Realizing Deemed Income From ‘Holey’ New Taxes

Worried about what new taxes are coming? You’re not alone.

The bipartisan infrastructure bill (H.R. 3684) released August 1 confirms that Democrats are saving their tax code proposals for an upcoming year-end reconciliation bill like the Tax Cuts and Jobs Act.

On August 9 Senate Democrats released their long-awaited budget resolution, which includes reconciliation instructions on $3.5 trillion in new spending and tax cuts.

Reconciliation requires creative taxes to satisfy revenue limitations, and the amount of spending desired by Congress is higher than ever before. Provisions for realization upon death or gifts, expansive mark-to-market regimes on publicly traded assets, and a wealth tax could push the boundaries of the constitutional power to tax.

If you’re familiar with the strategies successfully used to realize the TCJA’s reconciliation revenue goals, I’m sure you expect some last-minute surprises. The holes in general tax principles through new taxes will likely grow, and I see no end in sight.

The TCJA paved the way for almost unlimited possibilities on taxes that defy general tax principles that can be enacted through the reconciliation process. It would be naïve to think that similar expansions won’t occur in upcoming legislation.

One gaping question is whether there are any limitations that can help narrow the scope of what taxpayers can expect from new taxes. I believe the limitations are few and that new taxes will stretch historic norms.

The constitutionality of the TCJA’s transition tax is inescapably tied to the constitutionality of recently proposed wealth transfer and mark-to-market taxes. One argument for realization of income under section 965 stems from the corporate U.S. shareholders who transitioned to the benefits of the section 245A territorial deduction. But noncorporate U.S. shareholders didn’t benefit from that transition.

Section 965 is unlikely an unconstitutionally unapportioned direct tax because of its deemed repatriation mechanism. It’s probably not an unconstitutionally retroactive wholly new tax. But its deemed realization approach for generating new revenue while avoiding status as a wholly new tax will assuredly be followed in creating more new taxes that fail to comport with general tax principles.

Before we know it, the holes that these onslaughts of new pay-fors create could be so vast that the thought of having any reasonable expectations for wholly new taxes will be laughable.

I agree with Jasper L. Cummings, Jr., that taxes “are almost always constitutional.” If reconciliation can be used to impose tax on three decades of earnings deemed distributed to individual U.S. shareholders of foreign-majority-owned corporations, decades of appreciation on assets can be taxed upon realization at death.

If a global minimum tax can be imposed to eliminate offshore deferral, President Kennedy’s original goal before Congress chose to focus on what it perceived as tax-motivated abuses, then a yearly mark-to-market regime on derivatives and publicly traded assets can be implemented to eliminate deferral.

Income Redefined

The TCJA’s mandatory repatriation tax flies in the face of the Supreme Court’s three-pronged definition of realized accessions to wealth in its seminal 1955 Glenshaw Glass decision. However, while perhaps the most flagrant, section 965 represents merely a continuation of the congressional trend to create new taxes based on indirect measures of “income.”

Section 965 violates three criteria that have historically been used to identify income. It can impose tax when there is no accession of wealth. This can happen when, for example, the shareholder holds loss shares (because section 965 is based on earnings and profits and section 312(f) does not take into account unrecognized losses).

Second, under section 965 there is no realization, or change, in legal entitlements for individual shareholders who can’t benefit from the section 245A territorial deduction.

Finally, U.S. shareholders may have no dominion or control over any accession to wealth because a U.S. shareholder with loss shares will have no accession to wealth at all; also, section 965’s application to deferred foreign income corporations (DFICs), as opposed to controlled foreign corporations, means that all U.S. shareholders of foreign-controlled DFICs may be in the minority and be unable to compel or control anything.

Section 965 imposes a one-time transition tax on post-1986 untaxed foreign E&P of DFICs owned by U.S. shareholders. While it was enacted as part of subpart F to give the appearance of taxpayer notice and thus evidently address constitutionality concerns, its tax on the E&P of DFICs represents a stark contrast to the CFC regime in which U.S. shareholders were required to own more than 50% of the corporation and thus could generally compel a distribution and were only taxed currently on “bad” (for example, passive) earnings from a single year.

