Canada’s rental market is entering a more complicated phase. The question for multifamily lenders, mortgage insurers and investors is no longer simply whether a market needs more rental housing. It is whether a particular building, with a particular unit mix and rent position, can achieve the occupancy and cash flow assumed in its underwriting.
In its June 2026 rental update, Canada Mortgage and Housing Corporation (CMHC) reported softer asking rents in major markets, growing competition from new rental supply and investor-owned condominiums, and longer lease-up periods for some newly completed properties. Yet average rents paid on occupied units continued to rise. Both observations can be true—and that distinction matters for credit decisions. Source: CMHC, 2026 Mid-Year Rental Market Update.
The implication is straightforward: asking rent alone is not a reliable proxy for effective income, and a citywide vacancy rate is not a substitute for building-level underwriting.
One rental market, several different realities
At least four measures are often blended together in market discussions, although each answers a different question:
- Asking rent is the price advertised to prospective tenants. It may not reflect concessions or a final executed lease.
- Contract rent is the amount specified in an executed lease, subject to its actual terms.
- Effective rent reflects the economic value of the lease after relevant incentives over a defined period. It depends on how concessions are allocated and whether they recur.
- Collected rental income depends on occupied, paying units, collections, turnover, and operating timing.
A newly completed tower advertising $2,600 monthly rent with two free months on a 12-month lease does not generate the same first-year base rent as a comparable occupied unit collecting $2,600 every month. The annualized effective rent in that simple example is $2,167 per month, before other operating considerations. A promotional concession may be temporary, but its effect on initial cash flow is immediate.
It would also be incorrect to apply that promotional effective rent forever. Underwriting should separately model the initial lease period, subsequent renewal assumptions, vacancy and turnover, and the possibility that incentives persist.
What the 2026 evidence actually says
CMHC observes that rental markets in major Canadian centres have moved toward more balanced conditions as supply increased and demand slowed. Newer, higher-priced buildings have experienced greater pressure than some older stabilized buildings and family-sized units. CMHC also reports that landlords have increasingly used parking discounts, move-in credits, gift cards and other incentives, with some concessions reaching multiple months of free rent. These are market observations, not a claim that every project or city faces the same pressure. CMHC, June 2026.
The market-wide vacancy rate can be particularly misleading. CMHC’s preliminary work indicates that the vacancy level consistent with balanced conditions differs by metropolitan area and over time. In its 2025 comparisons, CMHC reported a 3.0% apartment vacancy rate in Toronto and a 3.7% rate in Vancouver, but these figures alone do not establish the expected performance of a newly delivered, premium-priced building. Local submarket, age, unit configuration, pricing and competing supply all matter.
CMHC’s fall 2026 supply reporting similarly points to rental-market normalization rather than universal oversupply. The important underwriting task is therefore segmentation, not a blanket bullish or bearish conclusion. CMHC, Housing Supply Report.
A practical five-layer underwriting framework
1. Establish building and unit identity
Start with a verifiable inventory: building address, unit count, unit labels where available, bedroom and bathroom mix, floor area, building age, completion phase and property type. Reconcile permits, marketing materials, rent rolls and available public records. A building-level average obscures whether the vacant inventory is concentrated in studios, large family units or a premium floorplan that has a narrower tenant base.
Each source needs a provenance marker and observation date. Missing attributes should remain missing; they should not silently become precise model inputs.
2. Separate market rent from realized economics
For each unit type, distinguish comparable advertised rents, signed contract rents, concessions, occupancy, turnover and actual collections when available. Normalize utility inclusions, parking, furnishing, lease duration and incentive terms before comparing properties.
For example, a month of free rent can be treated as a year-one cash-flow reduction while the continuing contractual rent remains visible for the stabilization case. A lender may also test a downside case where similar concessions are required at each turnover.
3. Model the path to stabilization—not just the destination
Stabilized occupancy is only one assumption. Lease-up speed determines months of forgone income, carrying cost pressure and the timing of debt-service coverage. An assessment should show monthly unit absorption, expected concession cost, achievable effective rent, operating expenses and the point at which coverage reaches the institution’s required threshold.
