2026-09-02
There is a question every single-name REIT has to answer before anything else matters, and almost no analysis asks it. If you can buy the whole Canadian REIT sector in one ticker, for a fee measured in basis points, why would you take the risk of owning one building portfolio instead of nineteen?
So the hurdle here is not an abstract valuation target. It is two exchange-traded funds.
The iShares S&P/TSX Capped REIT Index ETF (XRE) tracks the S&P/TSX Capped REIT Index, holds roughly 17 names, charges a management expense ratio of 0.61%, and distributes somewhere in the region of 4.5% depending on the measurement date. The Vanguard FTSE Canadian Capped REIT Index ETF (VRE) tracks the FTSE Canada All Cap Real Estate Capped 25% Index, holds roughly 19 names, charges about 0.38%, and distributes closer to 4.2%. Note that VRE is not a pure REIT index, since the benchmark includes real estate operating companies as well as trusts.
A single REIT has to clear both funds on two counts to earn a place in a portfolio:
- It must yield more than the higher of the two, which sets the bar at roughly 4.5%.
- It must be safer than a diversified basket of its own peers.
The first test is arithmetic. The second needs a definition, because in the narrow sense a single security is never safer than a nineteen-name index. That is not what matters to an owner. Safety here means the durability of the distribution and the probability of permanent capital loss, not the wobble in the quote. On that definition a single trust absolutely can be safer than the index, and the reason is concentration.
XRE is market-cap weighted, which makes it top-heavy: CAPREIT, RioCan and Granite together account for more than 30% of the fund. VRE has roughly half its weight in five holdings. More to the point, both funds hold whatever the index tells them to hold. They own the office REIT paying out 104% of adjusted funds from operations. They own the trust with twelve times debt to EBITDA. An index cannot decline to buy a yield trap, and a person can. That is the entire argument for doing this work.
The fee is worth one line of arithmetic too. XRE’s 0.61% is a permanent, compounding drag that a direct holding does not carry. Over sixteen years, 0.61% a year costs you roughly 9% of your terminal capital. That is not nothing, but it is also not a reason on its own. It is a rebate you collect for accepting concentration risk, and the rest of this piece is about whether any of these names is worth accepting it for.
How the Price Levels Are Derived
Buy levels solve for the price at which a sixteen-year hold produces a 9% annualized return. The model takes the 2026 FFO or AFFO base, grows it at a stated rate, pays out a constant share of it as distributions, and exits at a terminal multiple in year sixteen. The internal rate of return on that cash flow stream is set to 9% and the equation solved for price.
Trim and sell levels use a shorter lens, because a trimming decision is not a sixteen-year decision. They solve for the price at which the forward five-year return falls to 6% and 4% respectively. Six percent is roughly where you would rather own the index. Four percent is roughly where you would rather own a bond.
The assumptions are below. They are the load-bearing part of this analysis and every one of them is arguable.
| REIT | 2026 base | Metric | Payout | Growth assumed | Terminal multiple |
|---|---|---|---|---|---|
| Granite | 5.50 | AFFO | 65% | 5.0% | 18x |
| RioCan | 1.61 | Core FFO | 72% | 4.0% | 14x |
| Primaris | 1.875 | FFO | 47% | 5.0% | 13x |
| Boardwalk | 4.70 | FFO | 38% | 5.0% | 15x |
| CAPREIT | 2.50 | FFO | 62% | 3.0% | 15x |
| SmartCentres | 2.32 | FFO | 80% | 2.0% | 12x |
| Choice | 1.09 | FFO | 72% | 2.0% | 14x |
| Chartwell | 1.16 | FFO | 56% | 5.0% | 17x |
Growth rates are set below near-term guidance in every case where guidance is elevated. Granite guides to 7% to 8% FFO growth for 2026; the model uses 5%, on the view that a 20% renewal spread narrows as Toronto industrial vacancy sits at an eleven-year high. Primaris targets 4% to 6% over three years and the model uses 5%. Chartwell has delivered twelve quarters of double-digit growth and the model uses 5%, because occupancy recovery from 94.3% toward 95% has limited runway left.
Two of these deserve a flag rather than a footnote, and both are covered in their sections below.
