Subdivision, shared lots and housing affordability on Denman Island

Findings

Additional dwellings can provide cheaper housing. The evidence does not establish that removing ordinary subdivision permission is necessary to obtain that benefit, or that Denman's proposed combination will improve affordability overall. The strongest support is for smaller homes, reuse of existing space, shared infrastructure and some below-market family or landlord arrangements. The additional policy claim about restricting subdivision requires its own comparison.

This distinction matters because a rental cottage, a cooperative membership and a small separately titled home meet different needs. A low monthly rent does not create ownership; a lower purchase price does not house someone unable to obtain a mortgage. A family dwelling can provide substantial benefit without becoming available to an unrelated renter. None should be counted as another type of outcome.

The narrower claim that additional dwellings seem cheaper is plausible and supported in several settings. It should not be recast as a claim that subdivision restrictions are necessary. This study evaluates that further policy question separately because the proposed package combines both changes.

The local record supports the seriousness of the housing problem and identifies plausible mechanisms for improvement. It also records financing barriers and uncertainty about delivery. Outside studies support additional dwellings as a useful source of lower-priced housing, while challenging categorical claims that market housing or smaller-lot development cannot help. No reviewed study measures the net effect of Denman's particular proposed policy package.

The practical implication is to evaluate additional dwellings and subdivision restrictions as separate interventions, then compare their combined effects. An evidence-based position can support more homes on existing lots while reserving judgment on an island-wide loss of ordinary subdivision options.

1. Define the comparison

“Cheaper” has at least five meanings: lower construction expenditure; lower total monthly housing payment; lower price for equivalent accommodation; affordability relative to a household's income; and lower financial barriers to secure tenure or ownership. The study must report each separately. Comparing a small rental suite with a large house purchase confounds building size, land, financing and rights.

“Market housing” is not synonymous with a subdivided lot. A privately rented suite can be market housing, while a subdivided parcel transferred to a nonprofit can support restricted affordable housing. A cooperative can contain substantial member equity, a large land gift, restricted resale value or none of those features. These are distinct arrangements, not interchangeable labels.

The first comparison should hold the underlying parcel, total dwellings, bedroom capacity, quality, tenure and environmental constraints constant. Compare a rental with a rental under the two title arrangements; assess ownership and cooperative entry separately. This isolates the direct cost of title structure. A second comparison should allow realistic differences in size, density, finance, family discounts and occupancy under the competing policy packages: these may be mechanisms of policy, not differences to remove statistically. Otherwise, the analysis either confounds title with housing product or assumes away the effect that policy is meant to produce.

There are two legitimate land-cost accounts. An incumbent owner may need no new land purchase to add a dwelling; that is a real cash-flow advantage. An economic comparison should also recognize the land's opportunity cost, a donation, or the benefit conferred by existing ownership. Pricing land at zero for one scenario and at full acquisition value for the other does not isolate a construction saving caused by keeping one title.

2. Denman's proposal and its stated rationale

The current hearing binder records second reading of Bylaws 260 and 264 on August 24, 2026; adoption remains blank. Its proposed subdivision regime combines a 64-hectare general minimum with restrictions on permitted purposes. Specified community, conservation, Indigenous and affordable-housing provisions, family-subdivision provisions and a bylaw-amendment route remain. It is therefore accurate to describe removal of ordinary subdivision permission, but inaccurate to describe every form of subdivision as impossible. Passing an area calculation alone would not authorize ordinary residential subdivision.¹

The rules also distinguish suites, detached secondary dwellings and cluster housing. An internal suite is subject to size limits and, outside the ALR, a covenant preventing separate strata subdivision. The cluster policy requires rezoning. Its listed criteria do not include the general rent/price agreement that appeared as an option in the staff report; the separate nonmarket multi-unit policy does require a housing agreement. Accordingly, these provisions do not make every new additional dwelling permanently below-market housing.¹

The January 2026 discussion paper argues that conventional subdivision generally produces market-valued housing, encourages speculation and higher land values, and creates fragmentation and servicing concerns. It estimates substantial shoreline exposure among potential lots and presents more than one restriction option. The paper also notes that dwelling capacity can already exist without subdivision. These are policy reasoning and spatial estimates; the passages reviewed do not provide a matched cost study or predicted change in rents caused by the restriction.²

