One Investor's Shares Linked Three Cambridge Companies to the CRA

The situation

By the time Dilshan came to us, the CRA had already reassessed him once, he had already tried to settle it once with different representation, and the settlement had made things worse rather than better. The corporation he owned, a small custom fabrication business he had built up over nearly a decade, had accepted a proposal from an earlier advisor that conceded far more than the underlying facts required, apparently on the theory that a quick resolution was better than a drawn-out fight. Dilshan came to us afterward asking whether the first settlement could be reopened, because the numbers still did not make sense to him.

The root problem had started years earlier and had nothing to do with Dilshan's own business decisions. Ishara, a police sergeant, had put a modest inheritance into two small local businesses as a passive investor: a minority, non-voting-sounding stake that she and both business owners understood as a quiet, hands-off arrangement. One of those businesses was Dilshan's fabrication company. The other was a physiotherapy clinic owned and run by Kasia. Ishara held shares in both, plus a small stake in a third, unrelated corporation she had invested in separately.

None of the three business owners knew about each other's arrangements with Ishara until a CRA audit of the physiotherapy clinic, triggered by an unrelated bookkeeping issue, pulled the corporate ownership records for all of Ishara's holdings as a matter of routine. The auditor noticed that Ishara held enough of an interest, combined with certain voting arrangements attached to her shares, to potentially bring all three corporations within the CRA's definition of associated corporations for tax purposes — a status that has nothing to do with whether businesses actually work together, share customers, or even know of each other's existence, and everything to do with who controls them.

Corporations found to be associated must share a single small business deduction limit among themselves rather than each claiming the full amount separately, which reduces the amount of active business income each corporation can tax at the lower small business rate. For three previously unconnected corporations that had each been filing as though the full limit applied to them individually, the retroactive effect was a meaningful reassessment across all three, with Dilshan's corporation bearing the largest share given its size relative to Kasia's clinic and the third company.

What the law actually said

Corporate association rules exist to stop business owners from multiplying the small business deduction by splitting one business into several corporations that are really controlled by the same person or group of people. The rules look past who technically holds voting shares and ask a broader question: who, in substance, controls the corporation, whether directly, through related persons, or through arrangements that give someone influence beyond their nominal shareholding. A shareholder who is genuinely passive, with no real say over how a company is run, is not the target the rules are aimed at, but the test does not simply take a shareholder's word for it — it looks at the actual rights attached to the shares and any agreements around them.

The first advisor's settlement had conceded that Ishara's arrangement with all three corporations met the control threshold, without meaningfully testing whether her shares carried the kind of influence the association rules actually require. That concession, once made to the CRA, was difficult to walk back cleanly; reopening a settled matter is possible, but it requires showing the earlier position was based on an incomplete or mistaken understanding of the facts, not simply that Dilshan had changed his mind about the outcome.

Reviewing the actual share terms, we found meaningful differences among the three arrangements that the earlier settlement had treated as identical. Ishara's shares in Dilshan's corporation were genuinely non-voting on ordinary matters, with no right to appoint directors or veto decisions — a passive investment in substance as well as in name. Her arrangement with Kasia's clinic was different: an early handshake understanding, never properly documented, had given her an informal veto over major clinic decisions in exchange for her investment, which was closer to real influence even without a formal voting share. Her third holding was smaller still and, on the facts, clearly passive.

This meant the blanket association the CRA had assumed, and the earlier settlement had accepted, did not hold up evenly across all three companies. Dilshan's corporation had a real argument that it was not associated with the other two at all, based on the actual rights Ishara held in it specifically, separate from whatever informal influence she may have had over Kasia's clinic.