Section 965 is effectively a mark-to-market tax that ignores loss assets on U.S. shareholders of DFICs who would have had no reasonable expectation of taxation under subpart F. As a one-time tax whose impact is in the rear-view mirror for most, its ongoing implications could vastly accelerate the expanding holes in the definition of income.

Intent of Subpart F

Although Kennedy proposed to end tax deferral in all countries in 1961, Congress chose only to eliminate deferral in tax haven countries through the enactment of subpart F in 1962. Congress focused on tax havens because it was believed that they provided inappropriate tax incentives for U.S. businesses to move capital abroad and engage in inappropriate tax planning.

To address tax haven deferral, subpart F was designed to curb tax avoidance goals. Regarding tax haven countries, Kennedy explained that the “undesirability of continuing deferral is underscored where deferral has served as a shelter for tax escape through the unjustifiable use of tax havens such as Switzerland.” 

The legislative history clarifies that the “bill is beneficial in that it will stop the drain on our investment which is artificially induced by the low tax rates it is possible to obtain through the use of tax haven subsidiaries.”

Kennedy’s proposal to end tax deferral in developed countries was not adopted because Congress was focused on perceived tax abuses. Even Kennedy himself acknowledged that many “American investors properly made use of this deferral in the conduct of their foreign investment” and that his unadopted proposal to end deferral in all countries “implies no criticism of the investors who utilize this privilege.”

In light of subpart F’s more narrow focus, there was no certainty that untaxed E&P would ever be taxed because positive E&P and deficits fluctuate. Congress specifically chose to retain deferral generally upon the enactment of subpart F, and it’s unlikely that taxpayers could have reasonably foreseen that the antiavoidance regime would be modified to retroactively tax 30 years’ worth of untaxed earnings.

In fairness, Kennedy’s original proposal to eliminate all deferral could be viewed as an indication that something akin to section 965 was a reasonable possibility and, particularly in light of higher historic tax rates, would have imposed a higher tax burden on most U.S. shareholders.

That said, subpart F focused on CFCs, so it’s hard to apply that logic to the DFIC regime found in section 965, which unlike subpart F requires no foreign corporation to be controlled by U.S. shareholders. The broader applicability of section 965 was likely designed to increase revenue to allow the corporate tax rate to be reduced to 21% while still complying with the revenue limitations of the reconciliation process.

The failure to connect section 965 and DFICs to the subpart F regime seems to represent an understanding that the Court’s threshold for unconstitutionality has been continuously lowered over the years and would tolerate the mandatory repatriation tax (and thus nearly any other major proposal currently on the table for reconciliation legislation likely to be enacted around Thanksgiving).

Thus, while it seems possible for the Court to conclude that section 965 is a wholly new tax that is unconstitutionally retroactive, continually decreasing thresholds for unconstitutionality have made it more likely that the tax will not be found unconstitutional.

And clearly Congress understands that: The taxes it has enacted have strayed further and further from historic tax principles surrounding the concepts of income and realization. Still, section 965 does not embody the abuses that had generally led to these expansions.

Is Unconstitutionality Insurmountable?

Charles Edward Andrew Lincoln IV describes the state of affairs regarding unconstitutionally unapportioned direct taxes as follows:

“Today, a direct tax is on income a person earns or on property, and an indirect tax is on a transaction. Indirect taxes often take the form of sales taxes, such as GST or VAT. However, these ideas were not the definitions as the Pollock case uses them. In the Pollock case, a direct tax is a tax levied on property.”

In Pollock, the Court struck down the 1894 income tax as an unapportioned direct tax because taxing a company’s dividends, royalties, and rents was tantamount to taxing the underlying real estate and personal property itself. Thus, Pollock expanded the view of a direct tax beyond capitation, also known as a poll tax or head tax, and real property, to include personal property.

