British Columbia told local governments to zone for small-scale multi-unit housing. A local trust committee is not a local government.

The five provisions, quoted

• Local Government Act, Schedule — Definitions

  “local government” means (a) the council of a municipality, and (b) the board of a regional district

  Two bodies, and a local trust committee is neither. The same Schedule separately defines “local trust committee” as “a local trust committee within the meaning of the Islands Trust Act”, so the Act plainly knows what one is.

• Local Government Act s.481.3(4) — small-scale multi-unit housing

  A local government must exercise the powers under section 479 to permit…

  This is the 2023 duty to zone for townhouses, triplexes and secondary suites — the medium density the islands are short of. The obligation is placed on a local government.

• Local Government Act s.585.1 — housing needs reports

  “local government” includes a local trust committee

  ★ The hinge. For the duty to MEASURE housing need, the Legislature expressly extended the term to cover local trust committees. It would not have needed to say so if they were already included — and it did not say so for the duty to zone.

• Islands Trust Act s.29(1)(b)(i) — added by Bill 44 (2023), s.47

  Part 14 [Planning and Land Use Management] of the Local Government Act, other than the following: (i) section 481.3 [zoning bylaws and small-scale multi-family housing]

  ★ The carve-out we first said did not exist. A local trust committee gets a regional district board's Part 14 powers except the ones listed, and the very first exception is the small-scale multi-unit duty. Bill 44 added this subparagraph in the same Act that created the duty, so the omission was decided by the Legislature, not left to interpretation. Also excluded: s.482.7 (affordable and special needs housing zoning) and Division 23 (transit-oriented areas).

• Islands Trust Act s.29(1)

  Each local trust committee has, in respect of its local trust area, all the power and authority of a regional district board under the following enactments…

  Power and authority — not duties. The Islands Trust Act hands an LTC a regional district’s planning toolkit; it does not hand it a regional district’s obligations.

• Islands Trust Act s.3 — the object of the trust

  The object of the trust is to preserve and protect the trust area and its unique amenities and environment for the benefit of the residents of the trust area and of British Columbia generally…

  Housing is not in the object clause. It is not there in 1974 and it is not there now. Housing reaches the Trust through a reporting regulation, not through its purpose.

Check them: Local Government Act, Part 14 · the Act’s Schedule of definitions · Islands Trust Act

This is not a claim that the Trust has broken a law. Non-application is not non-compliance, and a local trust committee that chooses to allow medium density is free to. We asked the Ministry whether leaving local trust committees out of s.481.3 was deliberate; the statute has since answered that itself (see the correction below). The question that remains is why the islands with the worst measured housing need were given the weakest obligation to do anything about it.

⚠️ This is a reading of the statutory text, not legal advice. Every provision is quoted above and linked so it can be checked. If a lawyer or the Ministry shows that s.481.3 does reach a local trust committee, this section is wrong and we will say so here.

⚠️ Corrected 2 September 2026. We first wrote that there was “no carve-out” and that “nobody appears to have decided”. Half right. The duty never reaches a local trust committee — but not by accident. The carve-out is not in ss.481.3 or 481.4 of the Local Government Act, where we looked; it is in s.29(1)(b)(i) of the Islands Trust Act, which Bill 44 itself amended to list s.481.3 among the powers a local trust committee does not have. That is an exclusion the Legislature decided, and the Minister of Housing gave the reason in the House on 20 November 2023: the Trust’s “preserve-and-protect mandate”, and water. The Trust did not ask for it: the Salt Spring committee wrote to the Minister asking to be included as if it were a municipality, and an amendment to include it did not pass.

We have asked both bodies that could correct this

Written on 2026-08-31 to the Ministry of Housing and Municipal Affairs and the Islands Trust (Executive Office). Whether section 481.3 of the Local Government Act reaches a local trust committee; whether leaving local trust committees out of it was deliberate; and whether any local trust committee has adopted small-scale multi-unit provisions regardless.

Both replies will be published in full and unedited. If either body shows this reading is wrong, the page will be corrected and the correction will say so.

No reply yet. If neither has answered by 2026-09-30, that will be recorded here too.

What they are required to do is count

The Trust commissioned the assessment the regulation demands, and it is a serious document. It found the islands short 2,028 homes within five years and 6,361 within twenty.

Method. The assessment’s own Key Findings table. Ballenas-Winchelsea is 0/0 because it is expressly exempt from the Housing Needs Report Regulation. Bowen is absent because it reports as a municipality. Source. Islands Trust Housing Needs Assessment 2025, prepared by Urbanics Consultants Ltd, April 2025 — study area total 2,028 homes in five years and 6,361 in twenty (2025-04).

Island by island, the assessment states the same implication in the same words: the dwelling stock must grow by roughly 15–20% in five years and 43–61% in twenty. Galiano 19% and 61%. Gambier 20% and 61%. Lasqueti 18% and 52%. Mayne 15% and 43%. North Pender 16% and 50%.

