If you have watched banks and crypto circle each other for a decade, you already know the usual script. A memorandum of understanding. A sandbox. A quiet burial six months later. This one feels heavier, at least on paper, because the crypto side is no longer a single spot venue. It is a parent company that has spent the past year buying regulated pipes, and the bank side already runs a deposit-tokenization program aimed at collateral and margin. Whether those two facts become a deal is another question. Perhaps the most interesting aspect is how little either side needs a flashy announcement to change the way institutions actually settle.
Why A Custody And Payments Talk Matters Now
Kraken’s parent, Payward, has entered discussions with BNY over a potential partnership covering crypto and payments, according to two people familiar with the matter. The proposed work could run through Payward Services, the Wyoming platform built for banks, exchanges and asset managers. One of those people said parts of the conversation resemble the infrastructure piece of Payward’s recent arrangement with a major exchange operator. Talks are ongoing. There is no guarantee they close.
That last sentence should sit on the desk of anyone tempted to trade the headline. Private negotiations leak for all sorts of reasons. Sometimes a banker wants a marker in the market. Sometimes a founder wants a valuation story. Sometimes both sides are simply testing whether compliance teams can stand in the same room. I have found that the useful question is rarely “will they announce?” It is “what would have to be true for this to be more than a tour of each other’s slide decks?”
On the banking side, BNY is the custody, asset-servicing, clearing and wealth name that institutions already use when they do not want to explain the counterparty. Its parent is The Bank of New York Mellon Corporation, listed in New York under the ticker BK. Its digital cash program already touches institutional clients and market-infrastructure firms. On the crypto side, Payward’s business-to-business shelf includes spot trading, derivatives, tokenized equities, custody, staking, payments and traditional securities. The overlap is not subtle. It is the whole point.
Six Lines Of Business, Not One Product
The people describing the talks did not frame them as a single coin listing or a white-label wallet. They named a spread of work.
- Crypto products aimed at clients who already sit inside a bank relationship
- Custody, where the argument is less about novelty and more about who holds the keys and the liability
- Wealth management, which is where tokenized assets either become a sleeve or stay a curiosity
- Trading, including the regulated derivatives stack Payward has been assembling
- Payments, the dull pipe that decides whether a stablecoin is a slogan or a settlement tool
- Financial infrastructure, the unglamorous layer of connectivity, surveillance and post-trade
Read that list twice. A partnership that only covers one of those lines is a vendor contract. A partnership that even tries to cover several is an operating model. Banks hate operating-model changes. They also hate being late when a peer figures out collateral that moves after the cash window closes. That tension is the whole story.
What “No Deal Yet” Actually Protects
Nothing in the account says terms are agreed, exclusivity is granted, or a launch date exists. Speaking anonymously because the negotiations are private, one person described an arrangement that remains under discussion. Good. Early leaks that pretend otherwise tend to age badly.
Still, the shape of the leak is informative. It points at Payward Services rather than the retail app most people associate with the Kraken name. That is a tell. Retail crypto partnerships make headlines. Institutional ones make revenue, and they take longer because every control function gets a vote. If you have ever sat through a bank’s third-party risk review, you know the vote is not a formality.
The Parent, The Brand, And The Pipes
Most readers meet this company through the exchange brand. The legal and strategic actor in these talks is the parent. That distinction matters more than branding trivia. A bank does not integrate a consumer app. It integrates a services entity with a regulatory file, a custody control map, and a way to explain the relationship to examiners.
Payward has spent the past two years making that file thicker. A U.S. derivatives acquisition closed in May after being announced in April for up to $550 million in cash and stock, with final terms undisclosed at closing. The acquired structure includes a designated contract market, a derivatives clearing organization and a futures commission merchant. Those three registrations cover exchange, clearing and brokerage under the commodity regulator’s framework. Separately, the retail futures platform NinjaTrader came in during 2025 for roughly $1.5 billion. A stablecoin payments business, announced around $600 million, closed on July 1. Wallet infrastructure from an asset purchase, not a full company buy, was said to have supported more than 60 million wallets and over $10 billion in stablecoin volume. Terms on that wallet deal were not disclosed.
Put those pieces on one page and the BNY conversation stops looking like a random coffee. It looks like a firm that bought the missing U.S. permissions, then went looking for a balance-sheet partner that already speaks collateral.
