Weekend Food For Thought WFFT
On Today's Menu: USA vs China, Top 15 Economies, The Multi-Anchor Currency Regime, Deep Dive into Korean Tech, Keys to Learning, and much more topics..
Hello from Lisboa,
I hope you had an interesting and productive week.
Claude Sonnet states that:”Intelligence failures are almost never failures of information alone. They are failures of system design — broken feedback loops, missing connections, rigid mental models, organizational silos, and insufficient diversity of thought.”
Take Note…Go explore the world with flexible systems level thinking…
1 Getting Visual
2 If You Read One Thing Today - Make Sure it is This
3 Consequential Thinking about Consequential Matters
4 Big Ideas
5 Big thinking
6 The Purpose of Knowledge
1 Getting Visual
The Big Picture: Change is the Only Constant - Exhibit 1: Twenty-five years ago, the United States was the world’s dominant trading power. Today, China has overtaken America as the top goods trading partner for most countries globally. n 2000, only 33 countries traded more with China than the United States. By 2025, China had become the top goods trading partner for most countries worldwide.
The Big Picture: Change is the Only Constant - Exhibit 2: How Economic Power Shifted in the Last 10 Years - The global economic order has shifted dramatically over the last decade, with countries reshuffling positions amid inflation shocks, geopolitical tensions, pandemic disruptions, and the rapid rise of AI-driven industries.The U.S. remains the world’s largest economy at $32.4 trillion in 2026 forecasts, while China crossed the $20 trillion mark. India posted one of the fastest growth rates among major economies, while Japan became the only G20 economy to shrink over the decade. The period from 2016 to 2026 saw major reordering among the world’s top economies, with Mexico overtaking Spain, India overtaking France, and Russia leapfrogging both Brazil and Canada.
The Big Picture: Change is the Only Constant - Exhibit 3: The changing face of the US economy and it’s top ten companies…
Spotlight: The US Government Finances Are Not Ready for a Recession: “During US recessions, the budget deficit typically widens by around 4% of GDP as unemployment benefits surge and tax revenues collapse. That would be manageable if the US were entering a potential recession from a position of fiscal strength. It is not. In fact, the US has never entered a recession with this little fiscal buffer. The investment implication is clear: do not expect lower interest rates to bail out valuations. The standard recession playbook that growth slows, the Fed cuts, rates fall and multiples expand breaks down when the sovereign borrower is already stretched. In the front end, inflation driven by higher energy prices, tariffs and immigration restrictions is proving stickier than the Fed expected, constraining how aggressively it can cut. At the long end, the fiscal trajectory is structurally bearish for bonds. Treasury is already funding record deficits almost entirely through T-bills to avoid putting upward pressure on long yields, a strategy that cannot continue indefinitely. When coupon issuance eventually has to increase, the supply shock will push long yields higher, not lower. And in a recession, the deficit blows out further, requiring even more issuance at precisely the moment when market appetite for duration is most uncertain. The bottom line is that rates are staying higher for longer across the curve, and the traditional path to value creation through multiple expansion is largely closed. Value will have to come from the hard work of operational improvement, i.e., earnings growth, margin expansion and cash generation, and not from the discount rate doing investors a favor.” - Apollo Research
Spotlight: Potential Circle Jerk Economics at Massive Scale: “The AI “hyperscalers” reported bumper earnings for the first quarter, with both sales and profits beating expectations. But if you look more closely at the P&L, some of them had a lot of help from an interesting line item. Alphabet, for example, booked $37.7bn of “other income” in just the first three months of the year, accounting for over half the company’s net income over the period. Accounting nerds and other clever readers will probably have guessed from the pattern what constitutes “other income” in this case: the ebb and (mostly) flow in the valuations of their sizeable private investments in companies like OpenAI and Anthropic. Alphabet is the biggest investor in Anthropic, and Amazon is one of the biggest. Thanks to their slugs of money, Anthropic’s valuation has vaulted from $183bn in September to $380bn today, according to PitchBook (and there’s another mammoth funding round in the works right now). This has then allowed Alphabet and Amazon to mark up the value of their existing stakes. Amazon reported “other income” of nearly $16bn in the first quarter, up from $2.7bn in the same period last year. That was nearly half its overall net income for the three months. As Goldman Sachs analysts said in a note last week: “The hyperscalers’ earnings growth this quarter was boosted by an unusually large contribution from equity stakes in private companies. Alphabet and Amazon generated “other income” totalling $53 billion in Q1 2026, which accounted for nearly 60% of those two companies’ income in Q1 and 34% of the total $155 billion in income this quarter across the five largest hyperscalers. This represents the group’s largest collective share of earnings attributable to “other income” in at least a decade. Of this $53 billion in “other income,” $49 billion was explicitly due to equity stakes in private companies.” This is another sign of just how comically codependent the AI tech industry has become. Not only have private investments and increasingly engorged funding rounds become a meaningful driver of the hyperscalers’ aggregate earnings, but the money the hyperscalers have pumped into the likes of Anthropic and OpenAI has allowed the AI companies to sign huge computing deals with Alphabet’s Google Cloud, Microsoft’s Azure and Amazon Web Services. In fact, The Information has crunched the numbers, and OpenAI and Anthropic now make up about half of the entire cloud computing order books at Oracle, Alphabet, Amazon and Microsoft.” - Robin Wigglesworth on the alarming circularity of the AI-big tech boom.
