← All reports
Verification and currency in Canada’s beneficial ownership registers
Lionel J. Perez, BA, JD, LL.L · CorpCentre Research Programme
Register extract: 4 August 2026 · preregistered sample of 5,000 · full-population census of 276,172 records (7 September 2026) · 31 pages
Canada’s beneficial-ownership transparency regime rests on a register whose content enters by self-declaration and is not verified by the registry at entry, and whose currency this report measures directly. The documentary record establishes the architecture: registry-side verification obligations attach to the actors operating the filing channels, and none attaches to the information passing through them; the machine channels omit the beneficial-ownership block entirely; and Quebec’s statute shows that a legislature can route identification documents to the registrar for one class of person while leaving verification of another to the registrant — a different allocation of duty, applied to directors rather than ultimate beneficiaries, and one whose operational effect the statute does not state. A preregistered sampling study of 5,000 active federal corporations — five strata by annual-return condition, every fetch successful, every residual adjudicated — then measures the consequence. ISC information is fresh exactly where returns are filed (94.1% listed among current filers) and absent where they are not: 95.9% and 99.6% of the two lapsed strata show no information ever filed, because the confirmation obligation rides on the annual return and lapsing across the regime’s January 2024 start leaves the block empty until a return or change filing occurs. ISC information appears in only 2.00% [1.47–2.71] of lapsed corporations’ records, and more than half of those records arrived with annual returns filed in the ten days between extract and fetch; the share consistent with the 15-day change channel is 0.86% [0.54–1.38]. Weighted to the 508,470-corporation cohort: ~24,300 [18,500–31,900] declare no qualifying individual, ~4,350 [2,300–8,300] are statutorily excluded, and ~2,300 [1,000–5,500] formally declare themselves unable to identify their own individuals with significant control. The last two rest on small cells and are correspondingly wide. A subsequent full-population census of 276,172 corporation records (§7) counts, rather than estimates, the corporations that have never had beneficial-ownership information appear at all: 17,973. That count is complete for the strata in which the quantity occurs, which the census enumerates in full; the unobserved remainder of the register is current filers, where both the sample and the census observe the condition at zero. Reporting entities subject to s. 138 must, when required to verify a CBCA corporation’s identity, obtain and take reasonable measures to confirm its beneficial-ownership information; FINTRAC identifies this register as one reasonable measure for doing so, and since 1 October 2025 the Regulations prescribe its consultation, with discrepancy reporting, for clients assessed as high-risk. For the empty stock, consultation returns nothing to compare against; whether an empty block is itself a reportable discrepancy the Regulations do not say, and a report of the absence cannot supply what was never filed. Who verified the beneficial owner before the information entered the federal register? The registry’s own architecture does not establish that anyone did.
Identity is verified for registered intermediaries and, progressively, for online filers; no provision verifies the ISC content itself at entry. Verification of content exists after entry — a risk-based audit the mutual evaluation documents — and it selects on declarations, which is why it cannot reach a corporation that has filed none.
Neither the bulk datasets nor the Federal Corporation API carries the ISC block; the data exists page by page only. Obtaining the block for a population required 276,172 individual page retrievals over four weeks. Consultation at the scale of a client book — repeatable, documented and auditable — has no official channel.
94.1% of current filers show ISC information; 95.9–99.6% of corporations whose last return predates the public regime show none. 76.0% of never-filed corporations show ISC information — almost entirely supplied once at incorporation (709 of 760 listed records are dated 2024) and never confirmed since. The share of lapsed records consistent with the 15-day change channel is 0.86%. Of corporations with no contact with the register since the obligation commenced, 98.62% show no individual listed, against 6.97% of those with contact.
~24,300 no-qualifying-individual declarations, ~4,350 statutory exclusions, and ~2,300 corporations declaring themselves unable to identify their ISCs, all weighted estimates; the fourth gap — corporations that have never had ISC information appear at all — is counted at 17,973, complete for the strata in which it occurs.
Across all 219,243 corporations whose federal existence began on or after the obligation commenced, 94.68% name an individual with significant control and only 484 — one in 450 — have filed nothing at all; on the same records at the same moment, 71.41% are current on the annual return the confirmation duty rides on.
Perez, L. J. (2026). Who Verified the Beneficial Owner?: Verification and Currency in Canada’s Beneficial Ownership Registers. CorpCentre Research Programme, Report 12. Published 1 November 2026. https://www.corpcentre.ca/research-report-12-who-verified-the-beneficial-owner
Disclosure. Prepared by Lionel J. Perez, BA, JD, LL.L, CEO of CorpCentre, which provides corporate compliance services to Canadian businesses. Its findings bear on the market in which the firm operates. Every figure is computed from publicly available data, and the method is set out in Appendix A in sufficient detail for any reader to reproduce the analysis independently or to challenge it. Readers are invited to do both. CorpCentre is a registered intermediary with Corporations Canada. In September 2025 it was invited to participate in the Department of Finance’s targeted consultation on company service providers. This report alleges no wrongdoing by any corporation: the conditions it measures are administrative states recorded by the corporations themselves, and no filed record is alleged to be false.
Attribution. Contains information licensed under the Open Government Licence — Canada (Corporations Canada, Federal Corporations datasets and the public corporation record pages). Quebec-derived material draws on the Registre des entreprises open dataset (Registraire des entreprises du Québec, via Données Québec), used under CC BY-NC-SA 4.0 as non-commercial research with attribution.