The Custody Wars: How Tokenized Funds Are Forcing Wall Street to Build a Bridge It Never Wanted

Last November, a mid-sized pension fund in the Midwest wanted to move $50 million from a money market fund into a short-term Treasury product. The trade itself took minutes. The settlement? Three business days. During that window, rates moved, the fund’s NAV drifted, and the treasury team spent six hours on phone calls confirming wire instructions with three different intermediaries. This is the plumbing of modern finance—rusty, slow, and absurdly expensive to maintain.

Now imagine the same trade executed on Ethereum at 2 a.m. on a Sunday. The tokens settle in twelve seconds. The yield, freshly accrued from overnight reverse repos, auto-distributes to the pension fund’s wallet every four hours. A smart contract handles the compliance check against accredited investor rules before the transaction even hits the mempool. No phone calls. No T+2 waiting game. No custodian’s custody of your custodian.

This isn’t science fiction. It’s the operational reality that BlackRock’s BUIDL fund, Franklin Templeton’s OnChain U.S. Government Money Fund, and a growing cohort of tokenized asset platforms are already building. And it’s creating a pressure cooker for traditional fund administrators, transfer agents, and custodians who suddenly find themselves caught between two incompatible worlds: the regulatory fortress of SEC-registered infrastructure, and the composable, always-on logic of decentralized finance. The institutions that figure out how to bridge this gap will capture trillions in migrating assets. Those that don’t may find themselves holding expensive licenses for markets that no longer need them.

What Tokenization Actually Means in 2024

Let’s clear up the terminology swamp. “Real-world asset tokenization” has become a buzzword so overused that it now describes everything from luxury watch fractionalization schemes to legitimate Treasury-backed securities. For this article, I’m talking specifically about the tokenization of regulated financial instruments—money market funds, Treasury bills, private credit, real estate debt—where the token itself carries legal claim to an underlying asset and operates under existing securities law.

This is distinct from the 2017 ICO era, where tokens were often speculative instruments with no enforceable rights. It’s also distinct from stablecoins, which (ideally) hold reserves but don’t pass through yield or represent ownership in a managed portfolio.

The current wave emerged from two converging forces. First, post-2022 interest rates made money market and Treasury yields genuinely attractive again—4-5% on short-term government paper versus near-zero for years prior. Second, the infrastructure matured. Ethereum’s shift to proof-of-stake, layer-2 scaling solutions like Arbitrum and Base, and institutional-grade custody from firms like Fireblocks and Coinbase Prime made it possible to put serious money on-chain without betting the firm on blockchain reliability.

The result: tokenized Treasury and money market products grew from roughly $100 million in early 2023 to over $2 billion by mid-2024, according to data from RWA.xyz. BlackRock’s BUIDL fund alone crossed $500 million in assets within months of launch. These aren’t retail crypto degens. These are institutional treasurers seeking yield, liquidity, and operational efficiency they literally cannot get through traditional channels.

The On-Chain Proof-of-Reserve Imperative

Here’s where custody gets interesting—and tense.

Traditional fund administration works on a principle of periodic attestation. A fund’s custodian (say, State Street or BNY Mellon) holds the assets. An administrator calculates NAV daily or weekly. An auditor checks the books quarterly. Investors receive statements monthly. Everyone trusts the chain of regulated entities because regulators audit them, and because the system, while slow, has functioned for decades.

Tokenized funds break this model in three ways that matter.

Continuous verification replaces periodic attestation. When BlackRock’s BUIDL holds reserves in cash and Treasury bills, those reserves don’t sit in a black box between quarterly reports. The fund’s smart contract architecture, built with Securitize, allows for on-chain verification of reserve composition—either through direct oracle feeds or through cryptographic attestations that update far more frequently than traditional NAV cycles. Investors don’t trust BlackRock’s quarterly filing; they verify the wallet balances themselves, in real time.

Composability creates unregulated downstream exposure. A pension fund holding BUIDL tokens can, in theory, deposit them as collateral in a DeFi lending protocol, use them to provide liquidity in an automated market maker, or lend them in an over-the-counter repo market. Each of these “money legos” interactions happens outside the traditional transfer agent’s oversight. The SEC-registered infrastructure that issued the token has no visibility into how it’s being used two, three, four hops down the line. Yet the legal claim—and the regulatory liability—traces back to that original issuance.

Programmable distribution breaks the dividend machinery. Traditional funds calculate yields, accrue them to shareholder accounts, and distribute them on a set schedule—monthly, quarterly, annually. The operational overhead is substantial: calculations, tax withholding, regulatory reporting, physical checks or wire transfers. Tokenized funds like Franklin Templeton’s FOBXX (the OnChain U.S. Government Money Fund) can distribute yield continuously, automatically, to any wallet address globally. The fund’s smart contract handles the accrual logic; the investor’s self-custody or exchange custody determines when and how they realize the income.

