globalisation. This time it would enable a lot of developing countries in Africa who are not close to industrialising countries to develop as well.
Another important factor is the attitude of the government to be invested. These would also include law and order as well as dispossession towards FDI. For example, Latin America during their socialist movement was predisposed towards nationalizing those foreign industries, which was not done in Asia.
"After Asia is done developing, Africa will be the only region with low costs."
You are implying that this will result in development in Africa. But our regional history in the USA doesn't show such development without huge government subsidies. Do I misunderstand?
"while automation of manufacturing might make it less of a job creator for countries like India, it’s a huge jump to conclude that it’s therefore unimportant to the economy. It’s possible that robotic factories, tended by a few engineers, will be an important source of revenue generation for developing countries"
Absent the attraction of labor cost arbitrage went would anyone build factories in developing countries to supply rich countries? More geopolitical risk, more supply chain risk, more distance to consumers etc. And if you mean building these factories to supply the local market, 1. Market won't get truly big before country gets rich 2. These factories will kill current less efficient competition which will destroy employment and reduce consumption capacity 3. Factories will largely be financed and built by foreigners at first reducing your local halo effect argument. That's why past success stories have always been built on export models at first.
Poland and South Korea have a couple of very big advantages, access to big markets (EU and China/Japan respectively) and cheap(ish) labour. Some countries in Africa will be able to manufacture for that continent if they can quell unrest and corruption, Brazil should be doing it for South America.
Bangladesh, Vietnam, Thailand etc have taken over the less value add manufacturing from South Korea, again because they have access to huge markets on their doorstep, geography is a big advantage.
So, Rodrik and Stiglitz’s nontradable services plan strikes me as a recipe for resort towns. Is this off?
As an educator in Asia, I’m skeptical of any plan that suggests a developing country should educate their growing workforce into nontradable services. Maybe I’m missing something.
If anything, Noah’s solution rings much truer: First, upgrade your overall productivity by getting your workforce to that place where a few engineers can operate highly productive factories. Nontradable services will follow.
> It’s possible that robotic factories, tended by a few engineers, will be an important source of revenue generation for developing countries. That revenue will then be spread around locally via local multiplier effects […] and each manufacturing job will create some larger number of other jobs in exactly the kind of nontradable services that Rodrik and Stiglitz want.
Now, Noah seems to think this leads to a durable disadvantage in education:
> A subtler problem is learning. Productivity growth comes from learning new technologies and new business models. In a world where factories employ very few people, those factories won’t be as useful as schools for workers.
But I don’t think so. The “learning problem” solves itself. How can you get your developing country to a place where a small number of people operate highly productive factories without first getting a large number of people operating less productive factories (perhaps exporting to China) or an even larger number of people learning the skills and habits of timeliness, communication, and routine work?
I don’t see it. That’s like having the MLB but no Little League, or a Premier League but no youth football.
Great article Noah. Some questions.
1. Based on the book "How Asia works" isn't it clear that most of Africa should focus a lot more on increasing agricultural productivity until they are more-or-less self sufficient? That would allow them to expand their foreign currency earning portfolio besides mining ores or fossil fuels (if the country has them), increase national savings from farmers, and stop using up foreign exchange to artificially prop up their currency to import food? It appears having a base level of food and fuel self-sufficiency is a baseline for a country to pursue export led growth. You can't expect a poor country that depends on cheap food and energy imports, to depreciate its currency for manufacturing exports when that policy would make food imports dramatically more expensive.
Most of Africa is in the 1.75 tons of food per hectare. Kenya a relative "star" of Sub-Saharan Africa is even lower at 1.49, Ethiopia is one of the few Sub-Saharan African countries that massively boosted food fields from less than 1 to 2.79 which has contributed a lot to their growth. Bangladesh and Vietnam have had a massive productivity boom to 5 and 6 tons respectively.
https://ourworldindata.org/grapher/cereal-yield?tab=chart&country=BGD~VNM~THA~TWN~ETH~OWID_ERE~KEN
To me it doesn't appear that export-led growth has failed. It literally hasn't even been tried yet in many poor countries in Africa since agricultural productivity has been neglected.
2. Would you agree that part of the reason why Asia sucked up manufacturing was because they were near Japan and Japan was basically the first foreign direct investor in Taiwan, SK, Thailand, and then later China when it opened up? Meanwhile, at the same time, large swarths of Africa were nationalizing industries and the would be big investor - South Africa - was an apartheid state focused on destabilizing its neighbors rather than invest factories in them.
As a bangladeshi citizen Noah being optimistic on the future of Bangladesh makes very optimistic as well. I have my own theory on the future of my country. Below I'm sharing a reddit post I wrote a few days ago.