Spotlight - US Data Center Buildout Trends: “Signs of a potential slowdown ahead Data center construction continues to rise, accompanied by rising spending on power generation capacity. But is there a potential slowdown ahead? The latest analysis based on satellite images shows that over 60% of data center capacity planned for completion in 2027 has not begun construction with another 7% delayed. The culprits are typically related to permitting issues and delays in obtaining gas turbines, transformers and skilled labor. Of all 47 categories in the US PPI report, transformers and power regulators have seen the second highest level of inflation since 2020. During this period, the wait time for generation step-up transformers has roughly tripled, as shown in the fourth chart (these are transformers used to prepare power from large generation sources for higher voltage transmission grids). And on natural gas turbines, GE Vernova bookings have risen from $103 billion three years ago to management’s projection of $200 billion by 2028.” - JPM Research
Spotlight: Innovative Leaps Tend to Happen Against Constraints - China’s AI Build: “The export controls have become capability-generating – labs in China are forced to be ruthlessly efficient. Despite the three-year compute handicap, Chinese open-source models are only six to eight months behind the US frontier. (…) It’s important to separate compute capacity for training AI models from serving customers. China has a huge AI market. Doubao alone reaches 100 million daily active users. Token volumes are equally vast. By February 2026, we estimate Chinese token volumes had reached ~9 quadrillion tokens a month – compared to ~4 quadrillion across the main US/Western providers. Alongside datacenters in Malaysia and Singapore, a large part of Chinese compute infrastructure is going to serve customers through inference. If half of the compute is used to serve those customers, that reduces the available compute for training models. We might conclude, with low confidence, that by the end of 2025, Chinese labs had as much compute available for model training as American labs did in mid-2023. By that logic, the performance of models from Chinese labs should be at least two years behind American if labs in both countries are using the same approach – more computing and more data to build better models. The framework treats capability as a function of compute, holding training efficiency roughly constant. But we aren’t seeing a 2-3 year gap. The headline is that Chinese models are three to six months behind the US on benchmark performance, according to DeepSeek, and 8 months according to the Center for AI Standards and Innovation, a US government agency. In fact, Chinese labs appear to be keeping pace with, or perhaps even narrowing the gap in some ways, with US labs. The question for us then became: are the capability headlines wrong, or is something closing the gap that the computed numbers don’t capture? There is an additional wrinkle: market structure. In the US, five key frontier labs dominate training compute: OpenAI, Anthropic, Google DeepMind, Meta and xAI. In China, a thousand flowers are blooming. (…)The efficiency moat strikes back - These labs are clearly finding efficiencies in training performant models. The efficiencies are actually being passed through to the models’ inference. That is, when they’re being used to serve customers, because they are much cheaper than roughly equivalent American models. One shouldn’t put too much weight on AI benchmarks. They can be gamed, and they might not easily reflect how a models “feels” or works in practice. But they are one inadequate reference point. DeepSeek’s V4 Pro, their flagship model, is comparable to Claude’s Opus 4.6, in some ways. Opus 4.6 was released in Feb 2026, and is not Anthropic’s latest model. Cost-wise, though, you can see the difference. DeepSeek charges $0.43 per million input tokens and $0.87 per million output tokens. Opus 4.6 is 11 times more costly in input and 28 times more expensive in output. These are not promotional one-offs. Across the Chinese frontier, Kimi K2.6 sits at $0.95 input (among the cheapest models in the global top 10 by GPQA Diamond), and Alibaba’s Qwen models are priced in a similar band. The cost-to-serve inference is a function of three factors: the actual cost of serving the model, its compute complexity and energy costs, and the margin the provider is willing to give up. The margins appear largely healthy. Z.ai serves its GLM-5 model at $1.00 per million input tokens, which is 3x cheaper than Claude Sonnet 4.6, and 5x cheaper on output. Despite this, it boasts a 50% gross margin, and MiniMax enterprise margins sit at 70%, though we don’t know if this holds across the board. DeepSeek, for its part, ran for years on internal funding alone, only turning to outside capital this month. Finally, that efficiency shows up in how easily these models run on consumer hardware like laptops and phones. The leading local models in the world are almost all Chinese open-source, aggressively distilled down to smaller, lighter variants. A 5GB Qwen3-8B model runs on my Mac, as does DeepSeek R1’s 7B distilled variant, which has been pulled 85 million times on Ollama, the second-most-downloaded local model in the world. Cursor even built its Composer 2 model on top of MoonshotAI’s Kimi K2.5 model. The only US-based open-source model we run locally is Google’s Gemma 4. (…) Has this had a real effect? Today, we can show it has. To try to quantify it, we have used an efficiency multiplier. This is a calculation we’ve put together to compare how capable the models are compared to where they ought to be given compute constraints. We found that Chinese labs are extracting 4-7x more intelligence per unit of compute than naive scaling predictions would suggest. In time saved? This looks like 2-3 years of efficiency gains. The future shape of the AI economy - The shape of the AI compute workloads is changing. Less training, much more inference. We’re moving from development to deployment. While development won’t end, the labs will continue to train increasingly capable models. Training capabilities and inference capabilities will