These three shifts force a fundamental question: what is the role of a traditional custodian or transfer agent in a system where verification is automated, distribution is programmable, and usage is permissionlessly composable?

The answer emerging from the market is something called “on-chain proof-of-reserve infrastructure”—a hybrid architecture where regulated entities maintain their legal and compliance roles, but adapt their technical operations to provide continuous, cryptographically verifiable attestations of asset backing and regulatory compliance.

Case Studies: Three Models of Institutional Tokenization

The market is sorting into distinct architectural approaches, each with different implications for custody and compliance.

BlackRock BUIDL: The Wall Street Incumbent Model

BlackRock’s USD Institutional Digital Liquidity Fund, ticker BUIDL, launched in March 2024 through Securitize’s platform. It invests exclusively in cash, Treasury bills, and repurchase agreements. The structure is deliberately conservative: SEC-registered, 40 Act fund, with BNY Mellon as custodian of the traditional assets and Securitize Markets as the transfer agent for tokenized shares.

The innovation lies in the redemption mechanics. BUIDL offers 24/7/365 redemptions to U.S. dollar stablecoin (specifically USDC) through a dedicated smart contract, with Circle as the conversion partner. This means an institutional investor can hold tokenized Treasury exposure, earn yield, and exit to digital dollars at any hour—not just when the Fed wire system is open.

As of mid-2024, BUIDL had attracted roughly $500 million in assets, with reported yields in the 4.7-5.1% range depending on the Treasury curve. The investor base is heavily institutional: hedge funds, registered investment advisers, family offices. Notably, several DeFi protocols and on-chain treasuries have also allocated, creating the composability tension described above.

For custody, BUIDL maintains a split architecture. BNY Mellon holds the actual Treasury securities in traditional custody. Securitize’s platform tokenizes the fund shares and manages the on-chain registry. The proof-of-reserve question is handled through Securitize’s periodic reporting and BlackRock’s brand credibility—not through continuous on-chain verification of the underlying Treasuries themselves. It’s a transitional model: traditional custody for the assets, tokenized wrapper for the shares.

Franklin Templeton OnChain Fund: The Technology-Forward Incumbent

Franklin Templeton has been experimenting with blockchain for fund administration since 2019, longer than almost any major asset manager. Their OnChain U.S. Government Money Fund (FOBXX) launched on Stellar in 2021 and expanded to Polygon in 2023, making it one of the first SEC-registered funds to operate across multiple public blockchains.

The technical architecture is more ambitious than BUIDL’s. Franklin Templeton built its own proprietary blockchain, Benji, for internal record-keeping, but uses public chains for investor-facing token transfers. The fund processes subscriptions and redemptions through a smart contract system that integrates with traditional bank accounts via API connections.

Crucially, Franklin Templeton has pushed further on the proof-of-reserve concept. The fund publishes its portfolio holdings daily—not monthly, not quarterly—and has experimented with oracle integrations that could verify those holdings on-chain. As of mid-2024, FOBXX held approximately $400 million in assets, with yields comparable to traditional money market funds.

The custody model is still hybrid: traditional custodians hold the underlying assets, Franklin Templeton’s technology stack handles tokenization and transfer. But the firm has been more vocal than BlackRock about the eventual goal of fully on-chain fund administration, where smart contracts replace transfer agents for routine operations.

Centrifuge: The DeFi-Native Alternative

Centrifuge represents a fundamentally different approach—one that starts from DeFi primitives and works backward toward regulatory compliance, rather than starting from regulated structures and adding tokenization.

Founded in 2017, Centrifuge operates a protocol for tokenizing real-world assets into structured pools. Unlike BUIDL or FOBXX, which are single-issuer funds with familiar legal structures, Centrifuge pools are typically backed by private credit: invoices, real estate loans, revenue-based financing. The protocol has financed over $500 million in assets across its lifetime, though active pool volumes fluctuate significantly with market conditions.

The custody and verification challenge is more acute here. Centrifuge uses a network of “asset originators” who bring loans onto the platform, and “validators” who attest to the existence and quality of the underlying assets. Proof-of-reserve is handled through a combination of legal structures (special purpose vehicles that hold the assets), off-chain data oracles, and on-chain token representations.

