Corporate Tax Integration and the TCJA: How Near the Mark?

Based on the combined nominal tax rates, corporate earnings pertaining to shareholders in the two lowest tax brackets, 10%, and 12% bear a nominally higher tax burden (i.e., a lower after-tax amount) than the normative tax burden. The same is true for shareholders in the 32%, 35%, and 37% brackets. In the case of those three higher brackets, however, the nominal excess tax burden is relatively low, ranging from less than one percentage to point to less than four percentage points, and therefore the combined tax burden closely approximates—without accounting for the possibility of deferring the dividend tax by corporate retentions—the Integrationist Norm. It is also worth noting that the nominal excess tax burden is greater in lower tax brackets. For some individual shareholders, those in the 22% bracket and those in the 24% bracket not subject to the 3.8% tax on net investment income, the combined nominal tax rates, even without accounting for the possibility of deferring the dividend tax by corporate retention of earnings, produce a lower tax burden (i.e., a higher after-tax amount) on corporate earnings than the normative tax burden.

Even before addressing the question of effective versus nominal tax rates, a pair of variations needs to be addressed. The first is the qualified business income deduction, also known as the pass-through business deduction, that reduces the tax burden imposed on some business profits earned outside of corporate form. The second is that, under the TCJA, the section 1 ordinary-income tax rates incorporate a sunset provision; without further legislation, the current ordinary-income tax rates under section 1 will expire at the end of 2025.

B. Incorporating the Qualified Business Income Deduction

The second set of comparisons, also for taxable years beginning after 2017 and before 2026, adjusts the section 1 ordinary-income tax rates to account for the qualified business income deduction (the QBI deduction), which allows a 20% deduction for business income earned via a pass-through entity or as a sole proprietorship. As a threshold question, it is far from clear whether the normative single-level tax rates should be stated by adjusting the section 1 ordinary-income tax rates for the QBI deduction. First, the QBI deduction is available to many, but not all types of businesses. Second, the QBI deduction is limited, and therefore potentially less than a full 20% of qualified business income, for taxpayers in a tax bracket higher than 24%. It is difficult to estimate how much less than 20% is the appropriate reduction because the limitation is tied not to the individual’s income but to a fraction of the amount of W-2 wages paid in the qualified trade or business and a fraction of basis of qualified property used in the qualified trade or business.

Nevertheless, arguendo, Table 2 is premised on restating the normative single-level tax rates to account for a full QBI deduction. That is, they are 80% of their nominal counterparts, which in any case understates the normative tax rate for individuals in the 32% and higher tax brackets.

Table 2

The result, adjusting for the QBI deduction, changes significantly. Corporate earnings pertaining to shareholders in all tax brackets bear a nominally higher tax burden (i.e., lower after-tax amount) than the normative tax burden. The likely purpose of the QBI deduction is to allow businesses outside of corporate form to share in the tax cut that reduced the maximum corporate tax rate from 35% to 21%. Importantly, because the QBI deduction is potentially limited for taxpayers in the 32% and higher tax brackets, the after-tax amounts in the second column are potentially somewhat overstated for those three brackets. That raises the possibility that the comparisons of tax burdens could flip in those three brackets. Once again, from the perspective of the Integrationist Norm, it is not at all clear that the QBI deduction should be assimilated in defining the normative single level of taxation. Moreover, bear in mind that this analysis does not yet account for the possibility of deferring the shareholder-level tax.

C. Using Post-2025 Income Tax Rates

Both the current section 1 ordinary-income tax rate schedule and the QBI deduction are slated to sunset at the end of 2025. Table 3 provides a third set of comparisons premised on the post-2025 section 1 ordinary-income tax rates reverting to their pre-TCJA status and reflecting that the new 21% flat-rate corporate tax rate is not scheduled to sunset.

Table 3

Based on the combined nominal tax rates, corporate earnings pertaining to shareholders in all tax brackets will bear a nominally higher tax burden (i.e., lower after-tax amount) than the normative tax burden. Although in the 39.6% tax bracket, the nominal excess burden will be only two-tenths of one percentage point, in the 35% tax bracket it will be only nine-tenths of a percentage point, and in the 33% tax bracket it will be only 2.9 percentage points. For shareholders in the 33% through 39.6% tax brackets, the combined nominal tax rates, without accounting for the possibility of deferring the dividend tax by corporate retention of earnings, produce a combined tax burden on corporate earnings that is remarkably close to the Integrationist Norm.

D. Nominal Combined Rates Conclusion

Without taking the possibility of deferral into account, what conclusions can be reached in terms of achieving the goal of the Integrationist Norm? For most shareholders in most tax brackets, the combined nominal tax rates for corporate tax and dividends remain higher than the normative single-level tax rates. Nevertheless, for shareholders in the highest tax brackets, the nominal excess tax burden is quite small, and the nominal combined tax burden has possibly never been closer to the ideal of the Integrationist Norm. The next question is to address how deferral affects the analysis.