become more important. Inference capacity will become more important. One conclusion we drew from our discussions was that the cultural, technical, and management affordances we saw in Chinese labs may well benefit them as we transition to this new phase. (…) Heading to the edge - Over the coming few years, more and more AI will be consumed on the edge, that is, on devices closer to where the intelligence is needed. think about in robots or in autonomous cars or in devices in your home, just on your phone. Xiaomi already has an empire of 750 million devices, from thermostats to cars, that are starting to integrate AI into daily life. Obviously, you can’t run a titanic frontier model on an Edge device. Today, it takes about 6 to 8 months for Frontier Model Performance to become available as open-source and open-weight models. and from there, roughly another 6-12 months for techniques like distillation to shrink the models down enough to run on consumer devices, like laptops and phones. Most of the models that do this are the Chinese open source models: Qwen, DeepSeek and MiMo. Models built for scarce compute are already shaped for this emerging environment. And, at some point, those edge devices may be robots. Chinese firms are already shipping: Galbot’s humanoids are running autonomously in warehouses and in specially purposed pharmacies. (…( The efficiency moat - Chinese labs may be building a competitive moat, one that is designed around these principles of efficiency. Good moats last. And we hypothesise that Chinese labs may have be building an efficiency moat, where they can consistently train competitive models at much lower cost than rivals and then serve these models at a much lower cost per token. The practices of parsimony are leading to leaner but competitive models. Could this lead to long-term advantages? Possibly: if the market evolves to greater inference on heterogenous infrastructure and a more diverse set of customers.(…) The elastic effect - Export controls were designed to freeze China out of frontier AI by choking off access to high-end compute. It is undeniable that they have had a serious impact. But the constraints have fostered new capabilities built around efficiency. The labs we visited fostered a culture of compounding research. They’re open, selective about what works, and are optimising relentlessly in a way that fits the shape of the future. It’s hard to know whether they would give up those practices if a spigot of GPUs magically opened up for them. Of course, they’d likely lean into that new capacity, but I would wager they wouldn’t give up the unique characteristics that have kept them competitive until now. The difference is how that compute is used. In 2023, most American capacity was tied up in training, not serving customers. By contrast, in 2025, China’s compute stack, augmented by data centers in Malaysia andSingapore, was doing double duty – supporting model training and serving hundreds of millions of consumers, and a rapidly growing base of enterprises, through apps like WeChat, Doubao and Alipay.” - Exponential View, China Field Trip Notes
Spotlight: A look at the machines that turn sand into intelligence: “In 2004, a leading-edge lithography tool cost around $50 million and delivered roughly 85 million transistors per dollar of wafer cost. By 2023, after nearly two decades of sustained engineering progress through the EUV transition, the tool cost about $185 million but delivered 815 million transistors per dollar. The deal held and each generation of more expensive tooling bought substantially more transistor density per dollar spent. In 2025, High-NA EUV entered production. It cost $380 million per tool, delivering 562 million transistors per dollar. Less than 2023 and 2021. The 2026 node projects to 412 million. If it holds, it will be the first sustained decrease in transistors-per-wafer-dollar since EUV restored the curve in 2019. The deal – the specific economic bargain that made the digital economy possible – is bending. ASML’s challenge is that it’s running out of levers to shrink chip details, and every bit of progress costs dramatically more than the last. There’s a basic rule of chipmaking with light (the Rayleigh criterion) that says how small a line you can draw depends on three things: the color of light (its wavelength), the shape of the lenses and mirrors that focus it (numerical aperture) and the patterning tricks – computational lithography, masks, materials – that warp the pattern so it prints more cleanly on the wafer (process factor). All three are hitting hard limits. From the fab perspective, ASML’s machines still look like extraordinary investments, even as the wider economics deteriorate. TSMC and Samsung use each new machine to generate daily wafer revenue that has broadly kept pace with the tool’s growing price, cutting the payback period from more than forty days in the mid-2000s to less than a week for projected A16 chip production. Inside the fab, each tool pays back faster than ever – because a small group of customers will pay whatever it costs to stay at the frontier. Across the system, the bargain is bending: the cost reductions that used to come from the curve itself are now passed downstream to chip designers, to hyperscalers, to anyone buying compute. That fab-level economics rests on a narrow assumption: three customers willing to pay frontier prices for declining density gains. It has held so far. Whether it holds through a $500 million successor tool is the open question – and TSMC’s delay of High-NA is the first signal that the spreadsheet is weighing heavily.” - Exponential View
2 If You Read One Thing Today - Make Sure it is This
Kevin Xu explores the “ChinAI mood” from his recent field trip to China - read it here in full for some compare and contrast on the trends in China and the US - Do it here:
Some Takeaways
“Updating priors” has become one of those tropey phrases that more and more people say, like “thinking from first principles”. Poker bros started it. Finance bros popularized it. Now tech bros, who are all by default LLM bros, are spreading it more, if only to signal that they know at least something about how backpropagation works.
