Anchorage Digital Layoffs Reveal Crypto Downturn Strain

I keep coming back to a number that does not look dramatic until you sit with it. Seventeen percent. Not a rounding error, not a quiet trim around the edges, and not the kind of cut a company makes when the only problem is a bloated marketing team. If the headcount Anchorage Digital described earlier this year still held, that slice works out to something like sixty-eight jobs. People who went home one week with a regulated bank on their résumé and a crypto market that had already spent months refusing to cooperate.

Reports this week say chief executive Nathan McCauley told employees about the reduction against a long, grinding downturn. Bitcoin did manage a brief climb back above $87,000 on Friday. That sounds almost comforting until you remember the October 2025 peak near $126,000. A bounce is not a recovery. I have found that firms in this industry talk about cycles the way sailors talk about weather. Everyone claims they prepared for the storm. Fewer admit they are already taking on water.

Why a Regulated Crypto Bank Is Cutting Staff Now

Anchorage is not a trading app with a logo and a prayer. It sits in a narrower, more serious corner of the market: custody, staking, governance, settlement, and stablecoin issuance, wrapped in a national trust bank charter. That status is supposed to be the moat. Clients pay for supervision, segregated accounts, and the boring machinery of compliance. So a workforce cut of this size, arriving months after fresh capital and a reported $4.2 billion valuation, feels less like a startup growing pain and more like a stress test of the whole institutional story.

Perhaps the most interesting part is the sequence. Capital arrived. The valuation was talked about in large numbers. Product lines kept expanding, from cross-chain stablecoin rails to custody for tokenized uranium. Then the staff reduction. Growth and contraction in the same year is not impossible. It is just uncomfortable. Companies do it when revenue lags the story they sold, or when the cost of being regulated turns out higher than the pitch deck suggested.

What the Reported Cut Actually Implies

In congressional testimony in February, McCauley put the global workforce at about 400 people. Apply 17 percent to that figure and you land near 68 roles. That math only holds if headcount did not swell or shrink in the months since. It is a company-wide estimate, not a confirmed count of U.S. losses. Still, for a firm of that size, losing close to one in six colleagues changes the temperature in the building.

Layoffs at that scale usually hit more than one desk. Engineering might lose people who kept issuance systems honest. Compliance might lose the reviewers who slow a launch down on purpose. Client coverage might thin out just as institutions ask harder questions about reserves and operational controls. I would not assume the cut was cosmetic. Seventeen percent rarely is.

• About 400 employees were cited in February testimony, so 17 percent implies roughly 68 roles if that base held.
• The figure is global, which means the pain may not map cleanly onto one office or one country.
• The reduction was reportedly communicated by the chief executive this week, not leaked as a rumor and then denied.
• Weak crypto markets were the stated backdrop, even after a large equity check earlier in the year.

There is a human lag in these stories that market commentary skips. The people leaving are often the ones who built the controls investors later praise. A thinner team can still issue a token. It is less obvious that it can keep every attestation, every wallet policy, and every client exception as tight as before. That is the part I watch.

Capital Arrived, Then the Headcount Fell

Tether announced a $100 million strategic equity investment on February 5, framed as a deepening of an existing relationship rather than a rescue. Earlier coverage of the round landed a couple of days later. The check was real money by any normal standard. In crypto, though, $100 million can look modest next to a $4.2 billion valuation, and modest next to the cost of running a chartered bank through a sour market.

Fresh equity does not cancel a downturn. It only changes how long a firm can pretend the downturn is someone else’s problem.

Tether already worked with Anchorage on USAT, its U.S.-focused stablecoin. The investment announcement pointed to custody, staking, governance, settlement, and issuance as the services that mattered. Experience with the bank’s compliance stack on that token was cited as part of the reason for writing the check. In plain terms, the investor bought a stake in the issuer of its own domestic product. That alignment can be a strength. It can also concentrate risk if the product, the bank, and the market all wobble together.

I have a slightly unfashionable view here. Strategic investors sometimes fund the story they need, not the cost structure the company has. A stablecoin sponsor wants a credible U.S. issuer. A chartered bank wants a flagship token that proves the charter is useful. Neither motive guarantees that 400 people are the right number when trading volumes fade and clients delay mandates.