III. The Effect of Deferral

The analysis thus far shows that for most taxpayers in most tax brackets the TCJA rate structure results in a nominal combined tax burden under the double-tax regime that is very near to the Integrationist Norm, and for some shareholders in some tax brackets, it results in a lower nominal combined tax burden than the Integrationist Norm. The analysis thus far, however, is without taking into account the possibility of reducing the effective shareholder-level tax via deferral. The possibility of deferring the shareholder-level tax introduces the possibility of creating, and in some circumstances of enhancing, a “double-tax discount.” Under what circumstances is it possible to use the retention of corporate earnings to generate or enhance this kind of “inside shelter” of a lower combined effective rate via double taxation?

Under the Integrationist Norm, each dollar or unit of business profits is subject to the full individual level tax (p) as realized at the entity level, yielding after-tax earnings of (1 – p). Those earned through a taxable corporation are subject to the corporate tax rate (c) as earned, and a second tax rate (k) when distributed or realized as gain on the sale of shares, yielding after-tax earnings (1 – c)(1 – k). Under what circumstances is the tax burden of the Integrationist Norm’s single-level tax higher than the combined double tax on corporate earnings? Equivalently, what conditions make the following statement true?

(1 – p) < (1 – c)(1 – k)

If c ≥ p, the inequality never holds, but under the terms of the TCJA, that is true only with respect to shareholders in the lowest two rate categories. That is single individuals with taxable income of no more than $38,700 per year and married couples with taxable income of no more than $77,400 per year. For all other shareholders c < p, and in the higher shareholder-tax brackets significantly lower.

Then the critical question is k, the shareholder-level tax on corporate earnings. A policy of retaining earnings generally reduces the effective rate of shareholder-level tax because shareholders generally do not account for corporate earnings until they receive them as dividend distributions, or as gain upon the sale of shares.

As a consequence, it has almost always been true that k < p, even when dividends were taxed at full ordinary-income tax rates. First, the shareholder-level tax can be deferred almost at will. That deferral effectively reduces the shareholder-level tax, which means that the actual effective shareholder-level tax rate is frequently substantially lower than the nominal shareholder tax rate. Second, the disparity is further increased to the extent the retained earnings reflected in share value are not taxed until the disposition of the shares, or avoid shareholder-level income tax entirely by the step-up in basis at death.

This potential disparity is not necessarily a major consideration if corporations are subject to a tax rate very close to the maximum individual tax rate. In such a circumstance, even with substantial shareholder-level deferral, the effective combined tax is likely to be more, and often significantly more, than the single-level tax of the Integrationist Norm. In 2003, when qualified dividend income first became taxable at capital-gain rates, the maximum individual income tax rate on ordinary income was 35%. At the time, the maximum marginal corporate tax rate was 35%, and for corporations with taxable income of between $335,000 and $10,000,000, the corporate tax was effectively a 34% flat tax. Lower effective corporate tax rates applied only to corporations with annual taxable income of less than $335,000.

Under that rate structure there was the real possibility of taxpayers producing a lower effective combined double tax by deferring the second shareholder-level tax via retained earnings. Nevertheless, the disparities between corporate and individual income tax rates were generally relatively small. Therefore, the extent of that form of inside shelter was probably not a large one, and it generally required lengthy deferrals of earnings distributions to produce a combined effective tax less than the single-level Integrationist Norm.

With the TCJA significantly restructuring the corporate tax to a flat tax at a rate substantially lower than the maximum individual tax rate, the possibilities for inside shelter via the deferral of taxation at the shareholder level are significantly enhanced. The 16 percentage-point disparity between the maximum the TCJA section 1 ordinary-income tax rate of 37% and the 21% corporate tax rate means that a substantial fraction of the Integrationist Norm’s single-level tax can be deferred almost indefinitely. Because of that inside shelter, in terms of effective tax rates, the so-called double tax is actually lower in many cases than the Integrationist Norm’s goal of full pass-through taxation equivalence.

Another way to address the same point is to ask how long a deferral of the shareholder-level tax is needed to offset that tax entirely. The answer is not very long. Under ordinary conditions the additional after-tax earnings on the deferred shareholder-level tax more than offsets that tax within a few years. The next question is, under the TCJA regime, how likely are taxpayers to take advantage of corporate form for this purpose?

IV. How Likely?

In the short term, the many businesses that are not already taxable as corporations are unlikely to switch to corporate tax status for the benefit of inside shelter. The TCJA was adopted entirely along partisan lines, and over the next few years, the balance of power may easily shift in the new Congress and the new administration, and they might easily adopt a set of tax rates that would impose a real double-tax burden on corporations.

