Heretical Tax Thoughts: Should GILTI Replace Subpart F?

A perplexing attribute of the U.S. tax system is that successive reform efforts tend to layer new regimes on top of old ones without bothering to repeal the latter. It’s as if retention of the older regimes provides a sense of security, regardless of how out-of-date or flawed they might be. Familiarity brings comfort.

Familiarity aside, this layering of tax schemes can be duplicative. It adds complexity to the tax code for both taxpayers and the IRS staff charged with administering the system. It can also create gaming opportunities.

This week’s column indulges in acts of heresy by questioning the continued relevance of our elders — specifically, the almost-60-year-old antideferral rules. Has subpart F been reduced to the clutter of the tax code, or does it carry lasting benefits? My default position is that operating a single anti-base-erosion regime is preferable to running multiple regimes, provided they’re not accomplishing different things.

To that end, we channel the minimalist teachings of Marie Kondo. Kondo is the Japanese lifestyle guru who extols the virtue of decluttering your household, and by extension enhancing the quality of your life. Less is more. (Who would have thought there’s a fortune to be made in telling people to toss out their junk?)

Kondo’s method relies on a type of critical self-assessment: We ask whether each of our physical possessions delivers a sense of tokimeku, which roughly translates to a “spark of joy,” or “the quickening of your heart.” If not, then you must discard the item — but only after sincerely thanking it for having once served a purpose. There’s emotional closure in expressions of gratitude, even when directed at inanimate objects.

So, readers: Does subpart F still provide any sparks of joy?

Is there a legitimate purpose in having an antideferral regime (nominally, just a timing rule) when deferral itself has been repealed as a result of the Tax Cuts and Jobs Act? More to the point, do we still need subpart F when we have the global intangible low-taxed income regime?

How we answer that question could change soon, given that the Biden administration and key members of Congress are contemplating major changes to the GILTI rules. Across Washington, policymakers are asking themselves which of these rival proposals offers the best approach.

Allow me to float the following proposition: The optimal version of GILTI is the one that renders subpart F wholly obsolete, such that we may thank it for its service to the country and promptly retire it. The urge to declutter calls for no less.

Keeping Things Tidy

For the past four years, I’ve been quick to answer the above question in the affirmative. Yes, we still need subpart F in the post-TCJA environment. The sparks of joy might not be as plentiful, but they’re still there.

A current-year income inclusion at the statutory rate (even at 21 percent) is quite different from the outcome you might get under GILTI, with its 50% deduction for qualifying income. That will continue after the GILTI deduction drops to 37.5%, as scheduled, for tax years beginning after 2025. We need subpart F, in part, to police that rate differential.

I sense that many of you will not be convinced this is a compelling interest. Fair enough.

There’s more to it, of course. As income inclusion rules, GILTI and subpart F might be kindred spirits, but they cover different ground. Think of their structure: Subpart F is defined by the categories of income it consists of, while GILTI is defined by exclusion — the category of income it is not.

Also, subpart F primarily concerns passive income, while GILTI won’t be bogged down by the distinction between passive and active income. The fact that active income is fair game under GILTI arguably makes it the superior model for guarding against profit shifting.

What does hamper the reach of GILTI is the exclusion for qualified business asset investment. With the QBAI allowance, there’s a segment of foreign income that avoids income inclusion entirely and remains eligible for tax-free treatment under the dividends received deduction. When you think about it, that’s a remarkable departure from the way the U.S. tax code has operated for the last 100 years, with taxpayers taxed on income “from whatever source derived.” 

QBAI is a statutory class of foreign income that can avoid domestic taxation, absent some backstop mechanism, and the source-country taxation might not amount to much if the host country is a tax haven.

On the one hand, QBAI is what territoriality looks like — and the sales pitch for the TCJA did promise territorial results. On the other hand, nobody said territoriality would be free of guardrails. Having a controlled foreign corporation regime as one of those backstops accomplishes some positive good by protecting against the profit shifting that GILTI allows.

In this context, the modern purpose of subpart F is no longer to control whether a CFC’s earnings are taxed sooner rather than later, but to police whether those foreign earnings shall go untaxed and be eligible for the section 245A participation exemption when distributed to the U.S. group. That’s an important function, despite a taxpayer’s ability to plan around subpart F through the check-the-box regulations.

One way to appreciate subpart F’s importance is to illustrate what can occur when it doesn’t cover a subset of CFC earnings. Consider the high-tax exception of section 954(b)(4). Generally speaking, the GILTI regime imposes tax on U.S. groups based on CFC income that isn’t otherwise taxed under subpart F. That prevents the same dollars from being taxed twice.

An exception provides that for determine GILTI inclusions, CFC income will ignore any amounts that were excluded under the high-tax exception to subpart F. As a result, the group can avoid U.S. tax on such income under both subpart F and GILTI, and the distribution to the group is tax-free under the participation exemption.