And what a 570,000-character housing assessment never mentions

Method. Case-insensitive occurrence counts over the full extracted text of the assessment, 570,713 characters. Counted mechanically, not read for sense — a term could in principle appear in a heading or a footnote. The four zeroes are what matter and a zero cannot be a false positive. Source. Islands Trust Housing Needs Assessment 2025, full text (2025-04).

“rezoning”, “approval”, “completions”, “multi-unit” — not once. The document measures how many homes are needed and never asks how one would be permitted. That is not a criticism of the consultants; it is what the regulation asked for. It is a criticism of a regime in which the housing duty stops at counting.

The amount of housing permitted in the past will shape population growth and shape this model’s projection of future household growth. Because this data is so dependent on past policy outcomes, it should not be used solely to inform housing needs.

The Trust’s own consultant, warning that the projection partly launders past restriction into lower future need. If an island has been hard to build on for fifty years, fewer people moved there, so the model forecasts fewer households, so the measured need comes out smaller. The number the Trust is required to report is depressed by the thing being measured.

How the Trust prices its own housing work

Method. Local trust committee project lines in the Trust’s own budget papers, quoted as printed. These are the years budgeted to REVIEW the rules, before any rule changes. Source. Islands Trust Council, 2026/27 budget — Local Trust Committee Projects (2026).

Four years to review the housing rules on Denman. Five to review Gabriola’s plan and bylaw. These are the years budgeted before anything changes, and they are the Trust’s own figures.

You cannot tell from the Islands Trust’s own published record whether an application was ever decided.

Thirty-one open rezonings have not appeared anywhere in the Trust’s document library since before 2021, and fourteen not since before 2017. These are not stalled files being worked slowly — they are files that stopped moving a decade ago and were never closed out. The Trust publishes no decisions register, so there is no second source to check them against.

Method. Every rezoning in the Trust’s published applications register whose status is open, in progress, under review or ongoing, de-duplicated by file number, matched against the full text of all 7,167 documents we hold from the Trust’s own library. “Silent” means the file appears in the record and its most recent appearance predates the date given. Source. Islands Trust applications register and public document library, as held in this project’s archive (2026-08-31).

GB-RZ-2007.1 — 24 appearances in the Trust’s own meeting record, the last on 2014-07-17. A Gabriola rezoning with twenty-four appearances in the Trust’s own meeting record, the last of them in July 2014. It is still published as in progress. North Pender’s NP-RZ-2011.1 and .2 went quiet in September 2012; Hornby’s HO-RZ-2012.1 in May 2013.

That is a smaller claim than “the Trust is slow”, and a firmer one. It also means nobody — not the Trust, not a researcher, not an islander wondering what happened to a housing proposal — can measure how long the Trust takes, from what the Trust publishes.

This belongs with the rest of the Trust’s record-keeping problems — the accessibility audit covers the documents that are unreadable, and the ones that have been removed outright.

We withdrew a claim to publish this page

2026-08-31

Withdrawn: “Open rezoning applications on the Islands Trust islands average 7.1 years, and that is how long the Trust takes.”

The average age is real, the interpretation was wrong. Checking every open rezoning against the Trust’s own meeting record showed that many of them stopped appearing in it years ago — so the age of an open file measures how long the register has gone un-updated, not how long the work takes. The processing-time claim is withdrawn. We could not measure processing time at all, because the published record does not contain decision dates.

Two data-quality problems in the register weakened the original counts as well: the same file appears under several islands (NP-RZ-2011.1 is filed under North Pender, South Pender and Saturna), and near-duplicate rows exist (GB-RZ-2024-0091, GB-RZ-2024.91 and GB-RZ-2024.091 are one application). The 139 figure above is de-duplicated by file number; the earlier one was not.

The island that took local control did not build medium density either

Method. New homes registered under the Homeowner Protection Act. A null is the Province’s privacy suppression, applied where fewer than five homes were registered in a year — so a null means “between zero and four”, not “unknown”. Bowen is the only island in this dataset because it is the only island that is a municipality; the twelve Trust islands are invisible inside regional district totals. Source. Province of British Columbia, New Homes Registry 2016–2024 (2024).

★ This one goes against us. Bowen Island has run its own planning department since 2000. It is the island that took local control, and in seven of nine years it registered fewer than five multi-unit homes — the Province suppressed the figure for privacy. Six in 2017, six in 2019. Single detached houses carried on at roughly twenty-five a year throughout. If the implied promise of incorporation is that local control delivers medium density, Bowen is not evidence for it, and anyone making that promise should have to answer this table.

What this page cannot tell you

• We cannot measure how long the Trust takes to decide a housing application. Its published register records no decision dates and it publishes no decisions register, so the question is unanswerable from the public record. That is itself the finding, but it is not the finding we set out to make.
• The applications register contains duplicate rows and files attributed to more than one island, so any count taken from it is approximate and any per-island count is unreliable.
• The statutory reading is a reading. It is quoted so it can be checked, and it has not been tested by a lawyer, a court or the Ministry.
• New homes registered is not the same as homes built, and the Province suppresses any island-year under five, which on islands this size hides most of the signal. The twelve Trust islands do not appear in that dataset at all.
• Nothing here establishes that different governance would produce more housing. Bowen had different governance and built almost no medium density either.