The Capital Gap
Mike Cagney of Figure recently highlighted a fundamental constraint facing decentralized finance: the shortage of institutional capital willing to fund real-world assets (RWA) on public blockchains. As he noted, the entire stablecoin supply available for borrowing on Aave (the largest DeFi lending protocol) is under $10 billion. Figure alone could consume that capacity in months.
The problem isn’t technology. It’s plumbing.
Prime money market funds hold over $1 trillion in assets. Pension funds and endowments manage trillions more. These institutions want yield. They can invest in more than Treasuries. But they need a compliant path onto blockchain rails, one that fits their existing regulatory frameworks and risk mandates.
Securities lending can provide that path. As an example, let’s look at how retail brokers can provide revenue-generating services to high-net-worth accounts by lending from their fully-paid (not margin) portfolios; attract capital from prime money market fund managers; and, at the same time, satisfy the demand from correspondent brokers for securities to cover settlement of their short sales.
The Trade
Consider a structure that brings prime MMF capital into securities lending economics through a closed-loop financing mechanism.
Objective
Design a capital-efficient Fully Paid Lending (FPL) structure that satisfies collateral requirements under Exchange Act Rule 15c3-3 while enabling prime money market funds to participate in securities lending economics.
The Structure
A retail broker wants to borrow fully paid equities from its customers and on-lend them to the Street. SEC Rule 15c3-3 requires posting Treasury collateral to customers, but the broker needs cash to acquire those Treasuries.
The solution is a multi-leg circular trade:
| Leg 1: | Broker delivers Treasuries from its inventory to retail customers, borrowing their equities |
| Leg 2: | Broker on-lends equities to a Street borrower, receiving AAA corporate bonds as collateral |
| Leg 3: | Broker repos the corporates to a prime money market fund, receiving cash |
| Leg 4: | Broker uses cash to purchase replacement Treasuries, replenishing its collateral inventory |
The result: The MMF earns a repo rate. The brokerage firm removes the cash from its balance sheet, thereby reducing regulatory capital charges, while earning the spread between the equity lending fee and repo financing cost. Retail customers add interest to their portfolios and hold Treasury collateral per regulatory requirements.
Current Process: Sequential Settlement
In traditional (TradFi) markets, each leg settles independently across T+1 or T+2 cycles. This creates several problems:
Capital commitment. The broker must pre-fund a Treasury “box,” an inventory of unencumbered government securities to seed each transaction. SEC Rule 15c3-3 requires collateral delivery “upon the execution of the agreement or by the close of the business day.” The broker can’t wait for downstream legs to complete before posting collateral to retail customers.
Settlement risk. Temporal gaps between legs in TradFi create counterparty exposure. If any leg fails (a Street borrower doesn’t deliver corporates, the MMF repo doesn’t fund) the chain breaks.
Operational burden. Daily mark-to-market requirements across four legs. Margin calls in different asset classes. Reconciliation across multiple counterparties and custodians.
The box requirement means this structure isn’t truly “capital-light.” It is capital repositioned as working inventory rather than eliminated. But the mechanics work, and they’re compliant with current rules.
RWA Blockchain Process: Atomic Settlement
The same economic trade executes on a real-world asset blockchain (such as Provenance, developed by Figure) with all legs settling simultaneously through smart contracts.
A single atomic transaction either completes entirely or reverts entirely:
| IF AND ONLY IF all conditions are met simultaneously: 1. Retail customer transfers equity tokens to Broker 2. Broker transfers Treasury tokens to Retail customer 3. Broker transfers equity tokens to Street Borrower 4. Street Borrower transfers Corporate tokens to Broker 5. Broker transfers Corporate tokens to MMF (repo) 6. MMF transfers stablecoin to Broker 7. Broker acquires Treasury tokens from seller THEN: All legs settle atomically at t=0 ELSE: Entire transaction reverts, no legs execute |
What This Solves
No box capital required. If all legs are truly atomic, there’s no “first” leg. Treasuries are acquired simultaneously with delivery to retail customers. The circular dependency becomes a simultaneous equation that solves at t=0.
Zero settlement risk. No leg can fail independently. Either everything happens or nothing happens. The fragility of the rehypothecation chain disappears.
Automated operations. Smart contracts handle daily mark-to-market and margin calls. Continuous collateral verification replaces daily reconciliation.