I am as guilty of doing this as any.
But “updating priors” is a good principle to live by, even though in practice, it requires energy, humility, and commitment to finding the “ground truth” (another tropey phrase). It is especially hard to pull off on topics that other people think you know something about. That’s how I feel about AI in China, or “ChinAI”, as Jeffrey Ding’s inimitable and namesake newsletter calls it. “ChinAI” is a subject matter I treat with much humility, because China is moving fast, AI is also moving fast, and combined is moving faster than the sum of its parts. (Timestamping the title of a post on this subject is the least I can do.)
I recently had the good fortune to exercise this humility with a group of sharp, smart, and open-minded AI writers and researchers from the US and the UK. Some are China rookies, others are China veterans, I put myself in the middle of that spectrum. We spent nine days together, traversed four cities, and interacted with as many Chinese AI researchers from leading labs on the “ground” to discover some “truth”.
“Over the course of our trip, we chatted with researchers from almost all the leading AI labs in China – roughly 85% of that universe by my own estimation. Some were done in meetings inside large conference rooms that made us look like foreign dignitaries. Some happened more informally over lunches with Domino’s Pizza or dinners around a lazy susan. And some were quick chats in hotel lobbies, in order to fit something in between our hectic travel schedule and their more hectic work schedule. They’ve all got AGIs to build!”
“Our group’s intended focus was to talk to core researchers. And we got our wish for the most part. We met with many talented researchers, many of whom held the vaulted title of “interns”. This is for good reasons, because most of them are still PhD students. Except these internships are one year, sometimes two years long. They are also treated as full-time employees with full data, compute, and systems access to generate ideas and run experiments.
These “cracked interns” are the workhorses of China’s leading labs. The average age is in the mid-20s – whip smart computer scientists born around or after the turn of the century. They are terminally online (especially on X/Twitter) and can converse in English without much struggle on technical topics. Most of them hail from Chinese universities and have not studied abroad.
This stands in stark contrast to the lack of internship opportunities offered by leading western AI labs and the lack of “real work” an intern is even allowed to do if offered a spot, an observation that Nathan Lambert (and fellow traveler) shared in one of his posts about the trip. When we asked the labs about the underlying reasons, the explanation was ruthlessly practical.
For one, the labs don’t see more senior people, be it professors or PhD graduates with deep industry experience, as a good source of research talent, because the field of LLM and generative AI is too new and moving too fast. There is a premium placed on raw brain power and fresh minds swirling with new ideas from first principles (oops, I did it again). Less weight is assigned to track record and experience in adjacent AI disciplines. More than one lab has told me that going to academic conferences lately to scout for ideas has been a waste of time. The typical conference’s submission and admission schedule is way too slow to reflect the latest frontier AI research.
For another, coming from the PhD advisors’ point of view, there are simply not enough compute resources inside any university to allow the ideas of talented students to flourish and publish. Thus, seconding their students inside the labs equipped with more compute, while knowing the labs would allow their students to publish papers in partnership with their universities, is a win-win proposition.”
“There is a symbiotic chemistry and understanding between the academic institutions and the AI labs – the former filters for raw talent, the latter makes good use of them.
Strategically, this is an important observation, as this symbiosis may pave the way for a stronger homegrown AI talent pipeline in the long run.”
No Philosopher Kings
Among these twenty-something “cracked interns”, we found few who did much philosophizing about the societal implications and ramifications of what they were building day and night. When we asked them how they define AGI, more than a few researchers from different labs coincidentally answered in exactly the same way: “AGI is when the AI can replace me!”
