Glossary · Leasing

Renewal probability

Also called Renewal rate, Retention rate, Probability of renewal.

Renewal probability is the assumed chance that an expiring tenant stays rather than leaves, applied lease by lease. It blends the two very different outcomes of an expiration into one expected cash flow, and it is usually the single most valuable assumption in an office, retail or industrial model.

Updated August 6, 2026 · All terms

How it works

The two branches it blends are nowhere near each other. A renewal costs a modest improvement allowance, a reduced commission and no downtime. A vacate costs a full fit-out, a full commission, months of empty space and often free rent on top. On one suite the difference can run to several times the annual rent, which means a probability applied to it moves more dollars than most rent assumptions do.

It should be set tenant by tenant with a reason attached, not applied as a portfolio constant. A tenant that has renewed twice, has put its own capital into the space, has a use tied to the location or has expanded into adjacent suites is a different proposition from one whose parent has announced a consolidation. Market retention statistics are a starting point and say nothing about the particular tenant whose expiry is about to decide the deal.

A blended probability is also the wrong tool wherever one tenant dominates. Sixty-five percent of a single 25,000 square foot expiry is a cash flow that will never occur, because the tenant either stays or goes. On a concentrated rent roll the large expirations should be modeled discretely with both branches shown, and the probability weighting reserved for the tail of small suites where the averaging is genuinely doing statistics.

Formula

Expected cost of an expiration = probability x renewal cost + (1 - probability) x (new-lease cost + rent lost to downtime)
  • Renewal cost is the renewal improvement allowance plus the renewal commission
  • New-lease cost adds the full allowance, the full commission, free rent and the rent lost during downtime
  • The same weighting applies to the rent itself, since a renewal rent and a new-lease rent are rarely equal

Worked example

What ten points of renewal probability are worth

Illustrative. A 25,000 square foot expiry. A renewal costs $775,133 all in. A vacate costs $2,027,552 of leasing capital plus $600,000 of rent lost over nine months of downtime.

Renewal outcome, all in$775,133
Vacate outcome, all in$2,627,552
Difference between the two branches$1,852,419
Expected cost at 80 percent renewal$1,145,617
Expected cost at 65 percent renewal$1,423,480
Value of every ten points of probability$185,242

One assumption, adjustable in a single cell, worth $185,242 for every ten points. On a building with several expirations it is routinely the largest unevidenced number in the model, which is why lenders and buyers challenge it before they challenge market rent.

Conventions worth knowing

  • Renewal probability and renewal rent move together. A tenant kept by a rent concession renewed on economics the model did not assume.
  • Watch for renewal options. An option at a fixed rate below market is not a probability, it is a certainty that costs money.

The common mistake

Applying one probability to every lease in the building

A single 70 percent assumption across a rent roll treats an anchor tenant with eight years of history and a small suite on its first term as the same credit in the same space. It also conceals concentration: the expected cost looks smooth while the real outcome is a small number of large binary events. Set the probability tenant by tenant with a reason for each, model the largest expirations both ways rather than blended, and report what the cash flow looks like in the branch where the biggest tenant leaves.

Related terms

Every figure, traced to its source

Altyst computes these in exact decimal arithmetic, not with a language model, and clicking any number shows the formula behind it.