The June 2025 report and surrounding meeting discussion develop the later policy concept: reconsider inherited subdivision patterns, address unsuitable sites and explore clustered housing. They are not its first appearance: the February 2024 record already considered restricting subdivision to cases involving an affordable-housing land donation. Those earlier options should not be conflated with the later blanket approach or its eventual numeric threshold. The transcript's speaker attribution remains provisional. A broad reference to profit or family vacation lots is not a measured frequency, and a discussion about whether to remove subdivision is not an estimate of its affordability effect.^{3, 28}

The August 2026 staff report explicitly qualifies the expected outcome: actual affordable-unit delivery is difficult to predict, and the combined amendments reduce modeled buildout by about 100 units. That figure concerns theoretical capacity under the package, not 100 demolished homes or proof of the effect of subdivision restrictions alone. It must be evaluated alongside any gain from more usable, financeable or secure permissions.⁴

Staff had also acknowledged earlier that uptake depends on willing owners and financing, and that even small homes can be costly. The June 2023 report identifies delivery barriers outside land-use permissions. The fair criticism is therefore the absence of a demonstrated comparison for the specific package, rather than an assertion that planners never considered financing.⁵

3. Local evidence and actual delivery

Denman's 2023 Housing Advisory Planning Commission report includes historical sale observations: the ten cited 2022 sales below one hectare averaged about $902,000, and the five cheapest sales were below one hectare and below $700,000. These indicate a difficult ownership market while showing that smaller properties can be among the cheaper choices. They do not isolate subdivision's effect because house size, age, quality, location and servicing were not controlled.⁶

The same report cites a $670 median monthly rental shelter cost from the 2021 Census and warns that newer offerings were much higher. An incumbent-rental statistic from 2021 cannot serve as a 2026 asking rent or the revenue assumption for a new build. Its discussion appendices also include a presentation expressly not approved as HAPC recommendations; those statements should not be attributed wholesale to planners or the commission.⁶

Local proposals reveal affordability mechanisms worth studying, and limits that should not be hidden:

The hearing and application statements are evidence about participants' proposals and experience, not audited financial accounts or lender commitments. They warrant gathering actual entry costs, financing terms and ongoing charges rather than dismissing shared ownership or assuming it works equally well for all households.^{7, 8, 9, 2}

The nonprofit project record is similarly mixed. Denman Community Housing Society reports that The Ridge was created in 2015 through donated land, and that the proposed 20-unit Denman Green project is seeking construction funding. The earlier joint housing-association update records loss of the intended eight-unit Pepper Lane site. Counting all historic zoning as a current delivery pipeline would misstate the evidence. Equally, claiming nonprofit or shared-land housing has delivered nothing would disregard existing homes.^{10, 11}

These cases explain why the study needs a sequence for every project: permission, financing, construction, completion and occupancy. A lost site, conditional loan discussion, unbuilt permission and occupied home belong in different columns. The initial case register preserves these distinctions; it is not yet a representative sample or a measured comparison of title forms.

4. What outside research adds

The strongest geographically relevant implementation source is BC Housing's 2021 study of small and medium communities. Based on surveys, interviews and focus groups, it discusses infrastructure constraints and financial incentives. Its Cumberland case combines attached and detached additional dwellings with a reduced 325-square-metre minimum for infill subdivision. This documents that the two approaches can coexist; it does not prove either policy's causal rent effect. The report also considers strata ownership possibilities and acknowledges tradeoffs between affordability conditions and owners' willingness to build.¹²

Vancouver's 2018 survey received 308 owner and 304 occupant responses, with varying question denominators. It records strong rental use and perceived affordability relative to buying, but also household-size and space constraints. It reports lower laneway construction costs when built alongside a new principal home than when added while retaining the existing one. Thus, not every form of infill or retrofit is automatically the cheapest project. The voluntary survey lacks a matched subdivision control.¹³

A 2017 Portland-Seattle-Vancouver owner survey obtained 414 responses from 1,837 contacted addresses. Fifty-eight percent of responding owners reported below-market rents, including many tenancies outside family/friend arrangements. These are owner-reported comparisons, subject to response and successful-project selection. They support the usefulness of additional dwellings without establishing income eligibility, permanent affordability or a title-policy effect.¹⁴

