What it still lacks is the infrastructure that makes those products work at institutional scale: deep balance sheets, trusted counterparties, clear capital treatment, standardised lending documentation, liquid collateral markets and risk models that banks can actually underwrite.
That tension ran through the prime brokerage and lending panel at Crypto Finance Forum London, The panel moved quickly from lending rates and balance sheet constraints to tokenized collateral, DeFi risk, vaults and the role traditional banks may play as they move deeper into spot crypto.
Right now, balance sheet for large-scale crypto lending is coming almost entirely from crypto-native firms. They're doing the work, but there's a size problem. Once you get into the hundreds of millions, counterparty limits kick in and options dry up fast. The panel broadly agreed that traditional banks are the natural long-term answer, but the journey to get there is more structured than most people appreciate.
Daniel Elsawey framed the sequencing clearly, sketching out the journey banks must travel before they can offer serious credit products: first comes spot trading, then post-trade netting and end-of-day settlement, and only then does collateralized lending and leverage become viable. Without that foundation, risk teams have nothing to anchor their models to.
"When you get into the 100 million-plus range, you're tapped out from a counterparty risk standpoint - and that's really when you need the large sell-side banks to come in."
– Daniel Elsawey, Bitpanda
Miles added a regulatory dimension that underscores just how early-stage the framework still is. Under current Swiss Federal law, crypto held on a bank's balance sheet attracts an 800% risk-weighted asset charge. Across Europe and in markets around the world, regulators are still working out how to treat digital assets, and many landed on conservative numbers while that process plays out. AMINA's response is to sweep as much as legally possible off-balance-sheet, but the broader point is that the regulatory environment hasn't caught up with the asset class yet - and that mismatch is a big part of why the economics of crypto-backed lending look so different from traditional lending.
"A lot of regulators and frameworks applied very high percentages while they were still getting to grips with the asset - and that's a big part of why we see such a difference between how crypto collateral and ETF collateral are treated today."
– Myles Harrison, Amina Group
The implication for the industry is significant: until regulatory capital treatment improves - or until banks have enough spot trading experience to feel comfortable - the balance sheet needed to run a serious prime brokerage operation simply won't materialize from the traditional side. Crypto-native players will continue to fill the gap, but with real limits on how far they can take it.
Borrowing against spot Bitcoin costs roughly 200–300 basis points more than borrowing against a Bitcoin ETF, even though the underlying asset and its volatility are identical.
What's actually driving the premium is infrastructure, legal precedent, and decades of accumulated trust in the traditional securities lending market.
The ETF market has decades of court cases and standardized documentation behind it. Spot Bitcoin lending is still bespoke at every step. And until that changes, lenders will keep charging for the complexity.
"You build financial trust over years or decades. Borrowing against shares is a commoditized product. Borrowing against Bitcoin is still bespoke - there is no unified standard built through 40 years of court cases the way there is in traditional markets."
– Al Turnbull, Blockchain.com
James added another reason the gap persists: the market still lacks enough firms and operators who can move fluently between the crypto-native and traditional finance rails. In his view, spot Bitcoin lending and ETF-based financing may reference the same underlying asset, but they sit inside very different operating environments.
“The key thing you don’t have is people who can switch-hit between the two rails. The ones who can do both make all the money. But that fluency is still very rare.”
– James Harris, Tesseract
Until that fluency improves, the funding-rate delta is unlikely to disappear entirely. The cost is not just a volatility premium; it reflects the additional complexity, legal ambiguity and operational risk that still exists on the native crypto side.
Today's tokenized assets (T-bills, bonds, money market instruments) are largely representations of off-chain assets, not natively issued on-chain instruments. The underlying security still sits with a custodian. You still have to wait for market hours to settle. The secondary market for tokenized T-bills is relatively small. Strip away the wrapper and very little has changed.
Myles Harrison described the Catch-22 his team encounters daily:
"We have clients with T-bills posted as collateral, and we ask: do you want to use a tokenized T-bill product? And the question is: why? We really struggle on both sides of that argument to answer that right now."
– Myles Harrison, Amina Group
The panel agreed that the industry needs to get to natively issued tokenized assets, not representations. Until that happens, the efficiency gains that tokenization promises (real-time settlement, collateral mobility, 24/7 markets) will remain largely theoretical.
Daniel pointed to where the real opportunity lies in the interim: distribution. A bank that tokenizes its debt and lists it on a platform like Bitpanda suddenly has access to millions of retail users seeking yield - a customer base it has never historically reached.
"If a bank tokenizes some debt and lists it on Kraken or Bitpanda, with millions of retail users looking for yield but issued by a bank - this is a completely different user network than the bank normally has access to."
– Daniel Elsawey, Bitpanda
That also means tokenized collateral depends on more than the asset side alone. Myles argued that stablecoins are the necessary first layer, because there is limited value in tokenizing assets if settlement still has to move back off-chain. In his view, the sequence is stablecoins first, then tokenization, and only then lending markets against tokenized collateral.
