Senior Fellow at Carnegie China. For speaking engagements, please write to chinfinpettis@yahoo.com
12,176 followers 98 following 4,285 posts
view profile on Bluesky
Yicai: The bifurcation in China property markets will "intensify" over the rest of the year. Property prices are "likely to stabilize" in the first-tier and some second-tier cities, but excess capacity will continue to drive down prices in the rest of the country. www.yicaiglobal.com/news/chinas-...
2/2 The benefit for developing countries in diversifying away from the dollar is not just lower interest costs but, perhaps more importantly, less concentrated exposure to volatility in a single currency. This means less economic volatility and a better credit profile.
1/2 FT: "Developing countries are moving out of dollar debts and turning to currencies with rock bottom interest rates such as the Chinese renminbi and Swiss franc." www.ft.com/content/36f8...
9/9 This is what China has to avoid, and while there has been serious talk about replicating the adjustments that Tokyo identified in the mid-1980s, so far excess capacity has expanded, the consumption share hasn't increased, and the debt burden is rising faster than ever.
8/9 share of Japanese GDP dropped from 27% to 19%, investment growth went negative, and GDP growth dropped to just above zero. When manufacturing competitiveness is dependent on maintaining the growth model, you cannot change the latter without undermining the former.
7/9 But by then economic growth was too dependent on maintaining the growth model. When Tokyo finally tried to reign in credit in 1990-91, it took 17 years (1991-2008) for the consumption share of GDP to rise by ten percentage points. During this time the manufacturing...
6/9 By the mid-1980s Tokyo began talking urgently about increasing the role of domestic demand in the economy, reining in debt, downsizing excess capacity in manufacturing and infrastructure, and reducing its dependence on a large trade surpluses to keep excess capacity in check.
5/9 But while the technological advances were real, the process was sustainable only as long as the Japanese economy could continue to tolerate weak domestic demand, huge trade surpluses, and the surging debt needed to fund excess investment in manufacturing and infrastructure.
4/9 This is almost a perfect description of Japan in the 1980s, when Japanese companies became the world's technological leaders in most aspects of manufacturing, achieved at the cost of underpriced debt, high direct and indirect subsidies, and weak domestic demand.
3/9 This is not the first time we've seen an economic system that focuses on expanding production at all costs, hoping that the substantial technological advances that come from supporting manufacturing will eventually repay the substantial costs of this support.
2/9 As the article notes, China bears worry about the huge losses in the industry, even after substantial direct and indirect subsidies that come at the expense of domestic demand. China bulls focus instead on the rapid technological advances that characterize the industry.
1/9 Good article on strengths and weaknesses of China's approach to EV manufacturing. As the summary has it: "Beijing has run out of patience with companies slashing prices, and is urging restraint. But fierce competition is also producing a surge of innovation." www.nytimes.com/2025/09/02/b...
SCMP: "A strong push on Monday to create a development bank serving 10 Eurasian countries, including China, would help insulate the group from increasingly risky US dollar-dominated trade while accelerating key infrastructure work." www.scmp.com/economy/glob...
Caixin: "Chinese delivery giant Meituan has joined the crowded field of open-source artificial intelligence (AI) models, unveiling its own system." Even motorcycled delivery companies are piling into AI www.caixinglobal.com/2025-09-02/m...
Global Times: "Today, the relationship between China and the US stands as one of the most consequential bilateral partnerships. How these two powers interact directly determines whether the globe enjoys stability or suffers from discord." en.people.cn/n3/2025/0901...
Yes, although I think (I hope) we are starting to see a shift in thinking in parts of both parties.
Caixin: "What started as a trickle during the early days of the U.S.-China trade war has become a torrent. To retain access to U.S. markets, companies that have long powered China's export engine are setting up shop in Vietnam." www.caixinglobal.com/2025-09-01/c...
2/2 Restrictions on non-productive inflows would infuriate Wall Street (and most mainstream economists), but would be much more effective in supporting manufacturers and workers than the current approach, which just disrupts existing trade without affecting US trade imbalances.
1/2 If the courts rule out Trump's tariffs, it might force Washington to consider a more rational, less chaotic, approach to rebalancing US trade, for example by focusing on the role of the US capital account in absorbing global savings imbalances. www.ft.com/content/f3c7...
