The following article appeared in the latest edition of the Hamilton Law Association Law Journal.
Canada’s income tax system aspires to “neutrality” so that, in theory, an individual should be indifferent about whether she earns income directly or through a corporation. In the case of investment income, the theory actually comes pretty close to the reality. An individual who earns interest income or who realizes capital gains “through” a corporation pays only slightly more tax at the corporate and individual-shareholder level than the individual would have paid if the income had been earned by the individual directly.
The Income Tax Act (Canada) (the “Act”) attempts to achieve neutrality through two notional accounts of a corporation: refundable dividend tax on hand (“RDTOH”) and the capital dividend account (“CDA”). Let’s leave RDTOH for another day and focus on CDA.
The CDA ensures that certain receipts that would have been received by an individual tax free are earned by a corporation tax free and can be passed along to the corporation’s shareholders free of tax as well. The three most important of these receipts are the portion of capital gains that are not subject to tax, insurance proceeds and capital dividends received from another corporation. The Act contains a detailed formula for calculating the CDA of a corporation. For present purposes, let us consider the formula as it would apply to a corporation that had only gains and losses, insurance proceeds and capital dividends received from other corporations. The formula, then, can be paraphrased as follows:
CDA (at a particular time) = (A + B + C) ‑ D where
‘A’ is the amount, if any, by which the total of the non-taxable portion of the corporation’s capital gains realized before that time exceeds the non-allowable portion of its capital losses realized before that time;
‘B’ is the amount by which the proceeds of an insurance policy exceed the adjusted cost basis of that policy;
‘C’ is the amount of any capital dividend received by the corporation before that time; and
‘D’ is the total of all dividends paid from the corporation’s CDA before that time.
The CDA of a corporation is always calculated at a point in time. In addition, note two points about ‘A’:
- A cannot be negative because it is the amount, if any, by which gains exceed losses and because, in general, formulas in the Act cannot yield negative amounts. Accordingly, if a corporation’s capital losses exceed its gains, its CDA can still be positive if, for example, it has received insurance proceeds. In such a case, and assuming that C and D are nil, A would be nil and B would be positive so that the CDA of the corporation would be positive.
- The calculation in A only takes into account realized gains and losses. A corporation may have realized gains and significant accrued but unrealized losses that, if realized, would exceed the gains. A would still be positive in such a case because the formula ignores the unrealized losses. Moreover, the Act does not contain a specific anti-avoidance rule that would change this result.
With respect to ‘B’, the Act contains detailed rules for calculating the adjusted cost basis of an insurance policy. In general, ‘adjusted cost basis’ includes premiums paid on the policy. The calculation is quite detailed, however, and usually it is best simply to request the amount of the basis from the issuer of the policy. The issuer generally keeps track of the basis as a matter of course.
CDA Tricks and Traps
In calculating CDA, one must account for all events occurring from the beginning of the calculation period until its end, cumulatively, which means that the period can stretch over years or even decades. In addition, while in the formula above ‘A’ cannot be a negative number, that doesn’t mean that capital losses can’t have effects on other components of the formula.
For example, suppose that in the calculation period Holdco realizes net capital losses of $1,000 and receives a $500 capital dividend from another corporation. Holdco, even though it has net losses, can still pay a dividend of $500 from its CDA. On the other hand, if Holdco first realizes a $1,000 capital gain, then pays a $500 capital dividend, then realizes a $2,000 capital loss and finally receives a capital dividend of $500, it will not be entitled to pay another amount from its CDA immediately thereafter. In the formula, A is nil (and it will remain at nil until more gains are realized to offset the losses), C is $500 and D is $500. The CDA balance, then, is nil.
These results aren’t always intuitive, and so, when engaging in CDA planning, it is sometimes necessary to map out carefully the components that make up the balance of a corporation’s CDA to determine what can actually be paid from the account.
Paying a Dividend from CDA
The following steps must be completed to pay a dividend from CDA (per section 2101 of the Income Tax Regulations).
First, the corporation paying the dividend (Holdco) should calculate the amount of its CDA. Typically, the accountant for Holdco will perform this task using, among other things, the gains and losses reported by Holdco before that time on Schedule 6 of its T2.
Next, the directors of Holdco should pass a resolution declaring a dividend and designating the amount of the dividend as payable from Holdco’s CDA. Holdco’s lawyers should prepare this document.
