The Lease Expiration That Lenders Model Twice: How In-Place NOI Assumptions, Renewal Probability Haircuts, and Roll-Down Risk Diverge Between Debt-Yield and Stress-Case Underwriting

The Structural Paradox

A commercial real estate lender receives an underwriting package for a stabilized office or retail property with a key tenant rolling off in year five. The lender's debt-yield team models the property with the lease in place, applies a modest haircut to renewal probability, and delivers a debt service coverage ratio that clears the lender's floor. Separately, the stress-case team models the same lease expiration, but with a different renewal assumption, a different roll-down trajectory, and a different timeline for lease-up. Both models are "correct." Both are embedded in the same loan agreement. Neither one aligns with the other.

This divergence is not accidental. It is structural. It reflects how commercial real estate lending has evolved across cycles: the debt-yield framework prices the loan assuming a borrower's base case holds; the stress-case framework protects the lender if it does not. The problem is that neither framework is reconciled with the other, and the borrower inherits the gap as refinance risk, covenant pressure, and capital injection triggers that emerge years into the loan term when the actual lease renewal outcome becomes known.

Why Lenders Model Lease Renewals Twice

The debt-yield framework starts with in-place net operating income. The lender takes the current lease, applies the current rent, and calculates a stabilized cash flow. If the lease expires during the loan term, the lender applies a renewal probability adjustment. This adjustment is typically conservative but not severe: 85 to 95 percent probability that the tenant renews, with a modest rent step-down if renewal occurs (perhaps 5 to 10 percent below current rent, reflecting market softness). The resulting pro forma NOI is then used to calculate debt service coverage. This is the number that appears in the offering memorandum and justifies the debt-yield spread.

The stress-case framework does something different. It models multiple lease renewal outcomes: renewal at various rent levels, or non-renewal with vacancy and subsequent lease-up. The stress case typically assumes a lower renewal probability (60 to 75 percent), a sharper rent decline if renewal occurs (15 to 25 percent), and a longer and costlier lease-up period if the tenant does not renew (12 to 24 months of vacancy, followed by below-market rents for the replacement tenant). The resulting pro forma NOI is lower than the debt-yield case, and so is the debt service coverage under stress.

Both frameworks coexist within the same loan agreement. The debt-yield case justifies the loan pricing; the stress case determines the covenant thresholds, reserve requirements, and capital injection triggers. A lender might price a loan at 4.50 percent on the assumption of 90 percent renewal probability, then embed a DSCR covenant of 1.20x based on a 70 percent renewal probability. If the actual renewal outcome lands between these two assumptions, the borrower faces a covenant cushion that is thinner than the debt-yield analysis suggested but thicker than the stress-case analysis assumed. The borrower is caught in a gap that was never explicitly modeled.

The Roll-Down Risk That Emerges at Refinance

The structural divergence between debt-yield and stress-case frameworks creates a specific form of refinance risk that lenders rarely articulate and borrowers rarely anticipate. Consider a property with a key tenant lease expiring in year five of a ten-year loan. The debt-yield model assumes the tenant renews at 90 percent rent of current. The stress-case model assumes 70 percent renewal probability and 80 percent rent if renewal occurs. At year five, when the actual lease renewal outcome is known, the lender must reappraise the property for refinance purposes.

If the tenant renews at or above the debt-yield assumption, the property refinances cleanly. The borrower's cash flow supports the original debt-yield DSCR, and refinance occurs at or better than the original terms. But if the tenant renews below the debt-yield assumption, or if the tenant does not renew at all, the property's NOI is now lower than the debt-yield case assumed. The refinance lender reappraises at the lower NOI, and the borrower must either reduce leverage, inject equity, or extend the loan at higher rates to bridge the gap. The cost of this refinance risk was embedded in the underwriting but never explicitly priced into the debt yield, because the debt-yield framework assumed the higher renewal scenario.

The stress-case framework, meanwhile, provides no comfort. The stress case was designed to ensure that covenants do not breach if the worst case occurs. It was not designed to identify the cost of refinance risk in the middle scenario, where the actual outcome is worse than the debt-yield case but better than the stress case. That middle scenario is the most likely one, and it is the one that lenders have systematically under-priced across cycles.

The Covenant Cascade That Tightens at Renewal

The divergence between debt-yield and stress-case frameworks creates a specific covenant dynamic that emerges at lease renewal. The loan agreement typically includes a DSCR covenant tied to a stress-case NOI assumption. As the lease renewal date approaches, the borrower's actual NOI may diverge from the debt-yield assumption but remain above the stress-case assumption. The covenant is not breached, but the covenant cushion is compressed. If the actual NOI lands in this middle band, the borrower has less flexibility to absorb subsequent shocks: interest rate increases, capital expenditure needs, or tenant credit deterioration.

Lenders often respond to this compressed cushion by embedding step-down covenants, which require higher DSCR ratios if certain events occur (such as lease renewal below a specified rent level). These step-downs are rational from a lender's perspective: they protect against the scenario where the actual NOI converges toward the stress case. But they are also a form of implicit repricing that occurs after closing. The borrower did not price this repricing risk into the initial underwriting, because the debt-yield framework did not quantify it. The covenant step-down converts what appeared to be a fixed-rate, fixed-covenant loan into a variable-covenant structure that tightens when lease renewal risk materializes.

The Capital Injection That Bridges the Gap

When the lease renewal outcome is known, and the actual NOI has diverged from the debt-yield assumption, lenders often require capital injections to restore the DSCR covenant. These injections are not presented as repricing; they are presented as protective measures. But they are, in effect, a form of refinance cost that was embedded in the underwriting but never explicitly disclosed. A borrower who understood that the debt-yield assumption had a 10 to 15 percent probability of being wrong might have negotiated different covenant thresholds, or might have sought a different lender. Instead, the divergence between debt-yield and stress-case assumptions was treated as a technical matter of underwriting methodology, not as a pricing issue.

The structural reason for this is simple: lenders cannot price refinance risk directly. They can only price the initial debt yield based on stabilized assumptions, then protect themselves via covenants, reserves, and capital injection triggers. The gap between the debt-yield assumption and the actual outcome at renewal is a form of tail risk that the debt yield does not fully compensate for, but the covenants do attempt to manage. Borrowers who understand this dynamic can negotiate more carefully: higher initial leverage in exchange for covenant step-downs that are smaller, or lower leverage in exchange for covenants that are flatter across the renewal scenario. But most borrowers do not understand the divergence, because it is embedded in the technical language of "renewal probability haircuts" and "stress-case adjustments," not in the commercial terms of the loan.

What I'm Watching Next

• Whether lenders begin to reconcile debt-yield and stress-case frameworks by embedding explicit "mid-case" NOI scenarios in loan agreements, with corresponding covenant structures that reflect the most likely renewal outcome rather than the most conservative one.
• Whether borrowers start to demand sensitivity tables showing DSCR and leverage across multiple renewal scenarios, so that the cost of refinance risk is visible at pricing time rather than buried in the stress-case assumptions.
• Whether the next credit cycle, when lease renewals occur at rents significantly below the debt-yield assumptions, reveals how many loans were underwritten with a structural blind spot between the two modeling frameworks, and how many borrowers face unexpected capital injection or refinance pressure as a result.