California's 2021 survey demonstrates important variation. It retained 752 valid owner responses, a 4.8% response rate. In its Los Angeles sample, only 12% of income-generating ADUs met the study's affordability threshold at 80% of area median income for two people. Roughly half of respondents rented their ADU, while others housed relatives free or used the space for non-rental purposes. Those uses can be valuable, but are not all open-market rental supply.¹⁵

Portland's 2019 original study found lower average ADU rents per square foot than apartment comparators in most analyzed neighborhoods. It also distinguished substantially different costs for detached units and conversions. This is useful counterevidence to a claim that all apparent affordability is merely smaller floor area, although selection and quality differences remain. Historical US costs should not be imported as Denman construction estimates.¹⁶

Newer evidence strengthens, rather than eliminates, the need for careful definitions. A 2026 Israeli study analyzes a large rental-advertisement dataset from 2012-2019 and finds materially lower advertised ADU rents. Its neighborhood comparisons explicitly do not standardize for dwelling size; the stock includes informal internal partitions. It therefore measures the cheaper product mix being offered, not an effect of prohibiting subdivision or the cost of a new compliant island cottage.¹⁷

Other recent work is retained as a research lead rather than overstated. A 2025 ADU spillover paper examines neighboring property sale prices; its verified abstract is not a finding about ADU rents, and the full methods were inaccessible in this review. A December 2025 affordability-strategies review was accessible only through its abstract and references. Neither supplies a verified estimate of Denman's policy effect.^{18, 19}

5. Market housing, land values and competing explanations

An increase in the value of the original parcel does not necessarily mean that the entry price of each resulting home increases. Dividing a property, changing its permitted yield, altering the housing product and changing market-wide supply affect different prices. Land price per square metre, whole-property value, individual home price and rent can move differently. Speculation is a hypothesis requiring evidence about transactions and expectations, not another name for every land-value change.

Hamilton's 2024 study of Houston's smaller-lot reform uses a difference-in-differences design and finds no evidence that the reform increased assessed land values across its specifications. The setting is a large city with wastewater services; outcomes depend on the model and measure assessed land rather than all housing costs. It challenges a universal claim that subdivision necessarily raises land values, but does not predict Denman's result.²⁰

Asquith, Mast and Reed's study of new market-rate apartments finds nearby rents lower relative to comparison locations, consistent with added supply absorbing demand. This is evidence against the categorical argument that market housing cannot benefit other renters. Large urban apartment projects differ from rural lot splits, so neither the estimated percentage nor the timing should be transferred to Denman.²¹

Conversely, Stacy and colleagues' multi-city study associates less restrictive reforms with a modest increase in housing supply after several years, concentrated at higher rents; it finds no statistically significant lower-end rent improvement over the period examined. Its observational design and heterogeneous reforms limit causal interpretation. The finding warns against promising that zoning flexibility alone quickly reaches low-income households.²²

Both sides therefore face an evidentiary burden. The case for restriction must show benefits beyond those achievable by enabling additional dwellings alone. The case for subdivision must show realistic completion and access, rather than equating a market opportunity with affordable housing. Broad supply effects, local environmental suitability and enforceable low-income provision are separate questions.

6. Financing, tenure and household costs

Keeping a dwelling on an existing title may allow the established owner to borrow against the whole property. It does not automatically give an incoming household a mortgageable interest in that dwelling. A separate title can enable separate financing and sale, but adds approval, legal and servicing costs and does not guarantee lender approval. A cooperative or land-share interest has its own collateral, governance and exit conditions. The actual offer matters more than the label.

Current CMHC refinance provisions support qualifying existing homeowners creating legal secondary suites, with up to four units in total including existing units, up to 90% loan-to-value and a maximum 30-year amortization under specified conditions. The lending value or as-improved property value must be below $2 million. They require borrower/qualifying-relative occupancy and restrict short-term rental use. These provisions improve a financing route; they do not establish below-market rents, eligibility for every Denman project or an independent ownership route for the tenant.²³

The provincial Secondary Suite Incentive Program illustrates another distinction. It tied assistance to affordability conditions, but stopped accepting applications on March 31, 2025. A feasibility model cannot assume that discontinued program is currently available. Any replacement funding must be verified independently rather than inferred from an announcement.²⁴

