The contest in crypto digital banking is no longer centered on who can issue the best stablecoin card. The bigger question is who can build a lending model that actually makes money. The source article says monthly crypto card volume rose from about $100 million in early 2023 to more than $1.5 billion by the end of 2025, putting the market above an $18 billion annualized run rate. In 2025 alone, spending on stablecoin-linked bank cards reached $4.5 billion, up 673% year over year. Yet that growth has not been captured by the most DeFi-native or self-custodial products. On-chain card data shows Asia-based custodial platform RedotPay controls roughly 60% of the market, with volume around four times the combined total of the next 13 competitors.
Licenses are becoming the next battleground
The article points to a wave of moves between December 2025 and March 2026. Coinbase applied for a national trust charter. NuBank received conditional OCC approval for a U.S. national bank. PayPal applied to establish PayPal Bank. Revolut secured a full UK banking license and is pursuing one in the U.S. Kraken became the first crypto company with a Federal Reserve master account. Over the same period, 11 companies applied for OCC trust bank charters in 83 days, including Circle, Ripple, BitGo, Paxos, Fidelity, Bridge, Crypto.com, Morgan Stanley, Payoneer, Zerohash, and Protego.
Those filings all point to the same economic reality: payments alone are weak as a long-term profit engine. The source notes that 76% of traditional digital banks are unprofitable. The firms that did reach profitability built lending books and net interest income, rather than relying on card interchange. For crypto banks, that pressure is sharper because stablecoins compress FX spreads and settlement fees, making it harder to support a business with card spending and cashback subsidies alone.
Four models are emerging, but the moat is what matters
The article breaks the field into four archetypes. The first is crypto-friendly, banking-first digital banks with licenses. Nubank, SoFi, and Revolut are used as examples of institutions where credit remains the center of the business. Nubank reported $15.8 billion in revenue in fiscal 2025, with 85% coming from interest income. After getting a banking license, SoFi lifted quarterly net interest income from $94.9 million to $617 million over four years. In this model, stablecoins are treated mainly as payment or settlement rails, not as savings products.
The second group is made up of commerce and social super-apps such as MercadoPago, Grab, WeChat, and Alipay. Their edge comes from distribution and behavioral data, which can feed better credit decisions. The third is trading-first platforms including Robinhood, Coinbase, Binance, Kraken, Bybit, and OKX. These companies start with trading revenue and then layer on wallets, lending, payments, and licensing. That path is very different from stablecoin banks that begin with thin payment fees and try to add everything else later.
The fourth group is stablecoin-first, crypto-native platforms such as Ether.fi, Gnosis Pay, RedotPay, KAST, Holyheld, Bleap, Ready, Tria, Cypher, and Payy. Their proposition is clear: self-custody, DeFi yield, near-instant cross-border transfers, and global portability. Their weakness is just as clear. No player has yet broken through in scaled unsecured lending, and many are competing in the thinnest part of the stack while subsidizing user acquisition.
Shared infrastructure creates concentrated risk
A major point in the article is that many crypto digital banks are really just front ends built on shared rails. At the card-network level, Visa and Mastercard are close in project count, but Visa accounts for more than 90% of on-chain card transaction volume. That leaves the sector exposed to policy shifts, slower expansion, or fee changes at a single network. Issuers such as Rain, Reap, Baanx, and StraitsX act as regulated bridges between on-chain products and traditional payment systems.
Backend concentration is another issue. Rain supports Ether.fi, RedotPay, and Avalanche Card. If one infrastructure provider runs into technical trouble, regulatory pressure, or strategic changes, the shock would not be limited to one brand. The source cites Solus Partners, which analyzed 19 platforms and described infrastructure concentration and vendor dependence as systemic risk, comparing it to a “Synapse risk” for crypto digital banking.
Wallet-issued stablecoins are starting to absorb the value chain
The article also flags a competitive shift that can be easy to miss. By the end of the third quarter of 2025, MetaMask launched mUSD and Phantom launched CASH, each designed to fund spending inside their own debit card products. If users first convert assets into wallet-native stablecoins and then spend from there, the wallet captures channel fees, FX spreads, and reserve income that might otherwise have gone to an independent digital bank.
Early numbers show different trajectories. Phantom’s CASH rose from about $25 million in September to roughly $100 million by late December. MetaMask’s mUSD climbed to nearly $100 million in early October, then fell back to around $25 million, a 75% drop. For standalone crypto banks, that is not just another competitor. It means user ownership may shift upstream to the wallet layer.
The winning variables sit beyond payments
The source lists five factors that could reshape the field: on-chain credit scoring, full banking licenses for crypto-native firms, regulatory clarity on whether yield is allowed, agent-driven finance, and payment experiences that make on-chain activity almost invisible to the user. Credit is presented as the hardest and most important piece. Wallet history, DeFi behavior, repayment records in lending protocols, staking duration, transaction frequency, and protocol diversity could all feed into credit models. But the article says no crypto digital bank has done this at meaningful scale so far.
It does not declare a winner. What it does make clear is that the center of gravity is moving. Licensed digital banks already have a proven credit-led model. Stablecoin-first products still hold structural advantages in cross-border payments and emerging markets. Commerce-embedded players own distribution, though adding crypto comes with cost and regulatory dependency. The article’s conclusion is blunt in business terms: payments are only the entry point, while on-chain lending and license-driven balance-sheet power are where the real fight will be decided.