There is a number that should not exist, and it is 0.1%.
This is the "true tax rate" that ProPublica, working from leaked IRS records, attached to Warren Buffett for the years 2014 through 2018. Over that five-year window, the man's reported wealth grew by billions of dollars. His federal income tax, expressed as a share of that growth, rounded to almost nothing. A tenth of one percent.
Set aside, for one breath, whether the number is fair. Set aside whether it is useful. The number is not false. It is a computation. And the distance between a computation and a fact is where this entire story lives.
Robert Kiyosaki, the author of Rich Dad Poor Dad, picked up that number and used it as a load-bearing beam. His argument — posted in fragments to X and then aggregated by BeInCrypto into something that looked like a financial column — is that the American tax code punishes labor and rewards ownership. Earned income is taxed hardest. Portfolio income is taxed more gently. Passive income, in his telling, is sometimes treated so lightly that a diligent person can drift into the club that pays a tenth of a percent.
That framing is the interface. It is clean, motivating, and engineered to be shared. The protocol underneath it is messier, and the protocol is the only thing that matters if you are actually going to move money.
I have been reading protocols for a long time. In 2017, during the ICO fever, I spent six weeks disassembling the Gnosis Safe multisig contract at the assembly level and found a reentrancy path that the first release had not closed. I reported it privately to the core team before anyone could walk through the door. That habit — read the protocol, not the pitch — is what I bring to a story like this one, even though there is not a single line of Solidity in it. Because there is always a protocol. In personal finance, the protocol is the tax code. It has state. It has transitions. It has events that trigger those transitions.
Kiyosaki's piece never names the event. That is the gap I want to work in. The protocol does not lie; the interface does.
Context: What Actually Happened
Let me lay out the raw material before I take it apart.
Kiyosaki, a public figure with a book franchise that has sold tens of millions of copies across two decades, took to X with a series of statements. The substance: he has held Bitcoin since 2012 and Ethereum more recently. He groups these alongside gold and silver as "hard assets," the instruments he recommends to people who want to reduce their reliance on cash. He frames his financial thinking around three categories of income — earned, portfolio, and passive — and argues that the tax treatment of each is wildly unequal, with earned income suffering the worst of it and passive income enjoying the best. As evidence, he reaches for Buffett's alleged 0.1% rate.
BeInCrypto aggregated the posts. The outlet is a crypto-native publication, and the piece landed under a Web3 label, which is itself an interesting editorial decision, because the article contains no protocol analysis, no token economics, no architecture, no code. Bitcoin and Ethereum appear purely as line items in one man's self-reported portfolio. They are named, not examined.
I want to flag a metadata anomaly before I go further, because it bears on how much weight this content deserves. The posts carry a publication date that reads as September 24, 2026, yet the same body of reporting says Kiyosaki's "2026 target prices" remain far above the current price level. If the post were genuinely published in September of 2026, that sentence would be self-contradictory, unless it means targets for late 2026. There are two charitable readings. Either the date stamp is wrong and this is a historical opinion piece, or the piece is written from a vantage point where those targets have not yet been met — which would actually be evidence that earlier predictions failed. Either way, the timing is soft, and soft timing weakens any claim about freshness. I flag it and move on, but I do not forget it.
Here is the honest classification. This is not a blockchain article. It is a celebrity money-mindset article with a United States tax discussion bolted to the front. Its technical value is close to zero. Its tax-mechanics value is real, and its narrative value is real, and those are the two seams I intend to open.
The Tax Code Is a State Machine, and the Event Is Called Realization
If you want to understand why Buffett's rate can be a tenth of a percent, you do not need to understand Buffett. You need to understand a single concept, and that concept is realization.
Most people intuit taxation as a continuous process. Money comes in, money is taxed. This intuition is wrong, and it is wrong in a way that quietly reshapes every financial decision a person makes. The United States tax code does not tax wealth. It does not tax appreciation. It taxes events. It taxes the moment a gain is realized, the moment income is received, the moment a transaction closes. Between those moments, value can multiply indefinitely, and the tax system is, for the most part, blind to it.
This is the state machine. Your assets sit in a state called unrealized. When you sell, you transition to a state called realized, and that transition is what triggers the taxable event. The size of the tax is a function of the gain, the holding period, and your marginal bracket. Nothing else. The machine does not care how rich you feel. It cares whether you flipped the switch.
The protocol pays attention to one thing and one thing only: whether the realization event occurred.
Now apply that to a person whose entire strategy is to never flip the switch. Buffett does not sell Berkshire. He holds. His wealth compounds as unrealized appreciation, which means it never enters the taxable base. His actual reported income — largely dividends and a modest salary — is a rounding error compared to his net worth. ProPublica divided the tax he paid by the growth of his wealth and got 0.1%. That is not a tax rate in any sense the IRS recognizes. It is a ratio between two numbers that the tax code deliberately keeps in separate rooms.
