6 ms·

What do you mean when you say China fakes its GDP?

ram4jesus 4y ago
Not that I fully believe this, but the claims about Ghost Cities, for example. Infrastructure projects built to increase GDP and industrial activity without clear goals.
- eru 4y ago
  The Ghost Cities crop up every once in a while, but they aren't actually a problem. They usually turn into actual cities. (There are some other problems with Chinese GDP figures, but Ghost Cities aren't among them.)
LeapingLennie 4y ago
The Economist recently reported on a study that used the brightness of countries' lights at night as proxy for GDP[1][2]. The study suggests China and other autocracies have significantly exaggerated their GDP growth over the last twenty years. [1]: https://www.economist.com/graphic-detail/2022/09/29/a-study-of-lights-at-night-suggests-dictators-lie-about-economic-growth https://www.economist.com/graphic-detail/2022/09/29/a-study-... [2]: https://bfi.uchicago.edu/working-paper/how-much-should-we-trust-the-dictators-gdp-growth-estimates/ https://bfi.uchicago.edu/working-paper/how-much-should-we-tr...
- MikePlacid 4y ago
  One of the things that struck me as weird after landing in the US - is how wasteful the lightning here is. If I was an autocratic ruler, I would definitely scaled it back a couple of times.
  - papreclip 4y ago
    I would scale it back as an autocratic ruler even if hurt GDP. It would be nice to see more of the night sky
  - Ekaros 4y ago
    Or in general. Living in sparsely populated country with what in other places would qualify as towns. Driving motorway towards one of the at night. Like tens of kilometers out in middle forest you see a clearly lighted up horizon. The light pollution is real.
- veltas 4y ago
  I want to say "this could be a false correlation" and recommend against believing the study's suggestion, but the damage is already done. Most of the thousands of people that read this article and the handful of people that read your comment have come away being convinced of something that could very well be false as if it were scientific fact because it was in the Economist and it's a "study".
  - david-gpu 4y ago
    The authors of the study looked at several possible alternative hypotheses. I suggest reading it before dismissing it.
- leosarev 4y ago
  I could for example use a sales of buckwheat as a proxy of GDP! I think ut will be some interesting reporting about who is exaggerated their GDP!
  - chriswarbo 4y ago
    Interesting idea. How large are your sample sizes, and what are the 95% confidence intervals on your results?
dirtyid 4y ago
They're probably referring to the rudimentary Martinez study that's making PRC collapists convinced PRC is over reporting GDP. For reference, much more sophisticated light study by NBER (staffed by FED economists) concluded PRC growth was higher than PRC reported GDP. But more broadly, it's important to understand GDP in PRC is (and has always been) guestimate to appease foreign investers / media for comparison wank. It's well known PRC smooths GDP to show general trends, but internally they've used their own metrics like LKI index (proxy measurement using combination of rail freight activity, electricity usage, industrial activity etc) and total social financing to measure economic activity. A few years ago, the most rigorous western analysis of Chinese GDP I'm aware of was conducted by CSIS who comprehensively reverse engineered Chinese GDP reporting from ground up, sector by sector, and concluded China was (~1 trillion USD) larger than official numbers purported to be. Also remember PRC has incentive to underreport GDP to keep developing economy perks. Ultimately, GDP like all statistics, is political. It's why internally PRC leadership is focusing on quality growth and "comprehensive national power" , a nebulous but (IMO) more useful composite index to guide development.
- antman 4y ago
  Those things you say are well known are in general new to me at least. What are the main references/links for these arguments?
  - dirtyid 4y ago
    If one follows PRC econ news from subject matter experts (both PRC and western) who full time focus on PRC, they're common discussion points. Li Keqiang index, Context from wiki: >According to a State Department memo (released by WikiLeaks), Li Keqiang (then the Chinese Communist Party Committee Secretary of Liaoning) told a US ambassador in 2007 that the GDP figures in Liaoning were unreliable and that he himself used three other indicators: the railway cargo volume, electricity consumption and loans disbursed by banks. Total Social Finance - do a quick google for summary, another composite indicator because up until 2010s PRC just had shit tier data reporting without even factoring in central gov lack of trust of local gov reporting. LKI became mainstream when he became premier in 2013, both LKI and TSF gets reported in domestic and foreign analysis on PRC econ since. GDP smoothing: https://www.economist.com/free-exchange/2015/07/15/whether-to-believe-chinas-gdp-figures https://www.economist.com/free-exchange/2015/07/15/whether-t... >There is a difference between smoothing data and totally fabricating it. Evidence suggests that China is guilty of the former (the lesser charge) but not the latter (the more serious allegation). There's better academic writings in past decade that goes into why PRC smooths GDP, but TLDR circles back to GDP is more politically useful as setting a "target" for growth (for both local officials and to inform foreign media/investors) than actual "measure" for growth, which is better gauged with other indicators, that can still be gamed but less so than cells on a spread sheet. >NBER study from 2017: https://www.nber.org/papers/w23323 https://www.nber.org/papers/w23323 Also discusses LKI and other background in PRC gdp reporting. As well as a country specific analysis vs Martinez let's crudely apply blunt metrics to every country generalist approach. Note this was published same year as Martinez study. CSIS analysis: https://www.csis.org/analysis/broken-abacus https://www.csis.org/analysis/broken-abacus The reports seems to be removed/for pay now, but there's podcasts from author that summarizes the study and degree of potential underreporting. As for incentive to under report, that's more conspiracy theory but IMO well justified. PRC benefits structurally from being treated as developing economy much than transient PR of reaching high income per capita. If one believes PRC is under reporting GDP, or overreporting population, especially by some of the higher end numbers, then PRC per capita GDP may have reached higher income, which has been a contentious issue by other (developed) countries at the WTO who wants to strip PRC from benefits reserved for developing countries.
ravel-bar-foo 4y ago
Local government officials in China are promoted / demoted based on local GDP growth. This incentivizes them to increase local industry and also to fluff up estimates of local industrial output. Repeat across the entire country for 20 years at every level of government, and as a result it is now estimated that China's actual GDP is half of its official GDP.

