China August economic data: Retail sales (a nominal figure): 0.4% yoy YTD fixed asset investment (a real figure): -7.2%. Our domestic demand tracker has fallen to levels even below that during the Shanghai COVID-19 lockdown. “What is unsustainable will not be sustained.”

After reading this, I can’t shake the feeling that we may already be living in a Three-Body world—one in which humanity is gradually, and perhaps inevitably, dividing itself into three camps: the Adventists, the Redemptionists, and the Survivors. The Adventists want the Trisolarans to come. They see the arrival of a superior intelligence not as an existential threat, but as a potential replacement for a civilization they believe has reached its limits. The Redemptionists also believe the Trisolarans can save us—but they hope that the two civilizations can coexist, with superior intelligence serving as a means of transcending the flaws of human civilization rather than replacing it. The Survivors, by contrast, ask a much more primitive question: How do we survive the Trisolarans? For them, the preservation of human agency and human civilization takes precedence over the promise of whatever the future intelligence might bring. What makes the analogy increasingly unsettling is the rapid emergence of superintelligence. Perhaps we have already begun to fall, consciously or unconsciously, into these three camps. Some are eagerly anticipating the arrival of something greater than ourselves. Some believe AI will ultimately redeem humanity from many of the constraints that have defined our civilization. And some are increasingly preoccupied with a more fundamental question: whether humans will remain the ones making the decisions in the world we are creating. The most interesting question, then, may not be whether superintelligence will transform human civilization. That increasingly seems inevitable. The deeper question is which of these three visions of the future will ultimately prevail—and what happens when the people holding them begin competing for control over the intelligence that will shape that future? Perhaps the future of AI will be determined not simply by how intelligent machines become, but by the relative dynamics and power of the Adventists, the Redemptionists, and the Survivors among us.

This observation is in line with what I recently wrote in our report: "The problem is that China's high-tech sectors remain a relatively small part of the overall economy and have a high barrier to entry. The vast majority of the population therefore remains largely excluded from this new growth cycle." "For ordinary households working in traditional manufacturing and service industries—restaurants, transportation, construction, retail, and countless other parts of the domestic economy—the reality remains very different. These sectors continue to face deflation, shrinking margins, weak income growth, and rising bankruptcies. In other words, while the upper end of the K is enjoying an extraordinary boom, the lower end—which represents the overwhelming majority of the population—is dealing with a recession." "This is the fundamental political problem created by China's K-shaped development model: the gains are highly concentrated, while the pain is widely distributed. Ultimately, we believe the social backlash from the lower end of the K is likely to become increasingly difficult for Beijing to ignore."

Macro investors make the most money by betting against major trends when they become unsustainable. If you ask me which major macro trend looks the most unsustainable today, I would not point to rising US long-end yields or the BOJ continuing to stay behind the curve. I would point instead to the chart below: Chinese real investment growth has collapsed to a staggering -7%—a historical low outside of the initial COVID-19 shock. In my view, this is arguably the most unsustainable macro trend in the world today—and one that investors should pay very close attention to, and potentially position for a reversal, in the coming quarters.

The revaluation strategy backed by the threat of tariffs has already failed spectacularly. In my view, the decisive reason was that the Trump administration folded first in its trade war with Beijing, with American stores at one point essentially running out of goods. As a Beijing policy advisor told me a year ago, “We missed the best chance to shift China toward a more consumption-oriented growth model because the US proved too weak (怂).” Now that China has successfully weathered what was arguably the most aggressive trade war with the US—and its trade surplus continues to surge— I see little reason to believe that tariff threats can force Beijing to rebalance its economy. If anything, the experience may have reinforced the opposite lesson in Beijing: the US lacks the leverage to force China to change its economic model.

The year 2026 marks the official beginning of Chinese households’ deleveraging cycle. The definitive end of the household debt supercycle lies at the heart of the secular weakness in consumption—and, by extension, the prolonged decline in consumer-oriented equities.

Exactly. When Barry Eichengreen—the leading authority on the international monetary system and reserve currencies—raises concerns this serious about the US dollar's reserve status, we should pay close attention. The dollar's reserve status rests on a simple but fundamental assumption: foreign central banks are free to deploy their reserves as they see fit, especially by selling US Treasuries when circumstances require. If that assumption no longer holds because reserve managers fear political pressure from Washington, then the rationale for holding large dollar reserves inevitably begins to erode. Yes, the BOJ can access the Fed's FIMA Repo Facility. But what about those central banks that cannot? More importantly, should reserve managers have to rely on the goodwill of the issuer simply to access the liquidity of their own reserve assets? The US government's apparent anxiety over foreign central banks selling Treasuries may itself pose one of the greatest question marks on the dollar's reserve status.

Interesting chart comparing the ongoing KOSPI meltdown with China's 2015 stock market crash. The selloff in Korean equities appears even more severe at this stage, perhaps suggesting that a near-term bottom may be coming?

The success of Kimi-K3 once again suggests that the moat surrounding frontier AI labs may be far weaker than many investors believe. The rapid commoditization of AI services—driven by the swift catch-up of Chinese AI labs—is posing an increasingly significant challenge to the valuation narrative underpinning US frontier AI companies.

This post reminds me of a report I worte in late 2023, titled "China’s Oil Demand Has Peaked." The main takeaways from that report were: 1. China’s oil demand may have peaked on a secular basis. 2. The total number of internal combustion engine (ICE) vehicles currently on the road has declined for the first time in history, indicating a peak in gasoline demand. 3. Over the next five years, China’s annual gasoline demand may decline by an average of 35 million barrels, or 96K bbl/d. 4. Gasoline demand destruction will accelerate in 2027. As such, I suspect China's incentive to restock crude oil is much weaker than many investors anticipate, particularly given the recent deterioration in domestic demand and Beijing's continued reluctance to step up stimulus.

People tend to champion ideas like economic globalization when they are the ones benefiting from them. But when the tide turns and they begin to feel threatened, they almost invariably change their mind. At the end of the day, very few people are truly unafraid of competition.

China’s retail sales — a nominal figure — grew just 0.2% in April, marking the weakest reading on record outside the Covid era. Investor pessimism about consumption has become deeply entrenched: the underperformance of consumer-oriented sectors relative to the broader market