That expanded view of a direct tax created uncertainty about the ability of the federal government to tax without apportionment among the states. This ultimately led to the 16th Amendment, ratified in 1913, which gave Congress the power “to lay and collect taxes on incomes, from whatever source derived, without apportionment among the several states, and without regard to any census or enumeration.”

Although the logic of Pollock has been questioned, it has never been overruled. Income apparently means whatever Congress thinks, but that was not always true.

Eisner v. Macomber

In Macomber the Court relied on Pollock to hold that a tax on stock dividends was not a tax on income and was therefore unconstitutional as a direct unapportioned tax. The Court explained that income exists within the meaning of the 16th Amendment based on “substance, without regard to form.”

While focused on capital and labor, as opposed to the 16th Amendment’s “whatever source derived” terminology, the Court agreed that mere growth “of value in a capital investment is not income.” (This is what mark-to-market taxes have defied in limited circumstances, to date.) The Court provided that income is essentially “a gain or profit” received by a “taxpayer for his separate use, benefit and disposal.”

Because a stock dividend “adds nothing” and is simply a “mere readjustment of capital,” a tax on one is not a tax on income, and thus to be valid must be apportioned according to state populations.

While many believe the scope of this case has eroded over the years, it was initially taken seriously by Congress in determining what income could constitutionally be imputed to shareholders. In 1921 the Senate explained why the accumulated earnings tax was altered to be imposed on corporations instead of shareholders:

“Section 220 of the existing law provides that if any corporation is formed or availed of for the purpose of evading the surtax upon its stockholders, through the medium of permitting its gain and profits to accumulate instead of being divided, the stockholders shall be taxed in the same manner as partners. By reason of the recent decision of the Supreme Court in the stock-dividend case (Eisner v. Macomber, 252 U.S. 189), considerable doubt exists as to the constitutionality of this provision of existing law. Section 220 of the bill therefore proposes to amend section 220 of the existing law so as to impose upon corporations of the character above described a flat additional income tax of 25 per cent.”

The accumulated earnings tax, originally a 1% tax on shareholders in 1913, continues to be imposed as a penalty tax on corporations. Thus, while shareholders of foreign corporations were eventually taxed directly, although narrowly to appropriately correspond to abuses, Macomber has continued to influence the taxation of corporations.

In its 1961 report regarding Kennedy’s proposal to tax all foreign earnings as opposed to only those in tax havens, the Joint Committee on Taxation explained:

“The present personal holding company statute taxes the undistributed profit of foreign personal holding companies to the shareholders of such companies and this statute has been held valid (Eder v. Commissioner, 138 F.2d 27 (O.A. 2, 1943)). However, the foreign personal holding company statute deals with a relatively clear tax evasion area. Thus, there may be some question as to whether all the provisions proposed would be within the constitutional powers of Congress.”

Thus, while the foreign personal holding company rules were carefully tailored in 1937 to be no more drastic than required, Kennedy’s 1961 proposal was ultimately determined to be overbroad and was therefore limited to abuses.

This, of course, was not the case with the TCJA’s mandatory repatriation tax that required income inclusion for all post-1986 untaxed earnings of DFICs, including for those that didn’t benefit from the new territorial regime.

Glenshaw Glass

I agree with Tony Nitti that the Court’s decision in Glenshaw Glass “provides the most useful definition of income that exists to this day.” Unfortunately, the principles laid out in that case have continuously been ignored by the legislative branch.

While the TCJA may be viewed as delivering the definition of income its final blow, the section 965 mandatory repatriation tax was indeed a continuation of deemed dividend principles that began in the 1930s; although its leap beyond the abuses that justified ignoring Eisner v. Macomber’s requirements for a distribution in substance is huge.