The Screen
Yields are unchanged from the first version. Every yield below is approximate, and three of the reference prices are derived from published yield figures rather than confirmed quotes.
| REIT | Reference price | Distribution | Yield | Beats 4.5%? | Payout coverage | Safer than the index? |
|---|---|---|---|---|---|---|
| SmartCentres (SRU.UN) | ~28.00 | 1.85 | ~6.6% | Yes | 90.5% of AFFO | No |
| RioCan (REI.UN) | ~21.40 | 1.158 | ~5.4% | Yes | 73.8% of core FFO | Yes, marginally |
| Choice (CHP.UN) | ~15.97 | ~0.79 | ~4.9% | Yes | near 90% of AFFO | No |
| CAPREIT (CAR.UN) | 34.25 | 1.55 | 4.53% | Yes, barely | 59% to 62% of FFO | Yes |
| Granite (GRT.UN) | ~90.00 | 3.55 | ~3.9% | No | 66% of AFFO | Yes |
| Primaris (PMZ.UN) | 22.63 | 0.88 | ~3.9% | No | 48.8% of FFO | Yes |
| Boardwalk (BEI.UN) | ~65.00 | 1.80 | ~2.8% | No | ~38% of FFO | Yes |
| Chartwell (CSH.UN) | ~21.52 | ~0.65 | ~3.0% | No | below 60% target | Marginal |
Two clear both bars: RioCan and CAPREIT. CAPREIT clears the yield test by three basis points.
The four names that fail on yield are the four with the strongest distribution coverage. Granite, Primaris and Boardwalk yield less than the index precisely because they retain more cash. Boardwalk’s 2.8% is the safest distribution on this list, and it is safe because it is small.
The Verdicts
| REIT | Reference price | Buy below (9%, 16y) | Trim above (6%, 5y) | Sell above (4%, 5y) | Return at market | Verdict |
|---|---|---|---|---|---|---|
| RioCan (REI.UN) | ~21.40 | 23.36 | 25.96 | 28.33 | 9.9% | Buy |
| Granite (GRT.UN) | ~90.00 | 96.38 | 111.67 | 122.12 | 9.6% | Buy |
| Primaris (PMZ.UN) | 22.63 | 23.80 | 27.52 | 30.10 | 9.4% | Buy |
| SmartCentres (SRU.UN) | ~28.00 | 27.26 | 31.22 | 33.99 | 8.7% | Hold |
| Boardwalk (BEI.UN) | ~65.00 | 60.03 | 75.99 | 83.22 | 8.3% | Hold |
| CAPREIT (CAR.UN) | 34.25 | 31.01 | 39.60 | 43.26 | 8.1% | Hold, patient |
| Chartwell (CSH.UN) | ~21.52 | 18.52 | 21.97 | 24.03 | 7.7% | Hold, at trim |
| Choice (CHP.UN) | ~15.97 | 12.81 | 16.11 | 17.58 | 6.9% | Trim |
The Scores
Risk Score, where 10 is strongest and safest. Weights: Financial Stability 0.30, Earnings Volatility 0.20, Business Model Risk 0.20, Macro Sensitivity 0.15, Market Risk 0.15.
| REIT | Fin Stab | Earn Vol | Biz Model | Macro | Market | Risk Score |
|---|---|---|---|---|---|---|
| Choice | 8 | 9 | 8 | 6 | 8 | 7.90 |
| Granite | 9 | 8 | 8 | 6 | 7 | 7.85 |
| Boardwalk | 7 | 8 | 8 | 5 | 6 | 6.95 |
| CAPREIT | 8 | 7 | 7 | 5 | 6 | 6.85 |
| Primaris | 8 | 6 | 5 | 5 | 6 | 6.25 |
| RioCan | 6 | 7 | 7 | 5 | 6 | 6.25 |
| SmartCentres | 5 | 6 | 8 | 4 | 7 | 5.95 |
| Chartwell | 6 | 5 | 5 | 5 | 4 | 5.15 |
Opportunity Score, where 10 is most favorable. Weights: Growth Potential 0.30, Unit Economics 0.20, Competitive Advantage 0.20, Valuation Asymmetry 0.20, Catalysts 0.10.
