Regulators treat stablecoins as part of the financial system.
For years, the argument over stablecoins was whether governments would regulate them at all. That argument is largely over.
Singapore was amongst the first major financial centres to put a proper regulatory framework around stablecoins. Hong Kong has since gone further, introducing one of the most comprehensive licensing regimes anywhere for fiat-referenced issuers, covering reserve backing, redemption rights, capital requirements, and disclosure obligations in more detail than most jurisdictions have attempted.
Between the two, something has been settled. Regulators no longer treat stablecoins as a cryptocurrency curiosity sitting outside the financial system. They treat them as part of it.
That took years to resolve. The next question will take longer, and it gets far less attention. Getting a license tells you a coin is legal. It says nothing about whether it can actually move through the machinery that payments depend on, which has never been simple. Every transaction, regardless of what asset settles at the end of it, passes through identity verification, sanctions screening, anti-money laundering checks, foreign exchange conversion, settlement and reconciliation, and the reporting obligations that follow.
A stablecoin can replace the asset that moves at the end of that chain. It cannot remove the chain.
Consider a scenario that sounded speculative two years ago and no longer does. A software agent, acting for a business, sources a supplier in Vietnam, buys a content licence from a studio in Tokyo, and pays a contractor in the Philippines, all inside a single afternoon, without a person clicking approve.
Each of those decisions can happen in seconds. The payments cannot, because each one still has to clear a different currency, a different domestic clearing system ,and a different set of compliance checks before the funds land.
Take the Philippines contractor as an example. Even a straightforward stablecoin payment there still needs to pass through Bangko Sentral of the Philippines (BSP)-registered on and off ramps, satisfy source-of-funds checks under local AML rules, and convert into pesos at a rate the recipient can actually use. None of that disappears because the underlying transfer happened on a blockchain rather than through a correspondent bank. It just gets rebuilt around a new rail, with new points where something can go wrong.
This is where Asia’s payments landscape becomes relevant, and where a lot of commentary about the region gets it backwards. The usual framing is that Asia’s payments are fragmented: India has UPI, Singapore has PayNow, Thailand has PromptPay, Malaysia has DuitNow, Indonesia has QRIS, Vietnam has rapidly scaled VietQR, each built for its own market and none of them talking to the others.
That’s true as far as it goes, but “fragmented” undersells what these systems achieved. Individually, they are amongst the most successful retail payment networks built anywhere in the past decade. The problem was never poor design. It’s that nobody built them to interoperate, because nobody needed them at the time.
That’s changing. Governments across the region are now linking national QR systems to each other and piloting faster cross-border settlement mechanisms. Stablecoins are going to get folded into that same project, whether or not the people building them currently think of it that way.
What stablecoins probably won’t do is replace any of this. Banks, card networks, domestic wallets, and real-time transfer systems aren’t going anywhere. A stablecoin is more likely to end up as one more option sitting next to the others than a replacement for any of them.
This has practical consequences for who does well out of the next few years. Issuers who treat a stablecoin license as the finished product are likely to find limited demand, because most businesses don’t want a new currency, they want a payment that clears reliably wherever their customers happen to be. The firms better positioned are the ones already sitting across multiple payment rails, with the compliance and settlement relationships in different jurisdictions already built, who can add stablecoin settlement as one more option inside a system that already works.
Stripe took a similar route. After buying the stablecoin infrastructure firm Bridge in 2024, it rolled out stablecoin payment acceptance to merchants in more than 100 countries the following year. None of those merchants had to open a crypto exchange account or figure out how blockchain settlement works, the capability just showed up inside the payments system they were already using for everything else.
It’s a useful signal for where this is heading. The businesses that benefit first probably won’t be the ones launching their own stablecoin products. They’ll be the ones quietly wiring stablecoin settlement into infrastructure people have already adopted for other reasons.
There’s a pattern that shows up whenever a new payment technology arrives. It rarely wins by asking businesses to tear up how they operate. It wins by disappearing into how they already operate. Broadband didn’t replace the phone network so much as build on top of what was already there. Cloud computing didn’t ask companies to abandon their existing IT overnight, it absorbed it gradually, system by system. The unglamorous plumbing work, more than the headline technology, tends to decide which version of the future actually ships. Artificial intelligence is going through the same test right now, with compute infrastructure rather than the underlying models proving to be the real constraint on how fast it scales.
Digital money is likely to follow the same course. Its future depends less on the appeal of the technology and more on the systems that connect it to the financial networks businesses and consumers already rely on daily.
Hong Kong’s licensing regime settles a genuine and longstanding regulatory question. The harder one, which no license answers, is whether stablecoins can move through Asia’s existing payment systems rather than around them. That is the test worth watching over the next two years, not the announcements that got us here.