In Glenshaw Glass, the Court established a general principle and corresponding three-part test that the 16th Amendment grants Congress the power to tax any item that (1) increases the wealth of a taxpayer, (2) is realized, and (3) the taxpayer has control over unless specifically exempted by statute. In reaching that definition, the Court looked to the legislative history of the IRC of 1954, which explained that:

Section 61(a) provides that gross income includes “all income from whatever source derived.” This definition is based on the language of the 16th Amendment, and the word “income” is used in its constitutional sense.

While the importance of income and the 16th Amendment remain key to avoiding an unconstitutionally unapportioned direct tax, the concept of income and when it should be taken into account has continuously expanded throughout the IRC since Glenshaw Glass. New taxes have created holes in general tax principles, and limitations on new taxes have dwindled.

Wholly New Taxes

An income tax is excepted from the 16th’s Amendment apportionment requirement. Although income has been broadened to the point that whatever Congress says is an accession to wealth seems to be, a wholly new tax ostensibly has retroactivity limitations. In FSA 200242008, the IRS summarizes the threshold for finding that a tax violates due process:

“Rather, a court must consider the nature of the tax and the circumstances of its application to determine whether retroactive application is so harsh and oppressive that it violates due process. In this respect, retroactive application has been held unconstitutional only in situations in which a statute imposed a wholly new tax that could not reasonably have been anticipated by the taxpayer at the time of the transaction.”

Untermyer and Blodgett involved legislative gift tax provisions. The Court held that when applied to gifts made before the law’s enactment, the provisions were arbitrary and thus violated due process.

The Court held that the gift tax law enacted in 1924 was unconstitutional when it applied to gifts completed before its effective date. But the precedential value of these decisions is in doubt.

Importantly, deemed realization events may dispense with retroactivity concerns. For example, section 965 didn’t adjust prior-year subpart F inclusions to include all E&P but created a current deemed repatriation, which is the likely thrust behind arguments that the tax is not retroactive.

Of course, falling within the definition of a “wholly new tax” is nearly impossible given that creating a new tax a taxpayer could not reasonably anticipate is now a seemingly impossible hurdle after the countless ways over a century of congresses have creatively chosen to tax, penalize, and otherwise discourage behavior through the ever-expanding federal income tax system.

New tax proposals could be viewed as an adjustment to rates or the timing of income inclusion. Even realization at death and the elimination of stepped-up basis under section 1014 could be viewed as merely a timing difference; that is, realization at death triggers a fair market value basis before the property is inherited.

In the context of the retroactive application of federal income tax legislation, the Supreme Court has stated, “In each case it is necessary to consider the nature of the tax and the circumstances in which it is laid before it can be said that its retroactive application is so harsh and oppressive as to transgress the constitutional limitation.”

In Eder, the Second Circuit upheld the constitutionality of the foreign personal holding company rules notwithstanding liquidity concerns. The court recognized that the operation of the statutory rules to the facts at hand “may be harsh,” but also that “the congressional purpose was valid and the method of taxation was a reasonable means to achieve the desired ends.”

Carlton involved an amendment to estate tax deduction rules that applied retroactively. The Court held that the statute didn’t violate due process because the amendment was not illegitimate or arbitrary.

The Court unanimously reversed the Ninth Circuit’s decision and upheld the statutory correction. The majority adopted the following as the standard:

“Provided that the retroactive legislative purpose is supported by a legitimate legislative purpose furthered by rational means, judgments about the wisdom of such legislation remain within the exclusive province of the legislative and executive branches.”

It’s hard to counter Justice Antonin Scalia’s view that Carlton “guarantees that all retroactive tax laws will henceforth be valid.” He went on to explain that to pass constitutional muster, the retroactive aspects of the statute need only be “rationally related to a legitimate legislative purpose” and due process “does not prevent retroactive taxes” but guarantees “only (as it says) process.”

