The real test
A century eats almost everything
Pick any hundred-year window and lay it over history. Inside it you will find at least one major war, one currency that died, one government that fell, and one confiscation. The store of value that matters is the one still standing at the far end of all four.
This is a different question from the one markets usually ask, and it disqualifies most of the usual answers. A stock that compounds beautifully for thirty years is no help if the company – or the exchange, or the currency it's priced in – does not survive the fourth decade. The reason we underrate how hard this is comes down to survivorship bias: the indices and portfolios we admire are the ones that lived. The losers were quietly delisted from memory. Judge by the whole graveyard, not the handful of monuments, and the century test gets very strict very fast.
Thirty good years are easy.
It's the hundredth that does the judging.
We admire the portfolios that lived.
The losers were quietly delisted from memory.
The contenders
Grading the usual stores of value
Hold each candidate up to the same four hazards – inflation, default, seizure, and obsolescence – and see how it scores across a full century.
One honest caveat before the grading, because the choice of hazards is doing real work. This essay tests one specific property: the ability to carry wealth across a hundred years intact. That is not the only thing an asset can do, and on a different test the ranking inverts. Over a century, productive assets – equities and land – have produced real returns a sterile store of value simply cannot: a share is a claim on profits that compound, a field grows crops and rents, while gold and bitcoin sit there earning nothing and yielding nothing. If the question were "what grew the most wealth," the answer would not be a hard-money bearer asset; it would be the productive economy. So the shortlist that follows is short for this job – surviving as a store of value through war, default, seizure, and a dead currency – not short in general. A complete portfolio wants both: the productive engine for return, and the sterile anchor for the decade the engine seizes.
Click a store of value – the hundred-year verdict
Two finalists tend to emerge from any honest version of this exercise: things that are no one's liability and cannot be inflated. Gold has carried that flag for millennia, with the asterisks of weight and seizure explored elsewhere on this site. The new entrant makes the same claim, with the same two virtues and without the vault – but missing the one thing the others have and it cannot yet have: a long record. That is the real shape of the bet, stated fairly.
No one's liability.
Impossible to inflate. The shortlist is short.
The honest scorecard
What a century actually rewards
Strip the century test to its mechanics and four properties decide everything: no counterparty who can fail you, no issuer who can dilute you, the ability to cross a hostile border, and survival without upkeep.
On the first two, a fixed-supply bearer asset scores as well as anything ever has – better than gold on dilution, equal on counterparty risk. On the third, portability, it is in a class of its own: wealth you can carry across any border inside a memorised phrase is something no previous store of value could offer. The track record is the obvious gap, and it is real: gold proved durability over five thousand years; bitcoin, in roughly sixteen, has merely argued it. But there is a deeper, less-flattering version of that same gap, and an honest scorecard has to put it on the board.
Bitcoin's own fourth axis
The four hazards were chosen to be fair to every contender, so apply the fourth – obsolescence – to bitcoin itself, without flinching. Gold survives a century by doing nothing; an atom of it in 2126 is the same atom. Bitcoin is not an atom. It is a network, a protocol, and a key – three things that all have to keep working for a hundred years. The network has to be running and secured in 2126. The software has to still be maintained, and its assumptions still hold. And the holder, or an heir, has to be able to find and use a key generated a century earlier, on hardware and standards long obsolete. Lose any one of those and the bearer asset becomes an unspendable string. That is a genuine technological-survival risk, and pretending it away would be exactly the rigged scoring the other contenders deserve to be spared.
The honest answer is not that the risk is zero; it is that bitcoin is built to minimise the parts most likely to rot. The ledger is deliberately minimal – an ownership record so simple that any competent programmer can reimplement a client from the specification, which is open and copied across tens of thousands of independent machines, owned by no company that can fail or be acquired. There is no central server to switch off, no licence to lapse, no vendor whose bankruptcy ends it. The cryptography is the one part that genuinely ages – a hundred years is long enough that today's signatures may need replacing – but that is a known, bounded problem: the network can migrate to stronger schemes by the same consensus that runs it, the way the internet's own protocols have been upgraded under load for decades. And the key-survival problem is the holder's to solve, no different in kind from keeping gold findable across three generations; the difference is that a seed phrase can be copied perfectly, infinitely, and hidden anywhere, which gold cannot. None of this makes a century certain. It makes the obsolescence risk a problem of stewardship and engineering rather than a fatal flaw in the asset – which is the most any hundred-year holding can claim.
And the survivorship bias that disqualifies the others has to be turned on bitcoin too, or the scoring is rigged in its favour. The bullish line – "it survived every crash, every ban, every obituary, and it is still here" – carries its own version of the trap: we are admiring the one network that lived. Thousands of other coins launched on similar promises and are now dead or irrelevant; bitcoin is the survivor we get to study precisely because it survived, which tells us less about the next hundred years than the survival story implies. That it won the first round is real evidence – first-mover lead, the deepest security, the strongest network effect – but it is not proof it wins the next, and treating "it made it this far" as a guarantee is exactly the error we just refused to grant equities.
So an honest steward holds both facts at once: that on the properties a hundred-year asset needs, bitcoin scores extraordinarily, and that its weaknesses – an unproven record and a network that must be kept alive – are precisely the ones only time and care can close.
Why a small anchor wins the century
That is also why the long-clock portfolio is not all-or-nothing, and the reason is mathematical, not temperamental. Over a hundred-year horizon the dominant force is not return; it is survival. An asset that compounds at any positive rate but has a real chance of going to zero inside the window contributes, in expectation, almost nothing to the far end – the zero eats it. The asset that merely persists, even modestly, is the one still standing to be counted. So the question a steward actually faces is not "how much bitcoin" but "how much of my wealth has any exposure at all to the one candidate with all four properties." And here the asymmetry does the work: a small allocation can lose only what you put in – a bounded, known downside – while its share of a century-scale, supply-capped repricing is convex and effectively uncapped. A position small enough that its failure is survivable, in the one asset whose success is unbounded, dominates a larger position in anything whose best case is merely "did not die." Over a century, a modest anchor in the convex asset is not a hedge against the portfolio. It is the part most likely to define it.
The best properties on paper.
The shortest track record in the room.
The hundred-year question has only ever had a handful of honest answers. For the first time in a long while, the list has a new name on it – provisional, unproven by time, a network that must be kept alive as much as held, and built, on paper, for exactly this.
Still skeptical
The new name lacks a track record. Weigh the proven incumbent and the swings against it.
Gold's Long Reign →Volatility Is the Toll, Not the Trip →Curious
See the two hazards a century punishes most: dilution and the freeze.
The Debasement Tax →When States Freeze Money →Convinced
The man underwriting a hundred-year outcome, and the math of anchoring to it early.
The Steward's Wager →The Asymmetry →Sources & notes. The verdicts rest on well-documented long-run patterns: the US dollar's loss of roughly 96 to 99 percent of its purchasing power since 1913 (BLS CPI); sovereign defaults and bond losses across the 20th century; the near-total turnover of the public companies that made up century-old industrial indices; periodic land seizure and tax; and gold's multi-millennial persistence. Figures are approximate. Bitcoin's roughly sixteen-year history, and the fact that it must persist as a living network and not merely as held matter, are stated plainly as its central limitations; the claim about migrating cryptography by consensus is an argument from the protocol's design, not a guarantee. None of this is investment advice.