An ordinary self-contained rental also differs from sharing an owner's kitchen or bathroom. BC's Residential Tenancy Act excludes the latter arrangement under section 4(c); the same-lot location alone is not that exclusion. Section 4(a) also excludes accommodation rented by a not-for-profit housing cooperative to its member. Cooperative membership therefore requires its own assessment of governing law, rules and occupancy agreement. Assess tenancy protection, lawful owner-use provisions, sale or refinance risk and contractual duration without describing all additional-dwelling tenants as having no rights.²⁵

Legalizing an already occupied additional dwelling can improve residents' safety, tenure security and ability to remain, even when it adds no home to the physical stock. Those benefits deserve a separate outcome measure. Conversely, compliance costs or redevelopment may displace existing residents. A rental-cooperative membership also must not be counted as equity ownership; record occupancy rights, member capital, refund conditions and exit risk according to the actual arrangement.

For each scenario, a complete account includes land or contributed value; building work; water, wastewater and access; professional and approval costs; financing during construction; taxes, insurance, maintenance and reserves; vacancy; utilities; and any expected return on equity. For ownership, add down payment and exit/resale terms. An advertised low buy-in or monthly maintenance charge is only part of this account.

Illustrative finance sensitivity

The following is arithmetic, not a Denman cost estimate, rent forecast or current lending quote. It uses a hypothetical 25-year loan at 6% nominal annual interest divided into monthly periods, $300 monthly operating costs, 5% vacancy and $150 tenant-paid utilities. It excludes return on equity and assumes that the loan funds only the modeled additional dwelling. Actual Canadian loan conventions and terms must be substituted before a real feasibility decision.

Cash-covering rent equals monthly debt service plus operating costs, divided by one minus vacancy. At the conventional 30% gross-income benchmark, the middle illustration requires roughly $100,000 annual household income. Changing debt, subsidy, operating cost or the owner's required return changes the result. A low-cost building does not guarantee a low-income rent, and a generous owner may charge below a financial break-even calculation.

This arithmetic does not favor either title arrangement: it exposes assumptions that a comparative study must populate with evidence. Lower costs can improve feasibility or owner returns without being passed through to tenants; competitive supply, lease terms, subsidy conditions and enforceable agreements determine who receives the benefit. The reproducible sensitivity file includes alternative rates rather than pretending one assumed rate predicts the market.

7. Servicing and environmental constraints

Shared driveways, siting and services can reduce duplicated infrastructure and land disturbance. Their feasibility is site-specific. Island Health expressly addresses sewerage systems serving multiple structures on one parcel and systems serving strata/shared interests; increased flow or additional bedrooms can require reassessment or upgrades. A shared title does not create unlimited septic capacity, and separate titles do not necessarily require every component to be duplicated.²⁶

Groundwater licensing also depends on use and legal configuration, not simply the number of lots. The Water Sustainability Act's domestic-purpose definition includes qualifications and exclusions for some multi-family, strata and cooperative buildings. Obtain the applicable licensing and drinking-water requirements for each actual design rather than assume that clustering makes those obligations disappear.²⁷

The study should measure land disturbed, water demand, treatment capacity, access and long-term system liability per occupied household as well as in total. Compare the same bedroom capacity on suitable sites, then examine realistic occupancy. A reduction in titles is not an environmental outcome by itself; a reduction in feasible dwellings is not automatically an affordability improvement either.

Environmental and cultural protection can justify restrictions independently of price effects. If that is the deciding reason, the tradeoff should be stated directly, with targeted alternatives considered. An environmental preference should not be presented as a proven affordability result without the additional economic evidence.

8. Study design and the evidence needed next

The accompanying protocol defines five hypotheses before estimating a Denman effect. It begins with an inventory of identifiable subdivisions, additional dwellings and collective/nonprofit projects from 2015 onward, retaining earlier cases as history. Successful, withdrawn, stalled and unbuilt proposals are all relevant. Existing files establish much of the approval history; actual costs, lender terms and subsequent occupancy are the principal gaps.

Four policy packages should be compared: existing rules; more flexible additional dwellings with ordinary subdivision retained; the proposed combined package; and additional dwellings with a bounded, environmentally screened subdivision route. Nonmarket housing and explicit affordability mechanisms should be visible within each package. The study must not compare unrestricted large-home subdivision with subsidized tiny housing and label every difference a title effect.

Primary outcomes are completed year-round occupied homes, affordability at specified income and household-size bands, access for unrelated renters, entry to equity ownership, rental-cooperative access, total shelter payments and durability through turnover. Improved legal status and security for existing residents are reported separately from net new homes. Secondary outcomes include construction time, financing failures, replacement or demolition of existing homes, displacement, seasonal use and site impacts. Permissions and modeled buildout are intermediate indicators.

