Raoul Pal says only two groups of assets have shown the ability to consistently beat an 11% annual hurdle rate: technology assets and crypto assets. His argument starts from purchasing power rather than account balances, with the claim that nominal gains matter far less once currency dilution and everyday inflation are taken into account.
In the essay, Pal describes the familiar wealth playbook many investors were taught to follow: build savings, contribute to retirement accounts, buy index funds, and add property if the timing works. He says that framework belonged to a different macro era. Pal notes that he used to work as a hedge fund manager and later spent 21 years writing research for hedge funds and family offices, giving him a front-row view of how the environment behind traditional portfolio construction has changed.
He reduces economic growth to three drivers: more labor, higher output per worker, or more debt. For most of the last century, he writes, the first two did the heavy lifting. Now, with birth rates having fallen for decades and productivity growth slowing, economies rely far more on the third. Debt carries interest, and the politically viable way to manage that burden, in his telling, is monetary expansion.
That is why Pal focuses on global liquidity, which he defines as the total stock of money and credit in the financial system. He says it expands by about 8% a year. Add the 2% to 3% inflation rate people usually see in everyday goods, rent, and living costs, and the line that matters becomes roughly 11% annualized. In his framework, that is the real threshold between preserving purchasing power and losing it.
The 11% line as the new benchmark
Pal’s point is blunt: if an asset returns more than 11% a year, purchasing power rises; if it returns less, purchasing power falls even when the statement balance goes up. By that definition, earning 11% is not extraordinary. It only keeps wealth level in real terms.
That changes the way he thinks assets should be judged. The goal is no longer to assemble a basket that only looks safe or diversified on paper. The goal is to identify assets that can clear the 11% line over time.
How he reassesses bonds, property, gold, and equities
Pal starts with bonds. A bond, as he lays it out, is a loan: investors hand over capital, receive a fixed interest payment, and collect principal at maturity. The problem is that the coupon is typically set against ordinary inflation, not against monetary dilution. He gives the example of a government bond yielding 4% annually. If the currency used to denominate it loses value at 8% a year, the investor can receive every promised payment and still end up poorer in real terms by the time principal is returned.
Real estate, in his view, still retains some anti-debasement logic. Borrow at a fixed rate, buy a hard asset priced in a currency that keeps losing value, and over time the debt burden shrinks in real terms while the property price rises with money supply. He adds that he owns real estate himself.
Still, Pal argues that the trade which helped an earlier generation build major wealth was not real estate alone but the mortgage window behind it. Buying with leverage near the start of a 40-year falling-rate cycle created a powerful tailwind for valuations. He says that setup cannot simply be repeated now, after rates fell toward zero and then moved higher again. He also points to high price-to-income ratios and elevated mortgage rates as fresh constraints.
He adds another data point: since 2007, global liquidity has expanded much faster than house prices, pushing the ratio between the two lower over time. In nominal dollar terms property may still look more expensive, but in real purchasing-power terms it no longer offers what it once did.
Gold gets more nuance. Pal acknowledges a major rally. He writes that gold broke above $5,500 an ounce in January this year, setting a record high. By late August, when he wrote the piece, the price had slipped back to around $4,600, still up by about one-third for the year. Financial media, he says, labeled this run a "debasement trade."
Even so, he does not treat gold as a long-term compounding growth asset. His comparison is with central bank balance sheets over the past 15 years. On that basis, he says gold has largely kept pace with monetary expansion, which is exactly what it is supposed to do. Gold can preserve purchasing power and hedge against excess money creation. He is explicit that he is not dismissing that role.
His distinction is between preserving wealth and creating new wealth. Gold has no user-adoption curve, no business ecosystem growing on top of it, and no compounding effect tied to a rising network of participants. Its price, in his view, is driven by how anxious markets feel about the monetary system. That can push it sharply higher, but it is not the same thing as a decades-long compounding machine.
Equities do better in his framework. Pal says the broad indexes most people own have outperformed the prior categories. He puts the S&P 500’s annualized return over the past decade at about 13%, just above the 11% hurdle, though he stresses that this happened during what he describes as the strongest equity index bull market in financial history.
Why technology and crypto sit in the same bucket for him
Pal then lines up 10-year annualized returns across asset classes: about 12% for gold, roughly 13% for the S&P 500, around 20% for the Nasdaq 100, and between 58% and 70% for Bitcoin depending on the starting point and calculation method. Against an 11% hurdle, he says, the relative picture is clear.
But his case does not stop at the performance table. He argues that technology assets and crypto assets generate high compound returns because both are tied to user-adoption S-curves rather than the economics of traditional value assets. Drawing on Metcalfe’s law, he says a larger network raises the value captured by existing users. Adoption does not rise in a straight line. It starts slowly, accelerates sharply, and then levels off as the market matures.
As long as an asset remains in the steep part of that curve, Pal says, its return profile can outrun the pace of money creation for reasons rooted in network growth rather than short-term speculation. That is why, in his view, the key question is no longer whether technology and crypto can beat traditional assets. The real question is whether their adoption curves are already exhausted. His answer is no.
He goes one step further and says the next big wave of participants will not just be human users. That is the long-term variable he believes markets are still underpricing.