The trade-off is explicit: higher yields (often 8-15% for senior tranches, more for junior) in exchange for less regulatory protection and more complex verification. Centrifuge pools are not SEC-registered funds. Investors typically access them through DeFi protocols like MakerDAO, which has integrated Centrifuge collateral into its stablecoin backing, or through direct protocol participation.

What makes Centrifuge relevant to the custody wars is its demonstration that proof-of-reserve infrastructure can be built without traditional custodians at all—replaced by a combination of legal structuring, cryptographic verification, and economic staking. This is the competitive threat that keeps traditional administrators awake: not that they’ll be replaced tomorrow, but that an alternative architecture is proving viable for certain asset classes and investor types.

The Regulatory Arbitrage Problem

These three models—incumbent tokenization, technology-forward incumbent, and DeFi-native—are creating genuine regulatory arbitrage. Here’s how.

SEC-registered transfer agents operate under specific rules: Section 17A of the Securities Exchange Act, SEC Rule 17Ad-8, and a web of state-level requirements. They maintain official records of ownership, process transfers, handle lost certificates, and ensure compliance with various holding period and accreditation rules. For this, they charge basis points on assets under administration.

Tokenized funds need some of these services, but not all. A blockchain address doesn’t need certificate replacement. A smart contract doesn’t need manual transfer processing. What remains—accreditation verification, regulatory reporting, certain compliance functions—could theoretically be handled by lighter-weight structures or even, in some visions, by the smart contracts themselves with appropriate oracle inputs.

The arbitrage emerges because tokenized funds can offer functionality that traditional funds cannot—T+0 settlement, 24/7 liquidity, programmable yield—while operating under the same or similar regulatory umbrellas. Investors get superior operational features without sacrificing the regulatory protections of registered fund structures. Meanwhile, traditional funds are stuck with T+2 settlement, limited trading hours, and manual distribution processes because their infrastructure was built in the 1970s and patched since.

For regulators, this creates a dilemma. Clamp down on tokenized funds, and you push innovation to unregulated offshore venues or pure DeFi protocols. Let it proceed without adaptation of traditional rules, and you risk a two-tier system where registered transfer agents become increasingly irrelevant for an entire asset class.

The SEC’s approach through 2024 has been cautious engagement rather than prescriptive rulemaking. Staff statements have acknowledged the potential benefits of tokenization for settlement efficiency, while enforcement actions against unregistered crypto operations continue. The result is a gray zone where sophisticated institutions can navigate compliant tokenization, but clear standards for proof-of-reserve infrastructure, composability boundaries, and custodial responsibilities remain unwritten.

What Legacy Infrastructure Cannot Deliver

To understand why this pressure is existential for some incumbents, it’s worth cataloging what traditional fund administration simply cannot do right now, regardless of willingness or investment.

T+0 settlement with atomic delivery-versus-payment. Traditional settlement relies on correspondent banking networks, Fedwire operating hours, and batch processing. Same-day settlement exists for some instruments but requires manual intervention and carries cutoff time constraints. Atomic settlement—where asset transfer and payment occur simultaneously, with no counterparty risk window—is technically impossible in legacy architecture without blockchain or similar distributed ledger technology.

Continuous NAV and yield accrual. Traditional funds calculate NAV once daily, after market close. Yield accrues to shareholder records but isn’t realized until distribution. Tokenized funds can update token values continuously (or at least multiple times daily) and distribute yield as it accrues, creating different tax and reinvestment dynamics.

Permissionless composability. A traditional fund share cannot be used as collateral in a lending protocol, or combined with another asset in an automated market maker, without extensive legal and operational negotiation for each use case. Tokenized fund shares inherit the composability of their underlying blockchain—subject to legal constraints, but technically trivial to integrate.

Global 24/7 access. Traditional fund trading is limited to U.S. business hours, with additional constraints for international investors. Tokenized funds operate on blockchain time: always on, globally accessible, limited only by wallet infrastructure and regulatory eligibility verification.

These aren’t nice-to-have features for a subset of crypto-native investors. They’re operational requirements for modern treasury management, especially for global corporations, DAOs, and investment funds that operate across time zones and need immediate liquidity for opportunistic deployment.

Risks, Limitations, and Trade-Offs

The tokenization narrative often glosses over significant risks that any serious participant must understand.

Technical and Operational Risks

Smart contract vulnerabilities remain a genuine concern. While BUIDL and FOBXX use audited contracts from established providers, the attack surface expands with composability. A bug in a lending protocol that accepts BUIDL as collateral could cascade back to the fund’s investors. The 2022 Nomad bridge exploit, which drained $190 million through a simple smart contract flaw, illustrates how quickly on-chain value can disappear.

