Bad Debts?

By James J. Tobin, Esq.  

Ernst & Young LLP, New York, NY

*  

There has been lots of press over the last several years about multinational companies building up large cash reserves on their balance sheets. No doubt this is driven in no small part by concern over reliable liquidity in the debt markets based on companies' experiences around the world during the financial crisis. Decreasing reliance on external borrowing seems like a prudent global tax strategy approach as well, in light of the seeming global proliferation of proposals for new anti-abuse type rules targeting debt levels and interest deductibility.

Interestingly, the legislative attacks around the world focus both on limiting the amount of debt at a multinational parent company level and also on limiting the push down of debt to foreign subsidiaries. Thus, if this trend continues, many companies may find themselves challenged in obtaining a single tax deduction for funding costs, much less the multiple "dip" that some may have come to expect.

An example of a potential limitation on parent company debt is included in the recent proposal for a U.S. territorial tax regime released by House Ways and Means Committee Chairman Camp.  The Camp discussion draft includes a provision limiting the deductibility of U.S.-based funding costs based on concern that, in the new territorial tax world in which investments in foreign subsidiaries and branches produce predominantly tax-exempt income, there could be an incentive to focus leverage in the United States in order to deduct interest expense against U.S. profits. The provision is referred to as a thin-cap rule but it really is something of an anti-fat-cap rule in that it effectively discourages the use of "too much" equity in funding foreign subsidiaries.

As I indicated in an earlier commentary, there is a whole lot to like about the Camp proposal overall.1 In my view, it is high time the United States had a territorial tax system like almost all of our counterparts in the developed world. The general approach of the Camp proposal is a 95% dividend and capital gain exemption, rather than a full exemption, with no general expense apportionment/disallowance rules. This is a sensible approach that is consistent with many of the European participation exemption regimes. Indeed, Germany recently switched to a 95% exemption with no general expense disallowance because issues regarding the treatment of expenses under the prior 100% exemption regime were a significant source of complexity and controversy. In the case of the United States, the use of this kind of approach is especially welcome given the U.S. penchant for uneconomic and overzealous expense allocation rules. Consider the repeated proposal by the Obama administration to apply a §861/§864-type approach to allocate interest to the ownership of foreign subsidiaries - an approach which in a territorial tax system would result in almost all situations in a large amount of U.S.-based interest expense being deductible nowhere in the world. Chairman Camp and his staff wisely did not incorporate this particular idea from the administration in the design of their territorial tax regime.

The Camp drafters did, however, feel that there was a need for some enhanced protection from "over-leveraging" at the U.S. parent level. I'm not sure I agree there's a need for any special protection based on cross-border comparisons of debt levels, and would note that several other countries, including the United Kingdom, have not felt the need for such protection.2 But, nonetheless, the Camp proposal does include a thin-cap provision inspired by somewhat similar rules in Germany. In my mind, the provision which in its technical details and resulting effects is a good bit different from the German rules, needs a good bit of refinement.

The Camp proposal provides two alternative limitations in the computation of a group's allowable interest deductions: a comparative U.S.-versus-global debt-equity ratio test and a net interest expense as a percentage of adjusted taxable income test. The latter is based on existing §163(j) rules and refers to the definition of "adjusted taxable income" in §163(j)(6)(A). The percentage limitation for allowable net interest expense is not specified in the proposal and is still being considered. One would hope that given the use of §163(j) principles generally, the 50% limitation in that section will be incorporated. However, the German rules were something of an inspiration for the provision and the similar German percentage is only 30%, so suspicion that the chosen percentage could be closer to that level may be well founded. Whatever the percentage for this provision, there is no specific indication that the 50% threshold for §163(j) would be impacted. Nor do I think it should be, as the policy objectives underlying an earnings-stripping limit with respect to inbound investment seem quite different than the base protection concerns related to outbound investment in a territorial tax system.

The other limitation, the comparative debt-equity ratio test, is a newer concept and will be much more difficult to define and administer. The German rules that inspired this provision include a version of a relative leverage test, but only as a form of safe harbor or, as the Germans refer to it, an escape clause.  Under this test in the Camp proposal, interest expense attributable to excess indebtedness is potentially disallowed. "Excess indebtedness" is defined as the amount of indebtedness of a U.S. corporation (including all U.S. members of the group) in excess of what the debt would have been if the U.S. debt-equity ratio had been equal to the debt-equity ratio of the worldwide group. For purposes of calculating the U.S. debt-equity ratio, any interest in foreign subsidiaries is disregarded.  Thus, the test aims to compare the debt-equity ratio for the U.S. assets of the U.S. corporation to the worldwide group's debt-equity ratio.