That outcome (no taxation in the residence country) might be appropriate if the CFC is genuinely paying a high rate of tax in the source country. That’s the logic on which section 954(b)(4) is based, and it’s consistent with the concept of territoriality. But we’ve seen how the high-tax exception to subpart F can be manipulated by taxpayers.

So, yes, subpart F remains important in a post-GILTI universe — at least for now. That orientation is set to evolve in light of the various changes being discussed. By now, readers will be aware of the discussion draft released in late August by a trio of Democrats on the Senate Finance Committee: Committee Chair Ron Wyden, D-Ore., Mark Warner, D-Va., and Sherrod Brown, D-Ohio. Also, there’s a rival proposal from the other side of Capitol Hill, courtesy of the House Ways and Means Committee.

These proposals are quite different, and we don’t yet know which details will squirm their way into the final legislation. An overlooked detail is that there’s a nontrivial chance the forthcoming changes will significantly affect subpart F. This is apparent from a close reading of the House proposal, which one commentator describes as a prescription for “less subpart F” and “more profit shifting.”

That refers to section 138129(a) of the proposal, which would alter the definition of foreign base company sales income under code section 954(d), and section 138129(b), which would similarly alter the definition of foreign base company services income under code section 954(e).

The proposals would eliminate the so-called branch rule, which concerns situations in which a CFC carries on activities in a third country through a branch or similar establishment, resulting in the branch being treated the same as if it were a subsidiary.

The amended versions of section 954(d) and (e) would narrow the respective categories of subpart F income to transactions in which a U.S. resident taxable unit is positioned as the related person. It follows that income derived from foreign base company sales and service transactions occurring with non-U.S. resident taxable units would fall outside of subpart F.

That sounds like a statutory reversal of the U.S. Tax Court’s decision last year in Whirlpool, which applied the branch rule to foreign base company sales income — defined to include income from manufacturing branch activities.

The Mexican maquiladora sector doesn’t like the government’s interpretation of the branch rule and wants it scaled back. The Tax Court ruled in the government’s favor. The decision is on appeal before the Sixth Circuit.

Less of This, More of That

As things stand today, we need to keep subpart F. But if you were to ask me that question a year from now, there’s no telling how I might answer. If Congress waters down the GILTI rules, I suppose the need for subpart F will become more acute.

Conversely, if the Biden administration gets its way and the GILTI rules are strengthened, perhaps the case for subpart F will become less persuasive.

Today’s heresy is to suggest that the framing of the issue is backwards. Perhaps the repeal of subpart F should be the motivational driver for how we change GILTI. Where does that take us? Instead of less subpart F and more profit shifting, what if it delivered less subpart F and more GILTI?

Recall that the policy objective is to prevent CFC income from escaping U.S. tax under both the GILTI and subpart F, and then being distributed tax-free to the U.S. group. Because GILTI generally excludes CFC income taxed under subpart F, it should (theoretically) be safe to turn off subpart F so long as all the affected CFC income is then sucked up by GILTI. To emphasis the point, that’s not what occurs under section 954(b)(4).

The House proposal wouldn’t fit the bill because it retains QBAI, although the allowance would be halved from 10 percent to 5 percent. If subpart F were no longer around, it figures that QBAI would need to go away entirely.

The Senate proposal wouldn’t work either. It saddles the GILTI regime with its own version of a high-tax kickout — much like the one Treasury gave away last year while it was under different management.

Those regulations seem dubious given the lack of clear support in the statutory language. The Senate proposal would cure the infirmity of the regs by codifying the same result, but that’s the wrong direction if you’re trying to strengthen GILTI.

Directionally speaking, the Biden administration’s green book proposals could be a suitable starting point. They eliminate QBAI, apply GILTI on a per-country basis, and lessen the rate preference.

The White House would do all of that while increasing the statutory corporate rate to 28% — and without saying a word about repealing subpart F, which is not at all surprising. The result would not resemble a territorial regime as we’ve come to know it, and you’d have to wonder what would be left of the participation exemption once the above measures were fully implemented.

I wonder if there’s a grand compromise out there to be had, under which Republicans agree to the envisioned changes in the corporate tax base (doubling down on GILTI to stop profit shifting) and Democrats agree to a general lowering of the corporate tax rate (the opposite of what the Biden administration has proposed).

Probably not, because neither side would have confidence that their gains would prove durable. Subpart F has been systematically weakened over the decades, and there’s no reason to think a similar fate doesn’t await GILTI.

I’d like to hear from readers about what other enhancements are needed to address their concerns about profit shifting in the absence of subpart F. At some point in the future, these musing might not be hypothetical — especially if successive tax bills continue to chip away at subpart F. The prevailing winds already suggest that the combination of more GILTI and less subpart F could be a viable coupling.