Oracle reliability for proof-of-reserve is still maturing. Verifying that a custodian holds specific Treasury CUSIPs at a specific quantity requires either the custodian’s active cooperation in publishing attestations, or independent verification that introduces new trust assumptions. Chainlink and similar oracle networks are building this infrastructure, but coverage is incomplete and the economic security models are still being tested at scale.

Key management for institutional wallets introduces operational complexity that many traditional firms underestimate. The stories of lost seed phrases and hacked custody arrangements are too numerous to catalog. Multi-signature schemes, hardware security modules, and emerging multi-party computation protocols help, but they add latency and cost.

Regulatory and Legal Risks

The SEC has not provided clear guidance on how existing transfer agent rules apply to blockchain-based record-keeping. Firms like Securitize have received no-action relief for specific structures, but this doesn’t create generally applicable precedent. A change in SEC leadership or enforcement priorities could render current compliance strategies inadequate.

Cross-border complications multiply quickly. A German insurance fund holding tokenized U.S. Treasuries through a Singapore-based custodian, with yield distributed to a wallet controlled by a Cayman subsidiary, touches at least four regulatory regimes with inconsistent treatment of blockchain-based assets. The legal infrastructure for resolving disputes in this context is largely untested.

Tax treatment of continuous yield distribution, as opposed to periodic dividends, remains unclear in many jurisdictions. Automatic reinvestment through smart contracts may trigger different reporting obligations than traditional dividend reinvestment plans.

Economic and Market Risks

Liquidity in tokenized funds is thinner than headline numbers suggest. While BUIDL offers 24/7 redemption to USDC, large exits still depend on Circle’s conversion capacity and the depth of USDC liquidity pools. In a crisis, the on-chain redemption mechanism might face the same pressures as traditional fund gates, just expressed differently.

Yield compression is already occurring as more tokenized Treasury products compete for the same institutional dollars. Spreads between tokenized and traditional money market funds have narrowed from initial premiums to near-parity, suggesting that operational efficiency gains are being competed away rather than captured by early adopters.

Concentration risk in the tokenization infrastructure itself is notable. Securitize dominates institutional tokenization platform market share. A technical failure or regulatory action against this single point of failure would disrupt multiple major funds simultaneously.

User and Accessibility Risks

The “self-custody or trust” dilemma hasn’t been resolved. Institutional investors comfortable with traditional custody face a choice: maintain the familiar custodial relationship (sacrificing some composability), or hold tokens directly (accepting operational responsibility many aren’t equipped for). Intermediary solutions—qualified custodians for tokenized assets—are emerging but add back cost and complexity.

The user experience for non-technical treasury staff remains poor. Wallet interfaces, gas fee management, and transaction monitoring require skills that don’t map neatly from traditional treasury operations. Training costs and operational errors are real and underreported.

Practical Guidance for Market Participants

For readers evaluating how to engage with this transition, here are concrete considerations organized by role.

For Institutional Investors and Treasury Managers

1. Audit your current settlement costs. Calculate the full cost of T+2 settlement: counterparty risk capital, operational staff time, failed trade resolution, opportunity cost of delayed deployment. This baseline justifies tokenization investment even if headline yields are equivalent.

2. Evaluate custody options on three dimensions: regulatory protection (what happens if the custodian fails), operational capability (can they handle the tokenized workflow), and composability access (will they permit DeFi interactions or block them). Most traditional custodians score well on the first, poorly on the latter two.

3. Demand proof-of-reserve specifics. When evaluating tokenized products, ask: How frequently are reserves verified? By what mechanism? Is the verification cryptographic or attestation-based? Who bears liability if reserves are misstated? Vague answers suggest transitional infrastructure that may not survive regulatory scrutiny.

4. Start with liquidity management, not core portfolio. Tokenized money market funds are the lowest-risk entry point. Private credit tokenization, real estate, and more exotic structures require deeper due diligence on the specific verification mechanisms for those asset classes.

For Asset Managers and Fund Builders

1. Don’t build core infrastructure from scratch. The Securitize/Anchorage/Paxos layer of tokenization platforms has reached sufficient maturity that bespoke smart contract development for standard fund structures is usually unnecessary risk. Differentiate on asset selection and investor service, not tokenization plumbing.

2. Plan for composability boundaries. Decide explicitly whether your tokenized fund will permit DeFi integration, and if so, with what safeguards. This is a product design decision with regulatory implications, not merely a technical one.

3. Invest in oracle relationships early. The firms that establish reliable, audited proof-of-reserve data feeds will have competitive advantage as investors become more sophisticated about verification. This is becoming table stakes, not differentiation.