The initial analysis bands are gross annual household incomes of $30,000, $45,000, $60,000 and $90,000 in 2026 Canadian dollars, reported for one-, two- and four-person households where suitable accommodation exists. They are transparent analytical bands, not verified local income quantiles or program eligibility limits. Their 30% monthly shelter budgets are $750, $1,125, $1,500 and $2,250. Report who benefits and who loses across bands; more middle-market units cannot alone establish improvement for the lowest-income residents.

An observational comparison must control for age, size, bedrooms, location, condition, servicing and family discounts where the sample permits. A small island may not support a credible many-variable regression. If the available sample is small or selected, publish audited case histories, ranges and missingness instead of a precise causal coefficient. Comparison islands require compatible rules and pre-policy trends; nearby islands undergoing related reforms are not automatically valid controls.

A future adoption date would provide a monitoring baseline, not an instant experiment. Record outcomes over one, three and five years, separate concurrent funding and market changes, and retain censored projects still under construction. A post-policy rent movement alone cannot establish the policy's effect. The existing record does not yet support a numerical forecast for the number of low-income households helped or excluded by the proposed restriction.

The most informative next evidence is a matched local feasibility set: the same modest dwelling and site under alternative titles, supported by actual builder budgets, water/septic requirements and lender terms. Alongside it, assemble an occupancy and payment history for completed cases. No new interviews, survey solicitations or records requests have been sent as part of this assessment.

9. Questions a trustee candidate should be able to answer

1. Which outcome is the priority: lower rents, first-time ownership, secure family housing, or permanently income-restricted homes? Which household sizes and income bands?
2. What evidence shows that the subdivision restriction adds a benefit beyond enabling more dwellings on existing lots?
3. What lower-priced housing is available to people who do not already own land, have a donor or obtain family financing?
4. What assumptions connect permitted dwellings to financed, completed and occupied homes, and what happens if those assumptions fail?
5. Which affordability protections survive resale and tenant turnover, and which rely on the current owner's choice?
6. Could the same environmental goals be achieved with suitable-site rules, shared infrastructure and a bounded ownership route?

Supporting additional dwellings is consistent with the evidence. Treating that support as sufficient proof for removing ordinary subdivision is a further step that the reviewed evidence does not establish. The appropriate next decision is to require a comparison of actual feasible housing and access under the alternatives, with the tradeoffs stated openly.

Methods and limitations

This is a first-stage evidence assessment completed September 10, 2026, not a completed causal evaluation. It combines retained official local records, selected full transcript contexts, a fresh official hearing binder and original external studies. A broad automated screen covered 995 retained transcript files and 1,925,526 segments, finding 408 candidate subdivision/economic-language windows. These are retrieval counts; metadata inconsistencies prevent treating the apparent island labels as an independently verified meeting census. Selected relevant contexts were read, not every segment. Original audio was not independently authenticated.

The local report review used a targeted 520-document cache and selected project records. All 1,528 hearing-binder pages were text-extracted, but manual review focused on operative drafts and relevant reports; extraction is not full human review. External evidence ranges from implementation case studies to voluntary surveys and quasi-experimental urban research. Historical currencies, dates, selection biases and abstract-only access are identified. No study is treated as a direct Denman policy evaluation.

The report's analytical judgments and illustrative calculations are distinct from attributed findings. The linked sources control where summaries simplify legal or project-specific language. The report is independent of whether a particular official's motives are admirable: the empirical question is what the policy produces for households.

Sources

1. Islands Trust. Denman Bylaws 260/264 public hearing binder, file dated August 26, 2026. PDF pp.11,31-32,37,56,99-100; clean proposed bylaws. Hearing status: official project page.

2. Islands Trust. January 20, 2026 staff report and attachments, discussion pp.9-13; standalone PDF pp.14-18. Also retained in full agenda archive 1239, where pagination differs.

3. Islands Trust. June 26, 2025 staff report, pp.7-10; meeting transcript at 45:11, with surrounding discussion through 49:40 and 1:41:10-1:44:18. Machine derivative; original audio not authenticated here.

4. Islands Trust. August 24, 2026 staff report, pp.3,8-10, expected effects, modeled capacity and optional policy conditions.