That would be of little importance if businesses were free to enter and leave corporate status without tax cost, but that is not the case. Opting into corporate status under the check-the-box regulations imposes no tax cost. Under existing law, when a partnership is deemed to have become a corporation, it is treated as if the partnership transfers all of its assets (subject to its liabilities) to a newly formed corporation in exchange for the stock of the corporation, and distributes the stock to its partners in liquidation of their interests in the partnership. In other words, crossing the threshold into corporate status is treated as a tax-deferred incorporation transaction.

Movement in the other direction, however, does impose a tax cost. Under existing law, the conversion of an entity treated as a corporation into an entity treated as a partnership is deemed to be a corporate liquidation, which frequently entails taxation of gain at both the entity and shareholder levels. Given that the advantage of an actual discount from opting into double taxation might easily be gone in the near future, any flood of businesses into corporate tax status in the short term seems unlikely, at least not so long as it is uncertain that the current rate structure is likely to remain in place long enough to justify the possible tax cost of exiting corporate classification later.

In that case, it seems that the new opportunities for inside shelter are most likely to be undertaken primarily by businesses already in corporate form, and those obliged to become corporations for tax purposes because they are publicly traded. From the perspective of the Integrationist Norm, what are the potential concerns from the creation of an incentive to use corporate retained earnings to generate inside shelter in a way that effectively creates a tax discount for retained corporate earnings relative to the normative single-level tax?

V. Allocative Efficiency Questions

The disparity in the tax rates applicable to individuals versus corporations has three potential effects for allocative efficiency. First, it biases the decision whether to distribute corporate earnings. Second, it may create a bias in business entity choice. Third, it may create a bias in the balance of debt versus equity in corporate capitalization.

As already observed, the TCJA tax rate structure creates a bias in favor of deferring shareholder-level taxation via the retention of earnings. Is such a bias problematic? The answer depends on the nontax benefits that might be otherwise derived from an optimal dividend distribution policy.

A variety of possible explanations have been advanced to explain why investors prefer stocks that pay regular dividends, suggesting that a tax bias in favor of earnings retention might have negative effects. A commonly advanced thesis is that because asymmetric information exists concerning a corporation’s profitability, the payment of dividends serves to signal profitability to capital markets. Earnings distributions serve as effective signals because, the less profitable a corporation is, the greater the finance cost it must incur to replace the capital lost through such distributions. A sufficiently profitable corporation has no need to replace the distributed earnings. Hence, the dividend distribution is less costly for profitable enterprises than for unprofitable enterprises.

Another view is that dividend distributions serve to limit the agency costs associated with the ownership-control separation of the “Berle-Means corporation.” As Professor Michael Jensen observed:

Managers have incentives to cause their firms to grow beyond the optimal size. Growth increases managers’ power by increasing the resources under their control. It is also associated with increases in managers’ compensation, because changes in compensation are positively related to the growth in sales. The tendency of firms to reward middle managers through promotion rather than year-to-year bonuses also creates a strong organizational bias toward growth to supply the new positions that such promotion-based reward systems require. . . .

Free cash flow is cash flow in excess of that required to fund all projects that have positive net present values when discounted at the relevant cost of capital. Conflicts of interest between shareholders and managers over payout policies are especially severe when the organization generates substantial free cash flow. The problem is how to motivate managers to disgorge the cash rather than investing it at below the cost of capital or wasting it on organization inefficiencies.

The distribution of such free cash flows may serve as an effective signal of corporate management’s commitment to reduce agency costs.

A third possibility is that shareholders may not view the sale of shares as a perfect substitute for dividends, notwithstanding that finance theory indicates that they should be perfect substitutes. There are various possible explanations for such a phenomenon. Individuals may use strict rules as a means of limiting their own behavior. Therefore, they may refuse to sell any stock for current consumption because they fear that, once begun, they will not be able to limit themselves. They prefer stock with predictable dividends as a means of disciplining their own selection between current consumption and capital accumulation. Alternatively, over some ranges of outcomes, individuals may fail to integrate the component parts of risky activity, viewing the dividend distribution as “safe” income whose distribution is unconnected to changes in the underlying share’s value. Last, individuals may experience a greater degree of “regret” from the forgone possibility of share appreciation attendant with a stock sale than from the same forgone possibility attendant with a failure to reinvest cash dividends.

As a result, ceteris paribus, shareholders require a lower after-tax rate of return with respect to stock that pays regular dividends than for stock that does not pay dividends. As such, a corporation should attempt to match distributions to its optimal dividend payout ratio, at which the value of the corporate enterprise is maximized. If this view is correct, corporations will pay dividends up to a level at which the marginal net nontax benefit of dividend payout is equal to the marginal tax detriment, measured in terms of the excess tax burden on dividends. If the disparity in combined corporate tax rates relative to section 1 ordinary-income tax rates causes corporations to reduce their dividend payout ratios, it increases their cost of capital and results in a real decrease in corporate investment.