16/16 It is strange that the Trump administration is screaming that it wants a strong dollar because it is good for the US economy, and Trump's fiercest critics are screaming that Trump's policies will undermine the strong dollar, and that this will hurt the US economy. Confusion is bipartisan.
15/16 But it must nonetheless have an impact on the structure of the US economy, in which case we (and the British and the Canadians) really should be discussing the various other ways in which an economy can adjust to massive, unwanted net capital inflows.
14/16 But what they are more likely to find is that to the extent that the US has investment needs, these are not constrained by the inability of US businesses to access financing, in which case the dumping of foreign saving into the US economy will have no impact on investment.
13/16 They may still conclude that the US is in fact a saving-constrained developing country, in which US businesses have huge productive investment needs that they are unable to meet because they are struggling to raise the cash needed to build more production in the US.
12/16 Rather than assert that net capital inflows must lower US interest rates, and that this must lead to more investment and faster growth, economists should consider more carefully the various ways in which a real economy adjusts to changes in its external imbalances.
11/16 This hasn't been true for decades, which means that the other two ways, neither of which is good for the US economy (or that of the UK or Canada), are the more likely consequences of net capital inflows. carnegieendowment.org/china-financ...
10/16 If we assume that the first way is the only possible way, then yes, the US is better off with more net capital inflows. But this would only be the case if high US investment needs were constrained by the scarce availability of capital (i.e. if it is a developing economy).
9/16 These depend on underlying conditions. One way is indeed with lower interest rates and more investment. But a second way is with lower domestic interest rates and higher unemployment. And yet a third way is with no change in interest rates and higher domestic debt.
8/16 This is simply not true. Net capital inflows do change the structure of the recipient economy (any change in a country's external imbalance must be reflected by a change in its internal imbalances) but there are many ways the economy can adjust to absorb net capital inflows.
7/16 Sure enough, Bloomberg makes this argument without any discussion when it assets that "In a world where euro- or yen-denominated assets are more strongly vying for investor attention, borrowing costs for the US government would need to rise."
6/16 The standard explanation for how a country benefits economically from the widespread use of its currency is that net capital inflows automatically reduce domestic interest rates and raise investment. This seems like such a truism that it doesn't need to be discussed.
5/16 household transfers should naturally see the manufacturing share of their economies rise, and they should tend towards running trade surpluses. Their trade partners should naturally run deficits and see production shift from manufacturing to commodities or nontradables.
4/16 As a corollary, it also mean that these countries must allow manufacturing to move to countries that want to expand their share of global manufacturing. This shouldn't surprise as much as it does. Countries that subsidize manufacturing with explicit or implicit...
3/16 is that countries like the US (and the UK and Canada), whose currencies play a far larger global role than justified by the size of their economies, must run trade deficits large enough to balance net capital inflows, i.e. to accommodate the trade surpluses of other countries.
2/16 There is no evidence at all that a country's economy must benefit from the wide use of its currency. To the extent the global use of a currency primarily reflects the extent to which other countries want to acquire local assets in exchange, its only obvious economic impact...
1/16 While Bloomberg is right to say that "a less hegemonic dollar would [adversely] affect America’s geopolitical prowess," it is wrong to assume that it would also adversely affect the US economy, or that it would raise US interest rates. www.bloomberg.com/news/article...
SCMP: "China’s top banks face continued pressure, as a slowing economy, lower interest rates and weak loan demand weighed on first-half earnings." With nominal GDP up by a little more than 4% over the period, profits were down by 0.3%, with a net interest margin of 1.3%. www.scmp.com/business/chi...
Yicai: "China has issued a three-year action plan to optimize university majors in response to the accelerated technological revolution and industrial transformation to cultivate more high-level innovative and interdisciplinary talents." www.yicaiglobal.com/news/china-t...
At Beijing's School Club last week. Three guys who have been part of the Beijing music scene since ancient times.
Yes, in Wen Jiabao's 2007 speech he was already warning about low consumption.
11/11 If Chinese manufactures are truly efficient, rising wages should be a manageable problem. If they aren't, rising wages will be hard to absorb. I'd be more confident that the former were the case if, with such low relative wages, Chinese manufacturers had been a lot more profitable.
