Here are two things we know about tax reform, and one thing we don’t. First, a conference bill will be released tonight. Second, it will contain one or more significant limits on the ability of business taxpayers to claim deductions for interest expense. The unknown element is whether the final legislative language will adhere to the House or Senate versions. For the reasons explained below, the details matter.
For as long as anyone can remember, the Internal Revenue Code has given businesses every reason to leverage up. The ability to deduct interest expense stems partly from the matching principle. If lenders are obliged to report interest payments received as taxable income, it seems only fair that borrowers would be able to claim a deduction for the same payments. This treatment also functions as a major boost to the banking sector. Firms are more likely to take on additional debt when they crunch the numbers and realize that their post-tax cost of borrowing is favorable in comparison with the pretax cost.
Code section 163(j) provides some minimal limits on deductibility, but over the years it’s proven relatively toothless. Lawmakers have known for a long time that the rule needs fixing – or replacing altogether – but were hesitant to do it other than as part of fundamental tax reform. Tightening our interest rules outside a broader tax reform effort would translate to a major tax increase for businesses.
The House version of H.R. 1, the Tax Cut and Jobs Act of 2017, would limit net interest expense in two ways: (1) a fixed ratio that applies generally in the domestic context, and (2) a group ratio that applies in the cross-border context and is intended to prevent base erosion.
The fixed ratio is the more straightforward of the two. It can be calculated without regard to other related entities. It provides that net interest expense shall not exceed 30 percent of the taxpayer’s adjusted taxable income. Excess interest could be carried forward for up to five years. For these purposes, adjusted taxable income would be defined as EBITDA, borrowing a term from accounting speak. That’s earnings before interest, taxes, depreciation, and amortization.
It’s useful that lawmakers used a recognized gauge such as EBITDA. This approach is far preferable to dreaming up an entirely new metric, which would only add complexity to a tax system that’s already complicated enough. Note that the restriction applies to “net” interest expense. That means taxpayers would be allowed to use their annual interest income (assuming they have any) to offset their annual interest expense, before they apply the 30 percent restriction. The measure generally comports with the recommendations of action 4 of the OECD’s base erosion and profit-shifting (BEPS) project. That’s ironic, given how the United States has been lukewarm on all things BEPS.
According to official revenue estimates provided by the Joint Committee on Taxation, this provision of the House bill would raise $171 billion over the 10-year budget window from 2018 through 2027.
The Senate bill also contains an equivalent fixed ratio interest restriction, but the calculation is slightly different. It follows the same 30percent rule, but uses an alternate definition of adjusted taxable income that is less generous to taxpayers – earnings before interest and taxes, or EBIT in lieu of EBITDA. The difference between the two concepts may seem merely technical, but it results in very different revenue scores. The Senate version raises $307 billion over 10 years – far more than the House version.
Number crunchers on Wall Street routinely conflate EBIT with a company’s operating income. In contrast, EBITDA is often conflated with cash flow. Neither concept is entirely consistent with generally accepted accounting principles, but both are widely relied on by prospective investors who hope to gain a deeper sense of a company’s value. EBITDA is always greater than EBIT for any given company, meaning the interest restriction in the House bill is necessarily more permissive than the Senate version. (The tougher the ratio, the higher the revenue score.) That explains the higher revenue score on the Senate side. Around Capitol Hill, many people assume the Senate version will prevail over the House version, precisely because it brings in more revenue – needed to pay for unrelated tax breaks elsewhere in the bill.
Here’s the tricky part. As mentioned above, both the House and Senate bills contain a separate interest restriction that takes the form of a group ratio and applies in the international context. This would be contained in a new code section 163(n). Under the House version, a company’s net interest expense would be limited to 110 percent of its proportionate share of worldwide EBITDA, relative to the other companies that make up its international financial reporting group. This rule applies only to companies that belong to groups with annual gross receipts exceeding $100 billion.
Why does such a rule make sense? Think of it this way: Multinationals frequently engage in related-party lending, and they make darn sure the borrower is in the United States – where the interest expense gets more bang for the buck (especially when the statutory rate is 35 percent, as under current law). But who’s to say how much leverage is too much? The allocable share of EBITDA operates as a rough proxy for an allowable amount of debt.
Say the U.S. members, collectively, are responsible for a quarter of the group’s worldwide EBITDA. It looks fishy if those same members account for 90 percent of the group’s interest expense. The extent to which those fractions are out of alignment – that is, there’s a gaping difference between proportional interest expense and proportional EBITDA – is a convenient marker of excess leverage. Are interest expense and EBITDA always in lockstep with each other? No, but the point here is to find a workable benchmark that’s directionally correct and not too subjective. For purposes of protecting against earnings stripping, a group ratio is a suitable cap for restricting interest deductions.
These interest rules apply in the alternate. Taxpayers would be subject to either the basic fixed ratio (the 30 percent rule) or the group ratio, whichever results in the greater disallowance. The rule is silent as to where the taxpayer’s international group is headquartered, meaning domestic-based groups are treated the same as foreign-based groups. General Motors and Ford would have the same treatment as Volkswagen and Toyota.
Again, the rule relies on the concept of net interest expenses, meaning interest income could be used to offset interest expense. Perhaps multinationals will be tempted to restructure their lending so that more interest payments fall to U.S. recipients. As unlikely as that seems, it would boost the amount of interest that could be squeezed into either the fixed ratio or group ratio. That really would be a change from the status quo.
The Senate’s group ratio is conceptually similar, but structured somewhat differently. Rather than use EBITDA as a gauge for permissible leverage, it looks at debt-to-equity comparisons. For example, there would be trouble if the U.S. member’s debt-to-equity ratio were 15 to 1, while the ratio for the worldwide group were only 3 to 1. Again, the extent those figures are askew suggests excess leverage and a suitable interest cap for purposes of tax reform.
One problem here is that the Senate rule is awkward from a feasibility standpoint. Many practitioners have already pointed out that it’s impractical to assume that suitable balance sheets (prepared according to U.S.-recognized GAAP rules) will be available for all the foreign members of the taxpayer’s group. For that reason, the balance sheet test simply won’t work. There’s little point in subjecting taxpayers to a mathematical rule for which they lack the necessary information to plug into the formula.
While I certainly don’t have a crystal ball, my best guess is that the final conference tax reform bill will contain the Senate version of the fixed-ratio restriction (30 percent of EBIT), coupled with the House version of the group-ratio restriction for international groups (110 percent of EBITDA). Combining these two concepts would go a long way toward slowing base erosion, and help pay for the many tax cuts that are being promised.