Holding US Real Estate Through a Canadian Corporation: the Structure That Solved a 1990s Problem and Creates Five Modern Ones

This structure is cross-border tax's living fossil: the single-purpose Canadian corporation holding US personal-use real estate proliferated when US estate tax exposure was broad and the treaty's relief thin, and it persists in files — and in some advisors' instincts — decades after the problem it solved shrank and the problems it creates compounded. The original logic: shares of a Canadian corporation are not US-situs assets, so interposing the corporation removed the Florida house from the US estate tax base — real protection in an era of low exemptions. What changed: the exemption's growth and the treaty's pro-rated credit (the snowbird estate playbook's core) took most Canadian estates out of US estate tax exposure entirely, converting the structure's benefit from live tax savings to insurance most holders no longer need — while the CRA's administrative tolerance that once softened the structure's Canadian costs (the historical position not assessing shareholder benefits on single-purpose corporations meeting conditions) was withdrawn for structures outside the grandfathered past, leaving the modern costs standing alone. The five modern problems: the shareholder benefit — personal use of corporate property is a taxable benefit to the shareholder at the property's rental value or an imputed return on cost, assessed annually, converting every February in Naples into Canadian income tax on rent nobody paid — the flagship cost, recurring, and audited; the exit stack — a corporate sale of the property runs FIRPTA at the corporate rate plus the 1120-F process, potential branch profits tax on repatriating the proceeds, and Canadian corporate tax on the same gain with foreign-affiliate surplus mechanics deciding how badly the personal extraction stacks — the all-in exit cost reliably exceeding the personal-ownership version by wide margins; the rental penalty — if the property earns rent, corporate rates without the individual net-election's graduated structure, plus the eventual second layer extracting profits, price corporate landlording above personal for this asset class; the compliance stack — T2 returns, 1120-F filings, Form 5472 on the related-party flows (the shareholder's use is itself a reportable transaction), FBAR-adjacent questions for the corporate accounts, vacancy-tax-style analyses where the pattern reverses — a four-figure annual professional bill for the privilege of the other four problems; and the financing-and-dealing friction — US lenders dislike foreign-corporate borrowers, title and insurance complicate, and buyers' counsel treat corporate sellers as diligence projects. What the analysis looks like today: for new purchases, the structure is presumptively wrong — the estate worksheet almost always shows the treaty credits covering the exposure, and where genuine exposure exists, the modern levers (Canadian-fund asset location for the securities side, nonrecourse debt, insurance, and for the largest estates, properly-built trusts settled before purchase) address it without manufacturing shareholder benefits; for existing structures, the unwind analysis — distributing the property out of the corporation is itself a disposition (FIRPTA on the corporate transfer, Canadian corporate gain, and a shareholder-level dividend or benefit on the extraction — the exit stack arriving early) — so unwinds get modeled against the status quo's annual costs and the eventual-sale stack, with the timing levers (low-value years, loss positions elsewhere, the shareholder's bracket years, coordination with a genuine sale to a third party, which lets the unwind and exit share one tax event) deciding whether to dissolve now, at sale, or at death; and for the grandfathered old structures still inside historical administrative positions, the analysis respects the grandfathering's conditions precisely, because renovations, refinancings, and transfers are the events that end it. The advisory posture, condensed: nobody builds this in the current decade without an estate profile the worksheet actually flags and a comparison against the modern levers — and most existing ones are unwound at the next natural transaction, because the structure's annual rent (benefits, filings, friction) buys insurance against an estate tax the treaty already waived for its holder.

Key takeaways

• The benefit assessment is the flagship cost: personal use of corporate property = annual taxable shareholder benefit at rental value or imputed return — recurring Canadian tax on unpaid rent, and the audit category these structures headline.
• The exit stack compounds: corporate FIRPTA and 1120-F, potential branch tax, Canadian corporate tax, then the personal extraction layer — modeled once, the all-in exit reliably embarrasses the personal-ownership comparison.
• The original problem mostly dissolved: the exemption and the treaty's pro-rated credit cover typical Canadian estates — the worksheet, not nostalgia, decides whether any US estate exposure exists to insure.
• Modern levers beat the fossil: asset-location, nonrecourse debt, insurance, and pre-acquisition trusts for genuinely large estates — protection without benefits, stacked filings, or corporate exits.
• Unwinds are dispositions — timed, not defaulted: distributing the property out triggers the exit taxes early, so unwind decisions model now-versus-at-sale-versus-at-death, with third-party-sale coordination the cleanest single-event route.
• Grandfathered structures live on their conditions: the historical administrative positions that shelter old single-purpose corporations end with the transactions (renovation-scale changes, refinancings, transfers) that violate their terms — known precisely, or lost casually.