Not that I fully believe this, but the claims about Ghost Cities, for example. Infrastructure projects built to increase GDP and industrial activity without clear goals.

eru 4y ago
The Ghost Cities crop up every once in a while, but they aren't actually a problem. They usually turn into actual cities. (There are some other problems with Chinese GDP figures, but Ghost Cities aren't among them.)

LeapingLennie 4y ago

The Economist recently reported on a study that used the brightness of countries' lights at night as proxy for GDP[1][2]. The study suggests China and other autocracies have significantly exaggerated their GDP growth over the last twenty years. [1]: https://www.economist.com/graphic-detail/2022/09/29/a-study-of-lights-at-night-suggests-dictators-lie-about-economic-growth https://www.economist.com/graphic-detail/2022/09/29/a-study-... [2]: https://bfi.uchicago.edu/working-paper/how-much-should-we-trust-the-dictators-gdp-growth-estimates/ https://bfi.uchicago.edu/working-paper/how-much-should-we-tr...

MikePlacid 4y ago
One of the things that struck me as weird after landing in the US - is how wasteful the lightning here is. If I was an autocratic ruler, I would definitely scaled it back a couple of times.
- papreclip 4y ago
  I would scale it back as an autocratic ruler even if hurt GDP. It would be nice to see more of the night sky
- Ekaros 4y ago
  Or in general. Living in sparsely populated country with what in other places would qualify as towns. Driving motorway towards one of the at night. Like tens of kilometers out in middle forest you see a clearly lighted up horizon. The light pollution is real.
veltas 4y ago
I want to say "this could be a false correlation" and recommend against believing the study's suggestion, but the damage is already done. Most of the thousands of people that read this article and the handful of people that read your comment have come away being convinced of something that could very well be false as if it were scientific fact because it was in the Economist and it's a "study".
- david-gpu 4y ago
  The authors of the study looked at several possible alternative hypotheses. I suggest reading it before dismissing it.
leosarev 4y ago
I could for example use a sales of buckwheat as a proxy of GDP! I think ut will be some interesting reporting about who is exaggerated their GDP!
- chriswarbo 4y ago
  Interesting idea. How large are your sample sizes, and what are the 95% confidence intervals on your results?