There wasn’t much hint of worry or melancholy in the tone of their answer either. Instead of fearing to be replaced, they sparkled in wonderment as to whether a machine can indeed be built to be more capable than its maker. And if that is accomplished one day, then they would happily move on to doing something else.
This is another contrast to their western counterparts, many of whom actively worry about AI’s safety or societal implications privately. They also spent time philosophizing about these issues on X, on podcasts, or even engaging with lawmakers to lobby for their preferred policy outcomes. There are no equivalent “philosopher kings” in the Chinese AI ecosystem.
It is instead populated by researchers who prefer to focus on just building the tech with, as fellow traveler Afra Wangelegantly coined, “monastic intensity”.”
Less Compute But Still More Research
By cards dealt, I don’t necessarily mean the system of government they are born into. Compute constraint, or lack of GPU cards (yes those cards), made irreparably worse by US export control is the most immediate obstacle to scale.
This US-policy-led compute constraint and its impact is all too predictable and should not surprise anyone. I foresaw this outcome 18 months ago when it was obvious then as it is now that compute scaling is nowhere near hitting a wall. Don’t let anyone convince you otherwise. Most of these labs are still running on the NVIDIA chips they were able to procure before the “yard” got bigger. Some are renting capacity from different hyperscalers, which is still legal and above board. Some, perhaps, are making use of smuggled chips, though we did not challenge our hosts on this front when the fact patterns have been laid bare (we had no interest in being naughty guests).
What was more telling was how Chinese labs were making use of their limited resources to still support core R&D to improve model intelligence or increase efficiency and reduce cost. Some labs are splitting their total compute resources by a 3:1:1 ratio – as in 3 GPUs for research, 1 for pre-training, 1 for post-training. Others privately shared that they allocate about half of their total resources to research, the rest to other types of workloads.
The precise proportion is less important than the direction of travel.
Devoting more resources to research and experimentation makes sense, especially given the compute constraints, when yolo’ing on a big training run is infinitely more costly. Spending considerable compute to try, test, toss aside, or incorporate ideas that work, however small and incremental, increases the odds of success of an eventual large-scale training run, of which most Chinese labs simply can’t afford to run too many. Since core research tends to consume most of the resources to build any new model, guarding precious GPUs for research is the smart strategy given the constraints.
During one of our meetings, in the spirit of reciprocity after pumping our host for information for close to two hours, we shared with the Chinese researchers the rough GPUs per researcher ratio inside OpenAI. Their jaws literally dropped when they heard the number. Yet, we all know the OpenAI researchers, or every researcher in every western AI lab under the sun, would still complain about how little compute they have.
In so many ways, AI researchers in Silicon Valley don’t know how good they have it.
Meanwhile, their Chinese counterparts are forging ahead with what little resources they have and playing for the best outcome with the cards they are dealt.”
“During one of our meetings, when the question of open source came up, the leader of the lab in his most tiger-dad-like mannerism voiced his commitment to open source, because he wanted to keep his researchers honest, while his researchers looked on stoically. (“You said your model is good? Open source it and prove it to the community!”) In other conversations, the one lab that researchers feared the most from a competitive lens is Seed from ByteDance – ironically the one lab that is proudly closed source.
The diversity in opinion and execution in this open vs closed spectrum, I suspect, will only widen as this year chugs along. The divergence I witnessed on this trip is just the beginning.”
“The way each lab carried itself in front of our motley crew of AI writers and researchers revealed this personality split in subtle but meaningful ways. What intrigues me is that regardless of whether a lab chooses to be more “Western” or more “Chinese”, the aim isn’t just about winning the home market, but also (and always) to expand overseas. And both personalities could work!
Why a “Western” vibe would work in a “going global” strategy is rather obvious. What isn’t obvious is why a “Chinese” personality could also work, especially when you are going after the wallets of old, legacy enterprises in the Global South. One of the labs, who embodied this “Chinese” identity, shared with us that they are doing business with a large chicken and eggs farm in a Global South country. To earn more of these kinds of businesses, they go directly to the CEOs of backwater yet large industries to push their AI product from the top-down – choosing to go where their fancier Silicon Valley counterparts are too high-browed to touch.
CEOs and shotcallers of most ilk actually love an entourage shepherding him around a shiny showroom, then getting wined, dined, and pampered before signing a purchase order. The tech native bosses, who hate talking to sales and marketing people are the exceptions, not the norm, in most business dealings.
That’s why even OpenAI is now corralling a group of private equity funds, consulting firms, and system integrators to help with the unglamorous push for enterprise adoption with a so-called Deployment Company.
Top-down force-feeding still works like a charm in AI. And what could sound more stereotypically “Chinese” than that!”
“As the world barrels towards a Trump-Xi summit this week, potentially to be followed by three more such occasions in the remainder of this year, it is hard to not feel a touch of anxiety over how these two AI superpowers would manage the risks and uncertainties of an inherently uncertain, probabilistic piece of technology.
