It may not have been the collision of two worlds, but there is at least a fender bender to deal with. In Enmax Energy Corporation v. Alberta, the Alberta Court of Appeal (“ACA”) recently reversed the decision of the trial court that had allowed two subsidiaries of Enmax Corporation (“ENMAX”) to deduct interest paid on three inter-company loans at rates of 11.5 per cent, 10.3 per cent, and 9.9 per cent.1 Although both the trial court and the ACA spent a lot of time on the principles of interest deductibility related case law and legislation, the ACA went to great pains to describe its decision as being driven by regulatory as opposed to tax considerations.

The two subsidiaries, Enmax Energy Corporation, and Enmax PSA Corporations were exempt from federal tax by virtue of the operation of paragraph 149(1)(d.6) of the Income Tax Act (Canada) (“ITA”). However, both corporations were required to make payments to the Balancing Pool.2 Those payments were computed under the Payment in Lieu of Tax Regulation, Alta Reg 112/2003 (“PILOT Regulation”), as being equivalent to the amount of tax that would otherwise have been payable under the ITA and Alberta Corporate Tax Act. The Balancing Pool payments regime is established by the Electric Utilities Act, SA 2003, c E-5.1 (“Utilities Act”) and the PILOT Regulation.

ENMAX was also tax exempt for the purposes of the ITA, but it was not subject to the Balancing Pool payment regime. That regime only applied to a “municipal entity” which referred to a municipality or corporation that was a retailer of electricity, a holder of a Power Purchase Agreement (“PPA”) or an entity that owned a right derived from a PPA that allowed it to exchange electrical energy and ancillary services with a subsidiary of a municipality. There is no reference to a holding company in that definition, although that would be the logical way to organize the municipal ownership of regulated and non-regulated businesses.

The trial court decision3 made a number of taxpayer friendly findings:

  1. A 91 per cent debt to equity ratio was not offensive in the absence of a domestic thin cap rule. See Paragraph 271.
  2. Each of the inter-company notes had an “Inter-Company Debt Pricing Memorandum” which set out the borrower’s explanation and justification for the reasonableness of the interest payable under each note. Having this type of “pricing” memo to support the terms of the notes and the business rationale was a reasonable process that will generally lead to a reasonable amount.
  3. Reasonableness in the context of applying the interest deductibility test is a subjective consideration. It does not necessarily follow that paying an interest rate that exceeds market is unreasonable, if business considerations otherwise support that conclusion. Moreover, the reasonableness of the interest rate should not be conflated with an arm’s length standard — they are not the same test. As pointed out at paragraph 113, when the ITA requires an arm’s length standard — that standard is specifically described in the applicable legislation.
  4. A final interest rate which was 150 bps in excess of the arm’s length standard could still be reasonable, given the taxpayer’s desire to have the top holding company act as the exclusive source of external financing. See paragraph 267.

In reversing those conclusions, the ACA substituted interest rates of 5.42 per cent, 5.26 per cent, and 5.24 per cent for the stated interest rates in the three loan agreements of 11.5 per cent, 10.3 per cent and 9.9 per cent, respectively. The principal justification for this conclusion was the stated purpose of the legislative scheme under the Utilities Act and PILOT Regulation to level the playing field between municipal entities in the electricity sector and the private sector. The conclusion reached by the ACA was that the playing field was titled in favour of ENMAX and the City of Calgary because they used artificial stratagems which were dependent on them not being subject to the Balancing Pool payment and tax exempt under the ITA.

In other words, artificial stratagems — which themselves depend on the fact that Calgary and ENMAX are not subject to payment of Balancing Pool Payments or income tax — are not reasonable. Otherwise, corporate parents like ENMAX, which are not subject to income tax or Balancing Pool Payments, could lend funds to their municipal entity subsidiaries at disproportionately high interest rates. This would in turn allow the municipal entities to enjoy corresponding disproportionately high deductions against their Balancing Pool Payments while the corporate parent collects these interest payments free of both income tax and Balancing Pool Payments. Without an objective limit on the amount of interest deductions, the municipal entity would enjoy a competitive advantage over their private sector competitors. This would render the Balancing Pool Payments regime toothless. [para 8]

The “artificial stratagems” identified by the ACA seem to be a combination of the corporate governance policy which required all external loans to be sourced through ENMAX, and the structuring of the inter-company loans to have a 91 per cent debt to equity ratio without the robust roster of financial covenants you would normally see in an external loan. The effect of the latter was to make the loans appear riskier than they might otherwise be, driving up the interest rate costs. As an aside, it would have been interesting to hear evidence on what the advantages were, if any, of sourcing all capital requirements to a parent company.