Given the centralized treasury management approach of most U.S. multinationals, which is largely driven by the depth and breadth of the U.S. capital markets, this comparative leverage test will likely be failed for the vast majority of U.S. companies.  Which means that most taxpayers and the IRS too will be faced with the complexity of applying this new balance sheet concept, a concept that has limited precedent elsewhere in the Code. (Presumably the test would be based on financial statement numbers as no one computes a global consolidated balance sheet on a U.S. tax basis. So the IRS will need U.S. GAAP, and in the future IFRS, training, as will we all.) Add to this administrative complexity some seeming flaws in the formulation of the comparative test. One particularly stark example3 is that on-lending to CFCs, even in amounts sufficient to equalize the U.S.-versus-foreign leverage, doesn't appear to solve the problem because all "interests" in CFCs are disregarded for purposes of calculating the debt-equity ratios, which presumably includes debt interests as well as equity interests.

To illustrate, assume a U.S. multinational has 100 of U.S. assets, 100 of equity investment in a CFC, and 100 of external debt at the U.S. parent level. Under the test as drafted, the group would have excess debt in the United States of 50, which is computed by limiting the allowable U.S. debt to the worldwide debt-equity ratio of 50% multiplied by U.S. equity of 100.4 If the U.S. parent pushed down 50 of internal debt to the CFC, let's say through a leveraged dividend or return of capital, its "interests" in the CFC would still be 100 (50 of equity and 50 of inter-company debt), such interests would still be disregarded for purposes of the allowable debt calculation, and the amount of excess debt under this test would still be 50, even though the U.S. parent and the CFC now would seem to be equally leveraged.  The amount of interest expense disallowed would be reduced somewhat because the disallowance percentage of 50% would be applied to the U.S. parent's net interest expense, which now would be the interest paid on the 100 of bank debt reduced by the interest received on the 50 of internal debt to its CFC. But it is still an inequitable result.  It appears that the comparative leverage test as drafted would be satisfied only if the external debt levels of the U.S. parent and the CFC are equal. Thus, the debt push down would need to be of external not internal debt. Which is not very realistic from a group treasury standpoint! So bottom line, a problematic proposal that could seriously undercut the potential benefits of a U.S. territorial regime.

But, to be fair, the United States would not be the only country to have special limitations on the interest deductions of a parent company of a worldwide group. As stated above, Germany has its limitation based on 30% of EBITDA (earnings before interest, taxes, depreciation, and amortization). Australia has a thin capitalization limitation that excludes the value of foreign subsidiaries from the limitation formula. And other countries such as the Netherlands apply a tracing concept to disallow interest expense on borrowing that is used to fund equity in foreign subsidiaries, dividends from which are eligible for participation exemption. Therefore, an interest limitation provision of some sort may well be in our future. However, as I have said before, I am an optimist and the optimist in me would note that the territorial tax bill introduced earlier this year by Senator Enzi does not include a thin-cap provision. Still, it is better to be optimistic and proactive rather than just optimistic. Chairman Camp did release his proposal as a discussion draft and the staff is receptive to constructive comments. So now is the time to be discussing all the policy and technical issues that underlie consideration of a thin-cap rule.

With the prospect of limitations on interest deductions in the United States, U.S. companies may be forced to adopt a strategy of "pushing down" some debt to their foreign subsidiaries. What new issues will they be faced with in the foreign jurisdictions when they do so? The answer is a pretty unfriendly local tax environment rife with anti-debt-creation provisions and aggressive tax enforcement activity focused on debt levels, interest rates, and business purpose/GAAR-type attacks. On the legislative front, the past few months have been particularly active. In France, the last Amended Finance Bill for 2011 introduced an anti-avoidance provision to deny deductions for interest expense related to the acquisition by a French company of shares of an affiliated non-French company unless it can be demonstrated that the French group truly exercised direction and control over the non-French company. (Existing French law already limited interest deductions with respect to debt related to the acquisition by a French company of a French affiliate.) Similarly, in March of this year, the Spanish government released proposals which would deny interest deductions on debt used for an intra-group acquisition of shares by a Spanish company unless the taxpayer could demonstrate bonafide commercial reasons for the acquisition by the Spanish company, and would also cap allowable interest deductions at 30% of EBITDA. (The latter part of the proposal is similar to the German rule.)

