Taxation With Realization After Moore

The U.S. Supreme Court’s Moore decision¹ did not produce an all-out taxpayer victory that enshrines realization as a constitutional requirement for an income tax. But it is not the last word. The same organized groups that brought us Moore are likely to try again, and at some point, they may succeed in persuading five justices to constitutionalize realization as a way of preventing a mark-to-market tax on individuals.² Can Congress do something before that to protect the code from being eviscerated?

Next Challenges?

If realization is defined as the receipt of money or property in exchange for an asset held by the taxpayer, there are quite a few provisions of the code that could be challenged as unconstitutional.³ They include subpart F; the global intangible low-taxed income regime; subchapter K; the grantor trust rules; sections 305(c) (deemed dividends), 367(d) (transfers of intangibles), 475 (mark-to-market for securities dealers), 817A (mark-to-market for insurance companies), 877A (exit tax), 884 (branch profits tax), 1256 (mark-to-market for future contracts), 1259 (short-against-the-box transactions), and 1272 (original issue discount); and in some cases, elective provisions like subchapter S and passive foreign investment company tax provisions.⁴

In some of these cases, a good defense can be made on the basis that the tax is an excise tax and not an income tax and therefore not subject to the realization requirement. Those provisions are sections 305(c), 367(d), 475, 817A, 877A, 884, 1256, and 1259.⁵ The whole corporate tax was held to be an excise tax in Flint, and that holding protects subpart F and GILTI, but I am not sure it would survive a constitutional challenge.⁶

The elective provisions are unlikely to be the focus of a constitutional challenge because they are elective, although perhaps provisions that limit the taxpayer’s ability to choose or to reverse an election once made need to be modified.⁷ For example, taxpayers could be enabled to reverse a subchapter S, check-the-box, or section 1296 (mark-to-market for a PFIC) election.

The Fix?

In other cases, the fix is relatively simple because it does not involve actually changing tax results. The grantor trust rules can be fixed this way by making the grantor trust a mandatory disregarded entity so that any realization by the trust is attributed to the grantor.

The two problematic areas that remain are the international provisions and subchapter K.⁸ In both cases, a fix is possible, but it entails a more thorough overhauling of important code provisions. Arguably, however, in both cases, the fix is better than the current state of affairs.

The problem with subpart F and GILTI is that they rely on the deemed dividend concept. As far as I know, no other country that adopted controlled foreign corporation rules after the United States uses deemed dividends; instead, they all look through the CFC and tax the shareholder directly on the relevant income. The deemed dividend concept arose from a now-obsolete concern about imposing tax on the foreign-source income of a CFC, and it causes many other problems.⁹

The fix for subpart F and GILTI is to adopt the corporate alternative minimum tax rule of directly including all CFCs in a consolidated return with the parent. That is a significant change because it eliminates deferral for subpart F, but deferral was already eliminated for GILTI and the corporate AMT. If desired, Congress could impose a lower tax rate on foreign-source income included in a consolidated return or exempt some forms of foreign-source income on competitiveness grounds, although I do not find this argument persuasive.¹⁰ This reform eliminates the realization issue because the income is realized directly by the taxpayer.

The most interesting required fix is to subchapter K. If partners cannot be taxed on distributive shares, then the whole concept behind subchapter K is invalid. One solution is to treat partnerships as aggregates — as they mostly were treated before subchapter K was enacted in 1954 — but that eliminates all the flexibility inherent in the allocation rules and subjects all partners to a mandatory partner’s interest in the partnership regime. That reform may be attractive because the allocation rules are prone to be abused, but there are other aspects of subchapter K that would be drastically modified by an aggregate approach.¹¹

Jeremy Bearer-Friend has proposed a more interesting approach to fixing subchapter K: restricting it to general partnerships by taxing all limited partnerships as well as limited liability companies and other entities enjoying limited liability at the entity level. He writes that:

The US has taxed businesses as separate from their owners since the Civil War, when the income tax was applied to firms in addition to high-income individuals. While this approach continues to this day through the corporate income tax, the corporate tax base has substantially diminished as a share of federal revenue since its peak in the 1950s. Innovations in legal form, including the creation of limited liability companies (LLCs) in the late 70s, have chipped away at the corporate tax base. These new firms granted business owners protections from creditors and litigants while allowing firms to elect out of corporate tax treatment. Today, more than 50 percent of all business income is earned through pass-through entities, including LLCs. As commercial enterprises evolved, tax policy failed to catch up.

