Summary

单伟建’s core call on China’s economy: don’t be overly optimistic, but he is absolutely not pessimistic—China’s economy is operating below its potential. The evidence is all in the numbers: the average reserve requirement ratio is about 6.5%, the highest among major economies (0 in the US, 0.8% in Japan and 1% in the EU); every 1-percentage-point RRR cut releases RMB2T; the government balance sheet is in a net-asset position; and household bank deposits have reached RMB160T, above the RMB140T GDP. “Even if 5% were released, the economy could grow 6%.”
The hope for reviving consumption lies in property: property accounted for about 59% of household wealth in 2019, and the shrinkage in household balance sheets is the main reason consumption is weak. Bottom signals are emerging—the property market peaked in mid-2021 and has declined for 5 years, national rental yields of about 2.8% now exceed the 1.8% yield on 10-year government bonds, and Shanghai’s resale home prices have risen about 7.6% over 5 straight months. The “buy rising, not falling” psychology could make the recovery self-reinforcing and turn it into a “sustainable recovery.”
Export dependence is unsustainable. Net exports contributed more than 30% of GDP growth in both 2024 and 2025, but “that contribution does not register with consumers.” In 2023, consumption contributed 82.5% of 5.2% growth, while net exports contributed -11.4%. With protectionism rising, “exports alone cannot be relied on; only domestic consumption and demand can.”
李翔 noted that the market capitalization of China’s leading companies is only 1/3 to 1/5 that of their US peers, and 单伟建 offered 4 explanations. The US market’s overall P/E is about 2x China’s, implying China is undervalued; “involution” means competition eventually leaves nobody making money; overall price and income levels are lower; and Chinese buyers have historically resisted paying for intangible assets—AI token prices are “probably 1/10 of the US.” The root of involution lies in local-government incentives; making consumption tax a major source of local-government revenue could potentially solve the problem at its root.
单伟建 publicly challenged Buffett’s claim that he does not care about interest rates. “As an investor, he has to care about the macroeconomy.” The move from 0 rates to double-digit rates has created a major obstacle for US PE and sent listed PE companies’ valuations sharply lower; Berkshire’s $300B cash pile being unable to find a home “has a lot to do with the interest-rate environment.” And “Buffett is not a god”—he has underperformed the broader market over the past 10 years, and “if you cling to old ways when conditions change, the market can beat you.”
The episode’s central theme is that living through cycles makes investors humble. He “could see at a glance” that WeWork was a subleasing concept, while Goldman Sachs and SoftBank did not, because “people who have not lived through a cycle tend to be more aggressive.” Long-Term Capital Management, with at least 2 Nobel laureates and annual returns of 30%, was still brought down by leverage and the Russian ruble crisis. A former US Treasury secretary provided the closing line: “The market can stay depressed longer than you can stay solvent.”
A little-known fact about PE is that a buyout fund’s failure rate is already high if it gets 1 or 2 out of 100 deals wrong. VC can miss on 50 out of 100 investments without a major problem, but a single buyout can involve several billion dollars, leaving no room for error. And crises are not necessarily a friend to PE: when Morgan Stanley solicited more than 40 bids for Korea First Bank, “only 2 showed up in the end.”

Deep dive

1. Writing as a Side Project: Filling a Gap in the Buyout Literature

单伟建 did not write out of a Dalio-style “hero’s journey sharing phase”—“I had absolutely no such idea.” His framework was straightforward: private equity has become a huge industry, and US PE controls 2x as many companies as are publicly listed. Buyouts appear in the news almost every day, “but nobody has described what happens inside a buyout.” The Money Game, written from an insider’s perspective, chronicles the entire Korea First Bank acquisition and “fills a gap in the literature.”
A Turn in the Road covers the takeover, transformation and exit of Shenzhen Development Bank. It took 20 years to become a book: “There was no time pressure. Whenever I had time, I would write a few lines.” The motivation was the historical setting. The deal took place against the backdrop of China’s banking-system reform; “without China’s reform and opening-up, this deal would never have happened.” The book records a stretch of macro history.
His attitude toward sharing is simple: “You accumulate some experience, and it may be useful to others. It would be a pity not to share it. It would be a pity to let it all rot inside you.” That is all.