The most recent anti-debt-push-down provision comes from Canada in a brand new Canadian budget proposal pejoratively referred to as the "foreign affiliate dumping" rule.  This proposal would apply to the acquisition of shares of a non-Canadian company which will be a related group member after the acquisition.  However, the negative tax consequence chosen by the government in this case is not an interest deduction disallowance but rather a deemed dividend of the principal amount of the debt which will be subject to dividend withholding tax. The intent of the provision seems to be an extension of existing section 212.1 of the Canadian Income Tax Act. That provision, similar to §304 of the Internal Revenue Code, creates deemed dividend consequences when one Canadian subsidiary acquires shares of an affiliated Canadian company. Proposed new section 212.3 would expand that rule to any acquisition of a non-Canadian company that is an affiliate after the acquisition. Thus, it would appear to cover: (1) the acquisition of shares of an existing non-Canadian group company in what would likely be a §304-type transaction for U.S. purposes; (2) the subscription for new shares in a non-Canadian affiliate; and (3) even the acquisition of an unrelated non-Canadian target. As in the French and Spanish proposals, there is a business-purpose-type exception for situations where the investment decision and the direction of the target emanate from Canada. But this type of exception would be difficult to rely on with a high degree of confidence in any significant acquisition or restructuring regardless of the level of participation by local Canadian (or French or Spanish, as the case may be) management in the investment decisions. As a practical matter, there typically would always be some significant involvement or oversight from the management of the U.S. parent company as well, which could run afoul of the local direction and control standard.

While the Canadian proposal is somewhat less punitive than the counterpart anti-abuse rules in that there is no direct disallowance of interest deductions, an upfront dividend withholding tax of 5% or more on the principal balance of the loan represents a very significant one-time cost which would negate most of the ongoing tax benefit of the financing costs, particularly in the current low interest rate environment. It is even more troubling when one considers that Canada does not have an earnings and profits concept with respect to dividends that would limit the potential application of the dividend withholding tax to a finite amount of earnings. Thus, in theory, a Canadian subsidiary newly-formed with cash which borrows to acquire a non-Canadian target could be covered under the proposed rule, with the result that the entire principal of the debt is a deemed dividend even though no local Canadian earnings have yet arisen. Plus, dividends paid out of future profits would still be subject to full dividend withholding tax, with no relief or offset for the dividend deemed to have occurred at the time of the acquisition. Oh and by the way, one shouldn't take too much solace in the fact that the new proposal doesn't trigger an interest deduction disallowance as there is another provision in the Budget that proposes to decrease the Canadian thin capitalization limit from 2 to 1 down to 1.5 to 1.

So, all in all, a hostile and turbulent tax environment facing the corporate treasury department. One that bears a very close watch and, given that in some cases grandfathering rules can apply to existing debt, perhaps merits a fresh look at existing global debt levels now before even more change occurs.

This commentary also will appear in the June 2012 issue of the  Tax Management International Journal.  For more information, in the Tax Management Portfolios, see Isenbergh, 900 T.M., Foundations of U.S. International Taxation,  and in Tax Practice Series, see ¶7110, U.S. International Taxation: General Principles.


  The views expressed herein are those of the author and do not necessarily reflect those of Ernst & Young LLP. 

 1 See "Off to Camp?," 41 Tax Mgmt. Int'l J. 96 (2/10/12). 

 2 The United Kingdom does have its new worldwide debt cap rule which limits the allowable debt in the United Kingdom to no more than the worldwide group's total external debt - this rule has always been a little hard for me to understand from a policy standpoint as it applies to inbound investors but it is a very different, and much softer, limitation than the one in the Camp proposal. 

 3 Also pointed out by David Noren in his recent article on the Camp proposal ("The Ways and Means Committee International Tax Reform Discussion Draft: Key Design Issues," 41 Tax Mgmt. Int'l J. 167 (4/13/12)) and by Paul Oosterhuis in his testimony at the hearing on the Camp proposal held by the House Ways and Means Committee's Subcommittee on Select Revenue Measures on Nov. 17, 2011. 

 4 These numbers highlight another potential oddity in the formula. "Equity" is defined as assets minus debt.  In this case, U.S. equity would appear to be zero, if it is computed by disregarding the interests in the CFCs and by taking into account the actual debt rather than the allowable debt being solved for.