What Remains Unresolved
Atomic settlement solves the operational and timing problems. It does not automatically solve the regulatory ones:
SEC Rule 15c3-3. The SEC’s 1982 adopting release stated that the rule “will compel the firm to turn over the collateral physically to the lender.” The Commission hasn’t confirmed that tokenized Treasury delivery satisfies this requirement.[1]
Rule 2a-7. Prime money market funds operate under specific eligible security and liquidity requirements. Tokenized repo would need SEC guidance or no-action relief.[2]
GAAP and SLR. Accounting treatment looks at economic positions, not settlement mechanics. Instantaneous settlement doesn’t change whether assets and liabilities are recognized. The balance sheet relief from atomic execution may be more limited than hoped.[3]
These aren’t insurmountable obstacles. They’re engineering problems dressed as legal questions. Figure has already navigated similar issues with HELOC securitizations on Provenance. The path exists; it just needs to be walked.
Comparison: Current vs. Blockchain Settlement
| Criterion | Current Process | Blockchain Process |
| Bootstrap capital | Required (Treasury box) | Not required |
| Settlement risk | Present (T+1/T+2 gaps) | Eliminated (atomic) |
| 15c3-3 compliance | Clear | Awaiting guidance |
| MMF participation | Standard repo | Tokenized repo (needs 2a-7 clarity) |
| Operational complexity | High | Automated |
| Regulatory certainty | High | Medium |
The Bridge
Mike Cagney is right that DeFi needs institutional capital. The stablecoin supply on Aave won’t fund the migration of real-world assets to public blockchain.
But the capital exists. Prime money market funds. Pension cash pools. Corporate treasury operations. They’re already active in repo markets. They already understand securities lending. They’re looking for yield.
What they need is a compliant bridge: a structure that maps familiar instruments (repo, securities lending, Treasury collateral) onto unfamiliar rails (blockchain, smart contracts, atomic settlement).
Fully paid lending is that bridge. The trade works today under current rules, using traditional settlement. It can work tomorrow under blockchain settlement, once regulators provide clarity on tokenized collateral delivery.
The economics are identical. Only the plumbing changes.
A Note on Spread Economics
None of this matters if the trade doesn’t pay.
The broker’s profit comes from the spread between the intrinsic fee earned on the equity loan (Leg 2) and the cumulative financing costs across the other legs: the repo rate paid to the MMF (Leg 3), plus any cost of carry on the Treasury collateral, plus the retail customer’s share of lending revenue.
For general collateral equities (names with low borrowing demand), intrinsic fees are minimal. The spread may be zero or negative. The structure simply doesn’t work for easy-to-borrow names.
For hard-to-borrow names (high short interest, limited float, corporate action plays), intrinsic fees can be substantial: 3%, 5%, sometimes 20% or more annualized. These are the names that make the structure viable.
This has an important implication: the structure is inherently selective. It won’t absorb all retail equity holdings into a lending program. It will focus on the hard-to-borrow segment where intrinsic value exceeds financing cost.
That’s not a flaw. It’s a feature. The structure finds its natural level based on market demand for specific securities. When short interest rises, more names become viable. When it falls, the program contracts. The economics are self-regulating.
For the prime MMF, this selectivity doesn’t matter. The MMF’s exposure is to the repo, collateralized by AAA corporates. It earns its repo rate regardless of which equities are flowing through the upstream legs. The intrinsic value question is the broker’s problem, not the fund’s.
* * *
Securities lending is a $3+ trillion market. Prime money market funds hold over $1 trillion. The capital to fund RWA migration to blockchain exists. It’s waiting for regulatory rails.
This is one structure. There are others. The point isn’t that this specific trade is the answer. It’s that the tools to build bridges between institutional capital and blockchain infrastructure already exist in securities finance.
The conversation worth having isn’t whether DeFi needs capital. It’s how to build the compliant pathways that let capital flow.
[1] While the SEC’s Division of Trading and Markets issued a statement in December 2025 regarding the custody of “crypto asset securities,” that guidance primarily addressed how a broker maintains possession of customer assets. It did not explicitly bridge the gap for collateral delivery under Rule 15c3-3(b)(3).
[2] The SEC would need to grant regulatory permission for MMFs to treat these tokens as “daily liquid assets” counting toward its 25% daily liquidity requirement.
[3] The balance sheet relief is more limited than hoped because the SEC and FASB haven’t changed the definition of a “loan” just because the loan happens faster. The SEC’s Division of Trading and Markets has signaled that “atomic execution” does not automatically qualify as a “final” settlement if the broker-dealer (or a controlled affiliate) maintains the “operational ability” to reverse or modify the transaction on a private or permissioned ledger.