Why he thinks classic diversification has lost some of its protection
Traditional portfolio construction assumed that bonds, gold, real estate, and equities were different risk buckets. Hold all four, and one shock should not sink the entire portfolio. Pal says that logic held for a long time but changed after 2008, when liquidity became the dominant force behind asset pricing.
Under that regime, he argues, bonds are effectively a liquidity trade, gold is a liquidity trade, property moves through the liquidity cycle, and index funds are a cleaner wrapper around liquidity-sensitive assets. He is not dismissing diversification itself. In fact, he says he still diversifies. His complaint is with the raw materials inside the old portfolio mix. If three of the four major asset types fail to beat the 11% hurdle, and the fourth only barely does so in an unusually strong bull market, spreading money across them may provide comfort without preserving real wealth.
That shifts the focus from the number of positions in a portfolio to the amount of capital allocated to assets with genuine long-term compounding potential.
Where value settles in crypto, in his view
Pal does not claim every layer of crypto should be treated the same. He says application-layer protocols can produce very large returns, and a project that solves a real commercial need can outperform the base chain underneath it.
The difficulty is picking the winners in advance. Base-layer blockchains, he argues, carry the settlement activity of the whole ecosystem. No matter which applications come out ahead, value still settles into the infrastructure layer. That lowers the burden of prediction. Instead of forecasting the exact app that will dominate, investors only need to form a view that economic activity will keep moving onchain.
That line of thinking also shapes his rejection of standard equity valuation frameworks for crypto. He lists fee multiples, revenue growth, and total-value-locked ratios as examples of metrics the industry borrowed from stocks without first proving they fit blockchains. Pal says GMI tested valuation measures across 12 major public chains and found that none reliably predicted future returns. The only metric that showed predictive value, according to the essay, was whether incoming capital stayed in the ecosystem over time.
His explanation is structural. A public blockchain is not a company earning a spread by selling products. Its value comes from the applications built on top of it, not just from transaction fees. Valuing Ethereum on fees alone, he writes, would be like trying to value the entire internet in 1998 by looking only at email service charges.
AI agents as the next underpriced source of onchain demand
In the second half of the piece, Pal shifts to AI. He says most industry size forecasts still assume human users are the only participants in digital economies: people making a few transactions a day, occasional payments, and intermittent requests. He argues that assumption is close to breaking.
His definition of AI agents is software that can sense, decide, and act without direct human instruction. Instead of serving only as tools, they are about to enter the economy as independent participants. Citing industry forecasts he has seen, Pal says the ratio of non-human intelligent identities to human employees inside companies could eventually reach 80:1.
He argues that AI agents cannot open bank accounts, lack legal identity, cannot visit physical branches, and cannot operate smoothly in a system where settlement windows close at five in the afternoon and transfers take three days. What they need, he says, is programmable money running on payment infrastructure that never shuts down. Public blockchains already provide that in native form.
To show that this buildout is not theoretical, he points to several products already appearing in public: Anthropic’s open-source Model Context Protocol, Google’s Agent2Agent and its WebMCP preview, and Coinbase’s revived x402 protocol, which allows agents to pay each other over HTTP.
For Pal, those releases are early signs of rising settlement demand. Even without speculative inflows, he says, real business demand could keep growing.
His risk warning: do not go all in, and do not use leverage
Despite his clear preference for long-duration growth assets, Pal adds a direct warning against survivorship bias. Markets are full of stories about people who went all in on one asset and reached financial freedom. What goes unseen, he says, are the many investors who concentrated into a single position and were eventually wiped out. He recommends reading anonymous trading confession posts if someone wants a more representative sample of outcomes.
That is why he does not recommend putting all capital into one asset. In his framework, long-term outcomes are driven by two main variables: how much capital is allocated to real growth tracks, and how long the investor can stay invested.
Then comes what he calls a hard rule: no leverage. Not a little leverage, not "careful" leverage, and not leverage paired with stop losses. His reason is practical. Leverage strips investors of the ability to survive a 50% drawdown. An investor can be right on the 10-year thesis and still get liquidated during a violent two-week drop.
For most people, he favors a layered allocation. Keep part of the portfolio in traditional assets for safety and peace of mind. Put a meaningful amount into long-term growth assets, but size that sleeve so a 50% drawdown does not force a bad decision. The rest of life, he suggests, should not be consumed by constant trading.
Pal adds that after 13 years of watching crypto markets, the investors who achieved long-term compounding were often not the ones trading all the time. In the moment, a crash feels dramatic. Over a longer chart, it can look like a small move. Doing nothing, he says, can itself be a strategy, even if it is much harder in practice than it sounds.
What he calls the real opportunity cost
Pal closes by returning to the 11% hurdle. If portfolio returns compound below that line, he says, the wealth earned through labor buys less freedom each passing year. The payoff from understanding this framework is not only a larger account balance. It is more choice: not having to watch markets constantly, having the confidence to refuse work one dislikes, and being able to spend time in places and with people that matter while still young enough to enjoy it.
He does not end with a model portfolio. He says he does not know the reader’s debt burden, time horizon, or risk tolerance, and anyone giving exact allocations without those three variables is effectively guessing with someone else’s money. What he does offer is a filter: take the 11% annual hurdle and apply it to every asset you own. In his view, anything that cannot beat that line is consuming time and freedom, no matter how comfortable it feels to hold.
