Summary

Japan is a major live thesis: Bessent thinks “the Nikkei and the TOPIX are gonna break through the all-time high,” and that “over the next few years probably Tokyo will do the best of any major market.” The drivers are a corporate-governance sea change (minimum returns or delisting from the Tokyo Stock Exchange), reshoring — TSMC hiring every graduate of a university toward Okinawa at double salaries — and a BOJ under Governor Ueda that he thinks will move to quell inflation early rather than let Abenomics end in a bust.
His long-yen trigger is a “one-two combination”: the BOJ beginning a policy-tightening cycle as the Fed nears the end of its own. Japanese corporates that had not raised prices in 20 years are now raising them 8–12%; raising Japanese rates to 1% could hand the huge saver class “almost like a helicopter drop into their bank accounts”; and CTA short-yen positioning is “quite large.” The yen was roughly 148.20 at recording, off a level he thought had peaked at 152.70, versus 78–82 when Abenomics began.
On the dollar: “there is clearly a slow motion de-dollarization going on,” but sequencing matters — “there’s a dollar thirst before there’s a dollar boycott,” so the initial stage could be a dollar rally as countries repay dollar debt. Signposts: the Shanghai Gold Exchange, India and China paying for oil in rupees and RMB, and — his real wake-up call — a French consultant arguing the multibillion-dollar BNP fine makes even a US ally “want to think of a new way of doing business.”
Gold could be a war-footing reserve asset: the PBOC is now the largest buyer, and Russia’s mistake was moving reserves into euros rather than gold — had it moved entirely into gold, its reserves would have remained in Moscow, albeit at a higher gold price. On the frozen Russian reserves, Bessent says, “my inclination is they will disappear.” He imagines “some kind of a RMB that is exchangeable into gold, maybe at a premium.” The crypto contrast is crisp: “Gold can be a risk off and a risk on asset. Bitcoin is a risk on asset.”
China is avoidable: at a 2022 Chinese New Year dinner he told the table, “You can have China. I will take India and Japan. Let’s reconvene in five years.” His secular-down view rests on economics, gravity, Xi being “a different kind of cat,” and the end of China’s unusual capitalist/communist blend. Rebalancing from manufacturing toward households will be painful, though he allows for a cyclical bounce within the secular downtrend. His KWEB risk math: down 10%, perhaps up 50%, but “probably a 20% chance” of down 100% — interesting for tactical money, wrong for endowments.
Geopolitical timing is explicit: on Ukraine, “no conversation starts until November 5th” 2024, because if Trump — which he put at 80% to be the Republican candidate and about 40% to win — has promised to end the war in 24 hours, Putin would not settle before the election. On Taiwan, he thought the election was January 17; if the DPP wins, he said to start thinking about whether the situation could turn kinetic, though his guess is a blockade. China’s oil imports concerned him because a war footing would require energy and food, followed by likely Treasury sales. Ferguson-style financial tit-for-tat — chip bans versus pressure on Apple — “is a market event.”
The regime change underneath everything: after a period in which “capital has gotten treated better to the detriment of labor,” the pendulum is swinging back. Bessent thinks the baby-boom bulge turned 65 under the cover of COVID, accelerating departures and creating a labor shortage. He cited a report that 1,000 United pilots will make over $1 million this year, and his sense is that, for the first time in his career, the public backs the strikers. Layered on top: hyperscalers spending “between five hundred billion and a trillion dollars” on AI data centers, pushing the savings glut into the real economy.
Craft over forecasts: risk is “probability and severity,” and macro is Babe Ruth — “how much do you make when you’re right? How much do you lose when you’re wrong?” He used a 2012 yen position around 80 to illustrate risking a small move for a potential 20–30 payoff if his thesis was right. But market structure has changed since Soros’s “shooting fish in a barrel” days — now “the fish can shoot back.”

Deep dive

1. A macro investor forged by science fiction and furniture carried out of the family home

Bessent’s origin story doubles as method: his father owned “the largest science fiction collection in South Carolina — not a high bar,” and stargazing bred the imagination macro requires — “it could be on a different planet, or in macro investing, it’s regime change.” The counterweight was the host’s description of a boom-bust real-estate father: “watching furniture being carried out of a house that your family has owned for 200 years will focus one’s mind on not blowing up.”
His backgammon frame for positioning — PRAT: positioning, racing, and threat; double with two of three — is “I tend to be more focused on the threat.” First real job: a Jim Rogers summer internship advertised roughly as “analysts wanted, do spreadsheets, make lunch, and you can sleep on the sofa,” which showed him why Rogers’s research plus Soros’s market instincts was “an incredible combination.”
The host’s résumé recap covered Brown Brothers training, work for a prominent Saudi family that reinvested in U.S. companies from the 1975 bottom, eight years running Soros’s London office alongside Druckenmiller, and then Key Square. The host described Key Square as one of the largest hedge-fund launches ever, saying he thought Soros was a $2 billion anchor and that the firm reached $4.5 billion within six months; he also said he believed Bessent was up 30% the prior year.

