Spatial intelligence guide

Location Risk in CRE Models: What Goes Unmodeled

Black-and-white aerial view of dense city streets for location risk assessment

Quick answer: location risk in commercial real estate is the risk that the submarket around a site — its catchment area, transit access, competitive supply, and demographic trajectory — cannot support the rent and absorption assumptions in the underwriting model. Unlike capex or vacancy, location risk cannot be fixed after acquisition, yet in most pre-development models it is inherited from broker comps rather than modeled as its own variable.

The thing you can’t fix after signing

Most underwriting errors are recoverable. Capex runs 15% over budget; you adjust. Vacancy comes in 200bps higher than projected, so you extend the leasing timeline or renegotiate terms. These are execution problems, and execution problems have solutions. The submarket story is harder. If you bought into the wrong location, or overestimated what a location could support, the rest of the model is pulling against that from day one. You can renovate the building. The catchment area, transit access, and competitive supply pipeline stay. This isn’t unique to one asset class, and it plays out differently in retail, office, and logistics. But the common thread is that location-level assumptions usually don’t get modeled with the same rigor as financial ones, even when they carry more of the return.

Why most teams work with a single layer when there are at least three

When analysts do think about location, they tend to focus on the city level: employment trends, population growth, GDP. Most IC decks still rely on metro averages here, partly because submarket data is harder to pull and partly because the city story is cleaner to present. The macro picture matters, but it’s also the easiest thing to get directionally right. The harder work is at the submarket level. This is where rent growth actually plays out, and it varies a lot even within the same market. A submarket with 12% year-over-year growth in competitor foot traffic reads very differently from one with an 18% decline. That gap isn’t visible from city-level data. You need to go closer. Below that is the parcel itself. Zoning, overlay restrictions, utility capacity, ingress and egress. Teams often treat these as a checklist: permitted use confirmed, tick. They are not always modeled as variables with real cost and timeline consequences. A utility extension that adds six months and $800k to a project was technically “confirmed” in due diligence. It just wasn’t modeled. In practice, the submarket layer gets the least attention. The city story goes in the investment memo. The parcel constraints surface in legal. The block-level dynamics fall into the gap between the two.

What the post-COVID office split actually showed

The best recent example of location risk materializing at scale is the office market. A 2021 study published in the Journal of Urban Economics estimated that post-COVID, commercial rent gradients fell by roughly 15% in transit-dependent cities, and the premium for proximity to rail stations also declined. Columbia Business School research cited by NBER projected NYC office values 39% below 2019 levels a decade out, on the path where hybrid work persists. Meanwhile, PwC’s 2025 outlook notes that CBD buildings with strong transit access and modern amenities are among the clearest winners of the current cycle. At the same time, the overall CMBS delinquency rate for office hit 11.66% in August 2025, the sector’s worst-ever level. The split was largely determined by location. And the signals were in location-level data before leasing began. Transit dependency, submarket tenant composition, distance from residential density. These variables were available. Most teams just hadn’t modeled them as assumptions that could move. For what it’s worth: quantifying exactly how much of the value destruction was “location” versus “structural market shift” is genuinely hard, and not all analysts measure this the same way. But the directional story holds.

A few things that tend to show up in practice

Broker comps are the most common input for rent assumptions, and they’re usually fine as a starting point. The problem is when they become the only input. That happens more often than people admit. Comps reflect what was signed 6–18 months ago in buildings that may have a different profile, tenant mix, or access situation than yours. In a submarket that’s moving, that lag matters more than it looks on a spreadsheet. New supply pipelines get underweighted in a specific way: teams check what’s been permitted, but less often track what’s actively under construction or what a large landlord two blocks away has been quietly pre-leasing for the past year. We’ve seen deals where the submarket supply picture was already outdated at signing, not because anyone missed something obvious, but because that data lives across four different sources and nobody had aggregated it recently. Exit assumptions are where the location story tends to get the least scrutiny. Most models use a spread over going-in yield, or a cap rate benchmarked at the metro level. Whether the specific submarket has actually traded, at what volume, and with which buyers often doesn’t make it into the model. For assets in thinner markets, that gap can be meaningful at exit. Cushman & Wakefield’s post-COVID analysis found that employees living within a mile of their workplace return to the office at over 90% of pre-pandemic levels, while those more than three miles away sit at around 70%. Whether that’s a location variable or a tenant preference variable is a reasonable debate. But it’s the kind of thing worth having in a pre-development analysis of an office asset, rather than assuming it away.

Here’s a way to test whether the location view is actually modeled or just assumed: ask what happens to the deal if the submarket story is wrong by 20%. If the answer traces back to documented, submarket-specific data, good. If it traces back to “we used market rents from the broker,” the location hasn’t been modeled. It’s been inherited. This isn’t about adding complexity for its own sake. It’s about making the assumptions visible so they can be questioned, updated, and defended by the team, lenders, and an investment committee.

Aino pulls submarket vacancy trends, competitive supply, infrastructure status, zoning, and demographic data for any address — usually before an analyst would finish pulling the same data manually. If you’re working on a pre-development deal and want to see what that looks like for a specific site, you can run a free analysis here.

Frequently asked questions about location risk

What is location risk in commercial real estate?

Location risk is the risk that the submarket around a property, including its catchment, transit access, supply pipeline, and demographics, cannot support the rent and absorption assumptions in the underwriting model. It is the one input that cannot be fixed after acquisition.

Why are broker comps not enough for rent assumptions?

Comps reflect deals signed 6–18 months ago in buildings that may differ in profile, tenant mix, or access. In a moving submarket that lag hides risk, so comps should be a starting point that is validated with current submarket vacancy, supply, and demographic data.

How do you stress-test a location assumption?

Ask what happens to the deal if the submarket story is wrong by 20%. If the answer traces to documented, submarket-specific data, the location is modeled; if it traces back to broker rents, the location has been inherited, not modeled.

Analyze your locations now → https://app.aino.world

What practitioners say

“It exposed location risks our spreadsheet would never have caught before underwriting.”
— Marcus, Development Director