Trepp data show conduit CMBS cash-in pressure easing while leverage lags

A sources-and-uses disclosure answers something most conduit loan filings leave implicit: whether the borrower put money into a refinancing or took money out of it. Trepp's latest read of those disclosures describes a market that has climbed out of the defensive conditions of 2023 and 2024 without returning to those of 2021 and 2022.

Trepp's sample is defined tightly: it tracks sources and uses for the 15 largest loans in each conduit CMBS deal over a window running from May 2021 through July 2026, making the findings a read on the top of the conduit market, the big-ticket collateral that drives most deal performance, rather than on the whole of securitized commercial real estate debt; conduit deals outside that set, and smaller loans within it, are not in the tabulation.

The cash-in share carries the first part of the story: loans on which borrowers had to contribute rather than extract climbed from 14.4% of classified volume in 2022 to 42.5% in 2024, as a rate shock raised debt costs while property values reset and left sponsors to close the gap between what a lender would advance and what a maturing balance required. Trepp says that pressure has since eased. Leverage, though, has recovered more slowly than loan purpose, and that gap is where the recovery's remaining risk sits.

Two ratios do the work, and they measure different things: the cash-in share counts how much of classified volume needed borrower equity, a test of how many deals required the sponsor to plug a gap, while the debt-stack figures measure how much of the capital structure the borrower funded rather than the lender. Both worsened through the reset, and both have improved more slowly than the loans themselves.

Equity extraction now runs near 16% of the total debt stack, below the 20% to 23% range Trepp records for 2021 and 2022, while cash-in borrowers still contributed 9.6%, nearly double the 4.9% they put in during 2021. If equity extraction is the clearest window onto how much cash-out leverage lenders will write, the four- to seven-point shortfall against the 2021-22 range suggests how much they are still withholding. Trepp's own framing is a two-speed market, and the label holds up: purpose has normalized ahead of leverage.

The cash-in line is the more stubborn of the two: borrowers in that cohort are supplying roughly twice the debt-stack share they did in 2021 even as the broader market loosens, which suggests the improvement is not spread evenly across the book. Trepp's own account leaves room for that reading: stronger assets can again return capital to their owners while lenders stay conservative about leverage everywhere else.

Read at face value, the two-speed description is a statement about underwriting discipline as much as about stress: cash-in pressure measures the discipline still being imposed, and equity extraction is the reward for assets that clear the bar. A market can post improving cash-in numbers for as long as lenders keep leverage tight, and Trepp's data show that is close to where things stand. The gap between purpose and leverage is the part of this cycle lenders have not given back.

Where the recovery comes from matters as much as how far it has run: acquisition lending is still subdued in Trepp's data, and the conduit recovery continues to lean on refinancing of existing ownership rather than a broad increase in property sales. The distinction matters. A sales-driven market pulls new assets into collateral, new sponsors into the borrower base, and fresh marks against which everyone can price. A refinancing-driven one cycles the same buildings past the same owners, so the deals getting done are owners rolling existing positions rather than buyers establishing new ones.

That is the shape of the maturity wall this publication has argued is clearing through rolling recapitalization rather than a wave of distress, and the sources-and-uses numbers cut in the same direction while adding a caveat. If lenders will extend but hold the advance rate down, maturing loans are being resolved with more sponsor equity than the 2021-22 precedent implies. The wall can clear without forced sales and still leave the market a collateral pool carrying more borrower cash and less lender leverage than it did five years ago, leaving lenders better protected and owners with more of their own capital at risk.

PRED has reported a related split in Trepp's loan-level work: acquisition debt yields barely moved across the 2021-22 vintages even as refinance yields repriced sharply. Set beside the sources-and-uses figures, that points to a market whose refinancing and acquisition engines are running on different terms.

Trepp's full report breaks cash-in and cash-out trends out by property type; the excerpt released this week does not carry the sector splits. When those surface, the acquisition share of new conduit financing is the line worth watching. Until it moves, the leverage figure will be the first to show whether conduit lenders have actually regained their appetite.