Bangladesh economic prospects in 2041 (3 Possible Scenarios)
Scenario 1: Industrialisation stalls because of poor infrastructure investments. Unlikely since the recent government's actions and also because we're an easy country to build infrastructure for. We have huge population and a large network of waterways. Unlike Philippines which is a bunch of islands. Hence our infrastructure costs per capita is low and return on investment is also higher.
Per capita income stuck at 3.3k (modern Philippines)
Scenario 2: Unable to create globally competitive contract manufacturing because of weak electricity and gas supply. I think technology will solve some of this problem because of solar panels, batteries and industrial heat pumps. By the 2030s we should be privatising electricity generation and retail (with transmission in public hands). Gas pipelines will be largely irrelevant since both households and businesses will move to electric heating systems.
Per capita income stuck at 5.5k (modern Indonesia)
Scenario 3: We manage to become effective contract manufacturers but we can't innovate. Basically we can create Foxconn but not a Apple. Contract manufacturing is a low margins business, majority of the money is engineering design, marketing and after sales services. There is a reason Walton has their R&D labs in Korea and not in Bangladesh. This is because of weak legal institutions, political instability, and culture. A open society that enables innovation is an underrated thing. No one is entirely sure how something like this can be created.
Per capita income stuck at 7.7k (modern Thailand)
A falling fertility rate is also a big pain point. An aging society will place additional burdens on the next generation. Additionally there is less roam for innovation if the country is full of a bunch of old farts.
It’s not an accident Mexico has become the largest trading partner for the U.S. Maquiladoras established northern Mexico and industrial/manufacturing belt. Mexico also assembles/test high-end medical equipment, etc. ENPH (micro-inverters) was ahead of the curve, four years ago moving its manufacturing operation from China to Mexico. This moved it closer to its largest market (California), simplifying logistics and reaping significant savings in re shipping costs.
Why can something similar be tried/incentivized in Central America? A number of vendor business radiate from large manufacturing operations. Freight destined for the North American market creates truck driving and/or shipping port jobs and upgrades. This could help stem some of the migration from Central American countries (light from gang violence and lack of employment), a benefit to Mexico and the U.S.
There’s nothing wrong with working with India and Southeast Asian countries to develop economic ties, but that does little to ease migration. It seems more cost effective and politically beneficial to do more of this with our southern neighbors.
Lack of state capacity has been the primary bottleneck in Africa's development. Although technology is changing the need for governments to have an activity role in industrialisation. For example, for manufacturing you need at least three things, electricity, gas lines, railways or robust road network. Not having this stuff was a huge barrier to even basic labour intensive industrialisation. But in the coming decades, small scale manufacturers in Africa can benefit from solar, batteries and industrial heat pumps for their energy needs. Starlink for their internet needs. More speculatively airships as a replacement for a bad road/rail network.
I think both stories omit an even more important factor: state capacity and overall quality of institutions. You need a pretty effective government to be able to implement the kind of development path that Studwell, for example, advocates. First of all, you need to be able to reallocate land efficiently, rather than as a source of patronage. Then you need to allocate capital to where you might have a comparative advantage, and not to cronies. Finally you need to actually allow firms of wealthy and well-connected people to fail. I don't see how most developing nations (e.g. Chad, South Sudan) have anywhere near this level of state capacity.
This is my general problem with Studwell. Sure the success could be due to industrial policy, but it could also be due to an effective government that would have been successful regardless of the development path that it took. And even if industrial policy is optimal in the presence of high state capacity, it could be terrible for low state capacity countries because it provides so many opportunities for elite capture. I don't understand how you can even begin to talk about economic development without addressing this fact, and yet Studwell makes no mention at all.
Noah, I'm wondering if you've ever read Global Trade and Conflicting National Interests by Baumol. It's an older book and you may already know everything in it, but years ago it was my first exposure to serious, academic, non-neo-liberal, non-Ricardo-based trade theory. I still recommend it, and it has a lot to say about industrial policy.
“Utility-scale solar PV and onshore wind are the cheapest options for new electricity generation in a significant majority of countries worldwide.“
https://www.iea.org/news/renewable-power-s-growth-is-being-turbocharged-as-countries-seek-to-strengthen-energy-security
I am reminded of something Jane Jacobs wrote in one of her books on economics. I can't remember the exact details, but she talks about "appropriate technology". She said that economists from Western, developed countries expected the economies in poor, undeveloped countries to behave like their home ones, forgetting the process of industrialization that occurred. Therefore they were recommending developing countries embrace high levels of automation because in the West, the idea that it could cause mass unemployment is Luddism, but in the global south, their economies simply weren't growing fast enough and so they were actually reducing the opportunities in agriculture and manufacturing as the population grew.