dirtyid 4y ago

They're probably referring to the rudimentary Martinez study that's making PRC collapists convinced PRC is over reporting GDP. For reference, much more sophisticated light study by NBER (staffed by FED economists) concluded PRC growth was higher than PRC reported GDP. But more broadly, it's important to understand GDP in PRC is (and has always been) guestimate to appease foreign investers / media for comparison wank. It's well known PRC smooths GDP to show general trends, but internally they've used their own metrics like LKI index (proxy measurement using combination of rail freight activity, electricity usage, industrial activity etc) and total social financing to measure economic activity. A few years ago, the most rigorous western analysis of Chinese GDP I'm aware of was conducted by CSIS who comprehensively reverse engineered Chinese GDP reporting from ground up, sector by sector, and concluded China was (~1 trillion USD) larger than official numbers purported to be. Also remember PRC has incentive to underreport GDP to keep developing economy perks. Ultimately, GDP like all statistics, is political. It's why internally PRC leadership is focusing on quality growth and "comprehensive national power" , a nebulous but (IMO) more useful composite index to guide development.

antman 4y ago
Those things you say are well known are in general new to me at least. What are the main references/links for these arguments?
- dirtyid 4y ago
  If one follows PRC econ news from subject matter experts (both PRC and western) who full time focus on PRC, they're common discussion points. Li Keqiang index, Context from wiki: >According to a State Department memo (released by WikiLeaks), Li Keqiang (then the Chinese Communist Party Committee Secretary of Liaoning) told a US ambassador in 2007 that the GDP figures in Liaoning were unreliable and that he himself used three other indicators: the railway cargo volume, electricity consumption and loans disbursed by banks. Total Social Finance - do a quick google for summary, another composite indicator because up until 2010s PRC just had shit tier data reporting without even factoring in central gov lack of trust of local gov reporting. LKI became mainstream when he became premier in 2013, both LKI and TSF gets reported in domestic and foreign analysis on PRC econ since. GDP smoothing: https://www.economist.com/free-exchange/2015/07/15/whether-to-believe-chinas-gdp-figures https://www.economist.com/free-exchange/2015/07/15/whether-t... >There is a difference between smoothing data and totally fabricating it. Evidence suggests that China is guilty of the former (the lesser charge) but not the latter (the more serious allegation). There's better academic writings in past decade that goes into why PRC smooths GDP, but TLDR circles back to GDP is more politically useful as setting a "target" for growth (for both local officials and to inform foreign media/investors) than actual "measure" for growth, which is better gauged with other indicators, that can still be gamed but less so than cells on a spread sheet. >NBER study from 2017: https://www.nber.org/papers/w23323 https://www.nber.org/papers/w23323 Also discusses LKI and other background in PRC gdp reporting. As well as a country specific analysis vs Martinez let's crudely apply blunt metrics to every country generalist approach. Note this was published same year as Martinez study. CSIS analysis: https://www.csis.org/analysis/broken-abacus https://www.csis.org/analysis/broken-abacus The reports seems to be removed/for pay now, but there's podcasts from author that summarizes the study and degree of potential underreporting. As for incentive to under report, that's more conspiracy theory but IMO well justified. PRC benefits structurally from being treated as developing economy much than transient PR of reaching high income per capita. If one believes PRC is under reporting GDP, or overreporting population, especially by some of the higher end numbers, then PRC per capita GDP may have reached higher income, which has been a contentious issue by other (developed) countries at the WTO who wants to strip PRC from benefits reserved for developing countries.

ravel-bar-foo 4y ago

Local government officials in China are promoted / demoted based on local GDP growth. This incentivizes them to increase local industry and also to fluff up estimates of local industrial output. Repeat across the entire country for 20 years at every level of government, and as a result it is now estimated that China's actual GDP is half of its official GDP.