How to Read Canadian Building Permit Data

Every month, Statistics Canada publishes building permit values and unit counts for municipalities across the country under Table 34-10-0292, "Building permits by type of structure and type of work." The survey asks provincial and municipal permit-issuing offices to report the value and count of permits authorized that month, before construction starts. It is one of the fastest-moving indicators in the federal statistical system: permit data for a given month is usually released about six weeks later, well ahead of the housing starts or completions data that follows once ground actually breaks.

Because permits are counted at authorization, they measure approval to build. They do not measure finished housing or completed commercial space. A spike in one month can come from a single large apartment or industrial project clearing approvals. OpenStats city pages therefore show several months of history so one spike is easier to put in context.

What the survey actually counts

The permits survey groups activity into two broad categories, and each behaves differently from one municipality to the next.

• Residential permits cover single-family houses, semi-detached and row housing, and apartment structures. Statistics Canada also reports a separate count of new housing units authorized, which is the figure most often quoted in housing-supply discussions because it converts a dollar value into a unit count.
• Non-residential permits cover industrial, commercial, and institutional construction: warehouses, retail space, office towers, schools, hospitals, and similar projects. These permits tend to arrive in large, irregular values because a single hospital or distribution centre can be worth more than a year of small commercial permits combined.

A city with a large hospital expansion or a new data centre under permit will show a spike in non-residential value that has nothing to do with the pace of home construction. Read residential and non-residential columns on their own. A single combined total can hide that difference.

Residential vs non-residential: why the split matters

Residential and non-residential permits respond to different underlying drivers. Residential permit volume tracks population growth, mortgage rates, land availability, and municipal zoning capacity. Non-residential permit volume tracks business investment cycles, institutional capital budgets, and, in resource-producing regions, commodity price cycles that drive industrial construction.

When comparing two cities, matching category to category avoids comparing unlike things. A smaller city with a new distribution warehouse might show a higher total permit value in a given month than a larger city with steady but unspectacular home construction, even though the larger city has a stronger underlying housing pipeline.

How OpenStats city pages are built

The permits topic site pulls monthly records directly from the Statistics Canada table for census metropolitan areas and matches them to the corresponding municipality using the published geography codes. Where a municipality falls inside a tracked CMA but is not separately broken out in the underlying table, the page is marked as a regional proxy and uses the CMA-level figures as the best available approximation. Every page states which tier applies before showing the numbers, so a reader always knows whether a figure is municipality-specific or regional.

Each city page reports the latest available month, a running total of new housing units authorized, and the split between residential and non-residential value. Where a municipality also runs its own open permit portal, such as Calgary's development permit feed, that source is cited separately as a faster local signal. The federal survey remains the basis for cross-city comparisons on OpenStats.

Reading a monthly figure correctly

• Treat a single month as noisy. Look for a pattern across at least a few months before drawing a conclusion.
• Check which tier the page uses (direct, regional proxy, or seed) before comparing across cities.
• Read residential value and non-residential value on their own. A combined total can hide a one-off commercial project.
• Remember that permit value is nominal (not adjusted for construction cost inflation), so a rising dollar figure does not always mean rising unit counts.

Why permits lead the housing cycle

Because a permit has to be issued before a shovel goes in the ground, permit data leads housing starts by weeks or months and leads completions by a year or more on larger projects. Municipal planning departments, homebuilders, and local business groups use permit trends as an early read on whether construction activity is accelerating or slowing before that shift shows up in employment or starts data. The tradeoff is volatility: a permit can also be withdrawn, revised, or delayed after issuance, which the monthly survey does not always capture in real time.

For that reason, most practical analysis pairs the permits series with completions or starts data over a longer window. Permits alone are a weak basis for judging whether a market is building enough relative to demand.

Municipal comparisons in practice

When comparing permit activity between two municipalities, normalize by population where possible. A city of 100,000 issuing permits for 200 new housing units in a month is running at a materially different pace than a city of 1,000,000 issuing permits for the same 200 units. OpenStats city pages list population alongside permit figures specifically so this kind of per-capita comparison can be made without leaving the page.

It is also worth checking whether a city's figures are direct-tier or a regional proxy before treating two cities as directly comparable. A regional proxy borrows from a wider CMA and will understate or overstate activity in any single municipality within that region depending on where growth is actually concentrated.