USAT and the Domestic Stablecoin Bet

USAT entered the market on January 27. The pitch was a dollar-backed token built for the federal stablecoin framework under the GENIUS Act. Anchorage Digital Bank, N.A. was named the legal issuer, a distinction worth keeping straight. The bank issues. Tether supports the product and, after February, holds equity in the company behind that issuer. Those are related facts, not the same fact.

The launch language was careful in the way regulated products have to be careful. USAT is not legal tender. It is not backed or guaranteed by the U.S. government. It does not carry FDIC or SIPC insurance. Anyone who has watched retail users treat a stablecoin like a savings account should read those lines twice. The token can be well reserved and still not be a bank deposit. That gap is the whole point of the disclaimer, and it is easy to skim past.

Cantor Fitzgerald was named reserve custodian and preferred primary dealer. The first phase of distribution included Bybit, Crypto.com, Kraken, OKX, and MoonPay. A respectable list, on paper. Distribution is not the same as durable demand. A token can be listed everywhere and still sit quietly if traders prefer the incumbents they already trust with size.

January’s first reserve attestation, as described in later coverage of the bank’s reporting, showed about 17.5 million redeemable tokens outstanding and roughly $17.6 million in supporting reserve assets. That is a small book next to the giants of the stablecoin market. Small can be fine at launch. Small also means the fixed cost of a chartered issuer is being carried by a product that has not yet proved it can scale. Layoffs do not prove the product failed. They do suggest the parent could not wait for scale on the old cost base.

A Quick Map of the Year So Far

Read that table from the bottom up and the mood changes. The product news looks busy. The labor news looks tight. Busy and tight can coexist, but only for a while. Eventually one of them wins.

Cross-Chain Rails and the LayerZero Arrangement

On September 22, coverage detailed a partnership announced the day before. LayerZero said it would supply cross-chain infrastructure for stablecoins issued through Anchorage’s banking platform, with USAT named as the first token on that arrangement. The technology was described as the preferred interoperability layer for eligible stablecoins from the bank. Anchorage keeps regulated issuance. LayerZero connects deployments across supported networks.

The omnichain fungible token standard is the technical promise here. An issuer can aim for one unified supply across several chains, keep control of contracts, pick the networks, and set security assumptions for cross-chain messages. That is attractive if you believe institutions will not tolerate a different token address, and a different operational story, on every chain they touch. It is also another system to supervise. Interoperability failures are not theoretical. Bridges and messaging layers have a scarred history, even when the marketing says this version is different.

LayerZero’s infrastructure is said to reach more than 170 networks. Reach is not the same as a confirmed deployment of each stablecoin on each network. That distinction matters, and it is the kind of nuance that gets flattened in launch posts. A preferred rail is a commercial choice. It is not proof that every client asset already moves everywhere.

Other names in the issuance portfolio, mentioned alongside the September reporting, included Western Union’s USDPT, OSL Group’s USDGO, and Falcon Finance’s fUSD. A roster like that suggests Anchorage wants to be the regulated factory, not a one-token shop. Factories need volume. If several issuers are still early, the factory’s fixed costs show up in the headcount line before they show up in fee income. That, to me, is a plausible bridge between September’s expansion headlines and October’s cuts.

Tokenized Uranium and the Etherlink Add-On

A separate September 17 report covered custody support for seven assets on Etherlink, a Tezos layer 2. Institutional clients could hold them in segregated accounts at the bank, under existing policies rather than a brand-new setup. The list included xU3O8, wrapped XTZ, stXTZ, USDT, USDC, USDSM, and wrapped ether.

xU3O8 is the odd one, and the interesting one. It represents physical uranium through Uranium.io. Anchorage framed the addition as a way for institutions to hold tokenized commodity exposure inside the same custody stack they already use for digital assets. I like the ambition. I also think commodity tokens drag a different set of questions into a crypto custody conversation: storage of the underlying, redemption mechanics, and whether a bank’s digital-asset controls really map onto a metal that does not live on a chain.

Expansion into that territory while cutting staff is not automatically contradictory. A firm can drop experimental side projects and still add one flagship asset. It can also announce coverage that a smaller team must now operate. The risk is quiet. Support looks broad on a product page, then response times slip, and the client who wanted uranium exposure next to stablecoins discovers the operation is thinner than the press note implied.

The Charter, the OCC, and the Operating Agreement

The regulatory origin story goes back to January 2021, when the Office of the Comptroller of the Currency approved Anchorage Trust Company’s conversion from a South Dakota trust company into a national trust bank. The approval came with an enforceable operating agreement covering capital, liquidity, and risk management. That agreement was the price of the charter. It was also a public reminder that this was not a lightly supervised experiment.