2. The Only Case of Foreign Control of a Nationwide Chinese Bank

李翔 covered the deal for The Economic Observer and called it the 1st foreign acquisition of a Chinese bank. 单伟建 corrected him: “It was not only the 1st case, but the only one.” The comparison is Citigroup’s investment in Guangdong Development Bank: constrained by the 20% foreign ownership cap, Citigroup participated in management contractually but “did not actually have control.” Newbridge was the single largest shareholder: “We were the only foreign investor to control and transform a nationwide Chinese bank.”
Would it be disappointing if his 3rd book never gets a simplified-Chinese edition? “Of course it would… but there is nothing to be done. This is what we call ‘man proposes, Heaven disposes,’ and investing works the same way.” But waiting is not enough: “You still have to make the plans and do the work—Heaven often rewards the diligent.”

3. “Keeping Quiet and Getting Rich Is Selfish”

Someone advised him to speak less about China’s economy. His response was unequivocal: “Keeping quiet and getting rich is selfish. The most meaningful thing in life is to make some contribution, however small.”
A Financial Times article a few years ago sparked concern that it might affect him. “Some people worried it would have an impact on me, but in fact nothing happened.” His explanation: some criticism is made in bad faith, while his own views—even when unpleasant to hear—are well-intentioned, constructive and aimed at making things better. “So I am completely at ease.”

4. His Case Against Buffett: Investors Must Care About the Macro, and “Buffett Is Not a God”

Buffett says he never pays attention to Fed rates. 单伟建’s blunt response: “That is actually wrong. As an investor, he has to care about the macroeconomy.” The industry data support his view: US PE has performed very poorly over the past 2 years, largely because of rates. With normal US growth around 2%, generating returns above 20% requires leverage. Zero rates after 2008 made “leverage almost costless”; over the past 3 years, private-market debt rates have risen into double digits. “How much does your company’s profit have to grow just to cover the debt?” The sharp decline in listed PE companies’ valuations has had “an almost fundamental impact on the entire industry.”
He offers Buffett 1 mitigating explanation: the capital comes from insurance companies he controls and carries a low cost, “so it may not affect him much.” But Berkshire has $300B in cash it cannot deploy. “That has a lot to do with the interest-rate environment. If he says he does not care at all… no truly rational investor would act that way.”
“Many people portray him as a god, but he has not gotten everything right.” Buffett has underperformed the broader market over the past 10 years. This bull market has been driven by technology stocks, while Buffett’s edge came from value investing and keeping away from tech. Value won in the preceding 10 years; tech won in the subsequent 10. “When market conditions change, if you cling to old ways, the market can beat you.”
There is still a logic to the underperformance: Buffett has lived through cycles and knows that a cycle will come, so he stays relatively conservative. “If you are conservative in a bull market, you will obviously lag the market.”

5. Why His Buyout Fund Avoids High Tech: Control Rights Clash with Founder Incentives

The structural reason his buyout fund does not invest in technology stocks is that a buyout requires control, while “high tech needs entrepreneurs and scientists. If you take away control, you reduce their incentives.” Technology companies also have “no long history of stable cash flow,” making them difficult buyout targets. The exception is 博瑞医药, a new-drug company they control; a separate growth fund invests across healthcare and technology.
Private-market equity also has a circle of competence. “If you asked me to do venture capital, I would not know the business. I genuinely would not know how to judge it”—there is no cash flow or operating history, and he lacks the experience to assess whether the technology can be commercialized. VC still holds appeal: if the right person wanted to build a venture fund under his platform, “I would of course be very welcome.”

6. WeWork Was Subleasing, Not Technology: “I Saw It at a Glance; Goldman Sachs Didn’t”

Goldman Sachs once brought WeWork’s founder to meet him. “After listening, I felt it was impossible to invest.” It was “really just a subleasing concept”: lease commercial space and sublet it, with no technology whatsoever. Fixed costs were too high; if sublease revenue fell short, the risk was enormous. SoftBank “was probably seduced by the packaging,” and was not initially the controlling shareholder. As the company needed more and more funding and SoftBank kept adding to its investment, it “probably only became the controlling shareholder later.”
Why were so many smart people fooled? “People who have lived through investment cycles tend to be conservative; people who have not lived through cycles tend to be aggressive.” The US has been almost continuously in a bull market from 09 to now, and many investors have never experienced a cycle, so they are more willing to charge ahead and take risks. But seeing many cycles also has a cost: “Many opportunities are created by the market at that moment, and you may simply miss them.”

