(Big time) I’m on my way, I’m successful (Big time)
(A lot) I have to show it, yeah (a lot)
(Big time) Much larger than life
(A lot) I’m gonna watch it grow (A lot)
-Peter Gabriel
In May, the World Bank completed a price survey to “officially” determine purchasing power parity GDP, one of its regular International Comparison Program (ICP) assessments.
Like the university rankings, the league table for the world’s largest economy has seen no surprises and has shifted enough for geeks to notice. Harvard is Harvard, and it makes little difference whether Princeton ranks higher or lower than Yale this year.
For the obsessives, the gap between China and the US widened by 5.6%, India inched closer to China, Japan dropped slightly but maintained its position, Russia overtook Germany, France overtook the UK, Indonesia dropped two places, Brazil moved up one, and the top 10 remained in the top 10.
While Russia fans may rejoice at its 13% increase in GDP in purchasing power parity terms and Brits may be worried about its fall out of the top 10, the latest ICP did not produce any notable revelations, nor should it.
Periodic price surveys are necessary to calibrate and maintain the accuracy of the PPP adjustment, but if significant changes occur, either too much time has passed between surveys or the survey methodology is broken.
The ICP is a massive undertaking. According to The Economist, World Bank researchers visited 16,000 stores in China alone to collect price data. The latest ICP assessment collected data in 2021, four years after the 2017 survey. They concluded that China’s GDP is undervalued by $1.4 trillion, and that China’s GDP in 2022, measured in purchasing power parity terms, will be 119% to 125% of the United States’.
According to The Economist, China’s National Bureau of Statistics (NBS) was not impressed with the results, saying “we need to interpret the results cautiously and get a proper grasp of the global economic situation and the situation of each economy within it,” and downplayed them by emphasizing that China is still a “developing country.” If the NBS doesn’t like this modest upward revision of China’s PPP GDP, then you’re probably not going to like the rest of this article.
China’s GDP in purchasing power parity terms is 25% larger than the US? Are you kidding me? Last year, China produced twice as much electricity as the US, 12.6 times as much steel, and 22 times as much cement. Chinese shipyards account for over 50% of global production, compared to a tiny fraction of US production. In 2023, China will produce 30.2 million cars, nearly three times the US’s 10.6 million.
On the demand side, 26 million cars were sold in China last year, 68% more than the 15.5 million sold in the United States. Chinese consumers purchased 434 million smartphones, three times as many as the 144 million sold in the United States. As a country, China consumes twice as much meat and eight times as much seafood as the United States. Chinese shoppers spend twice as much as their U.S. counterparts on luxury goods.
While Chinese travelers will take 620 million flights in 2023, 25% fewer than the 819 million flights taken by Americans, they will also take 3 billion high-speed rail trips (and 685 million conventional rail trips), far surpassing Amtrak’s 28 million trips.
With the exception of luxury items, all of the above are quantity or unit measurements that require adjustments to make apples-to-apples comparisons of quality and features. It would be highly presumptuous of us to discount the 16,000 store visits conducted by the World Bank and accuse them of grossly underestimating China’s PPP GDP.
Yet that is exactly what we are going to do. On its face, it is absurd that Chinese production and consumption, many times the size of the United States, should be realistically discounted for lower quality and functionality to amount to just 125% of U.S. GDP in purchasing power parity terms.
It’s not that we think the World Bank has done a bad job. We just think that China’s National Bureau of Statistics has underestimated GDP for decades, contrary to popular opinion, and that the World Bank has to work within the bounds of the data reported by the National Bureau of Statistics. This was politically important a few decades ago because of WTO concessions, and is politically important today as China aims for leadership in the Global South to maintain its status as a developing country.
We believe that China’s GDP and PPP GDP are underestimated due to an incomplete transition from the Material Production System (MPS) of national accounts, which deliberately excludes services. The World Bank likely measures China’s consumption of goods as many times that of the United States, but services consumption as only a fraction of that of the United States.
The United Nations System of National Accounts (UNSNA) provides voluntary guidelines and clearly states that countries should base their national accounts on local conditions, which for Western countries means adopting all the UNSNA “innovations” that have been in place over the years.