5. Islands Trust. June 6, 2023 agenda, staff report pp.11-12 / agenda pp.78-79, financing and uptake constraints.

6. Denman Housing Advisory Planning Commission. Final report, November 3, 2023, pp.16-17 and54-62. Internal report date controls over erroneous archive metadata.

7. Islands Trust. March 17, 2026 agenda, pp.12-18, draft February 17 COHO/Triple Rock community-information/hearing record; archive1129. Participant statements, not lender commitments.

8. Islands Trust. July 15, 2025 agenda, pp.25-26, Snag applicant proposal; archive1476.

9. Islands Trust. June 23, 2026 agenda, pp.132-140, especially application p.7 / package p.139, 3Ravens; archive993. Applicant estimates, not audited accounts.

10. Denman Community Housing Society. Projects, retrieved September 10, 2026, The Ridge and Denman Green operator descriptions. Live HTML retained in the study evidence folder.

11. Denman Housing Association. Denman Green and joint housing-organizations update, January 2023 update, loss of intended Pepper Lane site.

12. Tomalty, Ray; Wilson, Dan; Ho, Cheeying; and Bhonde, Devarsh. BC Housing. Accessory Dwelling Units: Case Studies and Best Practices from BC Communities, 2021, pp.5-7,16,23-25,53.

13. City of Vancouver. Laneway Housing Survey Summary, January 2018, pp.1-4 and detailed results; voluntary owner/occupant survey.

14. Chapple, Karen; Wegmann, Jake; Mashhood, Farzad; and Coleman, Rebecca. Jumpstarting the Market for Accessory Dwelling Units, July 2017, pp.14-18, survey and limitations.

15. Chapple, Karen; Ganetsos, Dori; and Lopez, Emmanuel. Implementing the Backyard Revolution, April 2021, pp.5-6,14-16,27-28; 2020 fieldwork.

16. Portland State University, Northwest Economic Research Center / Institute for Sustainable Solutions. The Portland ADU Market: Conditions, Costs, Drivers, Incentives, May 2019; October-November 2017 survey and neighborhood comparisons.

17. Modai-Snir, Tal, and Alterman, Rachelle. The Hidden Stock of Affordable Housing, Journal of Planning Education and Research, first published July 7, 2026. Data and Methods, Housing Size, and Rishon LeZion analysis.

18. Tanrisever, Idil. Spillover Effects of Accessory Dwelling Unit Development, Regional Science and Urban Economics114, September2025,104136. Revised original abstract, September29,2025; full methods inaccessible in this review.

19. Pothukuchi, Kameshwari, and Guinn, Andrew. Beyond Zoning Reform: A Review of Strategies for Accessory Dwelling Unit Affordability, Journal of Planning Literature, first published December2,2025. Abstract/reference-list access only.

20. Hamilton, Emily. The Effects of Minimum-Lot-Size Reform on Houston Land Values, Cityscape26(3),2024, pp.159-186. Assessed-value difference-in-differences study.

21. Asquith, Brian J.; Mast, Evan; and Reed, Davin. Supply Shock Versus Demand Shock: The Local Effects of New Housing in Low-Income Areas, Upjohn working paper, December2019; later published in Review of Economics and Statistics105(2),2023, pp.359-375.

22. Stacy, Christina P., et al. Land-Use Reforms and Housing Costs: Does Allowing for Increased Density Lead to Greater Affordability?, March29,2023. Original authors' research summary of multi-city panel analysis.

23. CMHC. CMHC Refinance for Building Secondary Suites, current product terms checked September10,2026. Eligibility remains lender/project-specific.

24. BC Housing. Secondary Suite Incentive Program, application closure March31,2025 and program conditions; checked September10,2026.

25. British Columbia. Residential Tenancy Act, particularly ss.4(a),4(c) and applicable tenancy provisions; current text checked September10,2026.

26. Island Health. Sewerage & Subdivision, system types, increased flow and assessment requirements; checked September10,2026.

27. British Columbia. Water Sustainability Act, s.2 domestic-purpose definition and relevant licensing provisions; current text checked September10,2026.

28. Islands Trust. February 20, 2024 special-meeting agenda and adopted minutes, staff housing-policy options and LTC responses; archive documents 2358 and 2361. See the dated chronology entry for the surrounding options and subsequent changes.