For section 965’s purpose of transition to a quasi-territorial system under section 245A, Hank Adler rhetorically asks:

“Would the same taxpayers who owned minority interests in non-publicly traded (illiquid) CFCs in 1986 have retained that ownership if they understood they would be required to pay taxes on accumulated undistributed E&P, in many cases with no expectation of dividends or the means to require those dividends?”

Generally, a wholly new tax is one that could not reasonably have been anticipated by the taxpayer at the time of the transaction, while other tax changes, such as rate increases, are more properly characterized as amendments to an existing tax.

Given Kennedy’s original desire to eliminate deferral entirely, it seems plausible that the Court would view section 965 as foreseeable. The same could be said for realization at death and mark-to-market taxes that already have grounding in tax law.

Sections 475A, 877A, and 1256, which allow for mark-to-market taxation and ignore traditional principles of realization, are additional examples that have whittled down historic tax principles for the inclusion of income.

New Realizations

If section 965 creates a new deemed repatriation, is it not retroactive, notwithstanding that it taxes three decades’ worth of earnings? One cannot ignore the analogy to the realization that would occur at death under new proposals.

Lawrence A. Zelenak explained that “the income and estate taxes are distinct, both conceptually and practically” and that “there is no reason why appreciation transferred at death should not be subject to both taxes — to the income tax because it is gain, and to the estate tax because it is a gratuitous transfer.” 

The original policy for allowing capital gains to avoid tax at death is about as clear as a lack of realization at death. Would ending section 1014’s lock-in effect be as valuable as the TCJA’s end to the lock-out effect caused by offshore deferral — specifically through the lens of dynamic scoring used for such reconciliation bills?

While congressional attempts to avoid permanent deferral of tax have historically been short lived, a revival of interest in preventing the lock-in effect is once again on the table. Ending the lock-in effect, equity, and, of course, $110 billion of revenue have piqued the interest of congressional Democrats for reconciliation, which, like the TCJA, would most assuredly be accompanied by large expenditures like the corporate tax reduction.

Infrastructure could be supported by proposals like H.R. 2286, introduced by Rep. Bill Pascrell Jr., D-N.J., on March 29, and the Sensible Taxation and Equity Promotion Act of 2021, introduced by Sen. Chris Van Hollen, D-Md., which would tax capital gains at death, with an exemption for the first $1 million of gain.

It’s not clear to me that a whittled down definition of income can be a definition of income at all. Perhaps the good definition of income from Glenshaw Glass is a relic, and realization is rightly viewed as simply for “administrative convenience.” 

Perhaps timing is irrelevant. One could view the choice between cash and accrual method accounting to support that. Calendar and fiscal tax years, too. Retroactive check-the-box electivity even.

If so, that would eliminate two timing prongs from Glenshaw Glass, and whether income is clearly realized by a taxpayer who has dominion and control over it would become irrelevant. That’s the case with section 965, which has no antiabuse underpinnings, ignoring its location in the code, and which was truly a pay-for imposed on all U.S shareholders owning the newly created category of corporation, DFICs.

But if timing doesn’t matter, how can income be defined, and any remaining constitutional limits respected? Timing can eliminate any accession to wealth.

Pre-coronavirus gains and losses were flipped nearly overnight for entire industries. Mark-to-market regimes, like the one in section 965, effectively pull a taxpayer’s choices off the table. Liquidity concerns for illiquid taxpayers with large tax bills will become 8-year payment plans as taxable income arises whenever Congress chooses to spend more or give more to other taxpayers, as it gave to C corporations in the TCJA. 

Section 965 didn’t take into account losses for the U.S. shareholders who paid it, so why would income need to take into account any cost? Any basis for tax seems reasonably foreseeable after the TCJA. Therefore, it seems no transition tax based on “wealth” or “income” can be denied if the TCJA’s isn’t.

By extending subpart F’s antiabuse principles to tax three decades’ worth of any type of earnings through section 965, Congress has shown that its power to tax is absolute, even through reconciliation. As for what additional “holey” new taxes might be included in a reconciliation bill later this year, I believe the possibilities are endless.