| REIT | Growth | Unit Econ | Comp Adv | Val Asym | Catalysts | Opp Score |
|---|---|---|---|---|---|---|
| Granite | 8 | 9 | 7 | 6 | 6 | 7.40 |
| Primaris | 8 | 7 | 6 | 6 | 8 | 7.00 |
| RioCan | 7 | 7 | 6 | 7 | 7 | 6.80 |
| Boardwalk | 4 | 8 | 7 | 7 | 7 | 6.30 |
| CAPREIT | 3 | 8 | 7 | 9 | 6 | 6.30 |
| Chartwell | 9 | 6 | 6 | 2 | 6 | 6.10 |
| SmartCentres | 4 | 4 | 8 | 7 | 6 | 5.60 |
| Choice | 3 | 7 | 8 | 3 | 6 | 5.10 |
Granite (GRT.UN)
The cleanest set of numbers in Canadian real estate, and the only name scoring above 7 on both measures. 2026 guidance is FFO per unit of 6.30 to 6.40 and AFFO per unit of 5.45 to 5.55, raised and narrowed at Q2 to FFO growth of 7% to 8%. Committed occupancy is 98.1% across 145 properties and 61.5 million square feet, with same-property NOI up 8.3% on a constant currency basis.
The balance sheet is the best of the group: net leverage 32%, debt to EBITDA 6.6 times down from 7.3 at the end of 2025, a weighted average borrowing cost near 2.61%, interest coverage above 5 times, roughly a billion in liquidity, and an AFFO payout near 66%. Fifteen consecutive annual distribution increases.
The growth is contractual. Warehouse rents roughly doubled over the past decade while many Granite leases were signed years ago, so renewals land about 20% higher with no new buildings required. Bear case: Toronto industrial vacancy has hit an eleven-year high, so future spreads will be smaller. Magna is still 26% of rent, down from 93% in 2012. Short-dated debt makes refinancing an active exercise.
Fails the yield test at roughly 3.9%, and clears the return test comfortably.
Price levels: on AFFO of 5.50 growing at 5% with an 18x exit, buy below 96.38, trim above 111.67, sell above 122.12. At 90.00 the sixteen-year return is 9.6%. Even at a 4% growth rate the buy level is above 84.00, so the conclusion holds under a materially more conservative assumption.
RioCan (REI.UN)
Retail committed occupancy hit a record 98.8% in Q2, the fourth consecutive quarter of commercial same-property NOI growth at 4% or better, with blended leasing spreads of 23.1% and new leases at 37.73 per square foot. Management raised same-property NOI guidance to 4.0% to 4.5% while holding core FFO guidance at 1.60 to 1.62 per unit.
Nobody has built a shopping centre in Canada in years, tenants have nowhere else to go, and RioCan’s old leases roll up rather than down as they expire.
Balance sheet is adequate rather than strong. Leverage near 9.4 times debt to EBITDA is the highest of the eight, but only about 30 million matures for the rest of 2026, roughly 70% of debt is unsecured, liquidity is near 700 million and the unencumbered pool is 9.7 billion. Core FFO payout is 73.8% and falling toward a 70% target. The RioCan Living monetization, targeted at 1.3 billion, is nearly complete and management expects to exceed it.
Bear case: still around 85% retail, so a genuine recession reaches the tenant base even with a necessity mix.
Clears both screens and produces the highest modelled return of the eight.
Price levels: on core FFO of 1.61 growing at 4% with a 14x exit, buy below 23.36, trim above 25.96, sell above 28.33. At 21.40 the sixteen-year return is 9.9%.
Primaris (PMZ.UN)
The growth here is already under contract rather than forecast, which is rare.
Canada’s only pure-play enclosed shopping centre REIT, with a 5.2 billion portfolio of which roughly 70% has been acquired since the 2021 spin-out from H&R. 2026 guidance is FFO per unit of 1.85 to 1.90, cash NOI of 390 to 400 million, occupancy of 86% to 88%. Q2 rental revenue was 174.1 million, up 15.5%, with FFO per unit of 0.456.
In-place occupancy is 86.6% against committed occupancy of 91.1%. That 450 basis point spread is signed. Management points to roughly 1,000 basis points of total opportunity on the path to 96% stabilised, and has given visibility to approximately 52 million of incremental annual cash NOI commencing over the next three years. Three-year targets are 94% to 96% occupancy, 3% to 4% same-property cash NOI growth and 4% to 6% annual FFO per unit growth. Roughly 84% of former Hudson’s Bay space is leased or in advanced negotiations, against an estimated 125 to 150 million of HBC-related capital spend over several years.
