Sixteen percent. That is the only number in Visa and CoinShares' joint survey, released October 5, that describes what people actually did rather than what they say they might do. Over the past twelve months, 16% of consumers across Asia-Pacific moved money with a stablecoin. Real transactions. Real settlement. No intent, no aspiration, no survey-friendly hypothetical. Everything else in the report โ the 46% who "might" use stablecoins, the 85% of investors who "plan" to increase allocations by 2026 โ is aspiration wearing the costume of data. And here is the detail the headlines skipped: 66% of Asia-Pacific respondents said they were "familiar" with stablecoins, but only 6% could accurately explain how they work. Familiarity without comprehension. That is not adoption. That is a marketing funnel that has never been tested under stress. Every block hides a confession โ and this survey's confession is the gap between those two numbers.
The survey carries weight because of who commissioned it. Visa is not a crypto company. It is the settlement layer the world already trusts โ the network that clears card payments across 200 countries. CoinShares is a listed digital asset manager. When a traditional payment rail and a crypto-native asset manager publish adoption data together, the industry reads it as institutional validation. That framing deserves scrutiny before it becomes consensus.
The report covers seven markets โ the US, UK, Germany, Sweden, and several Asia-Pacific jurisdictions โ and splits into two narratives. The first is payment: 16% of Asia-Pacific consumers used stablecoins in the past year, 66% claim familiarity, and 38% cite fraud as their primary concern. The second is allocation: 70% of surveyed high-net-worth investors in the US hold digital assets, the average portfolio carries roughly 10% in crypto, 80% of digital asset investors hold Bitcoin, and 85% intend to increase exposure by 2026. Only 6% identify as short-term traders. On its face, this is the institutionalization story rendered in numbers: long horizons, real allocations, genuine usage. The bull case writes itself. But a survey is a data source, not a verdict, and the first duty of anyone who reads survey data professionally is to interrogate the sample before the conclusion.
Start with who was surveyed. CoinShares is an asset manager whose business depends on institutional capital flowing into crypto. Its population of "high-net-worth investors" is almost certainly drawn from people already adjacent to digital assets โ its own clients, prospects, and crypto-engaged networks. When the report says "70% of US high-net-worth investors hold digital assets," it does not mean 70% of wealthy Americans. It means 70% of wealthy Americans who answered a crypto asset manager's survey. That is a self-selection bias so severe it inverts the meaning of the number. Extrapolating it to the broader population would overstate real penetration by an order of magnitude. A survey's sample is its argument. Change the sample, and the conclusion collapses.
Now the stablecoin data. The report treats "stablecoins" as a single asset class. It is not. Fiat-backed tokens like USDC and USDT hold value through reserves of Treasury bills and cash. Crypto-collateralized tokens like DAI depend on overcollateralization. Algorithmic models like UST โ the one that vaporized roughly $40 billion in 2022 โ held their peg through reflexivity, which is another word for hope. Minted in hope, burned in regret. The survey never separates these. It asks about "stablecoins" and reports one adoption figure, which means the 38% who fear fraud are being asked to trust a category that contains both audited, Treasury-backed instruments and models that have already failed catastrophically. Conflating a Treasury-backed token with a reflexive algorithmic experiment is not simplification. It is the erasure of the exact distinction that determines whether a user's money survives a crisis.
And the largest stablecoin of them all deserves its own paragraph. USDT commands roughly 70% of the stablecoin market. In my experience reviewing reserve attestations for institutional clients, I have never seen a document that functions as a true audit. Tether publishes attestations โ point-in-time snapshots signed by an accounting firm โ not audits. There is a difference, and the difference is the whole game. An audit tests controls, traces liabilities, and renders an opinion on internal structure. An attestation confirms a number on a date. The market prices USDT as if it were interchangeable with a fully audited instrument. It is not. Liquidity flows, but integrity stagnates. This survey, by never distinguishing reserve models, quietly reinforces the fiction that all stablecoins carry equivalent risk.
Then there is the intent problem. The survey's headline-grabbing figures are all forward-looking. 46% of Asia-Pacific consumers say they might use stablecoins within five years. 85% of investors plan to increase allocations by 2026. These are stated intentions, and stated intentions have a well-documented decay rate. When researchers compare what people say they will do with what they actually do, intention-to-action conversion routinely falls below half โ and in financial behavior, far below that. The survey offers 46% intent against 16% actual usage. It offers 85% intent to add, against a current allocation of roughly 10%. It never tracks the gap. The most important number in the report is the one it does not compute: the conversion rate between what respondents say and what their wallets do.
This is where my own experience sharpens the read. In 2024 I consulted for a major Australian bank evaluating Bitcoin ETF exposure. I sat through the strategy meetings, the polished decks, the genuine enthusiasm. Then I pulled the risk models apart and found the same assumption buried in every one: that stated institutional demand would translate linearly into custody, compliance, and flow. It never does. The bank eventually adopted stricter frameworks precisely because intention data had been treated as pipeline. The lesson transfers directly here. A survey that measures desire is not a survey that measures demand. We chased the glow, not the ledger.
One more blind spot deserves a flag. The seven markets skew developed: the US, UK, Germany, Sweden. The places where stablecoins solve the most acute problems โ Argentina, Nigeria, Turkey, where inflation and capital controls make dollar-denominated settlement a survival tool rather than a portfolio choice โ are absent. The survey therefore captures stablecoin adoption as a convenience narrative for wealthy markets while ignoring the usage that actually reveals product-market fit. That is a distortion in the opposite direction from the bull case: it understates real demand where it matters most and overstates it where it matters least.
Now the part the skeptics miss. It is easy โ and lazy โ to dismiss this survey as marketing. But dismissing it entirely is its own analytical failure. The bulls got one thing profoundly right, and it is the thing that matters most over a long horizon: the structural shift is real, and Visa's presence is the proof.
Visa does not publish surveys about technologies it plans to ignore. It publishes them when it is evaluating whether to route settlement through them. Visa has already run USDC settlement pilots on-chain. A joint survey with a crypto asset manager is a reconnaissance mission, not a thought experiment. The traditional payment rail is studying the on-chain rail because the economics โ 24/7 settlement, no correspondent banking chain, no multi-day float โ are genuinely superior for cross-border flows. The convergence of a multi-trillion-dollar card network with dollar-denominated on-chain settlement is not a narrative. It is a capital allocation decision in progress.
And the 10% allocation figure, flawed sample aside, points at something durable. If digital assets are settling toward a 5โ15% band in institutional portfolios โ the same band as alternative assets โ then crypto has crossed from speculation into asset allocation. That transition is structural, slow, and largely irreversible. The bulls who frame this as the beginning of institutional adoption are directionally correct, even if the specific numbers are inflated.
So here is the honest reading. The survey is not wrong. It is unverified, and it is published by parties who profit from its optimism. The 16% actual usage is real and matters. The 46% intention is aspiration and does not. The 6% comprehension rate is the number that should keep every stablecoin issuer awake, because an asset class that 66% recognize but only 6% understand is one reserve crisis away from a panic no regulator can contain. History is written in hex, not headlines. The question for 2026 is not whether institutions intend to allocate more. It is whether the wallets โ and the reserves behind them โ can survive the moment intention meets a depeg.