The blended market leasing assumption gives the correct expected cost and a cash flow that will never actually happen.
Blend each market leasing assumption as the renewal probability times the renewal term plus the rest times the new-tenant term. At an illustrative 70 per cent renewal probability, downtime of 0 and 9 months blends to 2.7 months, and the expected cost of a 25,000 sq ft expiry is 1,421,667, or $56.87 per sq ft. That figure is right as an expectation, and wrong as a description of the year the lease actually ends.
Worked in full in Real Estate Financial Modeling by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
Every property model has to say what happens when a lease expires. The standard shortcut, used in ARGUS and in most spreadsheet models, is to avoid modelling two branches, renewal and re-letting, and instead run a single blended path with every assumption weighted by the probability that the tenant stays. It is a sensible shortcut for value. It is a poor one for anything that has a threshold.
| Term | Renewal | New tenant | Blended |
|---|---|---|---|
| Downtime, months | 0.0 | 9.0 | 2.7 |
| Rent-free, months | 2.0 | 6.0 | 3.2 |
| Tenant improvements, $/sq ft | 15.00 | 50.00 | 25.50 |
| Leasing commissions, $/sq ft | 9.00 | 18.00 | 11.70 |
The suite sits in a 100,000 sq ft building let at the same $40.00, so building rent is 4,000,000, NOI after 400,000 of non-recoverable costs is 3,600,000, and debt service is 2,600,000: a 1.38x DSCR when stabilised, against a 1.15x covenant.
Blended term = p × renewal term + (1 − p) × new-tenant term
Cost of the expiry = (downtime + rent-free) × monthly rent + (TI + LC) × area
In Excel: =Renew_Prob*Renew_TI+(1-Renew_Prob)*New_TI for each term. Monthly rent on the suite is 25,000 × $40.00 ÷ 12 = 83,333.
| Path | Lost rent | TI + LC | Total | Per sq ft |
|---|---|---|---|---|
| Renewal | 166,667 | 600,000 | 766,667 | 30.67 |
| New tenant | 1,250,000 | 1,700,000 | 2,950,000 | 118.00 |
| Blended | 491,667 | 930,000 | 1,421,667 | 56.87 |
Because every term is linear in the probability, the blended total equals 70 per cent of the renewal cost plus 30 per cent of the new-tenant cost: 1,421,667 either way. For a valuation, which discounts expected cash flows, the blend loses almost nothing.
Now look at the year itself. The new-tenant path has 15 months of lost rent, so 12 fall in the first year and 3 in the second.
| Path | Rent lost in year 1 | NOI | DSCR | Cash after debt and leasing costs |
|---|---|---|---|---|
| Renewal (70%) | 166,667 | 3,433,333 | 1.32x | 233,333 |
| New tenant (30%) | 1,000,000 | 2,600,000 | 1.00x | −1,700,000 |
| Blended | 491,667 | 3,108,333 | 1.20x | −421,667 |
The blended model reports a 1.20x DSCR, comfortably above the 1.15x covenant, and a cash need of 421,667. Neither will happen. With 70 per cent probability the building passes easily and generates cash; with 30 per cent probability it falls to 1.00x, breaches the covenant, and needs 1,700,000 of fresh equity in the year. The new-tenant cost is 2.08 times the blended figure, and the blended year is not a scenario at all: 2.7 months of downtime is an average of 0 and 9, a void neither path produces. Because the new-tenant path pushes 3 of its 15 lost months into the second year, the blended year even loads more lost rent into year one than the probability-weighted 70/30 average of the two paths.
The threshold to test. At the 1.15x covenant, NOI can fall to 2,990,000, so the building can lose 610,000 of rent in a year, about 7.3 months of the suite. The renewal path loses 2.0 months, the blend 5.9, the new-tenant path 12. The covenant question is a question about the new-tenant path, and the blend cannot answer it.
| Renewal probability | Downtime, months | Rent-free, months | TI + LC, $/sq ft | Expected cost | Blended DSCR |
|---|---|---|---|---|---|
| 50% | 4.5 | 4.0 | 46.00 | 1,858,333 | 1.11x |
| 60% | 3.6 | 3.6 | 41.60 | 1,640,000 | 1.15x |
| 70% | 2.7 | 3.2 | 37.20 | 1,421,667 | 1.20x |
| 80% | 1.8 | 2.8 | 32.80 | 1,203,333 | 1.24x |
| 90% | 0.9 | 2.4 | 28.40 | 985,000 | 1.28x |
Each 10 points of renewal probability is worth 218,333 of expected cost, a tenth of the 2,183,333 gap between the two paths. The blended DSCR crosses the covenant at about 60 per cent, which invites the wrong conclusion: that the risk is about whether the probability is 60 or 70. At any probability, the real exposure is the same 1.00x on the new-tenant path; the probability only says how likely it is.
The common mistake is to run debt sizing, covenant tests and equity reserves on the blended path because it is the model's only path. For value, keep the blend. For anything with a threshold, run the new-tenant case for each material expiry as an explicit scenario, and size reserves on it. The second mistake is to layer a general vacancy allowance over the expiry years as well, which counts the same downtime twice. How the renewal assumption also drives the value of the rent roll is worked in what a lease mark-to-market is actually worth.
Blending by renewal probability gives the correct expected cost, 1,421,667 here, and a year-one DSCR of 1.20x that describes no possible outcome. Use the blend for value and the two branches for covenants and cash. The free workbooks for this book carry a rent roll and lease events inside a complete model, where the new-tenant case can be switched on one expiry at a time.
Each term, downtime, free rent, tenant improvements and leasing commissions, is blended as the renewal probability times the renewal value plus the remainder times the new-tenant value. At an illustrative 70 per cent, tenant improvements of $15.00 and $50.00 per sq ft blend to $25.50, and downtime of 0 and 9 months blends to 2.7 months.
For total cost, yes, because the blend is linear: on the illustrative 25,000 sq ft expiry the blended cost, 1,421,667, equals 70 per cent of the renewal cost plus 30 per cent of the new-tenant cost. For anything with a threshold, a covenant, a cash reserve or a funding need, no: the blend sits between two outcomes and describes neither.
It should be justified tenant by tenant, and its effect shown. On the illustrative expiry each 10 points of renewal probability is worth 218,333 of expected cost, and the blended year's DSCR runs from 1.11x at 50 per cent to 1.28x at 90 per cent. A single building-wide figure hides which tenants matter.
This article is one calculation from Real Estate Financial Modeling. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: real estate investing, finance and fund management → · All 453 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.