Mortgage-backed securities are the instrument through which millions of individual home loans are transformed into institutional investment products. A bank originates mortgages from individual borrowers, pools them by loan characteristics (loan-to-value ratio, geographic distribution, credit score, term), and sells securities backed by the pool’s cash flows to institutional investors. The pool’s monthly mortgage payments flow through to MBS investors as periodic principal and interest distributions. This securitisation model — which dates to the 1970s — has made it possible for home buyers to access competitive mortgage rates by connecting their borrowing needs to global institutional capital, while providing institutional investors with diversified real estate debt exposure at various risk-return profiles.
The MBS Market: Scale and Structure
Market Size
The US MBS market is the largest fixed income market in the world outside of US Treasuries: SIFMA reported approximately $12 trillion in outstanding MBS as of year-end 2025, comprising agency MBS (guaranteed by Fannie Mae, Freddie Mac, and Ginnie Mae) and non-agency MBS (private label, issued without government guarantee). Global MBS markets add several trillion more in European covered bonds, Australian RMBS, and Asian residential mortgage securitisations.
The Agency vs Non-Agency Distinction
Agency MBS — backed by GSE guarantees — are the most actively traded, most liquid, and most institutionally held segment of the market. They represent approximately 70% of total US MBS outstanding and are held by the Federal Reserve, major banks, and institutional investors globally as one of the most liquid fixed income instruments available. Non-agency MBS — issued without government guarantee, typically from private label securitisation vehicles — carry the full credit risk of the underlying mortgage pool and offer higher yields reflecting that additional risk.
This distinction is critical for tokenized MBS development: non-agency MBS can be tokenised using existing private securitisation infrastructure. Agency MBS tokenization requires GSE participation — Fannie Mae, Freddie Mac, or Ginnie Mae — in the tokenized format, which has not yet occurred at meaningful scale but represents the largest potential tokenized fixed income market globally.
Live Deployments: Figure Technologies and the HELOC Model
Figure Technologies, operating on its Provenance blockchain, has tokenized over $1 billion in home equity loans — revolving credit facilities secured by residential real estate. The Figure model demonstrates the tokenized residential mortgage infrastructure at institutional scale: loans are originated directly through Figure’s lending platform, tokenised as they are created, and sold to institutional investors through Provenance’s institutional distribution infrastructure. The blockchain record serves as the authoritative loan register, eliminating the paper-based note and mortgage tracking that traditional mortgage securitisation requires.
The operational advantage Figure demonstrates is the elimination of the traditional securitisation ‘shadow’ — the gap between the loan note in one custodian, the mortgage in the land title registry, and the MBS certificate in a securities depository. In Figure’s model, the single blockchain record contains all three: the loan’s payment terms, the security interest, and the investment right. This unification of the loan record reduces both administrative overhead and reconciliation risk.
Australia’s Homeloans.com.au partnership with Helio Lending to create the first tokenized Australian MBS product demonstrates the model’s cross-jurisdictional applicability. Australian RMBS is a well-established institutional market; the Homeloans.com.au tokenization extends that market to blockchain-based issuance and distribution.
The Structural Complexity of MBS Tokenization
Prepayment Risk
Mortgage pools are subject to prepayment risk: homeowners can refinance their mortgages when interest rates fall or sell their homes early, returning principal to MBS investors ahead of schedule. Prepayments are beneficial when they occur at par but disruptive because they return capital when reinvestment rates are lower (in a refinancing wave driven by falling rates). Traditional MBS structures handle prepayment through tranching (CMOs allocate prepayments to sequential tranches) and prepayment modelling (PSA and CPR models). Tokenized MBS smart contracts must incorporate prepayment allocation logic that correctly implements the chosen prepayment model — a significantly more complex specification than a simple fixed coupon instrument.
Servicer Obligations
MBS payment administration involves a servicer — typically a bank or mortgage company — that collects monthly payments from individual borrowers, handles delinquency and default management, and remits collected funds to the MBS trust for distribution to investors. The servicer is a critical operational intermediary whose failure creates MBS administration disruptions. Tokenized MBS structures must either maintain a traditional servicer relationship (with the servicer reporting to the smart contract via oracle) or develop on-chain servicing capabilities — neither of which is a simple engineering problem given the complexity of mortgage delinquency management.