A September Marker On Valuation
In September, a venture arm tied to a major U.S. exchange operator agreed to invest $100 million in Payward. Reporting at the time put the valuation near $21 billion, citing people familiar with the transaction. The public description of that arrangement had three parts: the investment itself, further work on equity tokens, and a market-surveillance agreement. Wells Fargo was named as exclusive capital markets adviser on the transaction.
Under that surveillance piece, Payward is expected to adopt the exchange operator’s technology across crypto, equities, tokenized equities, futures and options venues. A digital liquidity networks unit leads the collaboration on tokenized market infrastructure. For the planned equity tokens, both sides have talked about a launch in the second quarter of 2027, covering distribution, trading and post-trade, tied into Payward’s xStocks ecosystem, with the claim that shareholder rights and issuer protections stay intact.
Rails that do not close, with shareholder rights intact.
Payward co-CEO Arjun Sethi, on the planned token infrastructure
I like the line. I also treat it as a design goal, not a delivered system. The exchange operator presented the 2027 date as an expectation and warned that forward-looking statements do not guarantee performance. Regulation and the ability to actually implement strategic initiatives were listed among the factors that could knock the timeline around. Anyone who has watched tokenized stock projects slip a year understands why that caveat is not boilerplate. It is the project.
How The Nasdaq-Style Work Rhymes With BNY
One person close to the BNY discussions said parts of them resemble the infrastructure work in that September agreement. That is the sentence I would underline. Surveillance technology, post-trade connectivity, and a shared language for tokenized instruments are not glamorous. They are also the difference between a demo and something a risk committee will allow near client money.
BNY does not need another retail on-ramp. It needs a way to offer institutional clients digital instruments without inventing a new control stack from scratch. Payward, after the derivatives and payments buys, does not need another press cycle about being “bank friendly.” It needs a counterparty whose name already sits on custody agreements. Each side is shopping for the thing the other already has. That is a healthier starting point than a partnership built on mutual curiosity.
BNY’s Tokenized Deposits, Already In Motion
In a January announcement, BNY described the first step of its deposit-tokenization program: blockchain records that mirror participating clients’ existing deposit balances. The starting use case was collateral and margin. The digital entries represent clients’ demand-deposit claims against the bank. The system runs on a private, permissioned chain, while balances continue to appear in traditional records for regulatory and reporting purposes. Existing risk, compliance and control frameworks govern the capability. Near-real-time, rules-based cash movement was framed as a later objective, not a day-one feature.
That design choice is easy to mock if you live on public chains. It is also the only design most custody banks will ship. A private ledger that mirrors a deposit is not a new currency. It is a faster index of a claim the client already has. For margin workflows, speed matters more than ideology. A clearinghouse that can see cash move closer to real time has a different conversation about intraday risk than one that waits on batch files.
The January notes included a statement from ICE’s clearing leadership about work toward supporting tokenized deposits across its clearinghouses, with an eye on 24-hour trading and tokenized collateral, while staying in touch with clearing members. Other firms, including Citadel Securities, Invesco and WisdomTree, commented on the digital cash capability. Those comments sit apart from the Payward talks. They still sketch the neighborhood. BNY is already collecting institutional names around tokenized cash. A crypto-market partner would be an extension of that neighborhood, not a detour into a different city.
Collateral Is The Quiet Prize
Payments get the headlines because everyone understands a transfer. Collateral is where the money actually sits still and earns its keep. If a bank can represent a deposit as a blockchain record and a trading venue can accept that record against margin, you have shortened a loop that currently involves cut-off times, correspondent banks and a lot of phone calls. If you cannot, you have a nice diagram.
I’ve found that pilots die at the legal description of the claim, not at the hash function. Is the token the deposit, or a pointer to the deposit? Who is the creditor in insolvency? What happens if the chain halts and the core ledger does not? BNY’s January framing, balances still living in traditional records, is an attempt to answer those questions before a lawyer asks them in a worse mood. Any Payward arrangement would have to live with that answer, not overwrite it.
| Workstream | What Payward brings | What BNY already does | Hard part |
| Custody | Crypto and tokenized-asset operations | Institutional asset servicing | Key control and liability split |
| Trading | Spot, derivatives, surveillance tie-in | Clearing and client access | Venue rules and best execution |
| Payments | Stablecoin payment services | Deposit and cash management | Redeemability and cut-off times |
| Wealth | Tokenized equities roadmap | Wealth and asset servicing | Suitability and issuer rights |
| Infrastructure | B2B connectivity for banks | Market plumbing and digital cash | Permissioned versus open rails |
The table is a sketch, not a term sheet. It is useful because it shows why a six-area conversation can stall even when both chief executives like the idea. Each row has a different regulator, a different client promise, and a different way to lose money.
