From fund packaging to the treasury layer: why asset-backed credit will become the final form of on-chain lending.

Written by: @BirchHill_io

Translated by: AididiaoJP, Foresight News

“Asset-backed credit is the only type of credit on-chain that can solve the adverse selection problem. Tokenized fund packaging cannot achieve this. Birch Hill addresses this issue at the treasury layer.”

Three Markets, Not One

When it comes to on-chain lending, it is essential to separate three aspects that people commonly refer to by the same name.

The first type is over-collateralized crypto lending. Aave, Morpho, Compound, Spark. Borrowers stake $1.50 worth of ETH to borrow $1 in stablecoins. The liquidation mechanism is systemic. Collateral is highly liquid 24/7. This category has developed into one of the most reliable yield primitives in DeFi, with current stablecoin supply rates ranging from 3.5% to 7%. This works, but it also cannot scale beyond the risks of crypto collateral.

The second type is unsecured lending. This has been the holy grail of DeFi since 2017, but has continuously failed under loose forms because the protocol layer cannot answer the three questions every lending business must address: who is the borrower, how to price their default, and what happens when they don’t repay. The history of this category, thus far, is a graveyard.

The third type is asset-backed credit. Loans secured by real, identifiable collateral, backed by off-chain legal claims, third-party evaluations, and recourse rights. This is the fastest-growing category and the only one that has credible answers to the adverse selection problem on-chain lending has always faced.

The label “asset-backed credit” has been growing on-chain, primarily in the form of packaging. Currently, on-chain is mostly tokenized fund equity: funds hold loans, tokens represent shares in the funds, and the structure of the funds determines whether the adverse selection problem is genuinely resolved. In most cases, the protocol layer merely passes through the embedded risks of the fund sponsors.

Asset-backed credit is the fastest-growing category of on-chain lending, but its current growth method—using tokenized fund packaging—does not actually resolve the problem. Fund packaging inherits the sponsor's structure. What we are realizing is to solve the problem at the smart contract level: evaluation, structuring, and recourse are encoded into the treasury itself rather than inherited from off-chain fund managers.

Where Growth Actually Is

The on-chain real-world asset (RWA) category grew from $5.6 billion in early 2024 to $25.96 billion on June 3, 2026. This represents approximately a 4.6x expansion over 29 months, almost entirely concentrated in the past 12 months. The retracement after the peak in May is relatively minor compared to the overall trajectory, consistent with normal category fluctuations.

It is worth putting this growth in context. Even with a 4.6x expansion achieved over 29 months, today the entire on-chain RWA market is still insignificant compared to the scale that the market believes this category could reach. Deloitte's April 2025 forecast indicates that by 2035, the tokenized real estate market alone could reach $4 trillion, with tokenized loans and securitization being the largest sub-segments, touching $2.39 trillion.

The composition is more important than the headline. The first wave is tokenized U.S. Treasuries: BlackRock’s BUIDL, Franklin Templeton’s BENJI, Ondo, Superstate, and now JPMorgan’s MONY and competing products from BNY Mellon. These are essentially money market funds wrapped in blockchain. They demonstrate that institutions can and are willing to hold on-chain, but they do not solve the credit issue; they merely transplant existing products.

The second wave we are currently in is tokenized private credit. On-chain private credit has grown approximately 180% year-over-year and is now the largest non-stablecoin RWA segment. Maple Finance has $3.17 billion in its USDC lending treasury, approximately $926 million in its USDT treasury, with a total stablecoin deposit of about $4.1 billion. Centrifuge has surpassed $1.38 billion TVL. New entrants continue to emerge, including Centrifuge’s USDS treasury with about $865 million.

These are credit funds with on-chain pipelines. Borrowers undergo KYC, loans come with off-chain legal documents, third-party assessments are conducted by credit committee members, and recourse follows the legal process like other private credit funds. The blockchain contributes transparency in distribution and packaging, as well as an increasingly rigorous regulatory containment layer.

Stablecoins are the demand side and are growing faster.

The total supply of stablecoins is now approximately $323 billion. USDT is about $190 billion. USDC is around $73 billion. Other yield-bearing and crypto-backed stablecoins (USDS, DAI, USDe, etc.) account for an additional approximately $65 billion. The overall pool of floating capital seeking compliant on-chain yield, in our estimation, far exceeds $250 billion and is growing at a high single-digit percentage month over month. The question is no longer whether there is on-chain capital looking for yield, but rather what these capitals are allowed to do.