Items such as estimated rent, legal costs, and research and development expenditure are all now included in GDP. The UK has enthusiastically embraced both illegal drugs and prostitution as part of GDP because… well, why wouldn’t they? The UNSNA 2008 guidelines explicitly recommend that illicit market activity be included in GDP.
China’s National Bureau of Statistics has stuck to its guns at the conceptual level: rightly or wrongly, the Leninist MPS considers services as a necessary cost of material production, not real value creation. When China first tried to convert its MPS to the SNA in 1985, it added a ridiculously low 13% to the MPS figures and called it China’s services GDP.
For many years, the World Bank has been pressuring China’s National Bureau of Statistics to moderately increase its services GDP, but with limited success.
The housing price crisis in the West, especially in the United States, is driven primarily by inflation not in manufactured goods but in basic necessities like rent, health care, education, and child care. These costs are also rising in China, but to a lesser extent, and in any case are largely excluded from GDP.
The 2021 ICP survey also fails to capture the price and service wars that have erupted across industries and products — a bane for businesses but a boon for consumers.
This is most evident in the Chinese automotive market, where OEMs are slashing prices (Hyundai Sonata from $42,000 to $17,000) and offering cutting-edge technology at bargain prices (BYD Q plug-in hybrid electric vehicle with a 2,000 km range for $14,000). Solar panel prices will fall by 50% in 2023 and continue to decline in 2024. CATL has announced plans to halve the price of lithium-ion batteries by the end of 2024.
Restaurants offer niceties like hot towels, lotion by the sink, and stylish décor. Hairstylists hand out bottled water and fruit platters. Technology companies have slashed the price of large language models to essentially free. The quality of service in China is hard to quantify, but it is now far superior to that in the West, and perhaps Japan.
Adherence to the UNSNA has undermined the meaning of GDP: as essential services take up ever larger shares of Western economies, their growth does not seem to translate into any tangible improvement in living standards.
Are US healthcare and universities twice as good as they were in 2000? If US households had failed to make significant improvements in healthcare, education, housing and child care over the past 20 years, inflation may have been systematically under-reported and GDP growth may actually have been less than 1% per year (instead of 2%), which is tantamount to stagnation given the 0.8% annual population growth. This may go a long way to explaining public anger and the collapse of US politics.
China’s materialistic GDP may be a better measure of the economy in terms of standard of living, especially considering the obvious insanity of UNSNA in officially recommending that drugs, prostitution, illegal gambling and theft be included in GDP.
Western defense analysts have a point in saying that Chinese defense spending is wildly inflated, but it is not Chinese defense spending that is underestimated; it is defense spending by Western countries, especially their Pentagons, that needs to be reevaluated.
Somehow, because of the $1 trillion a year the United States spends on defense (including intelligence and Department of Energy programs), the U.S. Navy is shrinking, while China is building the world’s largest navy by ship count with a budget of $236 billion.
Similarly, analysts who lament that China accounts for 30 percent of global manufacturing output but only 13 percent of household consumption are completely misguided: China accounts for 20 to 40 percent of global demand for almost all consumer goods, but many of the services it consumes are excluded from national accounts.
So how big is it? How big is the Chinese economy really? About six months ago, I estimated that China’s GDP would need to grow by 25-40% to get to UNSNA base.
But after buying a car, purchasing a domestically branded carbon fiber road bike for $2,600 (the equivalent of a $15,000 Trek), paying $7.65 for Bluetooth earphones (far better than the $250 PowerBeats Pro they replaced), renting a car for $20 a day, staying in a boutique hotel for $30 a night, purchasing (and promptly losing) a very strong and durable umbrella for $2.20, undergoing a series of unfortunate medical interventions (both major and minor) under my expat health insurance deductible, and receiving superb customer service on even the most minor purchases, Han Feizi’s perception of price and value has been shattered.
No, it’s not that the ICP did a bad job. They were hamstrung by initial conditions imposed by the National Bureau of Statistics of China. And the data released in 2024 was taken in 2021. This is ancient history in China. Recently, the ICP adjusted China’s PPP GDP relative to the US by a few percentage points, much to the surprise of the National Bureau of Statistics.
But in reality, the exact adjustment will be a factor of 1 or 2.