The Waterfall Payment Calculation
MBS payment waterfalls — the priority rules that determine how collected cash flows are distributed among different MBS tranches — are among the most complex calculation structures in fixed income. A standard CMO waterfall involves sequential allocation of principal, pro-rata interest distribution at different rates, overcollateralisation triggers, interest rate swap settlements, and deal-specific structural features that vary by transaction. Implementing a waterfall payment calculation in smart contract logic requires careful specification and rigorous testing — the kind of smart contract audit and specification review discussed in the challenges article elsewhere in this series.
The GSE Participation Question
The most significant open question for tokenized MBS at institutional scale is whether Fannie Mae, Freddie Mac, and Ginnie Mae will develop tokenized-compatible formats for their guaranteed securities. The GSEs’ January 2026 exploration of tokenized MBS as described in industry reporting remains at the exploratory stage — there are no confirmed production tokenized GSE MBS programmes. Without GSE participation, tokenized MBS is limited to the non-agency market segment that represents 30% of total US MBS outstanding — a large market in absolute terms but a fraction of the total institutional opportunity.
The political and regulatory pathway for GSE MBS tokenization is more complex than private-label tokenization: the GSEs are government-controlled entities whose major system changes require regulatory approval from their conservator (FHFA) and potentially Congressional engagement. The January 2026 SEC joint statement on tokenized security taxonomies addressed tokenized MBS in the context of existing securities law but did not specifically address the GSE participation question.
The Administrative Efficiency Opportunity
The MBS market generates an enormous volume of monthly administrative events: servicers collect and remit monthly mortgage payments from millions of individual borrowers; MBS trusts distribute those payments to institutional investors through complex waterfall calculations; paying agents process distribution instructions; and custodians reconcile holdings across multiple systems. The total administrative overhead of this process across the $12 trillion market is estimated in the billions of dollars annually — overhead that ultimately flows through to borrowers in mortgage rates and to investors in reduced net yields.
Smart contract automation of the payment waterfall, servicer remittance notification, and investor distribution could materially reduce this overhead for tokenized MBS programmes — with the most immediate gains available in the non-agency market where the GSE administrative infrastructure does not need to be maintained. For institutional investors in tokenized non-agency MBS, the combination of T+0 settlement and automated distribution provides both administrative cost savings and improved cash flow timing.
Disclaimer: This article is for informational purposes only and does not constitute financial or legal advice. Readers should conduct their own research and consult with qualified professionals before making any investment or business decisions.
Frequently Asked Questions
1. What is the US MBS market size?
Approximately $12 trillion in outstanding MBS as of year-end 2025 (SIFMA), making it the largest fixed income market in the world outside of US Treasuries. It consists of approximately 70% agency MBS (GSE-guaranteed) and 30% non-agency (private label) MBS.
2. What has Figure Technologies done in tokenized mortgage finance?
Figure Technologies has tokenized over $1 billion in home equity loans (HELOCs) on its Provenance blockchain, creating the most significant live tokenized residential mortgage product in the US. The Figure model unifies the loan record, security interest, and investment right on a single blockchain register, eliminating the paper-based note tracking of traditional mortgage securitisation.
3. Why is GSE participation critical for tokenized MBS at institutional scale?
Fannie Mae, Freddie Mac, and Ginnie Mae guarantee or issue approximately 70% of US residential MBS. Without GSE-compatible tokenized formats, tokenized MBS is limited to the 30% non-agency market. GSE tokenization requires regulatory approval from FHFA (their conservator) and potentially Congressional engagement — a more complex pathway than private-label tokenization.
4. What is prepayment risk in MBS?
The risk that mortgage borrowers refinance or sell their homes early, returning principal to MBS investors ahead of schedule. Prepayments disrupt yield calculations and reinvestment planning. MBS smart contracts must incorporate prepayment allocation logic (PSA and CPR models) — a significantly more complex specification than simple fixed coupon instruments.
5. How does tokenized MBS differ from traditional MBS for institutional investors?
Economically identical exposure with operational differences: T+0 settlement (vs T+2), 24/7 market access, automated waterfall distribution (vs manual calculation and instruction), and blockchain ownership records (vs depository book-entry). Administrative cost savings over the instrument’s life are the primary financial improvement for institutional investors.