Compare actual leasing against the original schedule. Three months of slower absorption is different from a permanent reduction in achievable rent, even if both reduce short-term income.
4. Stress market and asset assumptions independently
At minimum, examine a base case, a slower-lease-up case, and a combined downside in which both effective rent and stabilization speed deteriorate. Test interest and refinancing sensitivity separately from operating performance. Avoid treating a broad city rental index as though it predicts an individual project’s performance.
5. Preserve uncertainty and review triggers
Reports should disclose which values are observed, derived, model-estimated or supplied by the borrower. Assign review triggers such as persistent concessions, weaker renewal retention, missed absorption milestones, or diverging comparable rents. Independent validation and human review remain essential, particularly where data coverage is thin.
Illustrative case: the cost of slower lease-up
Consider a hypothetical 120-unit building with a headline asking rent of $2,500 per month for simplicity. This is an illustration, not a Canadian market benchmark or a Gnowise model result.
- Headline annual rent at full occupancy: 120 × $2,500 × 12 = $3.60 million.
- Scenario A: 95% occupancy, no concessions assumed, yields $3.42 million in annualized gross scheduled collections before bad debt, operating costs and other adjustments.
- Scenario B: 85% occupancy, with a one-month-free promotion applied to every occupied unit over a 12-month initial term, yields $2.805 million of first-year rent before those same adjustments.
- Difference: $615,000 in illustrative annual rental receipts, or approximately 18% below Scenario A.
These deliberately simplified scenarios isolate occupancy and a lease concession. They do not represent net operating income or a capitalization-rate valuation; they omit phased move-in dates, parking and other revenues, expenses, credit losses, rent regulations, tenant turnover and concession eligibility. A real underwriting exercise should model monthly cash flows and property-specific evidence.
The example nonetheless demonstrates why relying exclusively on the advertised rent and a target occupancy assumption can materially change the risk picture.
What lenders and insurers should ask
For a financing or insurance review, five questions help distinguish an attractive rental narrative from a defensible income assumption:
- How many comparable units are actually leased, at what effective rents, and as of what dates?
- Which floorplans account for current vacancy and what competing inventory is entering the same submarket?
- What incentives are currently required, and how would continuing them affect the first operating year and renewals?
- How sensitive are debt-service coverage, refinancing capacity and collateral conclusions to slower absorption or lower effective income?
- Which inputs are verified facts, external estimates or borrower forecasts—and who owns ongoing monitoring?
For appraisers, the same distinctions support the appropriate selection and adjustment of rental evidence. For asset managers, they support earlier identification of performance deviations. Municipalities and housing agencies can also benefit from distinguishing broad rental supply from the affordability and unit mix actually delivered.
From market commentary to decision intelligence
The strongest rental intelligence links several levels of evidence: national and metropolitan context, neighbourhood conditions, building characteristics, unit-level rent observations, and actual performance where available. These levels should be connected, but never conflated.
Gnowise’s valuation, rental and market intelligence work is designed around this broader property-level decision context. Institutions evaluating multifamily exposure can use a combination of well-documented rental evidence, property attributes, local market indicators and scenario analysis to support due diligence. Coverage, timing, verification and model limitations must be assessed for the specific geography and asset rather than assumed universal.
The next generation of multifamily underwriting will be measured less by the volume of data collected than by its ability to answer a precise question: what cash flow is supportable for this building, under these market conditions, with this degree of evidence?
Explore Gnowise’s property intelligence capabilities or contact the team to discuss rental and collateral analysis workflows for your institution.
Sources and methodology notes
- CMHC — 2026 Mid-Year Rental Market Update (June 9, 2026): asking rents, concessions, vacancy patterns and the preliminary balanced-vacancy analysis.
- CMHC — Fall 2026 Housing Supply Report: recent rental construction and absorption context.
- OSFI — Annual Risk Outlook 2026–2027: supervisory emphasis on current collateral valuations and proactive portfolio risk management. The example above is an editorial illustration, not an OSFI-prescribed calculation.