What Asset-Backed Credit Is and Why It Works

When we talk about asset-backed credit, we refer to something specific. A loan is asset-backed when (a) the borrower pledges identifiable real-world collateral, (b) the lender has perfected security interests in that collateral, (c) recourse does not require the borrower's cooperation, (d) the loan-to-value ratio, term structure, and recourse waterfall are documented in legal agreements and remain enforceable upon default. In traditional private credit, this is common. In on-chain credit, it is the only distinguishing feature between effective structures and failed ones.

Effective types of collateral in asset-backed credit have three attributes: they are identifiable (not fungible), have an active secondary market for liquidation, and the lien process is well-understood in the borrower’s jurisdiction. Examples include: trade receivables, equipment financing, real estate-backed bridge loans, invoice factoring, structured consumer finance. Examples that do not work (or only work at wider spreads) include: future cash flows of operating businesses with no specific assets pledged, intangible intellectual property, illiquid private equity interests. Discipline is collateral first, borrower second.

Real estate-backed lending deserves special attention because it is the largest sector predicted among the entire tokenized opportunity. Deloitte's breakdown makes this point clear: by 2035, tokenized loans and securitization are expected to dominate the tokenized real estate portfolio, about three times the size of tokenized private real estate fund assets, and far exceed the tokenized equity of undeveloped or under-construction projects.

The earning of spreads is a function of work completed before deployment. Each loan undergoes credit assessment, pricing of default probability, default loss, and recourse processes. Each loan has documentation of covenants, reporting obligations, and triggering events. Each pool has a curator or evaluation partner who has a financial and reputational stake. Each treasury has a clear waterfall: who gets paid first when assets are liquidated, what triggers pool freezes, and how losses are allocated. Mechanisms are doing the work; marketing does nothing.

This is also why spreads persist. When tokenized T-bills cluster around about 3.5%, over-collateralized DeFi clusters between 2.5% and 4.2%, well-structured ABC pools remain steadily in the 4-5% range, with billions of dollars of capacity. This premium is compensation for the credit work that experts must complete. The protocol layer does not perform this work; it is built to enable external experts to curate it.

The Adverse Selection Problem and Why Most On-Chain Credit Has Not Solved It

Anyone who has worked in private credit knows that borrowers self-select into your products can tell you almost everything. The cheapest capital flows to the strongest borrowers. What you end up getting in credit is the credit that was rejected by everyone earlier in the stack. This is the adverse selection problem, and it is the only truly important issue in private credit.

On-chain credit, in its loose forms, has historically inverted the solutions to this problem. Early protocols offered capital to anonymous wallets willing to pay the highest rates. No identities. No recourse. No evaluation layer beyond pool statistics. The pressure period of 2022 made the consequences clear: pools marketing themselves as “institutional-grade” found their borrowers correlated, undisclosed, and in some cases even insolvent. Recovery on defaulted loans was effectively zero because the loans were unsecured and jurisdictions were not clear.

The effective version—permissive pools with KYC, off-chain documentation, dedicated curators, and real evaluation layers—now dominate the landscape of surviving on-chain credit markets. The on-chain market leaders have converged on essentially the same answer: the credit work happens off-chain, results are encoded in the pool. This is a significant improvement over permissioned unsecured lending. But this is not the endpoint of the design space.

Here’s a key point that the market has not internalized today: when asset-backed credit is within fund packaging, and the fund decides on underwriting, recourse, and waterfalls, the adverse selection problem is elevated a level. If fund managers have incentive misalignments, leverage exposure, or insufficient evaluations, then the on-chain treasury is merely a more transparent way of delivering bad risks.

Regulatory Turning Point: GENIUS, CLARITY and Yield Issues

If market structure is the first reason we are building now, American regulation is the second, and it has changed in a way that most operators are still catching up to over the past twelve months.