For Policymakers and Regulators

1. Clarify transfer agent rules for blockchain records. The current no-action letter approach creates uncertainty and favors well-resourced firms that can afford extensive legal negotiation. Clear standards for when blockchain record-keeping satisfies Section 17A would accelerate beneficial innovation while maintaining investor protection.

2. Coordinate internationally on tokenized asset treatment. The current patchwork of national approaches creates compliance complexity that drives activity to the least regulated venues. The IOSCO framework for crypto asset regulation provides a starting point, but specific guidance for tokenized traditional assets is needed.

3. Consider sandbox approaches for atomic settlement. The systemic risk benefits of eliminating settlement failures and counterparty exposure are substantial. Regulatory sandboxes that permit controlled testing of T+0 atomic settlement for standard instruments could generate empirical evidence for broader reform.

For Technology Builders and DeFi Protocols

1. Build with institutional compliance in mind. Protocols that can demonstrate how they handle accredited investor verification, sanctions screening, and transaction monitoring will be the ones that attract tokenized institutional assets. Pure permissionlessness is a feature for some use cases; selective permissioning is a requirement for others.

2. Develop standardized proof-of-reserve interfaces. The ecosystem needs common standards for how tokenized funds publish reserve attestations, similar to how ERC-20 standardized token interfaces. This is infrastructure that benefits from coordination rather than competition.

The Next 12-24 Months: Scenarios and Trajectories

Looking ahead, the most likely trajectory is continued bifurcation followed by gradual convergence.

In the near term, expect more asset managers to launch tokenized versions of standard products—Treasury funds, short-duration bond funds, perhaps some equity index exposure. The technical playbook is now established; the regulatory path, while not entirely clear, is navigable. BlackRock’s BUIDL success will be studied and emulated. The total tokenized Treasury and money market market could reasonably reach $5-10 billion by late 2025, though this depends heavily on interest rate trajectories and regulatory developments.

The custody and transfer agent landscape will see more explicit strategic positioning. Some traditional custodians, like BNY Mellon’s cautious engagement, will build hybrid capabilities. Others may acquire tokenization platforms outright. Expect at least one major custody merger or acquisition in this space within 18 months. Conversely, some custodians may retreat to purely traditional asset classes, ceding tokenized markets to crypto-native infrastructure providers.

Regulatory clarity will likely remain partial rather than comprehensive. The SEC may propose specific rules for tokenized fund administration, but the legislative process for broader framework updates moves slowly. More probable is continued staff-level guidance and enforcement actions that collectively define boundaries.

The proof-of-reserve infrastructure will mature significantly. What is now largely manual attestation and periodic reporting will shift toward automated, cryptographic verification for standard assets. For complex or illiquid assets—private credit, real estate, infrastructure—the verification challenge will remain harder, and the market may split between sophisticated investors comfortable with residual uncertainty, and structures that maintain traditional audit functions.

The composability question is the most uncertain. Will regulators permit tokenized fund shares to integrate freely with DeFi protocols? Will they require walled gardens where approved smart contracts are whitelisted? Or will they push for a complete separation, where tokenized funds operate on permissioned blockchains with no DeFi access? The market is currently betting on a middle path—selective composability with compliance guardrails—but this could shift quickly with regulatory or market stress events.

What seems clear is that the underlying demand is genuine and structural. Institutional investors, having experienced tokenized settlement and programmable distribution, will not happily return to T+2 and manual dividend processing. The infrastructure being built now—however transitional, however imperfect—is laying groundwork for a fundamentally different operating model for fund administration.

The custodians and transfer agents that thrive will be those that stop defending their legacy position and start building the bridge. The ones that treat on-chain proof-of-reserve not as a threat to their role, but as an evolution of their verification function. The transition will be messy, expensive, and occasionally embarrassing for incumbents who move too slowly. But the direction is set. The only question is who pays for the rusted pipes to be replaced, and who captures the value when the new plumbing finally works.

---

What to Do Next

• Compare 2-3 relevant tools before choosing one.
• Validate fees, custody model, and jurisdiction support.
• Start small and track performance weekly.

Recommended Next Reads

• Crypto security basics: /category/cybersecurity/
• DeFi risk management: /category/defi/
• Blockchain technology explainers: /category/blockchain-technology/

Sources and Further Reading

• Chainalysis
• DefiLlama
• CoinGecko Research

FAQ

What is the main takeaway?

Focus on practical risk, utility, and execution rather than hype.

Who should care most?

Builders, active users, and investors exposed to the discussed sector.

What should readers do next?

Use the checklist, compare tools, and validate claims with primary sources.

Stay Updated

Subscribe to your site newsletter for weekly market breakdowns, tool comparisons, and risk alerts.