What we did

1. Requested the CRA's file and the terms of the earlier settlement in full. Before proposing anything new, we needed to understand exactly what the previous advisor had conceded and why, so we could identify which parts of the settlement rested on facts rather than assumptions, and which had simply been accepted without being tested. This review told us where the real leverage for reopening the file might exist, and where Dilshan's frustration, however understandable, was not going to be enough on its own to move the CRA.
2. Obtained and reviewed the actual share certificates and any side agreements for all three corporations. Rather than relying on how the arrangements had been described informally over the years, we asked Dilshan, and with his consent approached Kasia and Ishara, for the underlying legal documents, which revealed the real differences in voting rights and influence that the earlier settlement had glossed over.
3. Built a corporation-by-corporation control analysis instead of treating the three as one group. We prepared a separate written analysis of Ishara's actual legal and practical influence over each corporation individually, showing that her position in Dilshan's company genuinely differed from her position in Kasia's clinic. Separating the three mattered because the earlier settlement's core error was treating one uniform relationship as though it applied identically everywhere, and only a document that argued each corporation on its own facts could unwind that assumption.
4. Prepared a request to reopen the settled matter on the basis of new and more complete facts. We explained to the CRA that the original settlement had proceeded on an assumption of uniform control across all three corporations that the underlying documents did not support, and asked that Dilshan's corporation's position be reconsidered on its own facts. Framing the request around a factual gap, rather than disagreement with the outcome, gave the CRA a recognized basis to revisit a closed matter — something dissatisfaction alone would not have provided.
5. Negotiated directly with the CRA's appeals officer rather than accepting the file as closed. This required persuading the CRA that reopening was appropriate given the incomplete factual record the first settlement relied on, which took several rounds of correspondence and a formal meeting to walk through the share documents in detail. Persistence mattered because a first settlement, once accepted, is not something the CRA reopens lightly, and the appeals officer needed to be satisfied the new documents genuinely changed the picture rather than restating Dilshan's original objection in different words.
6. Recalculated the small business deduction exposure once association was narrowed. With the case for full association weakened for Dilshan's corporation specifically, we recalculated what portion of the original reassessment could reasonably still be attributed to a genuine, narrower association with the third, clearly passive holding, rather than the broader three-way association originally assumed. This recalculation gave Dilshan a concrete figure to weigh against the cost and delay of continuing to fight for a full reversal, turning an abstract legal argument about control into a number he could actually decide against.
7. Advised Dilshan on tightening the shareholder documentation going forward. Once the immediate dispute was addressed, we recommended Dilshan formalize, in writing, the terms of Ishara's investment in his corporation, so that any future review would not have to reconstruct informal understandings after the fact. Ishara's genuinely passive position in Dilshan's company was the reason the narrower argument succeeded at all, and writing that down now means the next reviewer will not have to take anyone's word for what everyone already understood.

The outcome

The CRA agreed to reopen the settled matter based on the more complete factual record, which was not a certainty going in, and accepted that Dilshan's corporation and Kasia's clinic were not genuinely associated with each other in the way the original settlement had conceded. That narrowed the small business deduction sharing considerably. It did not eliminate it: the CRA maintained, and the documented informal veto arrangement supported, that Ishara's involvement in the third corporation still gave rise to some level of association affecting Dilshan's corporation through her overlapping holdings, and the earlier settlement's concession on that narrower point was harder to unwind.

The final reassessment against Dilshan's corporation was reduced from the original figure of roughly $370,000 to approximately $190,000, reflecting a real but smaller association than the CRA had initially found. Dilshan paid the reduced amount along with the interest that had accrued while the matter was under review, rather than pursuing a further formal appeal, given the remaining exposure was modest relative to the cost and time of continuing the dispute.

The outcome was a hard lesson rather than a clean win. Dilshan lost real money on an association he had no part in creating and no visibility into until an unrelated audit surfaced it, and the earlier settlement's concessions meant the final number, while much improved, was never going to return to zero. What the reopened file did accomplish was containing the damage to a defensible figure grounded in the actual facts, rather than leaving Dilshan bound by a settlement built on an assumption that did not hold up once the documents were properly reviewed.

What you can learn from this