In 1970, the Government National Mortgage Association — Ginnie Mae — issued the first mortgage-backed security (MBS): a standardized, tradeable representation of pooled home loans backed by physical assets. The mortgage became a token. Fifty-six years before anyone used the word "tokenization," the US government invented it for home loans.
By 2026, the US MBS market stands at approximately $12 trillion in outstanding issuance — the largest fixed-income market in the world after US Treasuries. Every time a bank originates a mortgage and sells it to Fannie Mae or Freddie Mac, that loan is pooled with thousands of others and repackaged into a security held by pension funds, insurance companies, and central banks from Singapore to Frankfurt. The homeowner making a monthly payment is funding returns for investors who have never heard of their street.
Three Problems the Original MBS Never Solved
Settlement latency. A traditional MBS trade settles in 48-72 hours through a chain of custodians, clearing brokers, and the Federal Reserve's book-entry system. Blockchain settlement could achieve finality in minutes, freeing collateral immediately rather than locking it for days. For the largest institutional MBS holders — central banks and pension funds with hundreds of billions in positions — the collateral mobility improvement is material.
Pool opacity. When a traditional MBS is issued, the underlying loan tape is provided at origination and updated monthly. Investors have no real-time visibility into prepayments, delinquencies, or modifications on individual loans. A blockchain-recorded MBS could make the loan tape a live feed: every payment, every delinquency, every modification recorded on-chain and visible to token holders in real time. This is the transparency improvement the 2008 crisis made obviously necessary — credit analysts who had to request data from servicers and wait weeks would instead query a public ledger.
Minimum investment size. Agency MBS trade in $1 million minimum increments at the institutional level. A tokenized MBS with $1,000 minimums changes who can earn mortgage income. The yield — currently around 5.5-6.5% for 30-year agency MBS — is higher than comparable Treasury yields and carries the same government guarantee.
Who Is Building It
Figure Technologies has originated over $10 billion in home equity loans on the Provenance blockchain — a purpose-built financial ledger — and has securitized those loans as blockchain-native securities since 2019. Loans are originated, serviced, and settled entirely on-chain. The securitization process that takes weeks for traditional pools takes days for Provenance-based pools because the loan data is already standardized and machine-readable at origination.
The Federal Home Loan Banks — the cooperative system providing liquidity to mortgage lenders — have explored tokenized collateral for advance lending. If a member bank's mortgage collateral is tokenized, pledging it becomes near-instantaneous rather than requiring physical document delivery.
The 2008 Lesson Applied
The 2008 financial crisis was widely described as a failure of mortgage-backed securities. It was not. It was a failure of mortgage underwriting — the process of deciding which borrowers qualified for loans. The securitization infrastructure worked exactly as designed: it efficiently distributed risk to investors who wanted it. The underlying mortgages were the problem, and the risk ratings were wrong.
Blockchain tokenization does not fix bad underwriting. What it addresses is information asymmetry. In 2007, investors in the riskiest MBS tranches had static disclosure documents, ratings opinions, and monthly servicer reports. By the time delinquency data showed systemic problems, the securities were already widely distributed. A live on-chain loan tape changes that equation — not by eliminating credit risk, but by making it visible in real time to anyone who holds a token.
→ Tokenized Treasuries — the other government-backed yield product
→ The Dutch East India Company — the original financial instrument
→ How to evaluate an RWA project