Most rent rolls report one vacancy number, and it is the physical one: empty units divided by total units. That number is cheap to produce and it answers a narrow question. Economic vacancy answers the question an owner is asking, which is how much of the building's earning capacity turned into money.
The definition, and the formula
Economic vacancy is gross potential rent minus rent collected, divided by gross potential rent.
Gross potential rent is what the property would earn in a period if every unit were leased at current market rent and every tenant paid in full. It is a ceiling, not a forecast. Nobody hits it. The point of the metric is the size of the distance between that ceiling and the deposit that lands in the account.
A worked example with round numbers. Forty units, market rent of 1,900 a month, so gross potential rent for the month is 76,000. Two units are empty, which is a physical vacancy of 5 percent. But you also gave a half month free on three new leases, six renewals are sitting 150 below market, and one tenant did not pay. Collected rent comes in around 67,000. Economic vacancy is close to 12 percent. The physical number said 5.
Both numbers are correct. They measure different things. The second one is the one that pays the mortgage.
Why physical vacancy understates the loss
Part of the confusion comes from borrowing a market statistic and using it as an operating metric. CMHC's Rental Market Survey defines a unit as vacant only if, at the time of the survey, it is physically unoccupied and available for immediate rental, which explicitly excludes a unit where a new lease has been signed and a unit undergoing major renovations.
Read that twice, because two of your worst earners fall outside it. A suite gutted for eight weeks is not vacant by that definition. A suite leased on the 3rd for a move-in on the 1st of next month is not vacant either. Both collect nothing. CMHC is measuring the supply available to a renter in a city on a given day, which is the right definition for that job and the wrong one for yours. It is also worth knowing that CMHC collected an availability rate only between 2005 and 2017 and has since discontinued it, so there is no published national series that closes the gap for you.
Use the market number as a benchmark for pricing. For an apartment vacancy reference, CMHC reported 2025 rates of 3.0 percent in Toronto, 2.9 in Montreal, 3.7 in Vancouver and 5.0 in Calgary. Do not put any of those beside your own economic vacancy and conclude anything. They are not the same measurement.
The four gaps between potential and collected
Every dollar of economic vacancy sits in one of four buckets, and the split matters more than the total.
- Physical vacancy. Days a unit was available and unrented. This is the only bucket most reports show.
- Concessions. Free months, waived parking, moving credits, signing incentives. Real rent, given away, on purpose.
- Loss to lease. The spread between what your occupied units are paying and what they would rent for today. Fully leased buildings can carry a large one.
- Bad debt and non-revenue units. Rent billed and not collected, plus superintendent suites, model units and offices carried at zero.
A property at 12 percent economic vacancy because of loss to lease has a pricing and renewal problem. The same 12 percent driven by turn time is a maintenance scheduling problem. The same 12 percent from concessions is a demand problem you have been papering over. One number, three completely different weeks of work.
Where the metric gets misused
Using in-place rent as the denominator. If gross potential rent is built from what tenants currently pay, loss to lease vanishes and economic vacancy collapses toward the physical number. The denominator has to be market rent, refreshed at least quarterly.
Filing concessions under marketing. A free month is forgone rent, not an advertising line. Booking it anywhere else makes the property look healthier than it is. It can still be the right move, and often is, but only if you can see what it costs.
Reporting it once a year. Economic vacancy is a rate over a period. Quarterly is the longest interval that still lets you correct anything.
Treating it as a scoreboard. It is a diagnostic. On its own it tells you almost nothing. Split into the four buckets, it tells you what to do on Monday.
Run it on one property this week
- Pull the unit list with contract rent for each occupied unit.
- Set a market rent per unit type from what you are asking today on live listings, not from last year's budget.
- Multiply market rent by unit count for the month. That is your gross potential rent.
- Subtract the four buckets separately: vacant days, concessions granted, the in-place gap on occupied units, and uncollected or non-revenue rent.
- Divide the total by gross potential rent. Keep the four sub-totals visible next to it.
An hour of work for a small portfolio, and the split is usually the surprise. Scattered-site landlords tend to find it in turn time, because a house between tenants is unstaffed and nobody is counting the days. Newer lease-ups tend to find it in concessions.
What changes once you have it
Pricing decisions stop being one-sided. Dropping the ask closes physical vacancy and widens loss to lease for years, which is why a free month often costs less than a lower rent even though it feels more expensive. You can see both effects in one metric instead of arguing about them.
Turn time gets a dollar value, so a two-week delay on a paint crew stops being a scheduling annoyance and becomes a number the owner can compare against the cost of a second crew. And when a fully occupied building underperforms its budget, you can point at the bucket responsible instead of defending an occupancy rate that looks fine.
Full occupancy is not the goal. Collecting the building's earning capacity is, and you cannot manage that with a metric that stops counting the moment someone signs.