February 2026 OCC records listed the termination of that original January 2021 operating agreement. Termination can mean the bank graduated out of a specific constraint. It can also mean the supervisory relationship simply moved to a different set of expectations. Outsiders should not romanticize either reading. What remains is the charter itself, and the obligation to run issuance and custody as a bank, not as a growth-stage app.

Charters cut both ways in a downturn. They attract clients who will not touch an unregulated custodian. They also impose costs that do not flex down as fast as trading revenue. Compliance, model risk, audit, and liquidity management are not weekend projects. If markets stay soft, the charter becomes both the reason clients stay and the reason the payroll feels heavy. I suspect that tension sits somewhere under this week’s news, even if nobody put it in a slide.

A charter is a promise to supervisors and a bill to shareholders. In a cold market, both arrive on time.

Market observation, not a company statement

Bitcoin’s Bounce Does Not Pay the Payroll

Friday’s move back above $87,000 will get screenshots. It should. Traders live on those levels. Operators live on something duller: mandates signed, assets that actually arrive, and fees that clear. A price still more than 30 percent under the October 2025 high does not refill an institutional pipeline by itself. Custody revenue often trails price, because committees move slower than charts.

There is a habit in this market of treating any green week as the end of the story. I do not buy it. The firms cutting staff now are responding to the last several quarters, not to Friday afternoon. If bitcoin holds and volumes return, some of these cuts will look early. If the bounce fades, they will look late. Either way, the decision was made against a backdrop that still qualifies as a downturn, not a victory lap.

Other crypto employers have been making similar noises. Reports in September described Bitcoin Suisse weighing as many as 60 Swiss roles in an overhaul. Different firm, different country, same weather. When more than one custody and brokerage name reaches for the same lever, it stops being a single management story and starts looking like an industry margin story.

How Institutional Clients Should Read the Cut

If you custody with a firm that just reduced staff, the useful questions are operational, not theatrical. Who still owns key ceremonies? Who signs off on reserve attestations? Did the client-coverage team shrink, or was the cut concentrated in projects that never shipped? A 17 percent reduction can be a healthy reset if the remaining bench is deep. It can be a warning if the same people were already covering two jobs.

1. Ask which functions were reduced and which were explicitly protected, especially compliance, custody operations, and issuance controls.
2. Review incident response and key-management coverage for nights, weekends, and chain-specific exceptions.
3. Check whether attestation timelines for issued stablecoins stay on the published cadence.
4. Confirm that segregated-account terms did not quietly change with the headcount.
5. Treat new asset listings, including commodity tokens, as services that must be staffed, not just announced.

None of that is disloyal to the firm. It is what a fiduciary does when a vendor changes shape. Anchorage’s pitch has always been supervision and process. Clients who bought that pitch should test it now, while the process is under strain, not after something breaks.

Valuation Versus the Cost of Waiting

A $4.2 billion valuation earlier this year set expectations. Valuations are opinions with term sheets attached. They assume growth, retention, and a market that does not stay hostile. When the market stays hostile, management has two honest options: spend the new capital to wait, or cut costs so the wait is shorter. The reported layoffs suggest the second option won, at least in part.

That choice does not erase the Tether investment. It contextualizes it. One hundred million dollars buys time. It does not buy a bull market. Investors who entered at a strategic premium may care more about the USAT relationship than about near-term headcount. Employees do not have that luxury. Clients sit in between, hoping the time purchased is spent on controls rather than on another announcement.

A simple way to hold the story:
  Capital in: $100 million strategic equity
  Valuation talked about: $4.2 billion
  Workforce base: about 400
  Reported cut: 17 percent
  Implied roles: about 68
  Market: still well below the 2025 bitcoin peak

I keep those figures together because narrative tends to separate them. The funding story and the layoff story are one story. Pretending otherwise is how this industry keeps surprising itself.

What Stablecoin Issuers Learn From a Thinner Team

Issuance looks clean from the outside. A token, a reserve, a monthly attestation, a redemption window. Inside, it is reconciliation, banking rails, exception handling, and a compliance review every time someone wants a new chain. Cross-chain messaging adds another review. Commodity tokens add another. Each addition is reasonable alone. Stacked, they demand people.