There is, however, an important empirical question upon whose answer this analysis will ultimately depend. Do the TCJA’s changes in tax rates actually increase the corporate incentive to avoid dividend distributions? Under the pre-TCJA tax rates, corporations were already subject to an incentive to retain earnings in order to avoid the excess tax burden of the double-tax regime. Under the TCJA rates, the incentive for many corporations is to retain earnings in order to engineer a lower effective tax burden than the Integrationist Norm of the section 1 ordinary-income tax rates. A priori, there is no clear reason to suppose that the one creates a greater distortion in distribution policy than the other.

Next, there is the question of whether the tax structure distorts the choice of business entity. In an environment in which corporate business entities bear an excess burden relative to the Integrationist Norm, there is a real concern about a distortion in the choice of business entity. A business entity formed as a corporation has very limited ability to opt into pass-through taxation, the nearest practicable analog to the pure Integrationist Norm.

On the other hand, in as much as the TCJA tax rates create a bias in favor of corporate taxation, with its opportunity for deferring shareholder taxation, it creates a bias against pass-through taxation. Given the fact that the “check-the-box” regulations make tax classification optional for noncorporate entities, the bias in favor of pass-through taxation creates much less bias in the choice of legal entity. Because so-called “alternate” business entities, such as LLCs and LLPs, are perfectly free to opt into corporate tax treatment, the bias in favor of corporate tax classification creates no necessary bias between one form of legal organization over another. As the bias of the TCJA appears to be primarily one of tax classification rather than actual legal structure, it seems to be of little importance in terms of substantive economic decision-making.

The third question is whether the TCJA’s tax rates affect a corporation’s incentives in regard to its debt versus equity capitalization balance. In as much as the payoff to debtholders is deductible to the corporations and ordinary income to debtholders, the tax burden on debt—viewed as a fraction of operating income—is equivalent to the “Normative After-Tax Amount” stated in the second column of Tables 1 and 3 in Parts I.A and I.C. That is, the payoff to debtholders is subject to the full individual tax rate with no deferral. The payoff to equity is now subject immediately only to the reduced 21% corporate tax rate, and the second level of tax can be deferred almost indefinitely. Therefore, in comparison to the tax burden on equity in corporate entities, the TCJA creates a lower net tax burden on equity than with respect to debt.

In other words, contrary to much of the Code’s recent history, the tax balance is in favor of corporate equity over corporate debt. The primary cause of this is the new reduced corporate tax rate, 21%. It implies that switching from equity to debt capitalization purchases a much smaller tax shield than was the case in an environment in which the maximum corporate tax rate was much nearer to the maximum individual ordinary-income tax rate.

There are well-known economic harms arising from a tax system that biases raising capital via debt rather than equity. Likewise, bias favoring equity rather debt can create its own economic distortions. Leverage can have the beneficial effect of imposing discipline on corporate management and of signaling firm value to the equity market. As such, leverage levels below a firm’s optimal level can themselves be harmful by undermining these beneficial effects of debt capitalization.

Just how much of a distortion is created by that bias is far from clear. First, it is not clear how much of the bias derives from the relative corporate and individual tax rates and how much from the limitation on business interest deductions. Under the TCJA, business interest deductions are limited, in general, to no more than business interest income and 30% of business income. From a corporate tax perspective, the most important benefit of classifying securities as debt is the deductibility (i.e., the tax shield) of the payoff to investors. The TCJA places a strict limit on the availability of that tax shield and, therefore, the tax benefit of debt classification. Moreover, in a period of historically—and artificially induced—low cost of debt, the tax bias in favor of equity capital may well be swamped by the monetary policy in favor of debt. Therefore, whether the tax rate structure has a meaningful effect on the debt-equity balance—all else considered—is far from certain.

In short, the TCJA tax rate changes for businesses may have an effect on allocative efficiency, but it is far from clear whether that is the case. Encouraging the artificial retention of earnings has allocative distortions, but whether that distortion is greater under the TCJA than it was under prior law is unclear. The TCJA’s bias in favor of corporate-form taxation seems unlikely in the check-the-box environment to have a meaningful effect on the actual selection of a business entity’s legal form. Last, whether the TCJA’s tax rate changes impose any meaningful distortion on debt-to-equity ratios is also unclear.