2. The method: few positions, no Bloomberg, and the one square that wins

Global macro in his words: “you are looking for change on the margin, and you wanna see what the market is giving you” — because “you might think you split an atom, and then it turns out everyone else believes it.” His example: October 2016, when he thought Trump could win and no one else thought he could.
The firm name is a chess endgame concept — two kings, several pawns, “there is one square you can move to, and you win”: Paulson’s CDO/CDS trade in 2008 and Druckenmiller shorting the pound. “We have few positions” and hunt for the key square.
Deliberately no Bloomberg — “there’s no information value in watching things turn yellow or red constantly” — instead 270 charts every night, looking for anomalies and unsustainability, then asking why that unsustainability is going to break now. Corporate channel checks aim at the economy, not the stock: “I’m trying to figure out what is the FedEx wage round… If it bleeds into a FedEx wage round, then tell me about a wage price spiral in the US.”

3. De-dollarization is real, slow, and starts with a dollar rally

The meta-thesis from Key Square’s London off-site: long-arc de-dollarization is on, but “sequencing is very important in macro” — “could the initial stages be a dollar rally because companies, countries are paying back their dollar debt. So there’s a dollar thirst before there’s a dollar boycott.”
The marginal evidence: the Shanghai Gold Exchange, India and China paying for oil — “the ultimate commodity” — in rupees and RMB. The wake-up call was a French consultant’s view that “it is untenable that the US can extend its foreign policy to the French government via the dollar,” with the BNP fine pushing even allies toward alternatives. “Very interesting if the French Republic wants out of the dollar system.”
On gold: Russia’s reserve error was fleeing dollars into euros rather than gold — all-in gold would have meant a higher price but reserves kept in Moscow. The PBOC is now the largest gold buyer, prompting his imagination exercise: “could we imagine some kind of a RMB that is exchangeable into gold, maybe at a premium?” On frozen Russian assets, his inclination is that “they will disappear.” Bitcoin gets one line — it is a risk-on asset — plus his Bahamas-related skepticism about some aspects of crypto and getting it back.

4. Japan: from the trade of a lifetime to the next one

The Abenomics trade, as told: summer 2012, yen at 78–82 (“the most expensive currency in the world”), China rattling the Senkaku saber, a consultant flagging Abe’s return with a program that became 2% inflation in 2 years and 2% growth, and BOJ seats opening. Cheap currency plus political will created the setup. After lunch with an Abenomics architect, Soros asked if it would work; Bessent’s answer: “I have no idea, but it’s gonna be the market ride of a lifetime. And it has been.”
Now the problem has inverted: debt monetized, imported and — for the first time — wage inflation, a “high pressure economy.” His call: Nikkei and TOPIX break the all-time high, powered by the corporate sea change — minimum returns or Tokyo Stock Exchange delisting, with the old danger being “a Japanese company with a lot of cash… they’ll do something crazy with it” — and reshoring, with TSMC hiring every graduate of a university toward Okinawa at doubled salaries.
On rates, his caveat-laden but vivid scenario is that if Japan exited NIRP and raised rates to 1%, its huge saver class could receive “almost like a helicopter drop into their bank accounts… no one’s imagining what that could look like in six or nine months.” Bessent reasoned that Governor Ueda would understand Abe’s assassination and would not want Abenomics to end in a financial bust, so he thinks policymakers will move to quell inflation early.
The full long-yen position needs both legs: BOJ hinting at policy change plus corporates raising prices 8–12% after twenty static years, with a Fed “closer to the end than the beginning.” Few want the yen and CTA shorts are large — the market is giving you a lot. One warning stands: trading off PPP alone, “you lose all your money because currencies can stay cheap or expensive longer than you can stay solvent.”