Financially the most conservative retail name here: FFO payout 48.8%, average net debt to adjusted EBITDA near 6.0 times. Insiders have been buying, including a trustee purchase of roughly 15,000 units at about 22.29 in late June.
Bear case: enclosed malls are the hardest retail format in North America and secondary markets amplify that. Net income swung to a 13.7 million loss in Q2 from a 50.4 million profit.
Fails the yield test at roughly 3.9%.
Price levels: on FFO of 1.875 growing at 5% with a 13x exit, buy below 23.80, trim above 27.52, sell above 30.10. At 22.63 the sixteen-year return is 9.4%. The margin is about a dollar, so this is the name most sensitive to the occupancy recovery actually arriving. At 3% growth instead of 5%, the buy level falls to roughly 19.50 and the thesis breaks.
Boardwalk (BEI.UN)
This is the verdict that changed most, and the reason is worth explaining because it is a modelling issue as much as a price issue.
2026 guidance is FFO per unit of 4.60 to 4.80 and AFFO per unit of 3.94 to 4.14, with same-property NOI growth of 1.0% to 3.5%. Q2 FFO was 1.19 per unit, up 2.6%, first-half 2.33, up 5.0%. NOI rose 2.9% to 107.2 million at a 66.6% operating margin, with same-property occupancy at 97.0%.
The affordability position is the moat. Average occupied two-bedroom rent is 1,612 against a Canadian average of 2,159. When the rental market softens, the landlord charging 75 cents on the market dollar loses least. Boardwalk held occupancy by moderating renewals and cutting incentives rather than chasing headline rent.
The distribution is 1.80 annualised at 0.15 monthly, an FFO payout near 38%, comfortably the lowest of the group. Boardwalk has sold 2,081 suites for 492.0 million year to date against a raised 2026 disposition target of 400 to 500 million, and after quarter end finalised a co-ownership with the DGAM Canadian Private Real Estate Fund, vending a 50% interest in four communities in Calgary and Victoria totalling 328 suites at proportionate share for 146.0 million gross. Debt to EBITDA improved to 9.34 times from 9.99 at year end, debt to total assets 43.0%, liquidity 374.8 million.
Here is the problem. A 2.8% yield plus organic growth of roughly 1%, which is what the guidance midpoint of 4.70 against 2025 actual of 4.65 implies, does not get anywhere near 9%. The entire case rests on the 62% of FFO that Boardwalk retains being converted into per-unit growth through buybacks and asset recycling. At 13.8 times FFO, retiring units is genuinely accretive, but it has to do a great deal of work.
| Growth assumed | Buy price for 9% | Return at 65.00 |
|---|---|---|
| 3% | 46.91 | 6.3% |
| 4% | 53.04 | 7.3% |
| 5% | 60.03 | 8.3% |
| 6% | 68.01 | 9.4% |
| 7% | 77.12 | 10.4% |
Boardwalk needs roughly 6% per-unit FFO growth to justify 65.00. Guidance implies about 1% organic. The buyback and the recycling have to supply the other five points, every year, for sixteen years. That is a bet on capital allocation rather than on real estate, and it is not one the current numbers demonstrate.
Price levels at the 5% base case: buy below 60.03, trim above 75.99, sell above 83.22. Verdict downgraded from Buy to Hold. Excellent assets, excellent balance sheet, wrong price for the growth on offer.
CAPREIT (CAR.UN)
Diluted NAV per unit was 54.38 at 30 June 2026 against a unit price of 34.25 on 7 August, roughly 63 cents on the dollar of stated book value and a record discount for this name. Management repurchased 71 million of units in the first half at an average near 36.00, and has deployed 1.0 billion since 2022 at an average of 43.00.
The operating business is holding up better than the headlines: occupancy 97.5% against a 95.3% national benchmark, same-property Canadian NOI margin 64.2% year to date, operating costs flat, mortgages at a 3.4% weighted average rate with 4.2 years to maturity, leverage 41.2% of gross book value, FFO payout 59% to 62%.
The demand thesis is what broke. Ottawa cut immigration targets and capped students, national asking rents have fallen year over year for twenty straight months, and vacancy is rising for a third year. Diluted FFO per unit slipped 1.1% to 0.654 in Q2. Fifty-one percent of Canadian turnover came from tenants with under two years of tenure, who renewed 7.1% lower, improving from a 10.8% decline in Q1 and turning positive in July.