Numbers From The Summer, Read Carefully
By mid-August, second-quarter figures confirmed the Bitnomial close on May 1 and the payments close on July 1. Adjusted revenue was reported at $508 million, up 17 percent year over year. Adjusted EBITDA was $23 million. Platform transaction volume fell 18 percent to $310 billion. Asset-based and other revenue made up 60 percent of the total, against 55 percent a year earlier.
Volume down, asset-based share up. That mix shift is the sentence hidden inside the press numbers. A business that lives only on trading fees feels every quiet month. A business that also earns from balances, staking, custody and payments can talk to a bank without pretending the next bull market is the business plan. Twenty-three million of adjusted EBITDA on half a billion of adjusted revenue is not a software-margin fairy tale. It is a market-structure company still spending to bolt on licenses. Banks notice that. They would rather see a partner that knows its cost base than one selling a story about infinite operating leverage.
Derivatives As The Adult Table
The May close on the U.S. derivatives business is easy to skip if you only trade spot. Do not skip it. A designated contract market, a clearing organization and a futures commission merchant are the three doors U.S. institutions actually recognize. Payward said it would connect that stack across Kraken, NinjaTrader and Payward Services, so banks, brokerages and payment firms could reach regulated U.S. crypto derivatives through one integration.
“One integration” is a sales sentence. The operational version is messier. Clearing members care about default waterfalls. FCMs care about customer-fund segregation. A bank wealth desk cares about whether the product is even allowed in the mandate. Still, without those registrations, the BNY conversation would be about information sharing. With them, it can be about order flow and margin. That is a different meeting.
Payments, Wallets, And The Boring Rail
The July payments acquisition and the late-July wallet asset purchase point at the other half of the rumored agenda. Stablecoin volume is not interesting because it is large in a press release. It is interesting because treasurers have started to ask whether a dollar claim on a chain can sit next to a wire in the same cash forecast. Most of the time the answer is still no, for policy reasons rather than technical ones. A bank partnership is one of the few ways that answer changes, because the bank already owns the cash-forecast relationship.
Wallets are the unloved cousin in that story. Sixty million of them sounds like a consumer metric. For an institutional pitch, the useful part is the plumbing: key management, recovery, policy engines, the dull ability to say who approved a transfer. Banks will not white-label a retail wallet and call it a day. They might license controls. There is a difference, and it shows up in the contract’s limitation of liability, not in the demo.
What A Real Agreement Would Have To Cover
If I were sitting in the room with a pen, I would want five things written down before anyone used the word partnership in public.
- Who is the custodian of record for each asset type, and who pays if keys or instructions fail
- Which client segments are in scope on day one, and which are explicitly out
- How tokenized deposits, if used at all, map to margin without creating a new unsecured exposure
- What surveillance and reporting each side owes the other, and to which regulator
- An exit ramp that does not strand client assets if the commercial romance ends
That fifth item is the one founders dislike and banks insist on. Institutional relationships are long. They are also reversible. A clean unwind is a feature, not a lack of commitment. The equity-token language about shareholder rights intact is the same instinct applied to issuers. Rights that only exist while a vendor is happy are not rights. They are a service level.
Regulation Is The Weather, Not The Footnote
Any U.S. arrangement in this neighborhood now lives under a split sky. Commodity regulators cover a growing slice of crypto market structure, including the derivatives permissions Payward bought. Securities regulators still sit on anything that looks like a fund, a tokenized share, or a custody promise to investment advisers. Bank regulators sit on the deposit itself. A private chain that mirrors a deposit does not retire those seats. It gives them a new diagram to mark up.
The equity-token target of the second quarter of 2027 is a useful reminder that calendars slip when permissions slip. Market-surveillance technology can be licensed faster than a new product can be blessed. If the BNY talks lean on infrastructure first and products later, they are copying a sequence that already survived one public announcement. If they try to launch a client-facing crypto shelf before the control map is boring, they are choosing the sequence that usually ends in a delayed blog post.
I do not think the interesting regulatory question is whether crypto is “allowed.” In the U.S. it is, inside a structure that is narrower than marketing sites imply. The interesting question is which license carries the client, and which affiliate is allowed to touch the cash. Payward’s recent buying spree is an attempt to answer that with entities rather than with opinions. BNY has been answering it with entities for a very long time.
