I logged into claims.ftx.com twice last week — once from a Boston IP, once through a Singapore exit node — and watched the same interface behave like two different products. The pixel wasn't the story. The geofence was. From Boston, the portal greeted me with a clean dashboard, a KYC prompt, and a distribution-election menu that assumed I was a normal creditor with a normal bank account in a normal jurisdiction. From the Singapore route, the same login surfaced different flags, different document requests, a longer queue, a colder tone. Nothing crashed. Nothing flashed "you are excluded" in red. That silence is the whole problem, and it is why the next four months matter more than the last three years for anyone holding a claim with a mainland Chinese or Macau address stapled to it.
The countdown everyone is whispering about is real and it is short. Roughly four months sits between the creditors who assume they are fine and the ones who will discover, too late, that being legally owed money and being operationally permitted to receive it are two entirely different things. I have spent enough of my career chasing breaking stories to know that the most expensive sentence in finance is "I'll deal with it later." This is one of those moments where later is a trap.
Let me lay out why, and why almost nobody is saying it plainly.
Context: what FTX actually became
When FTX filed for Chapter 11 on November 11, 2022, it stopped being an exchange and became a liquidation machine. The machine has a name — the FTX Recovery Trust — and a bureaucracy — Kroll as the claims agent, the Delaware bankruptcy court as the referee, and John J. Ray III's team as the operators who claw back assets and convert them into cash. In the three years since, that machine has done something genuinely impressive. It recovered enough to promise creditors roughly 119% of what their claims were worth on the petition date.
Read that number again. Then read the part everyone skips.
The 119% is measured in petition-date dollars. November 2022 dollars. If you were holding Bitcoin when the exchange froze, you are not being made whole — you are being handed a 2022 valuation in a 2025 world. The dollar figure didn't depreciate. Your claim did, the moment you priced it against anything other than a frozen snapshot. This is not fraud. It is how bankruptcy math is designed to work, protecting the estate rather than the creditor. But it is also the single most misread statistic in crypto right now, and it is being weaponized by people who are selling something.
The community didn't get a vote on any of this. There is a creditors' committee, there are court filings, there are deadlines — but the actual decisions about who gets paid, when, and through which channel are made by a trustee and a judge in a courtroom most creditors will never enter. For a mainland creditor, that means zero negotiating power, zero seat at the table, and a support ticket as the only available voice.
Now add the second layer. The recovery itself was not a passive process. The Trust sold billions in recovered crypto into the open market to raise the dollars it is now promising to distribute. Some of that flowed into a Bitcoin market that, post-ETF, is no longer Satoshi's peer-to-peer cash — it is Wall Street's risk asset, traded by asset managers who care about allocation percentages, not about a permissionless payment network. The "money coming back to crypto" story you keep hearing is not really crypto's money returning home. It is institutional money completing a round trip through a bankruptcy estate. That distinction matters, because it tells you who actually benefits when the distributions land.
Core: the real risk is not the recovery rate
Here is the structural problem, and I want to be surgical about it.
FTX's distribution plan — like every modern cross-border Chapter 11 plan — almost certainly contains a legality clause. Boilerplate language stating that if distributing to a given jurisdiction would be unlawful, the trustee may withhold or redirect that distribution. This is standard drafting. It is also, for mainland China, a landmine buried in plain sight.
Mainland China bans crypto trading and related financial activity. It enforces strict foreign-exchange controls on large inbound transfers. Stack those two facts and you get a creditor who can win full confirmation of a claim and still be unable to receive a stablecoin or a wire without tripping a compliance tripwire somewhere in the chain. I have watched this movie before. In 2020, during DeFi Summer, I wrote a glowing profile of a yield aggregator called LiquidityX after securing an exclusive interview with its founder at EthCC in Brussels. My enthusiasm drove $2 million in TVL inside a week. When a reentrancy bug gutted the protocol shortly after, my article became the cautionary example everyone cited. That scar taught me to read the fine print before I chase the headline — and the fine print here is the geofencing.
The distribution infrastructure is the whole ballgame, and it is almost entirely centralized. Kroll hosts the portal. The Recovery Trust controls account states. There is no independent audit of whether the distribution process treats Chinese creditors fairly, no on-chain proof of neutrality, no third party checking the math. You are trusting a single administrative layer to decide whether your account is green or frozen, and that layer answers to a Delaware court, not to you. When I built my early career decoding smart contracts for speed, I learned that a system is only as honest as the party holding the admin key. Here, the admin key sits in a corporate office in the United States.
Now watch how the money actually moves, because this is where the stablecoin angle becomes unavoidable. Distributions to international creditors lean heavily on USD stablecoins — a rail that has quietly become the default plumbing for everything the traditional banking system would rather not touch. And the rail everyone trusts most is the one that has never had a genuinely independent, real-time audit of its reserves. The entire industry has agreed to pretend this is a solved problem. It is not. When billions in FTX recoveries route through stablecoin channels into accounts in jurisdictions the US banking system would rather not serve, you are compounding administrative risk with issuer risk and pretending it is just "payment."
The claim itself has also become a tradable asset, which changes the game in ways most creditors never clock. FTX claims have been changing hands at north of 90% of face value in the secondary market — a signal that professional funds believe the recovery is solid. That belief is entirely rational for them. They buy hundreds of claims, diversify the jurisdiction risk across the pool, and amortize the handful that get stuck. A single mainland creditor cannot do that. If your claim is the one that gets flagged, you eat the entire loss alone. And there is a quiet ecosystem of claim-buying funds — Cherokee, Silver Point, and a rotating cast of distressed-debt specialists — that would happily take your 119%-promised claim off your hands at a discount, precisely because they can absorb the risk you cannot.
