Deutsche Bank Launches Crypto Custody as Nomura Exits: The Web3 Battle Between Fee Collectors and Risk Takers

Stock News11-16

According to Woofun AI, on November 11, 2026, the strategic divergence among traditional financial giants in the Web3 space reached a critical tipping point: Deutsche Bank (DB.US) officially launched its institutional crypto custody service, covering Bitcoin, Ethereum, USDC, EURC, and the AllUnity euro stablecoin EURAU, while simultaneously signing a memorandum of understanding with its asset management arm DWS to explore tokenized products. On the same day, Nomura Holdings (NMR.US) was reported to be in talks with potential buyers including Marex (MRX.US) to sell its crypto subsidiary Laser Digital.

This simultaneous advance and retreat is not an incidental operational adjustment but a fierce collision between two fundamentally different Web3 business models in reality. One institution chose to build infrastructure and collect deterministic fees, while the other fell into financial trouble by bearing market volatility risk. This article aims to strip away market noise, dissect the underlying logic of these institutions, clarify the structural reasons behind who made money and who suffered heavy losses, and reveal how revenue models determine the life-or-death boundary for traditional banks in the crypto world.

In the winners' camp, BlackRock (BLK.US) and JPMorgan (JPM.US) demonstrated how to achieve low-risk expansion through tokenizing existing assets. In August 2026, BlackRock (BLK.US) launched digital shares for six UCITS money market funds in its European ICS series, with combined assets of approximately $311 billion. The key point is that these funds were not directly converted into cryptocurrencies but instead ran on Ethereum through JPMorgan's Kinexys platform, achieving tokenization at the entry point. This model did not reduce fee income by going on-chain; rather, it opened new distribution and settlement channels through on-chain transfers. BlackRock's core move was to connect existing assets to new interfaces rather than create new high-risk assets.

Meanwhile, JPMorgan's Kinexys platform has reached daily transaction volumes of $5 billion, with management targeting an increase to over $10 billion. In October 2026, JPMorgan (JPM.US) further expanded its JPM Coin deposit token (JPMD) to Ethereum. The success of this path does not depend on bull market conditions or whether tokens are classified as securities; its core logic is simply that settlement is cheaper than the current system. By reducing settlement costs and improving liquidity, these two giants solidified their positions as market infrastructure without bearing directional risk.

Citigroup (C.US), BNY (BNY.US), Goldman Sachs (GS.US), and Morgan Stanley (MS.US) built another profitable paradigm through fee competition and channel development. Citigroup (C.US) expanded Citi Token Services to Japan and the UAE in September 2026, and media reports indicated that several major banks are planning a shared tokenized deposit network. Such businesses often do not charge explicit fees directly; their returns are reflected in improved deposit retention rates and customer stickiness. This is a defensive revenue strategy aimed at locking in the most expensive deposit resources.

BNY (BNY.US) and Goldman Sachs (GS.US) launched tokenized money market fund solutions for asset management clients as early as 2025, with participants including BlackRock (BLK.US), Fidelity, and Goldman Sachs Asset Management. The essence is selling a complete service package that allows your funds to be held on-chain. Morgan Stanley (MS.US) launched its own Bitcoin ETF in April 2026, cutting fees to 0.14%, the lowest tier in the market at the time, attracting approximately $100 million in the first week. Its profit model is clearly visible: extremely low fees multiplied by massive channel traffic to achieve stable income independent of coin price direction. The common thread among these institutions is that their revenue sources all come from fees and traffic; none rely on holding token inventory for profit.

Deutsche Bank (DB.US) focused its strategy on custody positioning, entering the market by establishing a foothold first and expanding products later. The institutional crypto custody service launched by Deutsche Bank (DB.US) not only covers mainstream assets but also operates under MiCA and German regulatory frameworks, ensuring compliance. On the launch day, Deutsche Bank (DB.US) signed an MOU with its asset management arm DWS to explore tokenized products. This strategy of not trading, not doing proprietary business, and focusing on custody first typically reflects traditional banks' ultimate pursuit of risk control. By providing custody services, Deutsche Bank (DB.US) established long-term client relationships without bearing market volatility risk, laying the foundation for subsequent expansion into tokenized issuance, settlement, and other higher-value-added services. The advantage of this model is that its revenue is tied to asset scale rather than market volatility, achieving stability and predictability.

In contrast, Nomura Holdings (NMR.US) and its subsidiary Laser Digital in the losers' camp fell into trouble due to market-making risks and structural mismatches. Laser Digital was founded in 2022, covering institutional trading, market making, asset management, and venture capital, and achieved profitability in 2024. However, the turning point came during the flash crash on October 10, 2025, when altcoins fell 20% to 27% within minutes, with some dropping nearly 70% (AmberData data). As of the December 2025 quarter, Nomura (NMR.US) recorded approximately 10 billion yen in losses in the EMEA region outside its four major divisions, which the company attributed to its digital asset business; the CFO stated that the firm had tightened positions and risk exposure. On a pre-tax loss basis, Nomura's (NMR.US) European operations posted quarterly losses of 13.8 billion yen and 7.4 billion yen for the quarters ending March and September 2026, respectively.

Although Nomura (NMR.US) did not disclose the specific proportion of Laser in European results, and Nomura (NMR.US) as a whole recorded record revenues during the same period, the structural mismatch problem for market makers became evident: what market makers fear most is not price declines but the inability to hedge inventory during a downturn. When Laser's inventory contained a large number of illiquid small- and mid-cap tokens, market repricing within minutes turned earning spreads into passively taking on positions. Quarterly mark-to-market requirements, VaR limits, and parent company capital reviews for listed banks restricted the freedom of market makers to withstand volatility during drawdowns. This is a fundamental conflict at the structural level, not a matter of team capability.