VI. Distributive Equity Questions

The next question is whether the TCJA’s encouragement of inside shelter via the corporate retention of earnings generates problems of distributive equity. To the extent that the schedule of individual income tax rates reflects an implicit normative judgment regarding the best distribution of the tax burden, the inside shelter—like any other shelter that is not universally available—potentially undermines the validity of that implicit, normative judgment. From the perspective of vertical equity, the normative acceptability of the rate schedule is predicated on some set of assumptions about how that rate schedule affects the progressivity of the imposition of the tax burden. The existence of the inside shelter implies that the actual distribution of the tax burden is less progressive than the nominal tax brackets indicate. Because the inside shelter benefits some but not all taxpayers in any given income tax rate bracket, it is not readily susceptible to assimilation into the process of setting those brackets. As such, a valid set of tax distribution assumptions cannot be created by resetting the nominal tax brackets to account for the inside shelter. The entire discussion of distributive equity is further complicated to the extent that the uncertainty about incidence of the corporate tax shifts an indeterminate portion of the inside shelter benefit away from equity investors.

As previously noted, under the pre-TCJA rate structure, the corporate tax rate was very close to the maximum section 1 ordinary-income tax rate. Consequently, shelter had the beneficial effect, in terms of the Integrationist Norm, of mitigating the excess burden of double taxation in many cases and moving the effective tax rates nearer to the normative rates. Under the TCJA rates the picture is significantly different. With the corporate tax rate set so much lower than the maximum individual tax rate, inside shelter creates a real discount in terms of effective rates relative to the normative tax rates of section 1.

At the same time, there are points to be made against eliminating the benefit of inside shelter. A potential objection is that the so-called distortions of inside shelter serve to mitigate the distortions otherwise created by the existing income tax system. This is especially true in comparison to a regime that imposes a lighter burden on capital income, such as a consumption-tax regime. Many serious minds have advanced thoughtful arguments, on both distributive equity and allocative efficiency grounds, that serve to justify advancing toward a consumption-tax paradigm. Further supporting this view are the many economic studies that conclude that an optimal tax system would not include a tax on capital. From an allocative perspective, to the extent that the income tax creates a bias against saving and capital formation, the inside shelter serves to mitigate that bias, and the allocative benefits created thereby might well be greater than the allocative distortions created.

In terms of distributive equity, consumption-tax proponents might see inside shelter as a form of self-help correction to a tax system that excessively burdens savers relative to consumers. A normative view of consumption taxation implies, by definition, that unconsumed accretions to wealth should not yet be taxed. On the other hand, all consumed accretions to wealth should be taxed in the period consumed, even if the wealth is accessed via borrowing.

This presents a conundrum for the existing transactionalist system, which has resigned itself to its inability to conform to either idealized accretionism or idealized consumption taxation. The accretionist paradigm would impose tax on economic income, whose most familiar definition is “the algebraic sum of (1) the market value of the store of rights exercised in consumption and (2) the change in the value of the store of property rights between the beginning and end of the period in question.” Consumption taxation would include only the former in the tax base, but it would include all consumption. Thus, accrued wealth converted to consumption via borrowing must be included in taxable consumption.

Transactionalism allows some accretions to wealth to be consumed without yet being taxed, by borrowing against the appreciated value of assets. In this context, the concern is that the wealth increase accumulated via inside shelter can be consumed without triggering a tax via borrowing against the appreciated value of corporate stock. The result is that the shareholder-level tax on that appreciation is deferred not only relative to an income tax in conformity with the Integrationist Norm, but also relative to an idealized consumption tax. The consumed amount will generally be taxed when the sheltered earnings are distributed or there is a disposition of the shares, but even that might never happen. If the shareholder dies before either of those events happens, the basis of the stock is stepped-up, and the debt can be satisfied by disposing of the stock with no income tax at all on value accrued before the individual’s death. From a distributive equity perspective, a valid objection can be raised that the combination of inside shelter and the ability to consume without taxation is not a move toward the consumption-tax paradigm but a selective exemption from tax.

At the same time, it is worth noting that the Code is already rife with equivalent opportunities to borrow against as-yet untaxed appreciation in assets to fund consumption, and perhaps to do so ultimately tax free via the step-up in basis. The combination of the realization requirement for the taxation of appreciated gain and the nontaxability of borrowing, even nonrecourse borrowing, against appreciated asset value, enables that possibility. Likewise, many taxpayers enjoy tax-deferred wealth accumulation via qualified retirement plans while simultaneously consuming via nontaxable borrowing that effectively allows nontaxable consumption of savings and wealth accretion. How much more the TCJA undermines the distributive equity assumptions implicit in the normative tax rates than was otherwise the case is far from clear. Given that reality, the question of whether the TCJA’s encouragement of inside shelter should be corrected depends on how any correction might be accomplished and whether a particular correction creates more harm than benefit.

VII. Equalize Corporate and Individual Tax Rates

One possible resolution would be to reset the flat rate corporate tax rate to be equal to the maximum section 1 ordinary-income tax rate. This would have the effect of eliminating the possibility of inside shelter. If enacted in isolation, with no change to shareholder-level taxation, the Integrationist Norm would view this proposal as a major retreat from the progress that has been made in the last two decades.