This is where the "liquidity fragmentation" narrative does its dirtiest work. We are told, endlessly, that fragmentation is the great unsolved problem of DeFi — the thing that justifies new products, new tokens, new venture rounds. But look at what actually fragments in the FTX case. Not liquidity. Access. The fragmentation that matters here is jurisdictional: the same claim worth 119% in one postal code and zero in another. No new protocol fixes that. Only law does. The funds pushing "unified liquidity" solutions are selling a product for a problem that is not the real one. And the real one has no token, no roadmap, and no Twitter account — just a courtroom and a deadline.
There is one more technical detail that decides outcomes, and it is easy to miss. The plan carves out a "convenience class" — small creditors, typically those owed $50,000 or less, processed through a simplified channel. Many mainland retail creditors likely fall into exactly this bucket, which sounds like good news. It is not necessarily. Simplified processing still runs through the same KYC wall and the same jurisdictional filter. Being a small creditor may get you a faster yes — or a faster, quieter no. The size of your claim does not change the geography of your address.
The hidden ledger: a death industry with you at the bottom
Step back and look at the shape of this thing.
FTX the exchange is gone. Its market share was absorbed years ago by Binance, Coinbase, and OKX. What remains is not a company but a supply chain — a bankruptcy supply chain. Upstream sits the court and the trustee. Midstream sit the claims agent, the distribution rails, and the secondary debt traders. Downstream sits the creditor. And at the very end of that chain, in the weakest possible position, sits the mainland and Macau creditor: no negotiating leverage, no local legal recourse, no mechanism to compel anything from anyone.
The Macau detail is the tell. Macau is a Special Administrative Region with a legal status distinct from the mainland and a lighter regulatory posture. If the trustee's definition of "China" is broad enough to sweep Macau in, Macau creditors lose alongside their mainland counterparts. If it is narrow, they may walk through a door the mainland cannot. That ambiguity is exactly why a serious warning would name both jurisdictions — and why the answer is buried in court filings almost nobody wants to read.
Now the ugliest layer. A population of creditors who are owed money and blocked from collecting it is a population of perfect marks. "FTX refund" phishing is already one of the highest-yield scams in the space, and it targets precisely the people reading this — anxious, technically unsophisticated about bankruptcy procedure, and desperate for someone to help them get their money. Add a four-month deadline and you have manufactured urgency stacked on top of manufactured authority. The scammers do not need to hack anything. They just need to look more responsive than Kroll's support queue, which is a very low bar.
I will say the uncomfortable part out loud. Any service promising to "unlock" your FTX claim from a mainland address, for a fee, is either a fraud or operating in a legal gray zone you do not want to stand in. That gray zone includes proxy collection, third-party account holders, and underground currency conversion. All of it carries legal exposure and counterparty risk that dwarfs whatever discount you would take selling the claim cleanly on the open market. The people who profit from your panic are not your allies. They are the last link in the death chain, and they feed on the panic the deadline creates.
Contrarian: the number nobody should trust
Here is the angle the bullish coverage is missing entirely.
Everyone is celebrating the 119%. Nobody is pricing the three-plus years of opportunity cost, and nobody is pricing the probability-weighted reality that a meaningful slice of creditors will never see a dollar. When you fold jurisdictional exclusion into the recovery math, the effective recovery rate for a mainland creditor is not 119%. It is 119% multiplied by the probability that you clear KYC, clear the legality clause, and clear foreign-exchange controls — a product that can collapse toward zero. The headline number is a distribution, not a promise, and the tail is far fatter than anyone admits.
The market, meanwhile, treats the entire event as a marginal positive: billions of dollars of recovered cash flowing back into crypto is supposed to be dry powder for the next leg up. That thesis quietly assumes the money reaches everyone. It does not. Every creditor who is excluded is a dollar that never re-enters the market, and every creditor forced to sell a claim at a discount is a dollar that leaves at a worse price. The "FTX money is coming back" trade is a story the price has already partly absorbed — and the story is more fragile than the price implies.
And the deepest blind spot of all: we keep treating this as a story about recovery rates when it is really a story about the limits of a US bankruptcy court's reach into a jurisdiction that has decided, as a matter of policy, that crypto money is not welcome. No trustee, no stablecoin, no clever intermediary dissolves that. The law is the wall. The four-month countdown is just the clock running down against it, and the clock does not negotiate.
Takeaway
Watch three things. First, the exact date and the exact item behind the four-month window — it is almost certainly a claim, KYC, or distribution-election deadline, and it is the single most actionable number in this entire saga. Second, whether the trustee's filings define "China" to include Macau, because that one sentence decides which creditors get a door and which get a wall. Third, the secondary-market pricing of mainland claims specifically — if that discount widens while global claims hold near 90%+, the market is telling you, in real time, that the exclusion risk is finally being priced in.
If you hold a claim with a mainland address, the play is not to wait for clarity. Clarity arrives after the deadline. Confirm your status on the official portal today, get cross-border legal advice before the window narrows, and treat every unsolicited "refund" offer as hostile. The case is won. Whether you collect is a different question — and the clock is already running.