Chinese Economics Thread

[manqiangrexue]

Growth accounting suggests that productivity growth in China has declined substantially post-2008 and the growth was due to capital formation.

https://twitter.com/i/web/status/1816896857479458816

This is entirely consistent with capital formation driving the overwhelming majority of Chinese growth gains - capital deepening absent productivity enhancements - has declining marginal returns and thus results in slower growth; capital deepening with productivity enhancements results in constant growth rates (see for example, US growth rates being at 2% consistently for decades - there is a lack of a diminishing marginal product of capital) and consistent with the Asian tiger experience - following their capital deepening phases, they ran out of easy growth and thus turned to the slower productivity growth as the main growth driver.

whether the TFP decline in China reflects long lag effects of R&D or Chinese industry’s incapacity to translate research gains into commercial outcomes is a separate question altogether tho

What's that? You found another dude with a blog to back up your econ version of "chloroquine cures COVID" theory?

That's right, don't use common sense. Don't check anything against China's tech and research running the US over in several key fields, running the US down in all fields, overtaking the US in overall research, etc... Whistling past the graveyard is the way to go. Just tell yourself China will fade and wait for it. We Chinese have a saying for people like and and that's, "You won't even know how you died."

 

[gelgoog]

TFP improvement likely did decrease during the later Hu and early Xi years due to the housing bubble and rampant speculation. The late 2000s and early 2010s were not a time of high quality development in China, as the 2008 financial crisis in the US caused excess money to be pumped into mostly infrastructure & construction in China to preserve lost manufacturing jobs at the cost of the national debt. I specifically remember the stimulus rush at the time that caused housing prices to rise rapidly and predictably led to the later 2020s crisis.

But these factors are improving since the investment focus has shifted from real estate to advanced manufacturing. The Chinese government is very much aware of the problem and that’s why they put a stop to the speculation bubble.

Urban families typically have less kids overall. While in a rural environment kids provide labor, in urban environments kids are a burden. As China's population moved from the countryside to the cities you would expect birth rates to naturally fall. The One Child Policy made this even worse though.

 

[fishrubber99]

I've encountered the theory (which I believe has some merit) where Chinese GDP growth has actually been suppressed by China's heavy investment in infrastructure and housing in the past decade, since return on invested capital is much higher on business investment:

https://twitter.com/i/web/status/1765662591878082912

This is different than the typical East Asian tiger experience since they offset their infrastructure/housing investment to a different period of when their economics started developing rapidly, or they spread out the housing/infrastructure portion of their investment over a longer period of time compared to China. The suppressive effect to GDP growth is because China's composition of investment has been particularly housing/infrastructure intensive in the past decade, but that might change now that there is a shift to high tech manufacturing.

(e.g. East Asian tiger Taiwan never had the same degree of housing investment as mainland China, which is why housing quality in certain parts of Taiwan are incredibly poor)

Please, Log in or Register to view URLs content!

This is also why China's TFP (productivity growth) is seen as declining or stagnating for the past decade, the TFP is calculated through dividing output (GDP) by an average between labour and capital input. If the things you're using labour and capital to build has smaller immediate returns, then TFP will decline. But housing/infrastructure provides a larger return when we look at the cumulative returns over a longer period time (e.g. you can feasibly live in a house for a few decades, while the factory equipment purchased from a few years ago might already need to be replaced)

 

[chgough34]

Glenn had good threads about TFP

https://twitter.com/i/web/status/1752187020280266769

https://twitter.com/i/web/status/1813610151305449840

Glenn’s threads don’t actually dispute there has been a substantial TFP slowdown post-2008. He states that the TFP slowdown was due to greater resi. construction and that it is likely TFP will increase due to the greater investment in manufacturing structures. It’s quite convincing but it doesn’t dispute the TFP slowdown whatsoever (Adam Wolfe’s data doesn’t use the PWT; and if you were to depreciate assets over longer periods of time, that would increase the capital stock and push TFP down, not up)

 

[broadsword]

Glenn’s threads don’t actually dispute there has been a substantial TFP slowdown post-2008. He states that the TFP slowdown was due to greater resi. construction and that it is likely TFP will increase due to the greater investment in manufacturing structures. It’s quite convincing but it doesn’t dispute the TFP slowdown whatsoever (Adam Wolfe’s data doesn’t use the PWT; and if you were to depreciate assets over longer periods of time, that would increase the capital stock and push TFP down, not up)

Who's disputing that there has been a TFP slowdown since your post here Chinese Economics Thread in which you wrote

Growth accounting suggests that productivity growth in China has declined substantially post-2008 and the growth was due to capital formation.