There are multiple justifications for entity-level taxes. First, a principle of tax eclecticism allows tax rates to be lower on each specific taxable base by diversifying the risk of tax avoidance across multiple revenue baskets. Second, entity-level taxes serve as a backstop for the individual income tax, preventing wealthy taxpayers from sheltering their income in more tax favorable instruments. Third, the corporate tax also protects democratic values by serving as a regulatory device to counteract the unbridled powers of large businesses. Firms have enormous influence over our democracy and this influence is as true for LLCs as other business enterprises. Lastly, an entity-level tax reflects the benefits of limited liability enjoyed by corporate shareholders and LLC members; the benefits of limited liability should come with a price. Any one of these justifications is sufficient to embrace entity-level taxes on firms.

Critics of taxing limited liability often point to the effect on small businesses and workers, but taxing limited liability does not necessarily require higher effective tax rates on small businesses. First, many pass-through businesses are not actually small businesses. Private-equity firms, for example, are typically structured as pass-through entities and enjoy the tax benefits of pass-through tax treatment. We could also easily lengthen the rate structures for businesses to create different brackets for the genuinely small pass-through businesses that would now be under the umbrella of an entity-level tax.¹² [Internal citations omitted.]

Bearer-Friend then proposes two reforms:

This is a sensible proposal. The United States has a much narrower definition of legal persons subject to entity-level taxation than most other countries. The problem of restricting the corporate tax to publicly traded entities is that it creates a bias against initial public offerings, and perhaps as a result, IPOs have been dwindling. As a Treasury integration draft proposed in 1991, the solution is to broaden the scope of entity taxation.¹⁴

The current regime has been defended as a tax on liquidity, although there is no correlation between liquidity and corporate income.¹⁵ Another defense is that shareholders in nonpublicly traded entities can be taxed directly by looking through the entity, but that requires taxation without realization. Imposing an entity tax on limited liability solves the problem, and it has the added advantage of being defensible as an excise tax.

One objection to the proposal is that it exacerbates the double taxation problem. The Treasury proposal was to prevent this by exempting both dividends and interest distributions to shareholders from income. But Bearer-Friend correctly rejects this option because it means that taxation of rich shareholders is limited to the corporate tax rate, which is typically lower than the shareholders’ individual income tax rate as well as flat (although it could be increased and made progressive, as Bearer-Friend also proposes). The American Law Institute draft rejected exemption as an integration method for the same reason.

Integration?

I do not consider integration to be an important goal of the income tax. Entity-level taxes are about regulating entities, while shareholder/partner taxes are about achieving progressivity. They each have a distinct role and do not have to be integrated.

More technically, the traditional reasons for integration are problematic.¹⁶

Historically, there have been three reasons advanced for countries to adopt corporate/shareholder integration and thus overcome these biases in the classical system. They are:¹⁷

1. to conduct business in noncorporate forms, to avoid double taxation of corporate income (although this is mitigated if the individual rate exceeds the corporate rate, because in corporate form, the individual tax can be deferred);

2. to avoid dividend distributions and instead retain earnings, thus avoiding the double tax (this bias is exacerbated when the individual rate exceeds the corporate rate); and

3. to capitalize corporations with debt (producing deductible interest) rather than equity (producing nondeductible dividends).

None of these arguments is completely convincing, which may be a reason why the United States maintained the classical system from 1936 to 2003, and strengthened it in 1986 with the repeal of the General Utilities doctrine, which enabled corporations to avoid corporate tax on a distribution of appreciated assets.

Alleged Bias Against Corporate Form

The alleged bias against the corporate form is mitigated to the extent the top individual rate exceeds the corporate rate, as it generally did until 2003 and after 2012, and by the absence of strong provisions to prevent retentions in the domestic context. That is particularly true for the current rate structure (the 21 percent corporate rate versus the 37 percent top individual rate).