The IRS never assessed Buffett at 0.1%. The IRS assessed him on his realized income, and on that base he may well have paid a rate in the twenties or thirties. The 0.1% figure is what you get when you take the numerator from one accounting system and the denominator from another. It is a comparison, and comparisons are arguments, and arguments are not assessments.
Kiyosaki's article actually does the responsible thing here, and I want to credit it. It flags the controversy — it notes that the 0.1% number is disputed because unrealized appreciation is not normally taxable income. That single sentence is the most intellectually honest line in the entire piece. Silence before the block confirms the truth. The author paused, acknowledged the mechanism, and then kept going. Most writers of financial motivation never pause at all.
But the pause is where the piece should have ended, or where it should have pivoted into the territory that actually matters for its audience. Because the moment you understand realization, you understand two things. First, why the ultra-wealthy can post absurdly low effective rates. Second, why you, the reader, probably cannot.
What the 0.1% Actually Measures — and What It Hides
Let me be precise, because this is where most commentary goes to die.
ProPublica's methodology compares tax paid to wealth growth. It is a legitimate journalistic move. It reveals a real and underappreciated fact: the very rich can accumulate enormous paper gains while their taxable income remains modest. That fact is worth knowing. But the methodology produces a number that is not comparable to the rate on your W-2, and treating it as comparable is the error that turns insight into ideology.
Consider what has to be true for the Buffett structure to work. First, you must own assets that appreciate without throwing off much taxable income. Second, you must be willing to never sell them. Third, and this is the part almost nobody mentions, you must be able to borrow against them when you want spending money. That is the actual mechanism. You do not sell the appreciated asset and trigger the gain. You pledge it as collateral, take a loan, and spend the loan. Loans are not income. They are not taxable. You are, in effect, converting unrealized appreciation into spendable cash without ever crossing the realization boundary.
There is a name for the terminal version of this strategy, and it is grim but instructive: buy, borrow, die. At death, in the United States, the heirs receive a step-up in basis. The asset's cost basis resets to its market value at the time of death. The unrealized gain — the entire lifetime of appreciation — is extinguished for tax purposes in the transition to the next generation. It is never taxed. Ever. The state machine has a terminal state, and that terminal state is not a realization event. It is a deletion.
The step-up in basis is the single most powerful wealth-transfer mechanism in the code, and it is almost never explained in the plain language it deserves.
Now hold this against Kiyosaki's three-income taxonomy. He says earned income is taxed worst, portfolio income next, passive income best. As a rough directional statement, this is defensible. Long-term capital gains and qualified dividends get preferential rates. Some passive structures — real estate with depreciation, certain partnership distributions — can defer or shelter income in ways wage earners can only dream about.
But here is the trap. The taxonomy implies that anyone can climb the ladder simply by reclassifying where their income is booked. That implication is false. The preferential treatment is conditional, and the conditions are load-bearing. Capital gains rates require holding periods. Depreciation recapture claws back benefits when you sell. Passive activity rules under Section 469 exist specifically to stop high earners from using passive losses to shelter active income. The code is not a menu. It is a set of machines with guards, and the guards were installed by legislators who understood exactly what people would try.
So when a famous author tells you that passive income is taxed lightly, he is describing the output of a machine without showing you the guards. That is the interface talking. The protocol has security features.
The Layer Kiyosaki Left Out — and It Is the Most Important One for This Audience
The BeInCrypto article is published for a crypto audience, and yet it contains no discussion of crypto taxation. None. Bitcoin and Ethereum appear as portfolio line items, and then the tax conversation proceeds as if they do not exist. This is a remarkable omission, and it is the seam where I want to spend the most time, because for the reader holding digital assets, the mechanics are different in ways that matter enormously.
Start with the classification. In the United States, the IRS treats virtual currency as property, not as currency. This single decision cascades through everything. Because it is property, buying and selling it generates capital gains and losses, not ordinary foreign-exchange treatment. Because it is property, you can hold it long enough to reach long-term rates. And because it is property, the realization logic applies — but the events that trigger realization are not always obvious to the person triggering them.
Here is where it gets genuinely technical. On a chain, every transfer is a signed state transition. From the protocol's perspective, there is no difference between you selling a token for dollars and you swapping it for another token. Both are transfers of value. But from the tax code's perspective, only one of them is obviously a sale. The chain records a transfer. The tax code records a disposition. They are not the same event, and the gap between them is where investors get blindsided.
Consider a few of the transition types that trip people up. A token-to-token swap is not a tax-free exchange in the eyes of the IRS. You disposed of the first asset. You realized a gain or loss. The fact that you never touched dollars is irrelevant to the machine. A purchase of goods or services with crypto is a disposition of the crypto at its fair market value, generating gain or loss measured against your basis. Accepting payment in crypto is ordinary income at the moment of receipt, at the fair market value on that date, and then you have a new basis that will matter later.