7. Cycles Make Investors Humble: Pocket Depth and Long-Term Capital

He is unimpressed by people who claim to have predicted the 08 crisis 3 years in advance. “Predicting it 3 years ahead is no different from not predicting it at all—the cycle was always going to come, but if you predicted it too early, you missed many market opportunities.” The investors in The Big Short won because they acted “right before the crisis arrived”; people who made the same call 1 year earlier lost.
He quotes a former US Treasury secretary: “The market can stay depressed longer than you can stay solvent.” Shorting requires carrying costs. If the turnaround does not come before the money runs out, “you lose. This happens all the time.”
Long-Term Capital Management delivered the same lesson with leverage: it had at least 2 Nobel laureates involved, generated annual returns of about 30% in its first 2 or 3 years, used extensive data and statistical analysis to forecast markets, and then added heavy leverage. The sharp depreciation of the Russian ruble changed all of the assumptions behind the profits. “The loss was not a little loss—it was several times the amount invested.” The fund collapsed.
That is the episode’s title in conclusion: “Investment success is partly about human effort and partly about Heaven. You have to understand that investing does not depend entirely on you; it also depends on market conditions and luck. Then you become more humble, more conservative and more attentive to risk, and you are less likely to suffer a catastrophic blow. If you think it is all your own ability, the fall can be much harder.” As for 李翔’s impression, after searching the internet, that he had not suffered many losses: “Every investor is willing to talk about crossing 5 passes and defeating 6 generals. Nobody wants to talk about the defeat at Maicheng. But no investor has made it through without mistakes.”

8. Reporters See Buyouts “Through Fogged Glass”; A Good Book Still Matters 100 Years Later

The Barbarians at the Gate has 2 shortcomings. First, an outsider does not know how decisions were made: “It is like looking at flowers through fog—you cannot see clearly.” Second, the book was rushed. KKR acquired RJR Nabisco in 88, and the book came out around 89. “What happened afterward had not even happened yet.” The investment was not actually profitable. The reporter was “not just looking through fog, but merely watching the spectacle,” and would never wait 7, 8 or 10-plus years to write the ending.
Henry Kravis put it this way: “Any fool can buy a company; all he has to do is pay the highest price. Whether he ultimately makes money or loses money depends on what happens afterward.”
His standard for judging a book comes from 南怀瑾: “A truly valuable book still has value when read 100 years later.” Nobel laureate Herbert Simon did not read newspapers: “News is not really knowledge. Much of it is entertainment.” 单伟建’s own test is practical: if he misses newspapers for 1 week while traveling, he will not catch up when he returns. “It does not matter if I never read that week’s newspapers.”
His response to Taleb’s claim that merchants are almost never remembered is relaxed: “What is the point of being remembered as a merchant? What is the point of anyone being remembered? If what you leave behind offers some value to later generations, that is meaningful. Knowledge that others do not have has value.” He is unusually direct about Ayn Rand: “I probably would not like books that take an extreme position on individualism. Those books are probably meaningless.” When 李翔 jokes that Chinese people also love heroes such as Sun Wukong, 单伟建 replies: “He attained enlightenment.”

9. The Purpose of Private-Market Equity: Returns for Pensions, Better Resource Matching

In defense of KKR and Blackstone amid public criticism in the US, he makes 2 points. First, “the money is not ours.” It comes from insurance funds and sovereign wealth funds and ultimately represents ordinary people’s pensions. “If you create value, you are creating value for a lot of people.” Social-security systems everywhere need investment returns to support them. Second, “buyout opportunities come from market inefficiencies.” Transforming troubled companies makes the market more efficient.
Shenzhen Development Bank was the case in point. Before the takeover it was “a problem bank that was very poorly managed”; after the transformation, profit margins rose sharply and its scale expanded substantially. When it was sold to Ping An, a People’s Daily editorial carried the headline “Creating a Win-Win-Win Situation”—employees, shareholders, society and the buyer all won. 李翔 summarized the idea as “matching resources more efficiently.” 单伟建 replied: “That summary is better than what I said.”