If Anchorage is slimming down while keeping USAT, the LayerZero preference, and the Etherlink list, prioritization becomes the real strategy. Something will move slower. Maybe it is the next chain. Maybe it is a new issuer in the portfolio. Maybe it is the speed of client onboarding. Slower is acceptable. Silent failure is not. The firms that handle downturns well say out loud what they will not do for a while.

There is also a competitive angle. Other custodians and issuers will read this week as an opening. Some will pitch continuity. Some will pitch price. A few will pitch both and deliver neither. Anchorage’s remaining advantage is still the charter and the habit of publishing reserve figures. Those advantages survive a layoff only if the publishing stays boring and on time. Boring is the product.

Regulation Did Not Pause for the Market

The GENIUS Act framework is the policy backdrop for USAT, and it does not care that bitcoin is off its highs. Rules arrive on their own calendar. A domestic stablecoin built for that framework still needs issuer discipline, reserve custody, and disclosures that a retail user can misunderstand in one scroll. The disclaimers about legal tender, government backing, and the absence of FDIC or SIPC insurance are not footnotes. They are the product boundary.

I have watched too many cycles where a token’s legal status gets softer in conversation than it is on the page. If you hold USAT, you hold a dollar-backed claim structure described by its issuer and sponsor, not a Treasury bill and not a bank deposit. The reserve custodian role assigned to Cantor Fitzgerald is meant to make that claim more credible. Credibility still depends on operations continuing after the org chart shrinks.

Broader U.S. crypto policy keeps moving in parallel, from market-structure fights to custody questions for funds. A chartered trust bank is exposed to all of it. Layoffs do not remove that exposure. They reduce the number of people available to answer a supervisor’s letter. That is a less glamorous risk than price, and probably the more serious one.

A Pattern, Not a One-Off Headline

Crypto winters do not announce themselves with a single print. They show up as delayed deals, quieter prime-brokerage desks, and then, eventually, headcount. Anchorage’s reported cut fits that pattern more than it breaks it. The unusual piece is the combination of a fresh strategic check, a high valuation, and a still-expanding product menu. Usually the cuts come after the announcements stop. Here they arrived while the announcements were still warm.

That mix can be read generously. Management raised money, built the rails, and is now fitting the team to reality. It can also be read less generously. The story ran ahead of the revenue, and labor is the line that can be changed fastest. Both readings can be partly true. I lean toward the second only because seventeen percent is a large adjective for a company that wants to look unhurried.

Employees in this sector already know the script. Equity that might be worth something in the next cycle. A Slack message. A calendar invite that ruins a Thursday. The industry’s memory for those Thursdays is short once prices rise. It should not be. The next bull market will hire some of the same people back at a premium and call it culture.

What Would Change My Mind

A few developments would make this week’s cut look like discipline rather than strain. USAT outstanding supply moving well beyond the January attestation range, with reserves that still match. A clear statement that issuance, custody operations, and compliance were insulated from the reduction. Evidence that the LayerZero arrangement is live for more than a press description, on networks clients actually use. And a market that does more than tag $87,000 for an afternoon.

Absent those, the sober take is simpler. A regulated crypto bank took strategic money from the sponsor of its domestic stablecoin, carried a multibillion valuation into a weak tape, kept adding infrastructure and assets, and still decided that roughly one in six roles had to go. That is allowed. It is also information.

The Part the Market Will Underprice

Price chatter will dominate the next few days. It always does. The underpriced piece is operational continuity at the exact firms institutions use as a reason to stay in the asset class. If custody and issuance desks quietly thin out across several brands at once, the next stress event has fewer practiced hands. Tokenized uranium, omnichain stablecoins, and national trust charters are all real projects. They are also projects that fail in boring ways: a missed attestation, a slow redemption, a key process that lived in one person’s head.

I do not think Anchorage’s charter is decorative, and I do not think a $100 million check was a stunt. I do think labor is where strategy becomes visible. Cutting 17 percent while talking expansion is a choice about what the firm believes it can still do well. Clients, co-issuers, and the strategic investor will find out whether that belief was accurate. The rest of us will find out on a lag, which is how this market usually delivers its lessons.

Until then, treat the headline as a data point in a longer winter, not as a verdict on one bank. Seventeen percent is specific enough to respect. The open question is whether the remaining eighty-three percent is built for the market that exists, or for the one everyone is still waiting to get back.