In addition, under current rules, the classical system applies primarily to large, publicly traded corporations, while small, closely held businesses can avoid double tax even if they are in corporate form for nontax purposes (by choosing to be taxed as S corporations or by incorporating as LLCs, which are treated as passthrough entities for tax purposes). It is doubtful if there is sufficient substitutability between the two forms of business for the double tax to create much deadweight loss (DWL) from the bias toward noncorporate form. Most estimates of DWL from this bias are quite low. For example, Austan Goolsbee found that an increase in the corporate tax rate by 10 percent reduces the corporate share of firms by 5 percent to 10 percent and the corporate share of sales and employment by 2 percent to 6 percent.¹⁸ Goolsbee concluded that “the impact of tax rates is an order of magnitude larger than previous estimates . . . and suggests a larger DWL from corporate taxation, but is still relatively modest.”¹⁹ As Goolsbee says, previous empirical studies found much lower DWLs; arguably, the double tax is a price large businesses have to pay for access to the public equity markets and the liquidity that accompanies that access.²⁰

To the extent that the corporate tax can be shifted to consumers or to labor, the bias disappears, and Treasury, “Integration of the Individual and Corporate Tax Systems: Taxing Business Income Once” (1992) and many others have suggested that considerable shifting can take place. The bias reappears if noncorporate businesses can likewise shift the individual tax burden, but it seems plausible that the shifting potential of large multinationals is larger than that of small, closely held businesses.

Avoid Dividend Distributions

Second, the bias in favor of retentions is reduced when (as in 2003-2017) the individual rate on dividends is not significantly higher than the corporate rate. In addition, this bias was mitigated before 2003 by the ability of corporations to redeem shares from shareholders at the favorable capital gains rate through share repurchases, and by the fact that most shareholders are tax exempt or corporate and thus do not pay a full tax on dividends. Even when the tax rate on dividends is the same as that on capital gains (as it has been since 2003), capital gain transactions may still be preferred for the ability to offset basis and are exempt for foreign shareholders. That is why many U.S. corporations have adopted structured redemption programs. Other corporations retain all their earnings, but it is not clear that this is primarily tax motivated (corporations used to pay dividends under the same rules in the past, and some have recently started doing so).²¹

Finally, there is an unresolved debate among economists whether the dividend tax is capitalized into the price of the shares. If it is, then the retention bias applies only to new equity, but new equity is unlikely to pay dividends for nontax reasons.²²

Bias in Favor of Debt

Third, the bias in favor of debt and against equity is a general problem of the income tax, which should not be addressed only in the corporate tax area.²³ Moreover, even to address it just for corporations, it is necessary to make dividends not exempt, but rather deductible, a form of integration that is never adopted (in part because it would automatically extend integration to foreign and tax-exempt shareholders).²⁴ If integration takes the normal forms of imputation or dividend exemption, there is still a difference in treatment between interest and dividends that can be manipulated. For example, if interest is taxed at the corporate level but dividends are not, clientele effects will still exist (tax-exempt entities will hold bonds and taxable shareholders stock, and taxable investors will purchase the latter because it only applies to a narrow class of corporations earning mostly passive income).

To summarize, it is unclear whether there are significant domestic efficiency gains associated with integration. The presumed gains depend on assumptions regarding the incidence of both the corporate tax and the dividend tax that many economists regard as unproven.²⁵

From an international perspective, both dividend exemption and imputation create additional biases depending on the tax regime in the residence country of the payer corporation. Under the post-2003 regime of partially exempting dividends from both domestic and foreign corporations, there is a bias in favor of investing in domestic corporations to the extent foreign-source countries levy a withholding tax on dividends. In addition, foreign investors in U.S. corporations are disadvantaged compared with U.S. investors, either because of U.S. withholding taxes on dividends or because their country of residence taxes dividends in full (which the United States can do nothing about). Neither of these biases arises in a classical system with foreign tax credits for withholding taxes.

Thus, the Bearer-Friend proposal to expand entity taxation to all entities with limited liability not only solves the realization problem for subchapters K and S but also is a worthwhile reform on its own, without requiring an exemption for dividends or interest.