As usual, she didn't cite her sources or any statistics, but the similarities are striking.
What about these other economists’ encouragement to invest in green energy?
I’m thinking this
-lowers the cost of electricity
-makes EVs more affordable to operate
-reduces pollution & related healthcare costs, thus increasing productivity
-and reduces or eliminates variable fossil fuel costs, which lead to inflation and political instability.
Interesting article.... though I find that the discussion is done in a vacuum. Industrialization in general or even in a focused vertical market form is not a complete enough discussion. Much like individuals or companies, one has to get very specific about the competitive differentiation which can be projected toward SPECIFIC customers/markets. Poorer countries have natural advantages in price, but that may not be enough for particular market categories.
Also, one idea which is not discussed very often is protectionism to allow for the formation of a local industry until the domestic industry is competitive worldwide. One of the successes for China, South Korea, and to a smaller degree India has been some policies which explicitly have these ideas in mind. Economically this is very inefficient. But has been famously said. we are all communists at home and capitalists outside the home. You might wonder where these countries learned about these sorts of policies....perhaps it was from the US in the 1800s as the US was trying to catchup with the technology juggernauts in Europe. We can thank most of the mills in New England based on some IP theft from England. it is best not to be too pious about these matters.
I'm sympathetic to Rodrik's thesis. Manufacturing used to be 32+% of global gdp before the 90s and now it's around 22%. So it is definitely a shrinking pie. I'm with Richard Baldwin who thinks the future of trade is services. Although no country today has gotten rich from services exports primarily, the theory does hold some potential. Since the late 2008s, trade as a share of global gdp has stagnated around 50%. However, trade in goods has dropped (as percentage of global gdp) but tradable services was making up an increasing share of trade. So tele-education, telemedicine, and telerobotics might unlock another wave of globalisation. This time it would enable a lot of developing countries in Africa who are not close to industrialising countries to develop as well.
Another important factor is the attitude of the government to be invested. These would also include law and order as well as dispossession towards FDI. For example, Latin America during their socialist movement was predisposed towards nationalizing those foreign industries, which was not done in Asia.
Great article. But one thing bugs me:
"After Asia is done developing, Africa will be the only region with low costs."
You are implying that this will result in development in Africa. But our regional history in the USA doesn't show such development without huge government subsidies. Do I misunderstand?
"while automation of manufacturing might make it less of a job creator for countries like India, it’s a huge jump to conclude that it’s therefore unimportant to the economy. It’s possible that robotic factories, tended by a few engineers, will be an important source of revenue generation for developing countries"
Absent the attraction of labor cost arbitrage went would anyone build factories in developing countries to supply rich countries? More geopolitical risk, more supply chain risk, more distance to consumers etc. And if you mean building these factories to supply the local market, 1. Market won't get truly big before country gets rich 2. These factories will kill current less efficient competition which will destroy employment and reduce consumption capacity 3. Factories will largely be financed and built by foreigners at first reducing your local halo effect argument. That's why past success stories have always been built on export models at first.
Poland and South Korea have a couple of very big advantages, access to big markets (EU and China/Japan respectively) and cheap(ish) labour. Some countries in Africa will be able to manufacture for that continent if they can quell unrest and corruption, Brazil should be doing it for South America.
Bangladesh, Vietnam, Thailand etc have taken over the less value add manufacturing from South Korea, again because they have access to huge markets on their doorstep, geography is a big advantage.
So, Rodrik and Stiglitz’s nontradable services plan strikes me as a recipe for resort towns. Is this off?
As an educator in Asia, I’m skeptical of any plan that suggests a developing country should educate their growing workforce into nontradable services. Maybe I’m missing something.
If anything, Noah’s solution rings much truer: First, upgrade your overall productivity by getting your workforce to that place where a few engineers can operate highly productive factories. Nontradable services will follow.
> It’s possible that robotic factories, tended by a few engineers, will be an important source of revenue generation for developing countries. That revenue will then be spread around locally via local multiplier effects […] and each manufacturing job will create some larger number of other jobs in exactly the kind of nontradable services that Rodrik and Stiglitz want.
Now, Noah seems to think this leads to a durable disadvantage in education:
> A subtler problem is learning. Productivity growth comes from learning new technologies and new business models. In a world where factories employ very few people, those factories won’t be as useful as schools for workers.
But I don’t think so. The “learning problem” solves itself. How can you get your developing country to a place where a small number of people operate highly productive factories without first getting a large number of people operating less productive factories (perhaps exporting to China) or an even larger number of people learning the skills and habits of timeliness, communication, and routine work?
I don’t see it. That’s like having the MLB but no Little League, or a Premier League but no youth football.