The Yield That Doesn’t Compress

Three sources of onchain yield. Two of them fall when the cycle turns. One doesn’t.

The useful question about a yield is not where it comes from. It is why it persists, or why it disappears. Two yields quoted at the same number today can behave in opposite ways a year from now, because the force that sets each one is different. Onchain yield sorts into three tiers, and the tier decides whether your return survives the next turn of the cycle.

The rate

The first tier is tokenized treasuries: funds that wrap government paper onchain. The yield here is the risk-free rate, and that is both its strength and its ceiling. When the policy rate is 5%, the product pays close to 5%. When the central bank cuts to 3%, it pays close to 3%. A two-point cut removes roughly 40% of the yield, and nothing about the onchain wrapper changes that, because the wrapper was never the source. The yield is government policy, and policy is the one variable the holder has no claim on.

The spread

The second tier is tokenized credit. Onchain lenders earn a spread over their cost of funds, and that spread is the yield. Two forces compress it. The first is rates, because most credit is priced off a floating base, so the yield falls with the curve. The second is competition: as more capital arrives to lend against the same borrowers, the spread each lender can charge narrows. This is structural, and it is happening now. A credit yield is a function of how much capital is chasing how many borrowers, and that ratio moves against the lender whenever capital is plentiful.

The margin

The third tier is operational yield. The return is the operating margin of a physical business, captured as cash flow rather than lent against. A coin laundromat that grosses a 30% margin grosses a 30% margin whether the ten-year sits at 2% or 5%, because quarters going into a washing machine are not priced off the rate curve. The yield is set by the economics of the business: utilization, pricing, and the cost of running the machine. None of those inputs is set by the central bank, and none by the supply of lending capital. In this category gross yields run 15 to 25%, with a target effective yield of 13 to 15% after the liquidity buffer, and that level is a function of operations, not the cycle.

What this means for an allocator

This is not a claim that operational yield is safer. It carries operational risk that a treasury does not, and that is the trade. The claim is narrower and more useful: operational yield is set by a different variable, so it does not compress when rates fall or when credit spreads tighten. In 2026, with the policy rate volatile and tokenized credit spreads narrowing under inflows, the first two tiers are both giving yield back. A book built only on those tiers is a single position on the rate and credit cycle, however many issuers it holds. Operational yield is the line on the page that holds its level while the others decline. That is its job in a portfolio, and it is the reason to hold it alongside the other two rather than instead of them.

The risk, and what manages it

Operational yield is not safer than a treasury, and the distinction is worth stating plainly. A treasury does not break down, take on a weak operator, or sit idle when local demand falls. What operational yield offers is not less risk but a different risk, on a different axis, one that does not compress when the rate cycle does.

It is also a risk that can be managed rather than waited out. When an operator underperforms, the telemetry registers it first: cycle counts and uptime decline weeks before the cash flow does, which turns a potential default into an early-warning signal. The response is reassignment rather than liquidation. The operator is replaced, utilization recovers, and the cash flow resumes, because the position was never the operator’s creditworthiness. It was the machine.

Behind that sit several layers, each sized so the one beneath it is rarely tested. The operator’s equipment serves as collateral through leaseback and step-in rights. A 10% origination overcollateralization cushion absorbs underperformance. A liquidity buffer keeps distributions whole while an operator is replaced. And the portfolio spans more than 1,000 assets across laundromats, commercial air-conditioning units, and vertical farms, so no single machine determines the outcome.

At DualMint, that structure has produced twelve consecutive months of distributions across more than 1,000 assets on Arbitrum, Base, and Peaq, with zero operator defaults.

Why it matters

A yield number tells you what you earn today. The force behind the number tells you what you earn after the cycle turns. Treasuries pay the rate, credit pays the spread, and both move when the cycle moves. Operational yield pays the operating margin, which does not. For an allocator watching the first two compress, that distinction is the difference between a return that survives the year and one that quietly erodes with it.