If the corporate tax rate were set to the maximum individual income tax rate, the full normative tax, and more in many cases, would be collected at the corporate level. It would reintroduce all of the distortions of the double-tax regime. The apparently obvious resolution to this problem would be the abolition of shareholder-level tax in its entirety, with no tax on dividend distributions or on share sales. Even that would not fully eliminate the excess burden of double taxation without a refund mechanism for shareholders in tax brackets below the highest section 1 ordinary-income tax rate. The amount of the refund would be set by the excess of the corporate tax rate over their own individual tax rates with respect to their shares of corporate profits. Presumably this would happen at the time of distribution for reasons of administrability.

Even if politically feasible, such a scheme raises objections of its own. Applied literally—that is, exempting from taxation both dividends and gain from all share dispositions—such a program would largely repeal the tax on asset sales. Taxpayers could easily place assets into corporate solution and recognize their full economic value through the tax-free disposition of the holding corporation’s shares.

Of course, no need exists to grant the exemption that liberally. The Code already has mechanisms distinguishing bona fide business corporations from those that hold assets passively. For example, the personal holding company tax imposes a 20% tax (i.e., the maximum dividend tax rate) on undistributed personal holding company income. One characteristic of a personal holding company is that 60% or more of its income must be passive investment income. In a similar vein, the section 1202 exclusion of gain on the disposition of qualified small business stock is predicated on the satisfaction of the “active business requirement,” which mandates that at least 80% of the corporation’s assets (by value) be used by the corporation in the active conduct of a trade or business. These tests could certainly be adapted to serve the purpose under examination here.

In fact, the program of advancing the agenda of the Integrationist Norm by exempting stock dispositions from taxation is already in effect for a significant fraction of U.S. corporations by virtue of section 1202. It applies only to qualified small business corporations, which is limited to domestic corporations holding no more than $50,000,000 of gross assets at the time of the stock’s issuance. The stock must be acquired at original issuance or via tax-free transactions or in exchange for other qualified stock and must be held for more than five years. Sale of stock meeting the necessary requirements receives a 100% gain exclusion if issued in 2010 or later, and a partial exclusion if issued earlier. The suggestion here is to expand the exclusion to cover the disposition of all corporate stock of those corporations conducting bona fide businesses.

If such a program is not politically practicable, as a fallback position, the correction could forego the expanded exemption of share sales from taxation. It could instead be limited to exempting dividends from shareholder-level taxation and allowing a refund of the excess tax collected at the corporate level with respect to distributions to shareholders not subject to the highest section 1 ordinary-income tax rate. Implemented in this fashion, however, it would clearly be a retreat from the progress made thus far toward the Integrationist Norm because (1) excess taxation would continue to exist on a least some stock dispositions and (2) the excess tax imposed on some corporate earnings would not be refunded until the distribution of those earnings.

Moreover, the political feasibility of a proposal to exempt dividends and dispositions of corporate stock is hard to imagine. Democrats are unlikely to support any move to eliminate the shareholder-level tax on dividends, let alone on the sales of shares. At the same time, Republicans are unlikely to vote to raise the corporate tax rate to equal the maximum section 1 rate. The reduction of the corporate tax rate to 21% under the TCJA was a major accomplishment of their policy agenda. They are unlikely to be willing to reverse what they consider to be tax policy progress. In the end, the Integrationist Norm cannot advocate resetting the corporate tax rates that high without a correction at the shareholder level to eliminate or at least mitigate the excess burden of double taxation. Adopting one without the other is worse than doing nothing.

VIII. Undistributed Profits Tax

A possible resolution to the TCJA’s bias in favor of inside shelter is an undistributed profits tax. In such a scheme, a corporation would be subject to a second tax to the extent that its earnings for that year are not distributed that same year. On distribution, the undistributed profits tax would be a refundable tax credit to the receiving shareholders, and the amount of their taxable distributions would be grossed up to include their share of the undistributed profits tax. In effect, the second corporate tax would be an advance withholding tax with respect to the undistributed profits. In the alternative, the refund might go to the distributing corporation, and shareholders simply would be subject to their own normative tax rates on dividends.

The most likely candidate for the tax rate would be the maximum tax rate on the distribution of qualified dividend income, 20%. No shareholder, other than those subject to the tax on net investment income, would be subject to further taxation on the receipt of a dividend distribution. In the case of shareholders who would be subject to the maximum rate, the shareholder tax and the refundable credit would exactly balance each other out. For shareholders in lower brackets, there would be an excess of credit over tax, which would be refunded.