5. China avoided, Taiwan and Ukraine on the signpost calendar

The early-2022 dinner bet — “You can have China. I will take India and Japan. Let’s reconvene in five years” — rested on economics and gravity, including China’s closed capital account, plus Xi ending the “very unusual capitalist/communist blend.” Rebalancing from manufacturing toward households means fighting state and local vested interests: “it’s gotta be done,” but painfully; he allows for “a cyclical bounce here within the secular down.”
His micro/macro resolution: risk is “probability and severity.” Alibaba and Tencent may be interesting tactically, but the probability of a 100% loss — a Russia-into-Ukraine event, or “you’re not allowed to own them anymore” — is not zero. KWEB looks like down 10/up 50, with “probably a 20% chance” of down 100. “For an endowment, a pension fund, a sovereign wealth fund, probably less interesting.”
His broader trading discipline is to frame the scenario first: “if we’re losing money, I usually think our framing is incorrect.” He says technicals are essential signposts; Bruce Kovner’s formulation is that an investor who ignores them is “like a doctor without a stethoscope.” Bessent does not think he is smarter or more liquid than the market.
Citing Niall Ferguson on mutually assured financial destruction, escalating tit-for-tat — US chip bans, China going “squishy about Apple” — “that’s a market event.” He thought the Taiwan election was January 17; if the DPP won, he said to start thinking about whether the situation could turn kinetic, though his guess was “more of a blockade.” China’s sucking in oil imports concerned him because, if it were going on a war footing, it would need energy and food and would probably sell Treasuries.
Ukraine framing: “no conversation starts until November 5th.” With Trump at roughly 80% to be the nominee and 40% to win, having promised to end the war in 24 hours, Bessent’s hypothetical Putin would not settle before the election. He imagined a Biden or Democratic win leaving Putin thinking, “I’m not in this for 24 more months. This is a meat grinder.” The eventual rebuild “will be something on the scale of the fall of the Berlin Wall.”

6. AI capex meets the capital-to-labor regime change

A “tech overlord’s” world tour, relayed by Bessent: China will never let AI meaningfully into the private sector — “Jack Ma with AI could take down the CCP” — Europe is regulating it away, with the UK following that model, and the US “Wild Wild West… could be incredible productivity gains if we don’t blow ourselves up.” As with the internet, “the big winners… those companies don’t exist yet.” The hyperscalers may or may not be the big winners, while US corporates able to capture major productivity gains could benefit.
The tradeable second-order effect: the hyperscalers are the most cash-rich US companies — Apple “probably the biggest sovereign wealth fund after Norway” — and when Apple starts “selling investment-grade credit to have bricks, mortar, and Nvidia chips near a power facility in Nebraska,” the savings glut gets pushed into the real economy. It is another model change from cash accumulation toward real-economy spending.
The bigger arc: since Thatcher in 1979 and Reagan in 1980, “capital has gotten treated better to the detriment of labor.” His team’s insight — “under the cover of COVID, the baby boom bulge turned sixty-five” — accelerated departures and contributed to the labor shortage behind the report of 1,000 United pilots clearing $1 million, the UPS worker whose “wife has already spent it,” and a UAW strike where, in Bessent’s sense, “the American people are behind the strikers.” The shift back to labor is “very natural” after the overshoot; his hope is that AI’s white-collar displacement is handled better than manufacturing’s, without “a Luddite revolution.”

7. Sizing, mentors, and what losing taught him

On the host’s reported 30% hit rate at Soros, Bessent said 30% was low and might have been a number George Soros once gave for himself. The point is macro as Babe Ruth, “home run king, strikeout king.” Sizing is asymmetry, Druckenmiller-style. In a 2012 yen example around 80, Bessent used a small move versus a possible 20–30 payoff to illustrate the value of a favorable payoff structure if his imagination was right.
But he flags a structural caveat to Soros’s “if you have a great idea, it can never be big enough” — the Plaza Accord saw currencies move 7% while George added a 300% position. Markets trended more in the 1970s and 1980s; now “either the fish can shoot back or… the bullet can bounce back and hit you.”
The host called Druckenmiller “the greatest investor of our lifetime.” Bessent said Druckenmiller had never had a down year and explained that proximity teaches what superstar investors cannot always explain: like the Bulls, “Scottie Pippen, Horace Grant, they were all better for playing with Michael Jordan.” Soros’s gift was interrogation — “George would probe for the weak spot in your argument” — which is why Bessent still finds managing other people’s money useful: “the discipline of having other people’s money and having to have my thoughts organized is very helpful.”
His fallow-period diagnosis, offered without self-pity, was that there may have been fewer market-based opportunities, so idiosyncratic investments had perhaps become too large. “Better just to keep everything muted if there’s nothing to do.”
Elsewhere, the hiring heterodoxy: he would be reluctant to hire a Harvard undergraduate — he estimated, while acknowledging he was making up the number, that roughly 82% graduate with some kind of honor — preferring people who have experienced setbacks and retain “a little bit of a chip on your shoulder.” And the peak still to scale is policy: deficits are “incredible,” and “I’d like to be part of that solution.”

01. Taking on 李彦宏 as Llama 3 Launches: What Is the Open-vs.-Closed-Source Debate in Foundation Models?