Before the CDA dividend is paid or payable, Holdco must execute and file CRA election form T2054 so that the dividend will be considered to be paid tax-free from Holdco’s CDA. A certified copy of the dividend resolution and a copy of the CDA calculation must accompany the form.
After the T2054 is filed, Holdco can pay the capital dividend to its shareholders. Holdco need not issue T5 slips for the capital dividends it pays.
Accountants’ Mistakes: How You Can Help
It’s not hard to make mistakes when calculating the CDA of a corporation, and so pity the poor accountant who must get it right. What happens if the accountant makes a mistake so that the dividend paid exceeds the balance of the corporate payor’s CDA
The Act imposes a special tax (Part III tax) on the corporate payor of such a dividend. In addition, the corporation and its shareholders can file an election so that the excess is simply treated as a taxable dividend (which will avoid the Part III tax too).
A corporation will almost always wish to avoid Part III tax. Moreover, the election described in the previous paragraph is not always possible or desirable. Fortunately, lawyers can help through “rectification”. Tax practitioners will be familiar with Juliar v. Canada (Attorney General), 1999 CanLII 15097 (ON S.C.), in which the Court, by exercising its jurisdiction in equity, amended certain transactions with retroactive effect so that the parties to the transactions could avoid the adverse income tax consequences associated with the unamended transactions. According to the Ontario Court of Appeal decision in the case (2000 CanLII 16883), it didn’t matter that the only purpose for the amendment was to avoid or postpone income taxes. What mattered was that the parties had a continuing, common intention to effect the transactions in a tax-deferred manner, which intention was thwarted by the form of the transaction.
The Ontario Superior Court came to a different conclusion, however, in Binder v. Saffron Rouge Inc., 2008 CanLII 1662, because the parties did not have the requisite common intention. A corporation wished to issue shares to non-residents. It was advised, however, that issuing a particular number of shares would result in the corporation losing its status as a Canadian-controlled private corporation (“CCPC”) for the purposes of the Act. The parties were not aware that the loss of CCPC status would in turn affect the ability of the shareholders of the corporation to claim the capital gains exemption. The non-resident expressed indifference about the loss of CCPC status (the corporation was losing money anyway at the time), and there was no evidence that the non-resident cared or even knew about the loss of the exemption. Accordingly, the Court denied the request for rectification on the basis that the parties did not have a common intention about the tax consequences of the transaction.
In light of these cases, how can a lawyer help? Apparently, rectification applications to correct mistakes in CDA calculations have become somewhat routine, and in most cases where an honest mistake has occurred, the Department of Justice and the CRA will not oppose an application to correct the mistake. Justice and the CRA, however, must be satisfied that, among other things, an honest mistake was made (this usually requires the accountant to “fall on his sword”, if it was the accountant’s error) and that the corporation meant to pay a dividend only from its CDA.
An applicant will establish these elements through affidavit evidence. The accountant will swear an affidavit explaining how the mistake occurred. The “corporation” (perhaps one of its directors) will swear an affidavit establishing the common intention of the board to pay a dividend tax-free. The latter affidavit will be more convincing if it is supported by a statement of intention in the resolution declaring the dividend. That is, it would be helpful to include a statement in the recitals of the resolution itself that sets out the beliefs of the board about the amount of the payor’s CDA, that states the intention of the board to pay a dividend tax-free from the corporation’s CDA and that confirms that the dividend being declared by the resolution is that dividend.
The Act’s provisions governing the calculation of a corporation’s CDA are complex, and it is not difficult to get the calculation wrong. A lawyer can help with this problem, however. Usually, it is taken for granted that a lawyer’s role in paying a dividend from the CDA of a corporation is restricted to filling out a precedent. Indeed, some accountants don’t even bother to involve the lawyer at all: they fill out their own resolution precedent. The lawyer can add value, however, by including statements about intention in the recitals of the relevant resolutions and ensuring that the directors of the payor actually direct their minds to the intentions described in the recitals. Taking these steps could materially improve the chances that any mistake made in calculating the CDA of a corporation can be fixed later through rectification.
 See also QL Hotel Service Limited v. Ontario (Finance), 2008 CanLII 15226 (ON S.C.).
 This is according to Marie-Therese Boris, senior counsel with the Department of Justice, who made a presentation on rectification from the Justice perspective at a meeting of the Ontario Bar Association’s tax section on October 30, 2008.