A Note On Competition, Without The Cheerleading
Other custody banks have run digital-asset pilots. Other exchanges have signed bank distribution deals that later shrank to a co-branded page. Pretending this conversation is unique would be lazy. What is less common is the combination of a fresh $100 million strategic check, a 2027 tokenized-equity timetable, a just-closed U.S. clearing stack, and a bank that has already put tokenized deposits into a collateral conversation with market infrastructure firms.
Could a rival bank get there first with a different crypto partner? Of course. Could Payward decide the integration cost is higher than the distribution is worth? Also yes. The leak does not remove those branches. It tells you both sides thought the meeting was worth having after the September deal, not before it. Sequence is a kind of evidence.
Wealth Desks And The Sleeve Problem
Wealth management showed up on the list of possible workstreams, and it is the easiest place to overpromise. A tokenized equity that preserves issuer rights is, in theory, something a wealth platform can hold next to an ordinary share. In practice, advisers need a research process, a fee schedule, a tax story and a way to explain why the wrapper is not just a more expensive version of the thing the client already owns.
The 24-hour rail is the sales hook. The rights language is the compliance hook. Both have to be true at once, which is harder than it sounds. A share that trades on a weekend but cannot vote, or cannot be delivered back into the traditional transfer-agent world, is a different instrument. Sethi’s line about rails that do not close only works if the legal rights survive the weekend too. BNY’s asset-servicing muscle is relevant here in a way a pure trading venue’s is not. Servicing is how corporate actions happen when nobody is tweeting about them.
Surveillance As The Unsexy Glue
The September agreement’s surveillance component deserves more attention than the valuation line. Cross-venue manipulation is the objection every serious market regulator reaches for when crypto and equities share a plumbing diagram. Adopting an incumbent surveillance stack across crypto, equities, tokenized equities, futures and options is an attempt to answer that objection with a vendor the regulators already know.
If BNY discussions really do rhyme with that infrastructure work, expect surveillance, reporting and audit rights to show up early in the paper. Banks do not bolt a new market onto client flow and hope the alerts are fine. They ask who sees the alert, how fast, and who has the authority to halt a client. Glamorous? No. Disqualifying if missing? Yes.
A practical read of the talks: Distribution without controls is a headline. Controls without distribution are a cost center. The deal, if it exists, has to be both.
What Clients Would Actually Notice
Most end clients will notice nothing for a long while, and that is not an insult. The first visible change, if any, would likely be a collateral option inside an institutional workflow, or a payments rail offered to a treasury client who already banks at BNY. Retail users of the exchange brand should not expect a new button next week. The people describing these talks pointed at services for banks, exchanges and asset managers. That is a wholesale sentence.
Wholesale is where crypto either grows up or stays a parallel hobby. I say that without romance. Parallel hobbies can be profitable. They just do not reshape how margin is posted at a clearinghouse. The January deposit program, aimed first at collateral and margin, is a bank saying the same thing from the other direction.
Risks That Should Stay On The Page
A fair write-up has to leave room for the miss. Negotiations can end because of price, because of exclusivity demands, because a regulator asks a question nobody wants in writing, or because one side’s board decides the reputational tail is fatter than the revenue. Crypto firms still carry enforcement history in the minds of bank directors, even when the specific entity at the table has spent heavily on licenses. Banks carry their own history of digital projects that never left the innovation lab.
There is also a strategic risk that gets less airtime. A too-tight partnership can cap a crypto firm’s other bank conversations. A too-loose one gives the bank optionality and the crypto firm a logo with no flow. The people familiar with this matter have not described exclusivity. Until someone does, assume both sides are keeping other doors ajar. That is normal. It is also why “explores” is the correct verb.
How To Read The Next Headline
If a memorandum appears, look for scope, not adjectives. Does it name custody of a defined asset list, or “explore digital opportunities”? Does it mention payments corridors, or “collaboration on innovation”? Does Payward Services appear as the contracting party? Is there a client segment, a geography, and a review date? Vague language is not always bad faith. It is often a sign that legal has not finished.
If nothing appears, do not treat the leak as fiction. Plenty of real work never earns a press release, especially when the first product is a collateral workflow that clients would rather not see described in public. Silence can mean the deal died. It can also mean the deal became operational and boring. Boring is the compliment, in this corner of finance.