Yet, if you practice mainstream consensus cleansing and narrative detoxing, as I regularly try to do, then things really aren’t that bad. Hanging out with researchers and chitchatting with the people doing the work is a good way to do this. You pick up on the humanity of the technology more rather than over-index on the abstraction of its hypothetical power.
That humanity comes through in seeing harsh realities. One researcher had a running fever yet felt obligated to entertain and engage with us for two days, while others worked overtime during a national holiday and likely slept on cots in the office hallway. It also comes through in joking asides and self-deprecating humor. One researcher from Moonshot AI half-seriously told us that AGI is the “friends you make along the way”. While inside Z.ai’s shiny showroom next to its reception desk, one display screen told all visitors in no uncertain terms that its journey to AGI is currently, and always will be, at 42%. (This is in reference to the number “42” in the Hitchhiker’s Guide to the Galaxy that is a familiar cult meme among all techies.)
The United States of America and the People’s Republic of China – as two abstractions of national identity, governing ideology, and technological modernity – may never become friends in my lifetime.
But on this trip at least, I witnessed friendships born and mutual respect formed between ordinary researchers, who subscribe to these two different identities, comply with these two different ideologies, yet still want to build and participate in a modernity that AI empowers, together.
As a humanist at heart, that is good enough for me.”
3 Consequential Thinking about Consequential Matters
Marco Papic and Mathieu Savary explores “the multi-anchor currency regime” and it’s potential implications for global asset markets in their recent BCA report - Go explore it in full here - Plenty of Consequential Thinking about Consequential Matters…
https://media.licdn.com/dms/document/media/v2/D4E10AQHcnXcevnx03A/ads-document-pdf-analyzed/B4EZ4DKZHYJoAQ-/0/1778169518818/bcagmac_br_2026_04_21pdf?e=1779386400&v=beta&t=nrnkhglPYSxbISqxr59BL2wmAqSv3bDXc4Ka3n0Dp3I
Some Takeaways
“Top Takeaway: The dollar’s reign is not ending. Its exclusivity is, and the investment implications are large. The international monetary system is shifting from a dollar-dominated framework toward a multi-anchor regime. This transition is structural and it is already repricing assets. Investors who wait for consensus will be late.
The dollar retains its dominance but is losing the monopoly on reserve functions.
Store of value, transaction utility, and funding currency are decoupling and migrating to different anchors at different speeds.
No single currency replaces the dollar. The question is which portfolio of assets replicates its functions.
Such a synthetic basket is already being constructed, implicitly, by the world’s most sophisticated reserve managers.
US term premia normalize upward as price-insensitive central bank reserve flows fragment across various currencies. European fixed income is a key beneficiary.
US equity valuations face a structural headwind as dollar seigniorage erodes. European equities face the opposite dynamic as capital market integration deepens.
Cross-asset correlations decline as dollar funding dominance attenuates. Portfolios are under-diversified for the world that is emerging.
Real assets, gold, and commodity-linked currencies are winners as they attract structural demand from multiple directions.”
In February 2022, the United States and its allies froze approximately $300 billion in Russian central bank reserves. The decision was unprecedented, geopolitically consequential, and financially instructive. It demonstrated that reserve assets held in the currencies of geopolitical adversaries are not safe assets in any unconditional sense. They are contingent claims: available until they are not.
The response was a systematic reassessment of what reserve safety means. Central bank gold purchases surged to record levels in 2022 and have remained elevated. Bilateral currency swap arrangements proliferated. The dim sum bond market in Hong Kong began its most sustained expansion in a decade. Individually, none of these constitutes a challenge to dollar dominance. Collectively, they mark the start of a structural shift that will extend well beyond the Ukraine war.”
What investors – but also policymakers, commentators, and regular folks living through this era – are witnessing is how chaotic and messy multipolarity really is.
However, it is also the norm for organizing states in the anarchic global system. To the surprise of most of our clients, a multipolar order has been the dominant geopolitical context since the late eighteenth century. It is only the recency bias of the Cold War and American hegemony that clouds the analytical judgment of most investors.
“Our assertion is that the world is multipolar and will remain so for quite some time.
By pivoting away from Magnanimity and towards Machiavellianism, the US has become “one of the powers,” albeit still the strongest one. It used its currency as a weapon against Russia – baring its teeth in a way that it had not since the advent of the American hegemonic order. Since then, the US has threatened to tax foreign owners of liquid capital, raised tariffs against partners while making trade deals with rivals, defended one ally militarily by threatening the world’s global oil supply, abrogated its responsibility to protect global shipping lanes, and questioned its almost century-old alliances. America has, in other words, behaved as any other country would. Not a hegemon. Not an enforcer of global rules and order. But simply a state pursuing its national interests.