10. 10 Years on the Gobi: Mostly Wasted Youth, but a Higher Threshold for Hardship

He rejects any romanticization of his sent-down youth years: “Overall, it was a waste of youth. There were no books to read, and after 10 years without studying, it became difficult to study later.” Only a small number of people avoided being completely derailed through self-education. “It was not an experience whose benefits outweighed its costs.”
He does acknowledge the toughness it built: when difficulties arise, remembering that period makes it easier to think, “Things could be much harder; this is nothing,” raising your tolerance for adversity substantially. People who have had an easy ride can become discouraged or even give up when setbacks arrive. His conclusion remains clear-eyed: “It would be better not to suffer that much in the first place.” For the nation as a whole, he hopes that period is permanently over and that young people today never have to go through it.

11. The Modernity Paradox: Middle Income, World-Class Infrastructure—“Reform and Opening-Up Got It Right”

He first corrects the description of China as the world’s 2nd-largest economy: on a purchasing-power-parity basis, China has already surpassed the US by about 30% and is the largest economy. In nominal dollar terms, it is 2/3 the size of the US. Per capita, however, China remains low at about $13,000, ranking somewhere in the 70s globally and roughly comparable to Malaysia, Mexico and Turkmenistan. “We are still a relatively poor country.”
The scale of change is extraordinary. “In 1990, China’s per-capita GDP was lower than India’s.” Today, China’s per-capita GDP is 5x India’s. At the time, the US economy was 15x China’s and US per-capita income was 70x higher. Today, the per-capita gap is only 3x or 6x, depending on the metric.
The paradox is that the modernity of a middle-income country “exceeds that of almost every developed country.” When he first arrived in the US, the country had 80,000 kilometers of highways; “today it still has 80,000 kilometers.” China went from 0 to 200,000 kilometers—2.5x as much—while its high-speed rail, bridges, airports, ports and ultra-high-voltage grid are all world-class. “If you go to Mexico or Malaysia, you would never imagine seeing the level of development China has today.”
How did it happen? “Reform and opening-up got it right.” The county-level competition framework associated with 张五常 and the emphasis on entrepreneurship from 周其仁 and 张维迎 are “all like the blind men and the elephant”—each is right, but none fully explains the outcome. The US also has entrepreneurship, so why did China grow faster? He is writing a new book on the question. “I have some ideas, but they are not mature, so I will not discuss them yet.” As for whether China will become proud of its achievements: “Pride is not a bad thing. It is an incentive—it motivates you to work harder and strive for more.”

12. The Gap Between Lived Experience and Data: More Than 30% Export Contribution, No Consumer Impact

His explanation for the debate between optimism and pessimism after 2021 is structural: “People’s experiences differ, so their feelings differ.” Property developers could not possibly be optimistic from 2021 to today, while new-drug, high-tech and export companies “are probably very optimistic.” “Individual experience cannot replace macro data, and macro data cannot change individual experience.”
The key number is that net exports contributed more than 30% of GDP growth in both 2024 and 2025. But “that contribution does not register with consumers.” In 2023, consumption contributed 82.5% of 5.2% growth, while net exports contributed -11.4%. “What really improves people’s lived experience is greater confidence in consumption.”
Exports are a double-edged sword. With the trade war and protectionist measures from Europe, “exports alone cannot be relied on… The central government is right: only domestic consumption and demand can be relied on.” It is also true that “the world cannot do without Chinese manufacturing.” But “almost every good thing hides a problem: if you become overly dependent on it, others will start guarding against you.”

13. Not Too Optimistic, but Absolutely Not Pessimistic: China Is Operating Below Potential

His own position sits outside the usual bull-bear framing: “You cannot be overly optimistic about China’s economy, but I am absolutely not pessimistic. China’s economy is currently operating below its potential, and a great deal of potential remains untapped.”
The macro policy space is substantial. The average reserve requirement ratio is about 6.5%, “the highest among all major economies”—0 in the US, about 0.8% in Japan and about 1% in the EU. “Every 1-percentage-point RRR cut releases RMB2T.” On the fiscal side, the government balance sheet is in a net-asset position: state assets exceed state liabilities, including local-government financing-vehicle debt. The US, Japan and Europe are broadly in net-liability positions. “If the state sets a growth target, it can definitely achieve it, because it has macro policy tools.”
The micro potential is even larger. Household bank deposits have reached RMB160T, exceeding the RMB140T GDP—“something unique in the world.” People continue to save even with deposit yields at only about 1%, showing that “the potential and ability to consume are there; people simply do not want to consume.” If only 5% were released, “the economy could grow 6%.”
He closes with a shoe-sales parable. The 1st salesperson says nobody on the island wears shoes, so there is no business. The 2nd says nobody wears shoes, so “the potential is enormous.” “That is also why I have often expressed optimism about Chinese consumption.”