A variation of the tables in Parts II.A and II.C demonstrate the comparison between the normative single-level tax burden and the pre-distribution tax burden of the combined corporate tax and undistributed profits tax. In the comparisons set forth in Tables 4 and 5, the undistributed profits tax (the upt) is assumed to be 20%. The first set of comparisons is premised on the tax rates on ordinary income adopted by the TCJA for taxable years beginning after 2017 and before 2026. The comparisons are set forth in Table 4.

Table 4

A second set of comparisons in Table 5 is premised on the post-2025 section 1 ordinary-income tax rates reverting to their pre-TCJA status, and reflecting that the new 21% flat corporate tax rate is not scheduled to sunset.

Table 5

In both tables, for taxpayers in the maximum tax bracket, the combined tax is still less than the normative single-level tax, although by a very small margin in the TCJA’s pre-2026 rate environment. For all other taxpayers the undistributed profits tax creates a current excess tax burden, relative to the normative single-level tax. This can be largely corrected by the distribution of corporate earnings. The refundable nature of the undistributed profits tax would bring the total tax into line with the combined tax burden in the absence of deferral as demonstrated in the tables in Part II.

At the same time, there are some drawbacks to such a structure. First, in the highest tax bracket, the combined tax burden would still be less than the normative tax burden. In the post-2025 tax environment the undistributed profits tax rate needed to equate to the 39.6% individual tax rate would be a shade higher than 23.54%. This would eliminate the inside shelter benefit in its entirety, but it would further increase the artificial incentive to distribute earnings prematurely.

Moreover, no operating business can conceivably be expected to distribute all of its earnings on a current basis because those earnings are retained to serve the capital needs of the corporation. True, in a perfect market with no transaction costs, the flotation of new securities would substitute for retained earnings. In the world in which those options impose significant costs, however, some significant fraction of earnings will need to be retained.

In terms of effective tax rates, based on the deferral of the refund of the excess of the undistributed profits tax over the normative shareholder-level tax, this option would reintroduce a significant excess tax burden on corporations relative to the single-level tax of the Integrationist Norm. In that sense, it would reintroduce the bias against the corporate form and the bias against equity capital, as well as the distributive equity problems of double taxation that the Integrationist Norm is intended to eliminate.

Further, there is a practical objection to the feasibility of this option. The undistributed profits tax concept has a déjà vu element to it. In 1935, the FDR administration proposed an undistributed profits tax to eliminate the possibility of inside shelter. Corporate management lobbied against the tax on the grounds that the proposal would artificially force excess and premature distributions of earnings. To avoid that consequence, corporate management yielded to end the prior law’s split-rate tax integration in favor of full double tax. A similar reaction would likely result in response to this proposal. Congress made an affirmative policy decision significantly to reduce corporate tax rates via the TCJA. If Congress is unlikely to repeal that action directly, it is just as unlikely to accomplish the same result via an undistributed profits tax.

IX. A Split-Rate Option

Another possibility is a split-rate system in which a portion of the normative tax is collected at each of the corporate and shareholder levels but in which the combined rate is somewhat higher than the single shareholder-level rate of the Integrationist Norm, as a means for compensating for the deferral of the shareholder-level tax via the retention of earnings. In fact, a system of this sort was in use from the War Revenue Act of 1917 until the Revenue Act of 1936.

Beginning with the Revenue Act of 1913, the individual income tax was split into two components: a normal tax of one percent of incomes in excess of an exemption amount of $3,000 for single filers and $4,000 for married filers and a surtax ranging from one percent to six percent on incomes in excess of $20,000. The surtax added a degree of progressivity to the individual income tax. Corporate taxable income (computed without a deduction for dividends paid) was subject to a flat one-percent tax. Dividends paid to individuals, although excluded from the one-percent regular individual tax, were subject to the one- to six-percent surtax. Thus, the corporate tax substituted for the one-percent individual shareholder tax, achieving a rough form of rate integration.

The War Revenue Act of 1917, however, raised the maximum surtax to 50%. Therefore, continuing to tax corporations at a rate equal to the normal individual rate and taxing distributions at only the surtax rate created a substantial preference, in present-value terms, for retained corporate earnings in comparison to earnings retained in a partnership subject to pass-through treatment. In fact, if—as was widely supposed—the surtax rates were to be reduced substantially after World War I, the ability to retain corporate earnings would have turned into a significant partial exemption from the surtax, in comparison to earnings via a partnership subject to pass-through taxation.

A proposed resolution considered by Congress was to impose a 15% tax on retained corporate earnings in excess of an exemption amount of 20% of corporate earnings. The full surtax would have continued to apply to distributed corporate earnings. Instead of moving in this direction of taxing retained earnings, Congress adopted a compromise position in which the combined corporate and individual tax rates would be more than the individual income tax rate on other income. Congress raised the corporate tax by two percentage points above the normal individual tax.