A country that refuses to bear the burden of global order is relieved of the benefit of exorbitant privilege. However, that process is not immediate nor linear. And it most certainly is constrained by the availability – or rather, lack thereof – of alternatives.
The dollar’s dominance rests on eighty years of institutional construction, network effects, and demonstrated crisis performance that no alternative can yet replicate.”
“The question for long-term investors is not whether the dollar remains dominant on a three-year horizon. It will.
The question is whether the structure of the international monetary system – i.e., the distribution of reserve functions, the architecture of global liquidity provision, and the pricing of safe assets – is undergoing a multipolarity-induced regime change that will reshape asset valuations, risk premia, and portfolio construction over the next decade.
The answer is yes, and the implications are large enough to warrant systematic analysis.”
“Gold sits outside the pyramid. It is the neutral anchor: belonging to no currency bloc, carrying no counterparty risk, and unable to be sanctioned. Its role in the emerging system is not as a competitor to any currency but as a hedge against the system itself. Bitcoin plays a similar role to gold, as it, too, is not a contingent claim.”
“Each reserve manager constructs their own version. Saudi Arabia’s basket will weigh commodity neutrality and RMB trade utility differently from Norway’s. Singapore’s will weigh Asia-Pacific transaction efficiency differently from Brazil’s.
The system is decentralized and therefore robust. No single political decision can restructure it. No sanctions regime can freeze it. No central bank can depreciate it by printing more of it.
This decentralization is also the source of the basket’s most important investment implication. Because no one is coordinating the shift, the transition will be gradual, uneven, and punctuated by episodes of dollar reassertion (such as today, amidst the war in the Middle East). But because each reserve manager is optimizing their own portfolio, not making a political statement, the aggregate trend is durable. It requires only that the alternatives continue to improve and that the cost of dollar concentration remains visible.”
“In a multi-anchor world, funding stress becomes currency-specific. A dollar funding squeeze does not automatically transmit to EUR, JPY, and RMB markets with the same force, because those markets have their own funding backstops: the ECB’s EUREP facility, the BoJ’s swap arrangements, the PBoC’s bilateral lines. The common funding factor weakens. Assets that appeared correlated because they shared a dollar funding base show their true underlying lower correlations as that common funding factor decreases.
The portfolio construction implication is large. Diversification becomes more effective during stress, when it matters most. The crisis correlations that are the most damaging input in any risk model decline structurally.”
“The practical implication for long-term investors is clear: the correlation matrices and risk models built on two decades of dollar-dominated data are overstating the correlation between non-dollar assets.
Portfolios constructed on those matrices are under-diversified relative to the multi-anchor equilibrium. The transition creates both a risk (correlation assumptions embedded in existing portfolios are wrong in ways that will be revealed during the next stress episode) and an opportunity (institutions that update their frameworks ahead of the consensus capture the diversification premium before it is priced).”
“Who bears the transition cost: The US is the structural growth loser.
The exorbitant privilege is worth an estimated 50-100 basis points on US borrowing costs across the entire stock of external liabilities. As that privilege weakens, the fiscal arithmetic deteriorates in a vicious circle: higher debt service costs widen deficits, which further erode the safe-haven premium on Treasuries, which raises funding costs further. This is not a crisis dynamic, but a slow, compounding drag on GDP that accumulates over a decade.”
Who captures the equilibrium dividend:
Europe is the conditional winner. The mobilization of €35 trillion in household savings into productive capital investment is a genuine growth catalyst if CMU and SIU deliver. The conditionality is real and the execution risk is non-trivial, but the direction is the clearest it has been since Maastricht.
EM commodity exporters with domestic capital market depth are the other structural beneficiaries: cheaper funding, reduced Fed cycle sensitivity, and growing reserve demand for their assets from diversifying CBs. The bifurcation within EM is the most actionable growth implication for investors as the asset class can no longer be treated as homogeneous beta to the dollar cycle.
China sits in its own category: controlled RMB internationalization delivers incremental trade and funding benefits without the capital flow volatility that full liberalization would bring. Over time, this change can assist Beijing in its goal to rebalance the Chinese economy toward consumption.
Finally, the Canadian and Australian economies also stand to benefit as their commodity-centric nature becomes particularly appealing to foreign investors when risks around commodity supplies stand front and center.”
Conclusion: The Portfolio Construction Rethink Is the Alpha Opportunity
The multi-anchor currency regime is not a forecast with a date attached. It is a direction, one with clear historical precedent, visible institutional momentum, and coherent investment logic.
Cohen’s pyramid is reshaping. The dollar is not falling off it. Its share of the top tier is eroding, and the patrician and elite tiers are expanding. The investment implications flow from that reshaping.