10/11 We'll see what happens. It is very good news for China that Beijing is forcing up minimum wages, but we shouldn't underestimate the amount by which direct and indirect wages must be raised, nor should we ignore the likely impact on the manufacturing sector.
9/11 That's a big increase, and it is unlikely that a manufacturing sector dependent on low wages, cheap capital, an undervalued currency, and other direct and indirect subsidies for its global competitiveness can easily absorb wage increases as high as that. carnegieendowment.org/posts/2024/0...
8/11 faster than GDP growth by roughly two percentage points. Broadly speaking, this implies that wages (along with other sources of household income) will have to increase by more than 6% annually if China hopes to maintain GDP growth rates of roughly 4%. carnegieendowment.org/posts/2024/1...
7/11 So what does this imply for Chinese wage policy? A little bit of arithmetic here is helpful. If we assume that China has ten years to raise the consumption share of GDP by 10 percentage points (the minimum amount needed), household income growth would have to rise...
6/11 from 27% (roughly where China is today) to 19%. This wasn't just a coincidence. As rising wages drove a higher domestic consumption share, they also undermined Japan's manufacturing competitiveness. This was one of the reasons Japanese GDP growth dropped to well under 1%.
5/11 Unfortunately most historical precedents suggest the latter is more likely. In Japan's case, for example, it took 17 years—from 1991 to 2008—to raise the consumption share of GDP by ten percentage points, and during that time the manufacturing share of Japan's GDP dropped...
4/11 Faster wage growth, in other words, will boost the consumption share of GDP, but it will also undermine China's manufacturing competitiveness. Either way China will rebalance, but one way is with a surge in consumption growth, and the other way is with much slower GDP growth.
3/11 It will also be an important test of the extent to which China can rebalance. China's low consumption share was largely driven by low wage growth relative to productivity growth in the 1990s and 2000s. This also drove the manufacturing sector's rising global competitiveness.
2/11 This is certainly good news. If raising minimum wages is done in a meaningful way, it will easily be the most effective of all the various ways in which Beijing has tried—without much success so far—to raise the household share of GDP and, with it, the consumption share.
1/11 Caixin: "A wave of minimum-wage increases is sweeping across China, with major economic hubs like Beijing and regions such as Hunan province announcing fresh hikes as part of a national push to lift incomes for the country’s lowest-paid workers." www.caixinglobal.com/2025-08-28/c...
2/2 The NBS credited the improvement to "a government-led campaign to curb excess competition." If they are right, this suggests that involution may be addressed by shifting at least part of the losses from the manufacturing sector to mining and utilities.
1/2 Industrial profits in July were down 1.7% year on year, an improvement on June's 4.3% decline. This was led by a 6.8% increase in manufacturing profits, up from 1.4% in June. english.news.cn/20250827/eee...
Washington's tariff policy for India subordinates US economic interests for a reformed and improved global trading system to its desire to use India to pressure Putin, which is the opposite Trump's promise to subordinate the global role of the US to its domestic interests. www.ft.com/content/bc3a...
2/2 costs, wiping out profits along the supply chain. And complicating the effort further, capacity keeps rising. Domestic polysilicon output increased 5.7% month-on-month in July, with production forecast to rise an additional 16% in August." carnegieendowment.org/posts/2025/0...
1/2 Caixin: "China’s solar power manufacturers, once the vanguard of the country’s green energy ambitions, are mired in a war of attrition. By mid-2024, prices for polysilicon, wafers, cells and modules had all fallen below production... www.caixinglobal.com/2025-08-26/c...
In my latest piece for the China Reform Imperative, I argue that this year's focus on "involution" is largely a restatement of China's decade-long problem of excess capacity and weak domestic demand. carnegieendowment.org/posts/2025/0...
15/15 Obviously no one wants to do the latter, and doing the former has proven difficult. But the problem will persist until Beijing either gets consumption to soar, lowers the GDP growth target, or decides to rein in runaway debt. carnegieendowment.org/posts/2025/0...
14/15 More importantly, what the Chinese economy must do, whether you call it underconsumption or overinvestment, is the same. If it cannot get consumption growth to rise, it must reduce the growth in production, i.e. GDP, until the two are in balance.