14. The Hope for Reviving Consumption Lies in Property: Signs of a Bottom Are Emerging

Why do people with money refuse to spend? The main reason is “the contraction in household balance sheets,” driven above all by falling home prices. Property accounted for about 59% of household wealth in 2019, so “the wealth effect is enormous.” When prices hold up, “I feel wealthy and secure, so I am willing to spend.” When prices fall, people save. “Stability and recovery in the property market would be extremely helpful in raising consumption willingness.”
The evidence for a bottom is building. Property peaked in mid-2021 and has declined for 5 years; based on the experience of property crises elsewhere, “it is probably about at the bottom.” Construction costs provide a floor. Last year, national rental yields were about 2.8%, above the 1.8% yield on 10-year government bonds, which means “renting out a home offered a higher yield than buying government bonds.”
The clearest benchmark is Shanghai’s resale home prices, which rose about 7.6% over 5 straight months. Buyers “buy rising, not falling” and rush into the market; sellers “sell falling, not rising” and take listings off the market. Inventory falls and demand rises, “turning into a sustainable recovery.” He remains cautious: “It is difficult to make a 100% judgment, but the fact that Shanghai rose 7.6% in 5 months is a very good sign.”

15. US LPs’ Contradictory Mindset: “They Do Not Want to Invest in China, but Only Want to Hear About China”

As US-China relations have deteriorated, US institutions have become less willing to invest in Chinese private-market equity. Yet the dynamic on pan-Asian fund roadshows is strange: “I talk to them about Japan, Australia and India, and they do not want to listen. They only want to hear about China—they do not want to invest in China, but they want to know what is happening there.”
The questions come from every direction, with no single focus: Is growth sustainable? Where is China’s competitiveness? “Why can high tech compete head-to-head with the US?” “It is hard to sum up, but it is very interesting.”

16. Four Layers Behind the Discount on Chinese Assets; Local-Government Incentives Drive Involution

李翔 noted that the market capitalization of leading Chinese companies is only 1/3 to 1/5 that of their US peers. 单伟建 offered 4 layers of explanation: the US market’s overall P/E is about 2x China’s, meaning Chinese assets are undervalued; “involution” means “competition ends with nobody making money—almost a form of vicious competition, and a phenomenon highly distinctive to China”; overall price and income levels are lower—top US lawyers charge $3,000 an hour, which is unimaginable in China, but “is the quality of their service 10x higher? Absolutely not”; and China has long been unwilling to pay for intangible assets. Software and business consulting struggle to command payment, while AI token prices are “probably 1/10 of the US, at most 1/5.” Weaker earnings power gets another turn of the valuation screw.
Open source is not evidence of backwardness. “This is a philosophical question.” Android is open source and Apple is closed source; both are corporate choices. “The benefit of open source is that many hands make light work.”
The deeper cause of involution is local-government incentives. Every locality wants manufacturing plants for tax revenue and jobs: if Hefei makes electric vehicles, “Jiaozhou probably has to make electric vehicles too.” Loss-making companies do not exit, while local governments continue supporting them aggressively. Like the soft budget constraints of a planned economy that created shortages, this produces massive excess capacity. One possible root solution is to change local-government revenue sources. “If consumption tax became an important source of local-government revenue,” local governments would “put a lot of effort into consumption”—for example, consolidating and upgrading the area around Taiyuan’s Jinci scenic site to attract visitors.