The ACA and the trial court were faced with widely varying estimates from expert reports of what a reasonable interest rate should be. They dramatically differed on: (1) whether an implicit parent guarantee of the debt should be considered; (2) the use of “midpoint” credit rating by the trial court with numerous adjustments to establish a reasonable interest rate; and (3) whether the reasonableness of an interest rate was the same or at least quite similar to establishing a market rate of interest.

The ACA identified seven different reviewable errors made by the trial court in coming to its conclusion, the more interesting of which we discuss below.

First, the interpretation of the reasonableness requirement in paragraph 20(1)(c)(i) ITA should have taken into account government policy expressed in the Utilities Act in deregulating the electrical power market.

The trial judge erred in concluding that the interest payable to ENMAX on the three inter-company loans was reasonable and thus fully deductible. Determining whether the interest paid on each loan was reasonable requires proper consideration of the legislative scheme under the Utilities Act and the PILOT Regulation, including its key purpose, namely to level the playing field. Not only must the interest rate itself be objectively reasonable, but the structure of each loan must also be objectively reasonable to the extent it affects the amount of the interest paid. [para 7]

The trial court had concluded that paragraph 20(1)(c) ITA should have the same meaning under the PILOT Regulation as it did under the ITA — it would be anomalous to find that Alberta’s legislature wanted to incorporate by reference the words of the ITA, but not the jurisprudence considering the provision. But, because of the government policy expressed above, the ACA felt itself free to conclude that this was not an income tax case, and it had flexibility in applying the interest deductibility rules. Specifically, ENMAX did not have the right to organize its affairs in a tax efficient fashion, consistent with the principles in the Duke of Westminster.4 Rather, ENMAX had to make the “right” payment to the Balancing Pool. That of course begs the question of what is the “right” amount, and whether every municipal entity understands what “competitive balance” means, in the context of complex legislation. Nevertheless, the ACA concluded the trial court was wrong:

Accordingly, the trial judge erred in law when he concluded that the reasonableness standard set out in s 20(1)(c)(i) of the ITA cannot generally be based on an arm’s length standard. Even if this were correct in the ordinary income tax context — and we do not agree it is — it is incorrect in law under the Balancing Pool Payments regime. The trial judge also erred in determining that ENMAX was unlikely to provide implicit parental support for loans secured by Energy and PSA. Thus, the trial judge’s conclusion that the interest payable on each loan was reasonable cannot stand. Therefore, we set aside the trial judge’s decision, affirm the Minister’s reassessments, and dismiss the appeals of those reassessments by Energy and PSA. [para 9]

Second, the trial court did not apply the reasonableness test correctly. The test needed to be applied on an objective basis, taking into account that reasonableness must be measured from the perspective of the borrower, particularly since one party to the financing was not taxable. The trial court had stated that reasonableness must be measured with reference to the legal transactions to which the borrower was a party, not other contracts it might have made. It supported this statement with reference to the foreign currency/interest deductibility planning in Shell.5 The ACA disagreed, and approached the reasonableness analysis by ignoring the parameters of the actual loans, which had a high debt to equity ratio, and lacked the standard financial covenants.