13/15 If you don't want to call this "underconsumption", that's fine, but then you have to call it "overinvestment". In either case it leads to excess capacity, whether that excess capacity shows up in investment in property, infrastructure or manufacturing (or all three).
12/15 Those who argue that China's real consumption is in fact "normal", and that this shows that China doesn't have a rebalancing problem, miss the point. China's consumption is only "normal" if China's GDP has been overstated by unrecognized overinvestment.
11/15 If, on the other hand, Chinese consumption is "normal" in real terms, then either China is running a huge trade deficit (it clearly isn't), or the value of Chinese investment must be overstated, which in turn means that China's GDP is overstated.
10/15 One one hand, if China's reported GDP is an accurate measure of the value of goods and services produced by the Chinese economy (i.e. comparable to other economies' GDP), then it is true that the Chinese consume far too little of what they produce. This is just arithmetic.
9/15 To put this in simpler terms, there are two very different ways in which we can express China's demand imbalance, but these are just different ways of expressing the same problem of excess capacity—and require the same set of solutions.
8/15 Kim notes that this is one obvious way to resolve the "underconsumption" problem: "Chinese underconsumption is possibly the product of undercounting consumption and overestimating GDP." Yes, except that consumption is only undercounted to the extent GDP is overstated. It's the same thing.
7/15 are capitalized rather than expensed, this means, among other things, that GDP growth has been systematically overstated every year by the amount of capitalized investment losses, as I explain in a recent essay. carnegieendowment.org/posts/2025/0...
6/15 But there is another way, to which Kim refers in his essay. It may be real GDP is much lower than reported GDP. If it is true that a significant share of investment is wasted, but that this waste isn't accounted for in the GDP data, mainly because investment losses...
5/15 In fact China has the highest investment share in history as well as the highest trade surplus in history. There are only two ways this can happen. One way, of course, is for Chinese consumption to comprise a very low share of GDP.
4/15 If Chinese consumption levels were truly "normal", than one or both of two other things must be true: either Chinese investment is "normal", or if it is higher than normal, China must be running a trade deficit. After all, Net exports = GDP - Consumption - Investment.
3/15 This misses the point. What matters in the macro context is not the absolute level of consumption but rather the gap between consumption and production. This gap must be balanced either via the trade account of via investment (or both).
2/15 The C40 essay illustrates some of the confusion that surrounds the issue of Chinese underconsumption. It argues that the Chinese don't underconsume because when corrected for prices or for income, Chinese consumption is more or less "normal". www.pekingnology.com/p/chinas-con...
1/15 This interesting essay by Oliver Kim on Chinese underconsumption was set off by a recent C40 Policy brief that argues, in the words of its title, that "China's consumption is not nearly as low as it appears." www.global-developments.org/p/does-china...
Very inspiring Caixin article about China's teen geeks. www.caixinglobal.com/2025-08-22/p...
9/9 This was the point Keynes made in his Bretton Woods trade proposal, and why he (and Harry Dexter White, too) were so opposed to the unfettered flow of global capital. It is surprising that it is taking us so long to understand why. www.commonplace.org/p/why-keynes...
8/9 in productive investment, whereas in the latter they don't. If global excess saving flowed less to the US, the UK, and Canada (roughly two-thirds of the total) and more to developing countries, global demand would rise, rather than shift from deficit countries to surplus countries.
7/9 But a world in which capital flows are largely driven by productive investment flows is very different from one in which capital flows mostly for other reasons. In both cases, domestic economies have to adjust to capital-flow imbalances, but in the former, they adjust to increases...
6/9 A few years ago, most economists thought it was crazy even to suggest that the capital account could drive the trade account, including in big economies like that of the US. Today it is much more accepted, even if many economists still don't understand the implications.
5/9 distorts global economies by forcing changes in external imbalances that have little to do with the productive flow of capital. Changes in external balances, of course, must be matched by changes in the internal imbalances of countries that provide or receive these flows.
4/9 It suggests, instead, that capital flows are driven mainly by currency and stock-market speculation, herd behavior, leverage, flight capital, and the pursuit of safe assets. There may be good reasons for investors individually to do these things, but collectively it...
3/9 The fact that it isn't suggests that international capital flows do not consist mainly of investment flowing from where it is less needed to where it is more needed, as the proponents of unfettered capital flows claim.