Much of DeFi's instability traces back to a single structural problem: bridges. When an asset moves from one blockchain to another, it crosses a bridge. The bridge holds collateral on one side and issues a synthetic representation on the other. DeFi markets have historically treated these two assets as equivalent, pricing them one-for-one, despite the post-bridge asset carrying meaningfully higher security risk.
The consequences of that kind of mispricing have sometimes been catastrophic: exploits such as Wormhole, Nomad and the Kelp DAO incident exposed how quickly risk can spread when DeFi markets treat wrapped or bridged assets as equivalent to their originals, despite different security assumptions.
"The market decided, in its infinite wisdom, to treat a pre-bridge asset the same as a post-bridge asset - one-for-one equivalent. That is the origin of every single bad thing that has happened in our space since 2022."
- James Harris, Tesseract
His prescription is counterintuitive but compelling: DeFi needs to concentrate, not disperse. Fewer assets, deeper liquidity in fewer highways, and genuine market-priced discounts for the additional risk carried by bridged assets. The industry has been too clever for its own good.
Al added a data point from his own client base:
"I have had zero client demand from traditional finance for DeFi products. The demand I have had is from super-aggressive traders who are looking to exploit arbitrages and do looping - incredibly fiddly stuff."
– Al Turnbull, Blockchain.com
For DeFi to become institutionally usable, bridge risk does not need to disappear entirely, but it does need to be reduced, isolated and understood. That means stronger bridge security, fewer fragmented wrapped versions of the same asset, clearer collateral backing, better monitoring, and more explicit pricing or haircuts for assets that carry additional technical and redemption risk. Institutions are unlikely to accept a market structure where different versions of an asset are treated as interchangeable while relying on very different security assumptions.
On-chain vaults function as transparent, programmable fund structures. Instead of receiving a monthly attestation of what a manager is doing with your capital, you can watch it happen in real time, directly from your wallet. For retail and institutional investors alike, that is a material improvement.
James pointed to a recent weekend exploit as an unplanned test of the model. Because Tesseract’s vaults operate 24/7, he argued that clients had a better outcome than those relying on human-managed portfolios – a useful example of why real-time credit management can matter in crypto markets.
"Our vaults were active 24/7, and as a result, our clients had a better outcome than the ones that were human-managed. I genuinely do think it is 10x better."
– James Harris, Tesseract
James structured the opportunity across three orders of magnitude: the 10x case is transparency (real-time visibility versus monthly statements); the 100x case has already arrived (the vault market grew from roughly $100 million to $20 billion in 18 months); and the 1,000x case is the long-term vision - a world where all assets are tokenized and managed as on-chain vaults.
"In 20 years, I think we will look at BlackRock and all of their products will basically be vaults. That is your 1,000x TAM – a quadrillion-dollar market opportunity."
– James Harris, Tesseract
The challenge he acknowledged openly: right now, the available universe of assets inside these vaults is almost entirely crypto-native (Bitcoin, ETH, stablecoins). Getting to the 1,000x case requires partners willing to tokenize everything else. The infrastructure is ready. The assets are not yet.
The panel closed on the same theme that ran through the whole discussion: traditional banks are finally moving, and the people building infrastructure on the crypto-native side are ready for it.
Daniel gave the most concrete prediction:
"In the next two years you will see serious tier-one, tier-two bank players offering spot crypto. In the background they will have to figure out the prime model for their higher-volume trading clients - and then you will have real innovation at that point."
– Daniel Elsawey, Bitpanda
Al, who has been in the space since 2018 and has experienced multiple cycles of banking instability, was perhaps the most personally invested in this moment:
"The quality of the banking partners that the industry is able to access seems to be getting significantly better. Having been here since 2018 and suffered a lot of banking ups and downs - the fact that we can actually now see tier-one banks looking to engage in transactional crypto trading is super exciting."
– Al Turnbull, Blockchain.com
And Myles offered the framing that best captured the panel's overall spirit - not a competition between two worlds, but a convergence of them:
"The future looks like a convergence of TradFi and DeFi - taking the best of both worlds. It is not TradFi or DeFi."
– Myles Harrison, Amina Group
Cryptio is the data transformation and ERP platform for regulated digital assets. Trusted by 400+ institutions including Circle, Société Général, Gemini and Paxos, Cryptio helps finance, operations and compliance teams turn fragmented on-chain, exchange, custodian, DeFi and internal system data into reconciled, auditable financial records.
Our platform supports core digital asset accounting, reconciliation, tokenization compliance, reporting and ERP workflows – giving institutions the data foundation they need to operate across increasingly complex digital asset activity.
For lenders, Cryptio’s Loan Management System provides end-to-end visibility across the loan book, from collateral tracking and real-time LTV monitoring to margin and liquidation alerts, automated reconciliation and GAAP/IFRS-ready reporting.