The existing-structure review

One meeting, four numbers: the annual carry (benefit assessments projected honestly, compliance fees, friction costs); the status-quo exit (the full corporate stack at projected sale values and dates); the unwind-now cost (the same stack at today's values, plus extraction); and the estate-tax insurance value (the worksheet's actual exposure — usually zero). The decision usually writes itself: zero insurance value plus five-figure annual carry points to unwinding at the next natural event; genuine exposure in a nine-figure estate points to re-engineering into the modern toolkit rather than keeping the fossil; and the grandfathered cases get their conditions documented and a do-not-touch list circulated to everyone who might innocently refinance the thing.

Worked example

A retired Oakville couple's file arrives with a 2011-vintage Ontario holdco owning their US$900,000 Marco Island house — built on their then-advisor's estate-tax logic, used personally twenty weeks a year, never rented. The review's four numbers: annual carry — a projected shareholder benefit near C$50,000 of imputed value (roughly C$24,000 of annual tax at their rates, unassessed so far but squarely within the CRA's current position, plus reassessment-window exposure for open years), C$6,500 of annual T2/1120-F/5472 preparation; status-quo exit at a projected US$1.1 million sale — the corporate FIRPTA-and-branch stack plus Canadian corporate tax plus extraction modeling to roughly US$140,000 more than personal ownership's version of the same sale; unwind-now — the same stack at today's value, softened by this year's circumstances (a capital loss position elsewhere in the corporation and the couple's low-bracket year); insurance value — their C$7 million worldwide estate sits far under the exemption: the worksheet shows zero US estate tax with or without the corporation. Decision: unwind now — the distribution's tax cost runs about C$85,000 against a status quo carrying C$30,000 a year plus the larger eventual exit; title moves to personal tenants-in-common with the standard playbook (wills updated, the estate worksheet noting the now-direct US-situs asset covered by treaty credits, insurance declined as unnecessary). The structure that was protection in 2011 had become a C$30,000-a-year subscription to a solved problem — and the file's closing memo says so in one sentence, for the next advisor who inherits the instinct.

Official sources

Under subsection 15(2), a loan to a shareholder is included in income unless "the loan is repaid within one year after the end of the tax year of the lender or creditor in which the loan was made" and the repayment "is not part of a series of loans or other transactions and repayments"; a taxable interest benefit arises where interest is not paid at the prescribed rate. — Canada Revenue Agency, Income Tax Folio S3-F1-C1, Shareholder Loans and Debts, https://www.canada.ca/en/revenue-agency/services/tax/technical-information/income-tax/income-tax-folios-index/series-3-property-investments-savings-plans/folio-1-shares-shareholders-security-transactions/income-tax-folio-s3-f1-c1-shareholder-loans-debts.html

"The disposition of a U.S. real property interest by a foreign person (the transferor) is subject to the Foreign Investment in Real Property Tax Act of 1980 (FIRPTA) income tax withholding." — Internal Revenue Service, FIRPTA Withholding, https://www.irs.gov/individuals/international-taxpayers/firpta-withholding

Practitioner note

The Canadian holdco over US personal-use property is the corridor's most durable fossil: built for an estate tax the treaty now waives for its typical holder, and billing annually in shareholder benefits, stacked filings, and a corporate exit that punishes the eventual sale. Our review reduces it to four numbers — carry, exit, unwind, insurance value — and the pattern is near-universal: zero insurance value, five-figure carry, unwind at the next natural event. The rare legitimate cases re-engineer into modern levers; the grandfathered ones get their conditions laminated.

See also: For Canada's principal residence exemption vs the US section 121 exclusion, see Canada's principal residence exemption vs the US section 121 exclusion; and browse every cross-border tax topic guide, organized by situation.

Next step

Fairlight prepares the holdco review and unwind — the four-number analysis, benefit-exposure quantification for open years, unwind timing against loss positions and sale coordination, and the modern-lever redesign where genuine estate exposure exists. See cross-border pricing or book a call.