Capital formation only? We replied to elucidate the cause.

 

[didklmyself]

Glenn’s threads don’t actually dispute there has been a substantial TFP slowdown post-2008. He states that the TFP slowdown was due to greater resi. construction and that it is likely TFP will increase due to the greater investment in manufacturing structures. It’s quite convincing but it doesn’t dispute the TFP slowdown whatsoever (Adam Wolfe’s data doesn’t use the PWT; and if you were to depreciate assets over longer periods of time, that would increase the capital stock and push TFP down, not up)

Growth with TFP slowdown isn't a flaw(which is your line of thinking), it is a characteristic of every fast urbanizing country. Glenn's thread is a response to idiots like you who bring up low TFP as something to be concerned about. There are multiple areas of Chinese economy that need improvement, this isn't one of them.

 

[broadsword]

Growth with TFP slowdown isn't a flaw(which is your line of thinking), it is a characteristic of every fast urbanizing country. Glenn's thread is a response to idiots like you who bring up low TFP as something to be concerned about. There are multiple areas of Chinese economy that need improvement, this isn't one of them.

It was his loud silent insinuation.

@chgough34 , define TFP and whether it is good or bad for countries like China and America.

 

[LeeDaChu]

Growth accounting suggests that productivity growth in China has declined substantially post-2008 and the growth was due to capital formation.

https://twitter.com/i/web/status/1816896857479458816

This is entirely consistent with capital formation driving the overwhelming majority of Chinese growth gains - capital deepening absent productivity enhancements - has declining marginal returns and thus results in slower growth; capital deepening with productivity enhancements results in constant growth rates (see for example, US growth rates being at 2% consistently for decades - there is a lack of a diminishing marginal product of capital) and consistent with the Asian tiger experience - following their capital deepening phases, they ran out of easy growth and thus turned to the slower productivity growth as the main growth driver.

whether the TFP decline in China reflects long lag effects of R&D or Chinese industry’s incapacity to translate research gains into commercial outcomes is a separate question altogether tho

Empirically Total Factor Productivity (TFP) is the residual when a theoretical model is applied to a sample, that is when everything is accounted for, what is left is Total Factor Productivity, basically it is saying so according to my model the GDP should be this, but we have gotten another number instead so lets attribute the difference between the two to technological growth and give it a fancy name. That means it is extremely sensitive to the data, the model specification and the estimation method. I have seen neoclassical economists being laughed at for bring up TFP, and this dude actually used a Solow Growth Model, a very primitive model taught to undergrads, where the only inputs are GDP, labour and physical capital, one can easily imagine how much can a model like this explain... At least amongst serious economists today, at the very least a Real Business Cycle Model augmented with various stuff is used for this kind of work.

There are generally only 3 things that increase overall productive capacity (productivity improvements, capital deepening, and increases in the labor force). Within productivity - it’s going to be either technical catchup, newfound innovation/productivity, or improvements in governance. There was a fairly large debate on what caused China’s post-1978 growth spurt: if it was due to secular increases in productivity, the growth rates shouldn’t have decreased as per capita gdp have increased (for example, in the U.S., gdp per capita growth has been 2% for decades and decades); but if it was due to capital deepening, China’s growth rate would decline as gdp per capita increased since capital has a declining marginal product. Since China’s growth rate has declined precipitously, it was clear that it was capital deepening doing a lot of the legwork. And in the Polish case - which China is now mirroring - it was about removing market distortions and technical catch-up in the post-Soviet environment that led to its growth, not productivity. If China was uniquely productive, it would have higher growth rates than poland

no, if you did progress in your education beyond an undergrad degree then you would know how ignorant that sounds. There is a host of other theoretical models and extensions that incorporate different stuff like innovations, pollution, human capita, etc. And they all make logical sense, whether or not they can be tested reliably is another matter. Human capital for example is empirically

Please, Log in or Register to view URLs content!

