All of it. Sellers are locked in, buyers are locked out, and two trillion dollars a year of rent reaches nobody. All three descend from one fact about how property is owned: it has no denomination. Not a shortage of fractional products, and not interest rates.
US Bureau of Economic Analysis, National Income and Product Accounts, series A2013C1A027NBEA. Verify at fred.stlouisfed.org.
National accountants concluded long ago that a household living in a house it owns consumes housing services identical to those a renter buys, and that omitting them would make GDP depend on the tenure mix of the housing stock rather than on real activity. So the income is imputed, counted and published every year.
Nobody receives it. It is the largest income stream in the American economy that is measured and not collected, and it has grown at roughly seven percent a year for as long as the series has been kept.
The reason is not that the income is fictional. If the same household sold the house and rented it back, the rent would become an actual payment and an actual receipt, and the national accounts would barely notice. Same building, same occupant, same street. What changed is who holds title.
An owner faces a binary. Keep the asset and receive nothing, or sell it entirely and leave. There is no instrument between those two states, and the whole of the uncollected income sits in the gap.
This is not a call for fractional ownership, which has been sold since 1960. A REIT share, a syndication unit, a statutory trust interest: each is a fraction of a wrapper somebody assembled above the building, and every one of them needs a seller who is leaving. A denomination needs only an owner who is staying. That is the operation that does not exist.
The same defect produces a second, visible effect. An owner holding a mortgage below market cannot sell without simultaneously surrendering the financing, so two separate economic decisions are welded into one. The Federal Housing Finance Agency has measured what that costs.
If slicing the asset were sufficient, this would have been solved already. Tokenization has existed since at least 2018 and multiple funded platforms have launched.
As of March 2026, tokenized real estate onchain totalled roughly $439 million across 64 assets held by about 11,700 people. Tokenized treasuries stood near $10 billion. Across every category, about 93 percent of onchain asset value sits in non-transferable form, which means no price discovery is occurring at all.
Divisibility is necessary and demonstrably insufficient. Slicing an asset does not tell you at what price the slices sell, or to whom.
In conventional real estate an asset has a cap rate. A building yielding 7 percent is purchasable only by investors whose required return is roughly 7 percent. An investor with a 12 percent hurdle cannot buy it at any price the seller accepts. Every investor whose requirement fails to match the asset is structurally excluded from owning it, and that is true of every building on earth.
A senior tranche takes a priority position at a fixed floor. A subordinate tranche absorbs the residual at a higher clearing yield. Both are satisfied out of the same net operating income. The property no longer has a cap rate. It has a yield curve, and the buyer universe expands from a narrow band to essentially everyone with capital and a return requirement.
An unlock without an engine is a one-time liquidity event. Rent is collected in fiat, the property pays its expenses, and net revenue purchases the ecosystem token on the open market for distribution to that property's holders. Supply is permanently fixed.
When a listed company repurchases shares, capital leaves the operating business, which is why buyback-heavy firms are accused of underinvesting. Here the distribution and the growth expenditure are the same transaction: the value returned to a participant is what produces the behaviour generating the next unit of revenue. Occupants pay on time to accrue. Buyers stake to hold position. Sellers list because the pool is deep.
Swipe the diagram sideways, then tap a stage.
Hover a stage to trace what leaves it. Click to read what actually happens there, and click the curved return paths to see how participants re-enter as demand.
We distinguish these sharply, because the credibility of everything above depends on it. The property operating layer ran live on a real eight-unit building with occupants linking bank accounts and paying rent online. The token layer has never been connected.
| Component | Status | Evidence |
|---|---|---|
| Property management module | Production | Rent roll, operating statements, ledgers, work orders, governance history. Ran live on an eight-unit building. |
| Payment rails | Ran live | Plaid and Dwolla integrated; occupants paid rent online. Currently disconnected and would need reconnection. |
| Minting factory and tranche model | Demonstrable | Seller interface and live clearing calculator, end to end. |
| Onchain contract layer | Not built | Requires rebuild on an audited ERC-3643 base, identity registry, partition factory, and independent audit. |
| Token economy | Untested | Specified and modelled. Never run with live participants. |
| Regulatory qualification | Not filed | Regulation A, Tier 2. Form 1-A drafting not commenced. |
Eighteen pages, and a PDF if you would rather read it away from a screen. The anomaly quantified, why tokenization has not resolved it, the two mechanisms, the regulatory architecture, the economics for a capital partner, and six falsifiable predictions with the measurements that would disconfirm each one.
One prediction is testable by any rental operator without us: do occupants who accrue an ownership stake for paying on time actually pay on time more often? If that is false, we want to know.
We are looking for a capital partner able to deploy against property positions and take equity in the platform, and for operating counterparties whose revenue depends on transaction volume: title and settlement, mortgage origination and servicing, brokerage, property operations, and tokenization infrastructure.