The fixed income, equity, and correlation implications all follow this logic: a world where reserve functions are distributed across multiple anchors prices each asset more efficiently, compresses US structural advantages, and makes diversification more effective when it matters most.
Real assets and gold benefit from structural demand from multiple simultaneous directions: CB diversification, re-industrialization, SWF rotation from sovereign bonds, and the search for unconditional safety in a world of weaponized finance.
The synthetic reserve basket (the portfolio of factors that collectively replicates the dollar’s reserve functions) reframes the debate.
The question is not which currency wins.
The question is which attributes of the reserve function are underpriced because investors do not yet see their separability.
For a long-term investor operating on a 5-10 year horizon with global reach and mandate flexibility, the multi-anchor transition is a structural regime change to be positioned for in fixed income duration, equity geography, correlation assumptions, and real asset allocation. It suggests that the death of the 60/40 portfolio will continue and that investing in commodities, capex plays, and non-US equities has further to run.
The dollar’s reign is not ending. Its exclusivity is.”
4 Big Ideas
Grace Shao takes a deep dive into Korean teach and goes beyond just the chips with this interesting conversation with Ethan Cho, one of the leading South Korean VCs - go explore hardware, manufacturing, memory chips, consumer AI, enterprise innovation, defense tech, cultural exports in full here:
Intro:
In this episode, I spoke to a leading South Korea VC, TheVentures’ CIO Ethan Cho. He argues that South Korea’s low fertility rate and aging population put pressure on Korea to be one of the world’s fastest adopters of AI technology, similar to its rapid embrace of high-speed internet in the early 2000s. While not a leader in foundational LLMs like the US or China, Korea’s strength lies in application and adaptation, particularly in B2C areas like personalized agents and commerce, where cultural familiarity with chatbots and digital transactions lowers resistance.
The Korean startup capital funding landscape is shaped by three forces: Chaebols (Samsung, SK, Hyundai), the government, and VC firms. CVCs from Chaebols tend to reinforce existing semiconductor and hardware value chains rather than explore tangential innovation. To counter this, the Korean government has become a dominant LP through initiatives like “Everybody’s Entrepreneurship,” injecting capital to encourage novice founders. On sovereign AI, he believes the government’s push is less about global dominance and more about securing sensitive areas like finance and defense, though he warns that domestically-built software has historically struggled to scale beyond Korea.
Ethan is shifting focus from purely domestic champions to founders with global ambition but local execution, often Koreans educated abroad who return dissatisfied with traditional jobs. He wants to back ventures that change the world, not just build another food delivery app. He also recognizes key opportunity areas, including defense tech, K-beauty, fashion, and mental health, as society adopts AI at scale.
5 Big thinking
FS Blog shares some perspectives on Learning that are worth exploring and internalising - start by reading it in full here: https://fs.blog/learning/ - plenty of links to other reading within the roots and branches of the ‘learning tree’…
Some Takeaways
“Once you understand the keys to learning, everything changes—from the way you ask questions to the way you consume information.”
“Learning is the act of incorporating new facts, concepts, and abilities into our brains.”
“People can end up stuck with a static amount of knowledge because we don’t just passively absorb new ideas and information. Learning something new requires active engagement.”
“The greatest enemy of learning is what you think you know. When you think you know something, learning something new means you might have to change your mind, so it’s easy to think there’s no room for new ideas. But not wanting to change your mind will keep you stuck in the same place.”
“The best approach you can take towards learning is one that helps you go to bed a little bit smarter each day.”
“Learning requires time to reflect. It requires discussing what you’ve learned and letting your mind wander.
You need to let go of trying to look smart, and focus instead on trying to be smart.”
“Focusing is an art—through experimentation and creativity, you can build systems that let you give your full attention to whatever you’re learning.”
“Outside of your intellectual comfort zone is where you experience the greatest learning.”
“There are two main places we can learn from: our own experience and history, or the experience of others.”
“It is a mistake to think we have reached the endpoint of human knowledge, or that anything you learn now will be true forever. When you learn from history, you draw from lessons shaped by the perspective of the person who captured what happened. Thus, historical knowledge is something to continuously update as you learn both from what happened and how you choose to look at it.”
“Double loop learning is a way of updating your opinions and ideas in response to new evidence and experience. When you keep repeating the same mistakes, you’re using single loop learning. It doesn’t get you far. When you reflect on experiences, collect new data, and make an active effort to seek out feedback, you’re using double loop learning.
Reflection allows you to distill experience into learning. Don’t just “do,” think about what you’re doing and what you’ve done. High performers make adjustments based on both their successes and failures.”
“Learning isn’t something you do at the behest of someone else. You’re responsible for it.
According to the prolific author Louis L’Amour, all education is self-education.
If you don’t take charge of your learning, no one else will.”
6 The Purpose of Knowledge
Have a Great weekend when You get to that stage,
Sune