Clears both screens but not the return hurdle. At 3% growth the sixteen-year return from 34.25 is 8.1%, and the buy level is 31.01.
The important caveat is that this model deliberately ignores the NAV discount. It values CAPREIT on cash flow alone, which is the conservative treatment. If the gap between 34.25 and 54.38 closes even partially over sixteen years, the return is well above 9%. At 4% growth, which is what a normalised rental market would produce, the buy level rises to 34.72 and the name clears at today’s price. So this is a Hold on the arithmetic and a Buy on the asymmetry, and the honest framing is that you are paying 8.1% today for an option on the discount closing.
Price levels: buy below 31.01, trim above 39.60, sell above 43.26. The 52-week low of 33.15 sits just above the buy level.
SmartCentres (SRU.UN)
The best real estate on this list attached to the thinnest payout coverage.
201 properties, 35.5 million square feet, roughly 12.1 billion in assets, 98.1% in-place and committed occupancy, anchored predominantly by Walmart, and 3,500 acres of owned land. Premium Outlets run at 99% occupancy with Toronto Premium Outlets in the national top three for sales, and a nearly 100,000 square foot expansion beginning construction in Q4 2026. Diluted NAV per unit was 35.10 at 30 June 2026, so at roughly 28.00 the units trade at about a 20% discount.
The income statement is the problem. Q2 FFO per unit was 0.58, unchanged year over year, and adjusted FFO excluding non-recurring items was 0.54, down from 0.55. NOI fell 1.0% to 139.9 million on fewer townhome closings. Net loss was 147.0 million, driven by a 196.2 million fair value loss on investment properties and a 42.4 million loss on financial instruments. The distribution is 1.85 annualised, and the AFFO payout ratio is 90.5% on a rolling twelve months.
Flat FFO with an AFFO payout above 90% is the reason this fails the safety test despite a 6.6% yield. There is very little retained cash to fund the development pipeline that is supposed to be the growth, which means it depends on external capital.
Price levels: on FFO of 2.32 growing at 2% with a 12x exit, buy below 27.26, trim above 31.22, sell above 33.99. At 28.00 the sixteen-year return is 8.7%, which is closer to the hurdle than the safety screen would suggest. Almost all of that return is the distribution, and the distribution is the thing that is not covered. That is the whole argument against it in one sentence.
Choice (CHP.UN)
The defensive anchor. Loblaw pays roughly 64% of retail rent across around 560 grocery-anchored plazas, 83% of tenants are necessity-based, and the industrial portfolio of 124 properties is seeing renewals about 40% higher. Occupancy 97.7%. Debt to EBITDA near 7 times, BBB rated, roughly 2 billion in liquidity. When rates ripped in 2022 and 2023 the maximum drawdown was about 23%, against 46% for CAPREIT.
Safety is fully priced and there is almost no growth behind it. 2026 guidance is FFO per unit of 1.08 to 1.10 with same-asset cash NOI growth of 2% to 3%. Reported FFO per unit grew 0.8% in Q2 and AFFO per unit fell 6.1% to 0.217. IFRS NAV is 14.73 against a price near 15.97, so you pay a premium to book. AFFO payout sits near 90%.
The First Capital transaction, with Choice taking the grocery-anchored assets and KingSett the rest, is expected to close in Q4 2026. It dilutes Loblaw concentration and adds to the moat, but introduces integration risk and leverage not yet reflected in guidance, which explicitly excludes the deal.
Price levels: on FFO of 1.09 growing at 2% with a 14x exit, buy below 12.81, trim above 16.11, sell above 17.58. At 15.97 the sixteen-year return is 6.9% and the units sit fourteen cents below the trim level. Upgraded in candour and downgraded in verdict: this is a Trim, not a Hold. You would need it 20% cheaper to earn the hurdle, and the deal closing in Q4 is a reason to want a discount rather than to pay a premium.