The GENIUS Act, as the first federal stablecoin framework, was signed into law on July 18, 2025. It establishes a federal regime for payment stablecoin issuers, requiring 1:1 reserves, and most importantly for our theory, it prohibits stablecoin issuers from directly paying interest or returns to holders in any form. The OCC proposed implementation rules on February 25, 2026. The statutory deadline for final regulations is July 18, 2026. The complete operating regime is expected to be in place by January 2027 at the latest.

The GENIUS Act left a loophole: it prohibits issuers from paying yields but does not fully tackle the issue of exchanges or associated platforms paying yields on stablecoin balances. The Digital Asset Market CLARITY Act, which is moving forward in the Senate and has passed the House, fills this gap. The latest draft prohibits providing yield directly or indirectly on stablecoin balances. The White House’s April 2026 paper “Impact of Stablecoin Yield Prohibition on Bank Lending” clearly outlines the policy logic: Washington wants stablecoins to become a payment rail rather than a deposit substitute and wants yields to flow through clearly defined investment products, not via the stablecoins themselves.

This is the regulatory window that most markets are mispricing. The set of products that can legally provide yield on on-chain dollars is being explicitly narrowed. The set of structures that can survive (registered funds, adequately disclosed lending treasuries, tokenized credit products) is being implicitly enhanced. The next twelve months will be a classification event.

The Treasury is Important Architecture

If stablecoin issuers cannot pay yields and exchanges cannot pay yields on stablecoin balances, then the only legal way to convert on-chain dollars into income is through a discrete, identifiable investment product. On-chain, that product is the treasury. ERC-4626 and its subsequent versions have become the de facto standard for tokenized yield-bearing positions, and treasuries are now the chassis for nearly all compliant on-chain credit products in the U.S. market.

In traditional asset-backed lending, packaging is ancillary. You raise LP commitments, draw down to the fund, deploy to loans, and distribute.

On-chain, treasuries do far more work than their traditional financial counterparts. A treasury serves as an issuance mechanism (it mints shares representing underlying loan claims), a disclosure mechanism (its accounting is publicly verifiable), a distribution mechanism (anyone with a wallet can interact with permission), a recourse mechanism (waterfalls and triggers are encoded), as well as increasingly becoming a regulatory containment vessel (permissions can enforce KYC, accredited investor thresholds, and jurisdictional restrictions in ways stablecoins cannot). When yields are prohibited at the stablecoin level, the treasury layer becomes the place where compliant, regulated, transparent yields are delivered.

This is why we believe the design of the on-chain treasury, its permissions, its accounting, its disclosure standards, its compliance posture, and most importantly how it encodes credit work will become the most important architectural choice in this market over the next eighteen months. In traditional ABC, you can have a mediocre fund packaging and excellent credit work, and still produce great products. On-chain, in the post-GENIUS/CLARITY world, a poorly designed treasury becomes a regulatory liability. A treasury that is merely tokenized fund equity only repositions the adverse selection problem rather than solving it.

Our Position

We focus on asset-backed credit because it is the only credit category on-chain with structurally defensible answers to adverse selection, and the data clearly shows demand on the supply side.

We take a compliance-first posture towards the U.S. because we believe the post-GENIUS and post-CLARITY U.S. regulatory framework will define the standards for how global on-chain yields are delivered. The cost of retrofitting compliance onto non-compliant structures is far higher than building it in from day one.

We view the treasury layer as a first-order design issue, not merely a packaging exercise, because we believe the architecture will be the place where regulatory, technical, and operational complexities compound over the next eighteen months. We are not interested in becoming another tokenized fund equity. We are interested in encoding evaluation, structuring, and recourse into the treasury itself at the protocol layer in ways the current generation of products does not.

Next Twelve Months

By July 2026, the OCC’s final GENIUS Act rules will be in place. By the end of 2026, the CLARITY Act framework is likely to have been enacted or be close to enactment. By January 2027, the complete payment stablecoin regime will be operational. Somewhere in that window, a generation of yield-bearing products designed for the regulatory environment before 2025 will need to be retired, while those designed for the post-CLARITY environment will need to be ready.

The on-chain capital base is already present. Stablecoin floating capital has reached approximately $323 billion. On-chain lending demand is growing at a triple-digit rate. Regulatory frameworks are being finalized, and with appropriate warning, they are moving in a direction favorable to well-structured, compliant, U.S.-based, treasury-backed asset-backed credit.