Thus, distributed corporate earnings were subject to a higher combined corporate and individual nominal rate than other forms of income subject only to individual taxation, as compensation for tax deferral via earnings retentions. The rate structure took this form for most of the next two decades, through 1935. The excess of the corporate tax rate over the normal individual tax rate varied from as little as two percentage points to as much as ten percentage points over that time.

A similar approach could be applied to the present situation. The question is how to set the corporate tax rate to compensate for the deferral of the shareholder-level tax. That in turn depends upon what will be the effective shareholder-level tax rate based on the deferral of that tax.

One approach to this problem is as follows. First, set the following equality:

(1 – p) = (1 – c)(1 – k),

where p is the individual income tax rate, c is the corporate tax rate, and k is the effective capital-gains tax rate. Working with the post-2025 maximum personal tax rate of 39.6%, consider corporate tax rates of 25% and 30%, respectively. For the lower corporate tax rate, the target effective shareholder-tax rate, k, is 19.47%. For the higher corporate tax rate, the target effective shareholder-tax rate is 13.71%.

The nominal capital-gains tax rate can be converted into an effective capital-gains rate if the after-tax earnings with respect to the investment of the deferred tax, are equal to the difference between the nominal capital-gains tax and the target effective capital-gains tax, according to the following equation:

E(g – k) = gE(e^rT – 1)(1 – g)

where is E is an amount of after-corporate-tax retained earnings, r is the after-corporate-tax rate of return with respect to those earnings, T is the period in years from the time the earnings accrue until they are taxed in shareholders’ hands at capital-gains rates, and g is the nominal capital-gains rate. Given a maximum capital-gains tax rate of 20%, the maximum deferral period, T’, that still achieves each of the target effective shareholder-tax rates can be determined.

Based on c = 25%, and therefore k = 19.47%, the maximum deferral period that still achieves the target effective shareholder-tax rate is

T’ = 0.0326 ÷ r.

Based on c = 30%, and therefore, k = 13.71%, the maximum deferral period that still achieves the target effective shareholder-tax rate is

T’ = 0.3316 ÷ r.

Table 6 shows the maximum deferral, in years, that still achieves each of the two target effective shareholder-tax rates based on a variety of possible after-corporate tax rates of return.

Table 6

As can be seen from the table, with a corporate tax rate of 25%, and therefore, a target effective shareholder-tax rate of 19.47%, even very short periods of deferral cause the system to fail to achieve the target rate—only 1.63 years, less than 20 months, for r = 2% and only 0.21 years, about 2.5 months, for r = 15%. On the other hand, with a corporate tax rate of 30%, and therefore a target effective shareholder-tax rate of 13.71%, the target can be achieved even with significant deferral periods—up to 16.58 years for r = 2% and 2.21 years for r = 15%.

In terms of political practicability, is it worth the effort to move forward? The table shows that the smaller increase in the corporate tax rate, from 21% to 25% achieves very little because very short deferral periods are sufficient to leave the combined tax burden less than the normative single-level tax. The higher corporate tax rate of 30% can achieve the target effective combined tax burden even with substantial deferral periods, but is it likely to be adopted? Given the prior political balance, that seemed unlikely, but that may change in the new Congress.

Is the resolution worth pursuing? From the perspective of the Integrationist Norm, probably not. Because the computations are based on shareholders in the highest post-2025 tax brackets, its adoption would necessarily overtax earnings attributable to shareholders in lower brackets, as the following table demonstrates. Table 7 shows that the proposal creates a significant excess double-tax burden for shareholders in tax brackets below the highest section 1 tax bracket.

Table 7

Thus, this solution to corporate under-taxation would resurrect many of the problems of the double-tax regime. From the perspective of the Integrationist Norm, a refundable tax credit for the excess tax burden at the corporate level would be necessary at a minimum to correct the error. At that point, it would be just as well to proceed to a full tax integration program.

This exercise in curing the effects of inside shelter began with an observation of the distortions to allocative efficiency and distributive equity. In the absence of a more comprehensive tax integration program, the cure will create its own distortions on both of those fronts. It is a judgment call as to which set of distortions is worse.

X. Concluding Thought

In significant measure, the TCJA has achieved the central goal of the Integrationist Norm but has overshot the mark by making it extraordinarily easy through deferral to convert the ostensible double tax into a tax discount relative to the Integrationist Norm. A comprehensive program of corporate tax integration could bring the Code back on target. Such a program, however, seems less likely now than ever. Given a choice between corporate tax integration and a large cut in corporate tax rates, Congress chose the latter in the TCJA. In such an environment, and since the implications of any proposed corrections to the TCJA overshoot seem as least as problematic as the existing situation, the wisest course may well entail living with the TCJA’s flaws until a more comprehensive corporate tax integration program becomes politically feasible.