ASX's failure and the resistance faced by OKX/ICE (ICE.US) tokenized stocks further revealed the limitations of the selling risk model. ASX announced in November 2022 that it was abandoning its plan to rebuild its CHESS clearing system using blockchain, a project that lasted approximately seven years and cost about A$250 million. The failure was not due to technical infeasibility but rather the simplification of an extremely complex institutional engineering project into a single technology upgrade. As Reuters commented, what was burned was not just money but market trust.

In November 2026, OKX and ICE (ICE.US)-affiliated tokenized US stock solutions were scheduled to launch under the SEC's temporary innovation exemption framework but faced collective pushback from issuers, with the opposition list surging from 22 to 67 companies. IMF research shows that tokenized stocks have 1.5 times the volatility of the same underlying traditional markets. Although this route has no disclosed losses yet, it is being repriced by the market. More alarmingly, crypto-native market-making benchmark B2C2 (under SBI) is also in talks to be sold. Even the people who know this business best are looking for an exit, which shows that the narrative that traditional institutions lost to native teams does not hold. The real issue is that the selling inventory risk business model itself is being repriced as a whole. Attribution is not the variable; the business model is the key determinant of survival.

Breaking down business models, the essential differences among four revenue models determine institutional fates. Model one is management and custody fees, with asset scale as the base, represented by BlackRock (BLK.US), BNY (BNY.US), Goldman Sachs (GS.US), and Deutsche Bank (DB.US) custody. This model does not bear directional risk; the larger the asset scale, the more stable the revenue. Model two is channel and settlement fees, with traffic as the base, represented by Kinexys and Citi Token Services. Such businesses can even be free, with returns reflected in deposit retention and customer stickiness. Model three is spreads and inventory, with the balance sheet as the base, involving market making, proprietary trading, and OTC business. This model is the best business when markets are favorable but has no downside floor and carries extremely high risk. Model four is asset appreciation, with returns coming from cycles, involving crypto venture capital and token treasuries. This model has strong upside appeal but ugly downside performance.

The first two models earn tolls that others must pay you; the latter two models earn based on how much uncertainty you are willing to bear. This fundamental difference explains why institutions like BlackRock (BLK.US) can grow steadily while institutions like Nomura (NMR.US) fall into losses. Quantitative analysis further reveals the financial comparison between scale effects and inventory risk. Assuming a scale of $10 billion with fees of 5 to 10 basis points, annual revenue would be approximately $5 million to $10 million. The value of this model lies in zero directional risk, high stickiness, and the ability to layer on staking, settlement, and foreign exchange fee items. In contrast, the model where revenue equals AUM times management fee rate times revenue-sharing ratio depends more on distribution; only institutions with end-user channels can push fee wars to 0.14%, leaving latecomers with almost no room to fight back.

For the inventory model, assuming $1 billion in inventory with 3% average daily volatility, a one-standard-deviation single-day inventory P&L would be in the range of approximately $30 million. Capital adequacy, hedging speed, and risk limits determine whether market making is a money printer or a money shredder, and these three factors are precisely the elements that are hardest for the banking system to adjust flexibly. In one sentence: the first two revenue streams grow linearly with scale and have limited tail risk; the third revenue stream and its risk come from the same variable and cannot be separated. Woofun AI's compiled data shows that this difference in financial structure gives pipeline-selling institutions significant resilience against market volatility, while risk-selling institutions are highly vulnerable to being devastated by a single black swan event.

From a regional perspective, the pipeline barriers built by Hong Kong's regulatory system provide another successful paradigm for traditional banks. Hong Kong's choice is very clear: make institutions thicker, not leverage bigger. The VATP licensing regime, the Stablecoins Ordinance (Cap. 656) effective August 1, 2025, and the HKMA-led tokenized bond and settlement pilots together form a combination of official registry plus regulated custody plus local currency settlement. Under this article's classification, Hong Kong has essentially built selling pipelines into a main road: custody, settlement, issuance, and secondary trading must all go through licensed entities, with purely speculative paths compressed to a minimum. When BlackRock (BLK.US) connects $311 billion in money market funds on-chain, JPMorgan (JPM.US) extends deposit tokens to Ethereum, and Deutsche Bank (DB.US) turns custody into a formal product, the real test is making these pipelines willing to connect their Asian nodes to this institutional system rather than elsewhere. Hong Kong's regulatory framework attracts institutions seeking stable revenue by providing certainty and compliance, thereby building a unique competitive advantage in the Web3 space.

In conclusion, the most profitable players in Web3 have never been those who understand tokens best but those who are best at collecting tolls. The experience of the past five years can be condensed into two sentences: those selling pipelines are turning scale into a moat; those selling risk are turning volatility into lessons. The next dividing line is not about whose trading desk is faster but about whose registry is thicker, whose custody is more trustworthy, and whose settlement is cheaper. The reported sale talks by Nomura (NMR.US) are not a surrender but possibly a dignified retreat; Deutsche Bank's (DB.US) launch is not a declaration but a quiet takeover. History does not score righteous battles; it only prices business models. For traditional financial institutions, only by recognizing their own advantages and choosing revenue models that match their risk tolerance can they stand invincible in the Web3 wave.

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Comments

We need your insight to fill this gap
Leave a comment