2/9 I am not sure this is actually happening—it's not what the BoP data suggest—but if it were, it would be a very good thing. Capital should not flow from fast-growing, capital-poor economies to slower-growing capital-rich economies. It should flow in the opposite direction.
1/9 FT: "investors worldwide were already nervous about owning too many US dollar investments. This news only energised a growing conviction by them to seek other places to put their money, including the emerging markets." www.ft.com/content/85e5...
FT: "The world’s largest economies will lack the workers they need to power growth and keep prices stable in the coming decades unless they attract more foreigners, top central bankers warned." www.ft.com/content/8bfd...
Yes, widely available, for example here: www.ecb.europa.eu/press/econom...
15/15 We will eventually see this transformation, but I think Beijing can still postpone it for a few more years. Until then, I mostly expect excess capacity in the involuted sectors to shift mainly into other manufacturing sectors and partly into infrastructure.
14/15 What this means is that the only sustainable solution to China's problem of excess capacity is to engineer a transformation of the country's economic system and growth model. Other "solutions" just shift excess capacity from one sector to another.
13/15 That's because it is extremely difficult to pull off. Low consumption in China is not the result of an oversight that can be fixed administratively. It is fundamental to an economic system in which household transfers subsidize rapid growth in investment and manufacturing.
12/15 The third possibility, and the best by far, is that it shifts resources to boost consumer spending. This is what most analysts cited in this article have proposed. But of course more consumption has also been the best "choice" for the past 10-15 years, and hasn't happened.
11/15 Another possibility (on which they already seem to have started), is to increase spending on infrastructure. The problem is that China already has "excess capacity" in infrastructure, after the huge amounts that poured into infrastructure investment between 2010 and 2017.
10/15 The most likely outcome, in my opinion, is it shifts investment into other manufacturing sectors, those not yet suffering involution. This, of course, won't resolve the problem of excess capacity, but it will make it much less concentrated in a few key sectors.
9/15 As it is extremely unlikely that we will see property investment recover any time soon, for China to cut capacity in the most involuted sectors of manufacturing will require shifting debt and resources to create another source of growth.
8/15 But this means that unless Beijing is willing to give up the GDP growth target (which for now, it isn't), it cannot "resolve" involution—i.e. cut capacity sharply in the worst affected areas—unless it can engineer a corresponding upsurge in demand elsewhere.
7/15 The consequence was perhaps predictable: the creation of excess capacity in the property sector now shifted to the creation of excess capacity in the manufacturing sector, and especially in politically preferred sectors, like solar panels, EVs and batteries.
6/15 More manufacturing, in other words, was not driven by any perceived increase in Chinese or global demand. It was wholly a reaction to the decline in property investment combined with a political determination to continue meeting excessively high GDP growth targets.
5/15 But China's GDP growth didn't slow at all. Why? Because Beijing did not allow the decline in property investment to translate into a decline in overall investment. It "solved" the declining property investment by engineering a surge—almost RMB for RMB—in manufacturing investment.
4/15 This, after all, has happened after property busts in nearly every other country. GDP growth is largely equal to consumption growth and investment growth, and in China, investment (especially property investment) comprised a much higher share of GDP than anywhere else.
3/15 When investment in the property sector dropped sharply after 2021-22, what "normally" should have happened is that the sharp resulting decline in Chinese investment growth should have led to a sharp decline in GDP growth.
2/15 a problem in China for at least 10-15 years, it's latest manifestation was a direct consequence of the collapse of a property bubble which was itself set off by policies Beijing implemented in order to protect the economy from an earlier case of weak demand and deflation.
1/15 Very good Bloomberg piece on Beijing’s recent push to curb overcapacity through an “anti-involution” campaign. It is important to remember in this context that while excess capacity has been... www.bloomberg.com/news/article...
7/7 But losers tend to cut back on spending faster than winners increase it, and so what's good for the economy over the long term can still be painful over the short term. Outside top-tier cities, I expect property prices to decline for at least another few years.
6/7 Plummeting property prices, by the way, are not necessarily bad. They represent a transfer of wealth from owners of multiple apartments to those who need them, e.g. the young and the poor (and musicians and artists, I might add), and that is overall a good thing.