. And that Noah Smith graph doesn't even show anything, only that Poland and China GDP growth rates are largely the same for an unspecified time period since its x-axis is not even labelled. With a vague data like that you can make all sorts of arguments to explain the Polish growth, Eastern expansion of the Eurozone, investments from rich neighbours like Germany, increase in trade with the rest of Europe. But you can't compare China with Poland without serious economic modelling and econometrics, they are not the same in size, geographical layout, demographic, economic starting point...

Growth accounting suggests that productivity growth in China has declined substantially post-2008 and the growth was due to capital formation.


This is entirely consistent with capital formation driving the overwhelming majority of Chinese growth gains - capital deepening absent productivity enhancements - has declining marginal returns and thus results in slower growth; capital deepening with productivity enhancements results in constant growth rates (see for example, US growth rates being at 2% consistently for decades - there is a lack of a diminishing marginal product of capital) and consistent with the Asian tiger experience - following their capital deepening phases, they ran out of easy growth and thus turned to the slower productivity growth as the main growth driver.

whether the TFP decline in China reflects long lag effects of R&D or Chinese industry’s incapacity to translate research gains into commercial outcomes is a separate question altogether tho

This does not show a decrease in productivity growth, it shows that at least within the theoretical model used since 1988 less and less growth is unaccounted for (remember TFP is a residual) and more can be attributed to capital growth which is in line with China's strong growth in industrial output since 1988 (you can't make electric cars or LNG carriers in your own backyard, you need heavy and expensive equipments), and with more and more automated factories in the future you can expect that to continue to grow.

 

[CMP]

Empirically Total Factor Productivity (TFP) is the residual when a theoretical model is applied to a sample, that is when everything is accounted for, what is left is Total Factor Productivity, basically it is saying so according to my model the GDP should be this, but we have gotten another number instead so lets attribute the difference between the two to technological growth and give it a fancy name. That means it is extremely sensitive to the data, the model specification and the estimation method. I have seen neoclassical economists being laughed at for bring up TFP, and this dude actually used a Solow Growth Model, a very primitive model taught to undergrads, where the only inputs are GDP, labour and physical capital, one can easily imagine how much can a model like this explain... At least amongst serious economists today, at the very least a Real Business Cycle Model augmented with various stuff is used for this kind of work.



no, if you did progress in your education beyond an undergrad degree then you would know how ignorant that sounds. There is a host of other theoretical models and extensions that incorporate different stuff like innovations, pollution, human capita, etc. And they all make logical sense, whether or not they can be tested reliably is another matter. Human capital for example is empirically

Please, Log in or Register to view URLs content!

. And that Noah Smith graph doesn't even show anything, only that Poland and China GDP growth rates are largely the same for an unspecified time period since its x-axis is not even labelled. With a vague data like that you can make all sorts of arguments to explain the Polish growth, Eastern expansion of the Eurozone, investments from rich neighbours like Germany, increase in trade with the rest of Europe. But you can't compare China with Poland without serious economic modelling and econometrics, they are not the same in size, geographical layout, demographic, economic starting point...



This does not show a decrease in productivity growth, it shows that at least within the theoretical model used since 1988 less and less growth is unaccounted for (remember TFP is a residual) and more can be attributed to capital growth which is in line with China's strong growth in industrial output since 1988 (you can't make electric cars or LNG carriers in your own backyard, you need heavy and expensive equipments), and with more and more automated factories in the future you can expect that to continue to grow.

It might be helpful if you repeated your debunking in the original X tweet he posted.

 

[Jiang ZeminFanboy]

Glenn’s threads don’t actually dispute there has been a substantial TFP slowdown post-2008. He states that the TFP slowdown was due to greater resi. construction and that it is likely TFP will increase due to the greater investment in manufacturing structures. It’s quite convincing but it doesn’t dispute the TFP slowdown whatsoever (Adam Wolfe’s data doesn’t use the PWT; and if you were to depreciate assets over longer periods of time, that would increase the capital stock and push TFP down, not up)

I don't take TFP index seriously

https://twitter.com/i/web/status/1747250945594171418