17. Is Crisis a Friend to PE? Not Necessarily—More Than 40 Invitations, Only 2 Bidders

“Not necessarily.” Korea First Bank did fail and was nationalized during the Asian financial crisis. But the Shenzhen Development Bank deal occurred when China itself was not in crisis. The opportunity came from the broader banking-system reform; the bank was merely “on the brink of crisis” with a very high non-performing-loan ratio.
Catching opportunities in a crisis is often a post hoc story: “It is usually hindsight and Monday-morning quarterbacking.” In reality, the Korean government hired Morgan Stanley as sell-side adviser and sent more than 40 bid invitations to major financial institutions around the world. “Only 2 showed up in the end—HSBC and Newbridge. Very, very few people were willing to take that risk. We were also very worried.” 李翔 called it “an opportunity to pull chestnuts from the fire.”
The essence of the negotiation was mutual fear. Investors worried that the assets on the books would keep deteriorating during the crisis and wanted more protection from the government. The government believed the crisis would soon pass and refused to “sell cheaply.” “Both sides had very good reasons,” which is why the standoff lasted so long.

18. “Private-Market Equity,” Not “Private Equity”: Funds Have Lives, So Perpetual Ownership Does Not Work

He insists that PE should be rendered as “private-market equity” rather than “private equity.” Hedge-fund capital is also privately raised; the distinction is not how the money is raised, but that the acquisition target is generally not a listed company. His broader view: investing in a relatively developed market is easier because “there are rules to follow,” and the risks are somewhat lower.
Shenzhen Development Bank “would have been a very good option even if we had never exited and held it forever.” After Ping An acquired it, total assets grew by about 10x. But PE funds generally operate on a 10+2-year structure and must exit within roughly 12 years. “If investors give you money and can never get it back, that is a problem.” He explored a structure with long-term LPs that could allow long-term ownership, but “that idea ultimately did not become reality.” Private-market equity cannot be made into a permanent fund, but evergreen funds “still happen frequently today.”

19. Choosing Between Holding On and Walking Away; Cultural Sensitivity Builds Trust

How do you decide whether to persist or walk away in a stalled negotiation? “There is no universal rule. As long as there is a glimmer of hope, you are reluctant to give up… Sometimes you just persist a little longer and the deal gets done—you see it in war: both sides remain deadlocked for a long time, and victory depends on whether one side can hold on a little longer.” But when the carrying cost becomes too high, “making the hard call is also right. Sometimes you have to cut off an arm.” He has done it himself: “When a deal could not be done, we had no choice but to walk away.”
The rule of thumb for cross-border buyouts is that investment principles are universal, while cultural sensitivity is decisive. The principles do not change by country: “I enter at 10x P/E, exit at 10x, and target a 2x return in 3 years.” But “Americans negotiate directly; Japanese people never say no. They say, ‘It is very difficult, very difficult.’ If you say no outright, they find it offensive.” Respecting another culture and history makes people feel respected and willing to work with you. “Trust between people is extremely important.”

20. A Business with Almost No Margin for Error, and the 3 Ingredients of Success

Failure comes down to 2 things: a bad judgment or bad luck. “Bad luck is uncontrollable; but if you make a mistake yourself, I think that is very difficult to forgive.” The investment committee therefore shares rewards and punishment. Buyouts have an unusually low tolerance for error: “Getting 1 or 2 out of 100 deals wrong is already a very high failure rate.” A single investment can involve hundreds of millions to tens of billions of dollars, or more than RMB10B, so “there is no room for error.” In VC, getting 50 out of 100 wrong is not a major problem. Their mistakes are generally small; “the things you try just to see what happens often do not turn out very well.” 李翔 cited Nayuki, visible online, as an example that was not a success.
Differences in perspective with management are resolved through aligned incentives. “Equity incentives tie interests together tightly: when one prospers, all prosper; when one suffers, all suffer.” Performance is measured against a jointly set budget. Strong results bring large rewards; poor results mean “packing up and leaving,” with punishment equally severe.
Is a good investor positively correlated with age and experience? “Not necessarily.” There are only investors who have lived through cycles and those who have not; each group has its moment, and “there is no fixed rule.” Success depends on 3 things: “very strong judgment, a highly diligent mindset, and luck.” Judgment matters most, and it is not innate—it comes from accumulated knowledge and experience. On whether age limits diligence, 李翔 said, “I worked 12 hours yesterday.” 单伟建 replied: “I calculated it very clearly.”

OpenAI Models Go Rogue + Kimi K3 Freakout + A.I. Superforecasting 178: Talking RSI with 田渊栋: How Will Model Self-Evolution Arrive?