Recent statements from Canada Revenue Agency (“CRA”) confirmed a reversal of their position on the interplay between amounts deemed to be dividends under paragraph 84.1(1)(b) of the Income Tax Act[1] and a payor corporation’s ability to recover refundable taxes under paragraph 129(i)(a) of the Act. This reveals a pragmatic approach to promoting the integration principle in the taxation of corporations and their shareholders.

A brief summary of the two provisions at issue will be helpful:

  • Section 84.1 of the Act operates to prevent the removal of corporate surplus as a tax-exempt capital gain by transferring shares of one corporation to another in exchange for shares and/or debt of the purchaser. Where section 84.1 applies, the individual taxpayer disposing of shares is deemed to receive a dividend to the extent that the fair market value of any non-share consideration (e.g., a promissory note) exceeds the greater of the paid-up capital and the adjusted cost base to the transferor of the disposed shares. Note that the section includes no mechanics to designate that the deemed dividend has been paid on any particular class of shares or that the transferor in fact be a shareholder of the payor corporation.
  • The refundable tax regime that operates through numerous provisions of the Act prevents an undue deferral of shareholder-level tax on passive income earned by a corporation. The income-earning corporation pays an additional amount of tax on its aggregate investment income on top of its general corporate tax rate. The corporation can then obtain a refund of this tax pursuant to subsection 129(1) of the Act, upon the payment of taxable dividends to its shareholders. In other words, once the tax has been paid on the dividend in the hands of the shareholder integration is achieve only if and when the payor corporation has recovered the additional tax it paid on the same amount. It is, therefore, to the advantage of the corporation that it has an effective means of recovering the refundable tax as soon as possible to ensure that those amounts do not remain stranded, and inaccessible, within the corporation. The mechanics of the dividend refund are set out in subsection 129(1), and are triggered on the basis of taxable dividends paid by the corporation on shares of its capital stock in the year.

In an earlier 2002 technical interpretation[2] (the “2002 TI”), CRA considered the issue of whether a deemed dividend triggered, pursuant to subsection 84.1(1), could give rise to a dividend refund to the deemed payor corporation and, reversing an earlier position, took the position that a dividend refund could not be triggered by a paragraph 84.1(1)(b) deemed dividend. The 2002 TI stated that for subparagraph 129(1)(a)(i) of the Act to apply, a taxable dividend must have been paid by a private corporation on shares of its capital stock, and that a paragraph 84.1(1)(b) deemed dividend is not “paid, or deemed to be paid, on shares of the capital stock of the corporation.” This highly technical interpretation had the result that an otherwise taxable deemed dividend deemed to have been paid by a corporation that had an available refundable dividend tax on hand balance could not permit the corporation to recover refundable taxes. This meant that the integration principle inherent in the refundable tax regime would not operate properly.

This interpretation was likely technically correct. Considering the general principle as enunciated in section 8.1 of the Interpretations Act[3] that taxation statutes operate to attach consequences to transactions and events as they occur in accordance with provincial laws of property and civil rights, a dividend has not been “paid” by reason of the application of paragraph 84.1(1)(b). But since, in the immortal words of Bowman J., “Parliament has the power to deem cows to be chickens”, one questions why a paragraph 84.1(1)(b) deemed dividend, which is otherwise a taxable dividend, should have this result. Also, consider the economic realities of what occurs: the deemed payor corporation has paid the additional amount of tax that is intended to be a prepayment of shareholder-level tax and, by virtue of paragraph 84.1(1)(b) the deemed recipient shareholder has paid the shareholder-level tax. Without the dividend refund integration is not achieved.

At both the 2019 APFF Roundtable[4] and the 2019 CTF Roundtable[5], CRA confirmed that its  position as set out in the 2002 TI “no longer represents CRA’s position” and stated that, “the granting of a dividend refund to a corporation deemed by paragraph 84.1(1)(b) to have paid a dividend provides an outcome that is more in accordance with the integration principle embedded in the Act.” So, for now at least, CRA appears to have moved away from what might have been considered to be an overly technical approach to a more practical approach that recognizes the importance and primacy of the integration principle.

What the CRA’s revised position does not do, however, is to address the issue of whether a dividend can be deemed to be paid, and received, in the abstract. In this respect, it is interesting to note a key difference in the fact scenario described in the 2002 TI and in the 2019 Roundtable answers. In the 2019 examples, Mr. A., the transferor, remained a shareholder of Opco 2, the deemed payor corporation; therefore, it is possible that the dividend be deemed to be paid on the shares in Opco 2 held by Mr. A. In the 2002 TI example Mr. A was not, and did not become, a shareholder of Opco 2. Therefore, the requirement in paragraph 129(1)(a)(i) that the dividend refund be computed “in respect of taxable dividends paid by the corporation on shares of its capital stock in the year” is not met in the 2002 TI example. The 2019 Roundtable answers appear to acknowledge that: (i) a dividend can be deemed to be paid to a non-shareholder of a deemed payor corporation; and, (ii) a non-shareholder of a deemed payor corporation can, nonetheless, receive a dividend.

Given that CRA’s position on this issue has now been reversed twice, it will be interesting to see if taxpayers take sufficient comfort in the new position in dealing with the consequences of section 84.1, already a notoriously difficult and problematic section of the Act with which taxpayers and tax professionals alike must contend.


[1] Income Tax Act, R.S.C. 1985, c. 1 (5th Supp.), as amended (herein referred to as the “Act”). Unless otherwise stated, statutory references in this article are references to the Act.

[2] 2002-0128955, September 26, 2002.

[3] Interpretation Act, R.S.C., 1985, c. I-21.

[4] 11 October 2019 APFF Roundtable, Q. 1.

[5] 3 December 2019 CTF Roundtable, Q. 4.