A Longer View On Bank Rails
Step back from the two names and the pattern is familiar. Market structure firms want weekend and overnight rails. Banks want those rails to terminate in a deposit they already understand. Crypto firms want the deposit’s distribution. The argument of the past few years was about whether those wants could share a ledger. The argument now is about whose ledger, whose license, and whose client owns the relationship when something breaks at 2 a.m.
Payward’s answer, visible in the acquisition list, is to own more of the licensed stack and rent the balance sheet. BNY’s answer, visible in the January program, is to keep the deposit where it is and let a permissioned record move faster for collateral. A partnership would be those two answers sharing a workflow. It would not require anyone to agree on the metaphysics of tokens. That, frankly, is why it might be buildable.
The useful test of a bank-crypto deal is not the launch photo. It is whether margin can move on a Sunday without anyone inventing a new kind of money to explain it.
What The Valuation Does And Does Not Say
A reported $21 billion mark, attached to a $100 million check, tells you a strategic investor was willing to pay up for access and for a seat near tokenized equities. It does not tell you BNY will sign, or at what price, or whether Payward’s adjusted earnings can carry the integration cost. Strategic checks are often smaller than the headline valuation implies, precisely because the value is supposed to arrive later through distribution and product rights. Treat the figure as context for bargaining power, not as a forecast of the BNY outcome.
Bargaining power still matters. A firm that just sold a slice to an exchange operator, and that can point at a clearinghouse registration, walks into a bank meeting differently than a firm asking for its first institutional logo. BNY walks in with a deposit program already commented on by market-makers and asset managers. Neither side is auditioning from zero. That raises the odds of a serious term sheet. It also raises the odds of a serious disagreement, because both sides have alternatives.
Tokenized Equities And The 2027 Clock
The equity-token plan is not the BNY story, but it shares staff, vendors and political capital. Distribution, trading and post-trade, connected to an existing tokenized-stock ecosystem, with issuer rights preserved, is a large surface area. A bank wealth and custody partner would make that surface easier to explain to traditional holders. It would also import the bank’s questions about corporate actions, proxy, withholding tax and account structure. Those questions sink more tokenization projects than throughput ever does.
So if you hear that the BNY talks “include infrastructure,” translate it. Infrastructure here means the boring rights and the boring files, not a faster block time. The 2027 expectation already carries a caution that regulation and execution can slip. Adding a global custody bank to the diagram does not shrink that caution. It gives the caution a better address.
A Reader’s Checklist
You do not need a terminal seat to follow this. You need a short list and a little patience.
- Watch whether any future note names Payward Services, not only the retail brand
- Separate BNY’s existing digital cash comments from anything that actually cites this negotiation
- Treat adjusted revenue and EBITDA as mix indicators, not as a victory lap
- Ask, if products are named, whether collateral, payments or wealth is first
- Discount any date that is not tied to a regulator or a clearing member rollout
- Remember that a failed talk can still signal where both firms think the market is going
That last point is easy to forget in a headline culture. Even a walk-away teaches you something. If a custody bank and a newly licensed derivatives group cannot agree on margin language, the industry’s 24-hour story is further away than the keynotes suggest. If they can, a lot of smaller pilots start to look like rehearsals.
Where I Land, For Now
I do not know whether these talks become a contract. Nobody outside the room does, and the room has already said there is no guarantee. What I do think is that the ingredient list is more serious than the usual partnership rumor. A bank with a live deposit-tokenization step aimed at collateral. A crypto parent with fresh U.S. derivatives permissions, a payments close, a wallet infrastructure buy, and a strategic investor tied to equity-token infrastructure. A services entity built to face banks rather than retail sign-ups. Six named areas instead of one mascot product.
That is not a prediction of success. It is a reason to keep the file open. The next useful information will be narrow: a contracting entity, a use case, a control split, maybe a pilot client category. Until then, the grown-up posture is the one the leak itself adopted. Discussions. Possible services. No promise. In market structure, that is often how the real work sounds before it either disappears or becomes plumbing nobody bothers to tweet about.
If you work allocations, the practical takeaway is simpler than the rumor. Institutional crypto is arguing, in public and in private, about who holds collateral and who moves cash after hours. Payward wants to be the market pipe. BNY already is the cash pipe for a huge set of clients. A meeting between those two roles was always going to happen somewhere. It appears to be happening here, unfinished, and that unfinished quality is the part worth trusting.
]]>