Third, there was a failure to take into account the implicit parental support of ENMAX in measuring the reasonableness of the loan terms. The ACA was very critical of the trial court’s reasoning and concluded that ENMAX almost certainly would have provided parental support:

Viewed through this lens, it is reasonable to conclude that ENMAX would likely have provided parental support not just for any arm’s length Market Loans but also for the Parental Loans because parental support also buttressed its own position, not just that of Energy or PSA. In case of default, ENMAX would have two options: (a) let its subsidiaries fail and attempt to recover as much as possible; or (b) prop them up and recover as much as possible on a going-concern basis. It is clear on this record that, for the reasons given, ENMAX’s interests were so linked with those of its subsidiaries that implicit parental support, as reflected in the second option, was inherent in the contractual arrangements made for the Parental Loans. As a result, even if one ignored the arm’s length check — which we repeat would be an error — a court must nevertheless find a reasonable pricing for the Parental Loans as structured. Parental support was clearly implicit in that paradigm. [para 117]

Certain specific facts supported this conclusion:

  • One of the subsidiaries provided over 70 per cent of earnings for the group.
  • Given that size, an insolvency of the subsidiary would certainly impact ENMAX’s own credit rating, and that credit rating was important to ENMAX — there was also the possible ripple effect on the credit rating of the City of Calgary to consider.
  • ENMAX had the financial resources to provide the support to its subsidiaries — it paid significant annual dividends of $168M and $147M in the relevant years to the City of Calgary.

Fourth, it was not appropriate to adjust the interest rate upwards to account for “saved” transaction costs, particularly when the estimate of those transaction costs was premised on the loans being speculative investment grade securities which the ACA, described as the creation of an “artificial stratagem”.

The next to final paragraph of the judgment neatly summarizes the approach and thinking process of the ACA.

In the end, it comes down to this. Is it reasonable that a lender would consider a bond issued by PSA or its parent, Energy, which supplies electrical energy to Calgarians and others throughout Alberta and contributes the majority of ENMAX’s income, to be a speculative grade CCC bond — essentially a junk bond — that carries with it a risk of default of 78.81%, when its parent, ENMAX, had an A- credit rating which carries with it a risk of default of only 1.69%? To ask this question is to answer it. [para 156]

The thrust of the ACA’s concern was whether it was reasonable for ENMAX to borrow for investment in two subsidiaries and charge an interest rate that exceeded its own cost of funds by as much as 500bps. In other words, a policy driven interpretation of PILOT Regulation would suggest that you consider and measure reasonableness by looking at group financing costs.

The ACA rejected the argument that the question of reasonableness must be determined solely on the basis of federal income tax legislative provisions and related jurisprudence. Rather, the ACA felt able to read that provincial legislative policy into the interest deductibility rules to limit the creation of deductible financing expenses. That legislative policy underlying the Utilities Act and PILOT Regulation was to the effect that one of the purposes “was to establish rules so that an efficient market for electricity based on fair and open competition can develop in which neither the market nor the structure of the Alberta electric industry is distorted by unfair advantages of government owned participants”.

This creates some uncertainty, at least for the municipal sector, interpreting the interaction of tax and financing matters in the context of similar statutory regimes. The tax jurisprudence allowed a range of amounts to be considered to be reasonable and showed some deference to business judgement. If the PILOT regime is not a tax regime, it may no longer be correct to compute such payments as if the municipal entity were “liable to pay” the amounts otherwise computed under the ITA, subject to specifically described adjustments. A more subjective (and uncertain) policy driven approach may be required. In this case, the concept of “reasonableness” was looked at from the perspective of what a fair contribution to the Balancing Pool should be, having regard to the requirements imposed on other public participants.

Paragraph 20(1)(c) ITA requires that the amount of interest payable be reasonable, not just the interest rate. This would seem to open the door for an argument about a domestic thin cap rule, but no evidence was introduced on that topic, or previous legislative proposals involving pension funds. If the ACA is right, and the interest deductibility rules can be applied differently depending on the home jurisdiction of the borrower, does this thinking lead to new results for First Nations, crown corporations and pension funds?

Perhaps that statement should be scaled back in its application to other jurisdictions because establishing ENMAX as a non-taxable corporation is something of an outlier. In Ontario, for example, holding companies are subject to the proxy tax regime, so the type of easy tax arbitrage described in the case is not available. Moreover, Ontario also prescribes the debt equity ratio at an amount not exceeding 70 per cent and requires the borrower to establish a market rate of interest in order to obtain a rate recovery. On the other hand, perhaps the absence of those rather common sense limitations might also suggest that ENMAX had a freer hand in arranging its corporate finance structure than the ACA thought it had.