Now the specialized events, the ones that live in the cryptographic layer. Staking rewards. When you earn staking rewards, the prevailing treatment is that you recognize ordinary income at the fair market value when the rewards become available to you. That means you may owe tax on assets you have not sold, whose value may have fallen by the time you pay. Mining follows similar logic. Airdrops follow the same pattern, with an added complication: you may not even know an airdrop landed, and there is genuine debate about when, or whether, it becomes income if you never claim it. Hard forks raise a further question — did the new tokens on the forked chain arrive as income, or as a windfall with zero basis until sold? The answers have shifted over time and are not entirely settled.
And then the one that should stop every crypto reader in their tracks. The wash sale rule.
In securities, if you sell a losing position and buy substantially the same asset back within thirty days, you cannot claim the loss. The rule exists to stop people from harvesting losses while keeping their exposure. As it stands in the United States, virtual currency has generally not been treated as a security or a stock for purposes of the wash sale rule. Which means, historically, you could sell a losing position, realize the loss for tax purposes, and immediately rebuy the same token. A harvest. A reset. No thirty-day window.
For years, this was one of the few structural tax advantages crypto holders had that equity traders did not — and it survived on the technicality that a token is property, not a security. Technicalities are the most fragile things in any protocol, and they get patched.
This is the layer Kiyosaki's article never touches. It tells a crypto audience to hold hard assets and reduce cash reliance, and it says nothing about the fact that the very act of acquiring, swapping, staking, or spending those assets is a series of realization events with their own rules. The interface says "hold and relax." The protocol says "every transfer is a state transition, and some of them are taxable." A reader who takes the advice and ignores the mechanics may discover, at filing time, that their non-taxable holdings generated a taxable bill.
There is a deeper point here about the asymmetry of information. Buffett's structure works because he never transitions state. His wealth exists as unrealized appreciation in a small number of assets he never sells, and he borrows against them when needed. Most people cannot borrow against their assets at institutional rates. Most people need cash flow. The moment you need cash flow, you either sell — triggering realization — or you borrow, and borrowing costs money that scales with your collateral and your creditworthiness. The elite structure is not a strategy. It is a privilege of scale, dressed in the vocabulary of strategy.
The Hard Asset Basket Is Three Assets Wearing One Coat
Kiyosaki groups gold, silver, Bitcoin, and Ethereum into a single category he calls hard assets, and it is worth pulling that grouping apart, because the assets are not the same kind of thing, and the crypto arc of this grouping has grown long enough that sloppy framing is now genuinely costly.
Gold and silver are physical. Their supply grows slowly and they have centuries of monetary history. Whatever you think of them, they are internally coherent as a category.
Bitcoin is often described as digital gold, and the comparison has some force. Fixed supply, issuance schedule written into the protocol, no central issuer. Whether it functions as a hedge is a live empirical question, and I have watched it trade like a risk asset through more than one liquidity crisis. That is a fact about correlation, not a verdict on its design.
Ethereum is a different animal entirely, and this is where the conflation gets dangerous. Ethereum is not a store-of-value protocol in the way Bitcoin is. It is a settlement and computation layer with a monetary asset on top. Its supply mechanics changed with EIP-1559 and the move to proof-of-stake, its issuance is not fixed, and its value is tied to demand for blockspace and staking yield, not to a simple scarcity narrative. When someone holds Bitcoin "since 2012" and Ethereum "more recently," and then presents both as forms of the same hard-asset hedge, they are erasing the distinction that actually matters for anyone allocating capital.
I say this as someone who has spent years inside these systems. A fixed-supply bearer asset and a programmable settlement layer are not the same trade. Stapling them together because both are crypto is like calling a vault and a stock exchange the same thing because both hold value.
The grouping survives because it is rhetorically convenient. Hard asset sounds solid. It conjures safety. And the crypto names ride along on that emotional gravity, gaining the aura of the metals without inheriting their actual properties. To own the chain is to own the history — but the history of Ethereum and the history of gold are not the same history, and pretending otherwise is a service to narrative, not to the reader.
The Narrative Economics of a Billionaire's Portfolio Photo
Now to the part that is not about taxes at all, and may be the most important part.
Kiyosaki has been saying versions of this for years. He has held Bitcoin since 2012, by his own account, and has publicly favored gold and silver for far longer. This is not new information. It is a repeated performance of a consistent persona. And in market terms, a repeated performance is nearly pure noise.