Chartwell (CSH.UN)
Operationally the standout. Twelve consecutive quarters of double-digit FFO per unit growth. Q2 FFO per unit up 16.7% to 0.28, FFO up 34% to 90.5 million. Same-property occupancy up 320 basis points to 94.3%, every platform above 90%. Same-property adjusted NOI up 11.9%, NOI per occupied suite up 8.1%, operating margin up 110 basis points to 42.5%. Agency staffing costs down 43% year to date. Twenty-four consecutive years of maintained distributions, including through the pandemic.
The demographic case is the strongest in Canadian real estate. The 85-plus population is expected to roughly triple by the mid-2040s, the boomers hold most of the country’s wealth, and there is nothing like enough purpose-built supply.
Three things keep this at hold. Valuation: roughly 18 to 19 times FFO against 13 to 14 times for the rest of the group, the lowest yield here, and the largest premium to NAV. It is an operating business, not a passive landlord, with staffing and execution risk, and Covid showed the tail when units went from about 16.00 to under 4.00. And move-ins depend on seniors selling the family home, so a soft housing market slows the funnel exactly when you are paying a peak multiple.
Net debt to adjusted EBITDA is 7.0 times, interest coverage 3.5 times, and unit count is up roughly 15% with about 72.7 million raised through the ATM. Per unit growth has stayed ahead of that dilution, which is the metric to watch.
Price levels: on FFO of 1.16 growing at 5% with a 17x exit, buy below 18.52, trim above 21.97, sell above 24.03. At 21.52 the sixteen-year return is 7.7% and the units sit forty-five cents below the trim level. Note how much rests on the growth rate here. At 7% growth the buy level rises above 22.00 and Chartwell becomes a buy at market. The model uses 5% because occupancy has 70 basis points left to run to management’s 95% target, after which the double-digit growth has to come from rate and margin rather than from filling empty suites.
Conclusion
Two tests were applied to eight Canadian REITs. The first asks whether a single trust yields more than XRE at roughly 4.5% while carrying a more durable distribution than a market-cap-weighted basket. The second asks whether it returns 9% annually over sixteen years at today’s price.
Two names clear the first test. RioCan yields roughly 5.4% against a 73.8% core FFO payout falling toward a 70% target. CAPREIT clears the yield bar by three basis points, backed by a 59% to 62% payout, 41.2% leverage and mortgages at 3.4% for another 4.2 years.
Three names clear the second. RioCan returns 9.9%, Granite 9.6% and Primaris 9.4%. Only RioCan passes both.
The pattern in the screen is that the four names failing on yield are the four with the strongest coverage. Granite pays out 66% of AFFO, Primaris 48.8% of FFO and Boardwalk roughly 38%, and they yield 3.9%, 3.9% and 2.8%. A low yield is what a well-covered distribution looks like when a trust retains cash to grow. The two highest yields on the board run the other way: SmartCentres at 6.6% pays 90.5% of AFFO with flat FFO and NOI down 1.0%, and Choice at 4.9% pays near 90% with FFO growing 0.8%. Both sit inside XRE and VRE, bought automatically because the index says so.
Rebuilding the price levels on a return basis changed four verdicts. Granite’s buy level rose from 88.00 to 96.38 and RioCan’s from 20.90 to 23.36, so both are buys at market rather than names to wait on. Boardwalk fell from Buy to Hold, with a buy level of 60.03 against a market price near 65.00. Choice fell from Hold to Trim, sitting fourteen cents below the price at which its five-year forward return drops to 6%.
Boardwalk is the clearest illustration of why the method matters. Its 38% payout is a strength for distribution safety and a problem for total return. A 2.8% yield plus roughly 1% organic growth cannot reach 9%. The case rests on the retained 62% of FFO being converted into per-unit growth through buybacks at 13.8 times FFO and 492 million of suite sales. The sensitivity table shows the name needs about 6% per-unit growth to justify its price.
CAPREIT is modelled on cash flow alone at 8.1%, with the 63-cents-on-the-dollar discount to a 54.38 NAV deliberately excluded because no date can be put on it closing. At 4% growth rather than 3%, it clears at today’s price.
Four names that were investable a year ago no longer trade as REITs. InterRent was taken private in July at 13.55, Artis was absorbed into RFA Financial in February, First Capital is being carved up between Choice and KingSett, and H&R is being restructured with Blackstone and PSP. Private capital is paying prices the public market will not.
Disclaimer: This analysis is for informational purposes only and does not constitute investment advice. Always perform your own due diligence or consult with a financial advisor before making investment decisions.
