Ontario's 2027 rent cap is 1.9%. Your margin lives in operations now.

In short: Ontario set the 2027 rent increase guideline at 1.9%, down from 2.1% this year and 2.5% the three years before that. For rent-controlled units, revenue growth is now capped by law while insurance, wages, and per-door software fees keep climbing. That leaves an operator two levers: cut people, or recover hours. This is the case for the second lever, with real numbers from a property operation that found its missing margin in the phones.

Key takeaways

• The 2027 rent increase guideline is 1.9%, the maximum for most rent-controlled units, meaning units first occupied on or before November 15, 2018. Going above it requires an approved above-guideline increase from the LTB.
• Increases need 90 days written notice on form N1. For a January 1, 2027 increase, you can serve notice as early as October 3, 2026. Put that date in the calendar now.
• The guideline tracks Ontario CPI down. Your cost base does not follow it. That gap is the squeeze, and it is structural.
• The industry's default response, cutting coordinators and support staff, degrades exactly the responsiveness that keeps tenants paying and owners staying.
• The recoverable margin is in hours: the phones, maintenance triage, and owner reporting. Map where they go before you decide anything about people.

What the 1.9% guideline actually covers

The Ontario government announced the 2027 guideline on July 7, 2026. The number comes from a formula, not a policy mood: the average Ontario Consumer Price Index over the twelve months from June 2025 to May 2026. Inflation cooled, so the guideline fell.

The mechanics, briefly. The 1.9% cap applies to most residential tenancies in units first occupied on or before November 15, 2018. Newer units are exempt, and rent can be reset freely between tenancies for any unit. To raise a controlled rent by more than 1.9%, you need the Landlord and Tenant Board to approve an above-guideline increase, typically for capital work or extraordinary cost increases, and that is a slow queue to stand in.

The operational date that matters more than the percentage: notice must be served 90 days ahead on form N1. If you want increases effective January 1, 2027, the earliest valid service date is October 3, 2026. Operators who miss the window do not lose the increase forever. They lose months of it, quietly, across every door they were late on.

The squeeze is structural, not cyclical

Three years ago the guideline was 2.5% and asking rents in Toronto were climbing. An operator could be sloppy on costs and the market covered for it.

Both of those supports are gone at once. The guideline has stepped down two years in a row, and CMHC's mid-year update describes asking rents in Toronto declining while landlords manage renewals and refinancing with operating costs still elevated. Turnover pricing, the traditional escape valve on controlled stock, is worth less when the market rent you reset to has softened.

Meanwhile the cost side of a property management operation does not track CPI. Insurance has its own trajectory. Wages ratchet up and do not come back down. Software vendors added AI features and per-door fees to go with them. The revenue line is now legislated to grow 1.9% while several of the biggest cost lines grow faster.

When revenue is capped by law, margin is no longer a pricing question. It is an operations question.

Two levers left, and the industry keeps pulling the wrong one

Strip it down and an operator squeezed this way has two moves: reduce what the operation costs, or reduce what the operation wastes.

The default move is the first one, and it usually means people. Cut the coordinator, stretch the remaining team across more doors, let response times drift. On a spreadsheet it works instantly. A salary disappears and the per-door math improves that same month.

Then the second-order effects arrive. In property management the product is responsiveness. A tenant with a leak who cannot reach anyone becomes an angrier tenant, then an arrears file, then a turnover. An owner who waits two weeks for a straight answer starts returning calls from your competitors. The salary saved is real, and so are the doors that walk out the back while nobody is watching the phones.

We say this as a firm with a stated position: we do not build AI systems whose purpose is to cut headcount. That stance costs us deals. We hold it anyway, partly on principle and partly because in this industry the math genuinely does not favour the cut. The scarce resource in a PM operation is not salary budget. It is attention, and cutting people buys margin by selling attention.

Where the hours actually leak

The second lever is duller and better: find the hours the operation already loses and take them back. In every operation we have mapped, the leaks concentrate in three places.

The phones. One property operation we work with put an AI voice agent on its inbound lines. It now answers roughly 600 calls a month at current volume, on the first ring, around the clock. About 27% of those calls arrive after hours or on weekends, calls that previously went to a voicemail box and either became an emergency by morning or a lead for someone else. Average call length is about a minute and a half, roughly two thirds of the property-line calls resolve inside the call, and the rest land with the right person as a structured summary instead of a "call me back." Nobody on that team lost a job. They stopped being a switchboard.

Maintenance triage. The expensive part of a work order is rarely the repair. It is the three rounds of phone tag before anyone knows whether the AC is blowing warm air or nothing at all. Structured intake that asks the follow-up questions at the moment of reporting shortens the whole chain and kills the wasted truck roll. We wrote up the broader pattern in [where AI actually lands in a property management operation](/blog/where-ai-lands-in-property-management).

Owner reporting. Month-end in most shops is a person assembling numbers from three systems into emails that all look slightly different. It is days of skilled attention spent on assembly rather than judgment, and it is one of the most automatable workflows in the building.

None of this is exotic. That is the point. The margin the guideline took away is sitting in ordinary workflows, recoverable without touching the org chart.

What to do before October 3

Three moves, in order.

First, run the N1 calendar now. List every controlled unit, its anniversary date, and the earliest valid service date. A 1.9% increase you actually serve beats a 2.1% memory of better years.

Second, price your 2027 cost base honestly. Insurance renewals, wage commitments, every per-door software fee. Know the size of the gap the guideline leaves before deciding how to close it.

Third, before any conversation about people, spend one week measuring where the hours go. Track the calls, the interruptions, the month-end grind. Operators are consistently wrong about their own time, and one messy real week beats a year of gut feel. That measurement is the first thing [an audit answers](/blog/five-questions-every-audit-answers), and it is the difference between cutting attention and recovering it.

The guideline is set and the squeeze is real. But 1.9% is only a ceiling on your revenue. It says nothing about your hours. If you run rental stock in Ontario and want to know where yours are going, that conversation is what we do, and the first call is free.