I want to be precise about the mechanism, because "noise" is thrown around loosely. A signal moves prices because it update expectations about future supply, demand, or cash flows. A statement that you have held an asset since 2012 updates no expectations. It is retrospective. It tells us nothing about what this person will do tomorrow, and even if it did, one person's holdings are not the marginal flow that sets price in a market this deep. The 2012 disclosure is a credential, not a catalyst.
The only thing that would convert this from noise into signal is a specific, new, large action. If Kiyosaki announced a material purchase or a material sale, that would be news, and the direction would matter. A statement of continued holding is not. Markets have learned this. The aggregate response to celebrity HODL talk is close to zero, and it has stayed close to zero through every cycle I have watched.
And this is where I have to name the conflict of interest, because it is structural and it is not incidental. Kiyosaki sells books. He sells speaking engagements. He sells a brand built on the claim that conventional financial advice is a trap. An opinion that reinforces that brand is not free. It is inventory. Vested interest distorts the lens of analysis, and the distortion is not always conscious. It is built into the incentive structure before a single word is written.
This is why I treat celebrity financial content the way I treat unaudited code. Not as fraud. As unverified. The default assumption should be that the output serves the author's interests until the author demonstrates otherwise, and the demonstration is evidence, not confidence. Kiyosaki's track record does not help him here. He has, for years, predicted a historic crash, and those predictions have not arrived on schedule. That does not make him wrong about everything. It makes him a source with a low, well-documented hit rate on timing, which is precisely the dimension where his predictions would need to be right to be useful.
Certainty is a bug in a stochastic world. The loudest voices in finance speak with the confidence of people who have never been forced to price their own uncertainty. The protocol underneath markets is probabilistic. Anyone who tells you otherwise is selling something, and in Kiyosaki's case, what he is selling is often literally a book.
Contrarian: A True Principle Applied to the Wrong People
Here is where I want to push against the piece, and against most of the commentary around it.
The tax principles in the article are largely correct. Unrealized gains are generally not taxed. Earned income is generally taxed harder than investment income. The super-wealthy can achieve effective rates that look absurd when measured against wealth growth. None of this is in serious dispute, and the article does a fair job presenting it, complete with the caveat about ProPublica's methodology. On the level of description, the piece is honest.
The failure is not in the description. It is in the implied application.
The article takes a mechanism that operates at the top of the wealth distribution, describes it accurately, and then places it in front of an audience that mostly cannot access it, without drawing the line between the two. It shows you the vault, explains how the vault works, and lets you assume you can walk in. You cannot. The vault requires an existing stock of highly appreciated, low-yield assets, a willingness to never sell, and access to cheap credit against those assets. If you have those things, you did not need the article. If you do not have those things, the article is a description of a country you will never visit.
This is the defining blind spot of the genre: a true principle presented to the wrong audience as if it were available to them.
And the crypto layer makes it worse rather than better. The article tells a crypto audience to reduce cash and hold hard assets. But the crypto portion of that advice carries tax obligations the article never mentions and volatility the article never warns about. "Hold and relax" is the interface. The protocol says that every acquisition, every swap, every staking reward, and every spend is a potential realization event with a filing consequence, and that the assets themselves can halve in a quarter. The article whispers stability in the direction of the most volatile holdings its audience owns.
The counterintuitive conclusion is this: the tax content of this article is the part most worth reading, and the investment content is the part least worth trusting. The two are bundled together so tightly that the credibility of the first bleeds into the second. That is how confidence laundering works. You establish authority on a real principle and then spend that authority on a recommendation that the principle does not support. The protocol is honest. The interface uses the honesty of the protocol as cover.
There is also a blind spot in the other direction, and it belongs to us, the crypto community. We spend enormous energy on consensus mechanisms, on slashing conditions, on the difference between optimistic and zero-knowledge rollups. We spend almost none on the tax layer that sits on top of our own holdings, even though it is the layer that most directly determines whether the wealth we accumulate survives contact with the state. We audit contracts obsessively and audit our own realization events almost never. That imbalance should embarrass a community that prides itself on self-custody. To own the chain is to own the history — and part of owning the history is understanding which of its events the state has decided to count.
Takeaway
The next seismic shift in crypto will not come from a new consensus algorithm. It will come from the tax layer, and it is already moving. If the special treatment of digital assets under the wash sale rule gets patched away, a strategy thousands of holders rely on will evaporate overnight. If staking and airdrop guidance tightens, the arithmetic of "earning yield" changes at the ledger level. If any jurisdiction reaches for unrealized gains directly — and proposals exist — the entire premise that you can hold forever and owe nothing will be tested in public.
Watch for the event, not the interface. The number that should not exist, 0.1%, is a computation. The question worth asking is what computation will be possible for the next generation of holders, and which of its inputs the state decides to count.
I can see the machine. I can read the guards. The protocol does not lie; the interface does. And the interface, this time, was written by a man who sells books.
