It is well understood that corporate capital structure affects tax collections. Most basically, corporate interest expense is deductible. With each interest accrual, the corporate tax base shrinks. Thus, there is a broad range of circumstances in which corporate managers are encouraged by the Internal Revenue Code (the "Code") to load their corporate capital structures with debt. But there is little support for the proposition that Congress desires corporations to adopt such debt-laden capital structures.(1) Indeed, much tax legislation suggests congressional displeasure with the achievable degree of corporate self-integration.(2)

On the other hand, corporate equity has its charms: shareholders are able to defer their gains essentially forever. Thus, in some circumstances the Code encourages corporate managers to load their corporate capital structures with equity. Based on the numerous provisions in the Code that depress the relative tax cost of equity, it is probably safe to conclude that Congress is more sanguine about equity than it is about debt.(3) But periodically, Congress tempers its enthusiasm.(4) And academicians as a group find the feature of equity deferral -- the realization requirement -- quite troubling.(5)

Given the current tax rate structure -- where the marginal tax rate of some persons exceeds the corporate tax rate and the marginal tax rate of others is exceeded by it -- corporations are generally well advised to employ both debt and equity in their capital structures. The former will be held by low tax rate taxpayers and will serve to lower the effective aggregate tax rate(6) on the corporation's taxable income. The latter will be held by high tax rate taxpayers and will serve to keep low the effective aggregate tax rate on the corporation's unrecognized economic income (such as any increase in the value of corporate assets, including goodwill). From the vantage of the Fisc, this is, of course, the worst of all possible worlds.

This Article does not propose to do away with the infirmities of the current corporate tax regime by abolishing double taxation. For while Code [sections] 11 may be the step child of federal income tax theory,(7) there currently appears to be no realistic prospect to repeal it.(8) At least in the case of publicly traded corporations -- the most important class of double-taxed entities -- Americans tend to view them either as a free good, which can be taxed with economic impunity, or as a proxy for the faceless rich, who are undertaxed in any event. Perhaps this will change in time, as the proliferation of 401(K) plans turns the hoards of middle class taxpayers into capitalists. But a change seems to be yet a good way off. And in any event, as I argue below, integration--- at least in its commonly proposed forms -- would not necessarily cure all that ails the current corporate tax system.

Thus, this Article takes double taxation as a given and as a challenge. It asks how, if at all, a double tax regime can be designed so that economic actors are powerless to use capital structure to influence tax collections. The linchpin to the answer, set forth in Part VI below, is that the Code cannot allow any nontrivial corporate deduction with respect to any returns earned by any corporate capital providers. In particular, and merely as one example, the corporate deduction for interest expense must be abolished.(9)

One immediate objection to such a suggestion is that it cannot adequately take into account the special needs of corporate financial intermediaries whose business it is, at least in part, to own highly leveraged interests in other corporations. Without interest deductions, so the argument goes, financial intermediation can not be profitable, and without such intermediation, capital deployment in the economy will become less efficient, with calamitous results. Although I leave to a subsequent article the full exposition of this problem, the short answer is that it need not be so. The question is indeed no different from any other question of consolidation -- that is, the proper treatment of corporate capital instruments held by other corporations -- and has, when so viewed, several relatively straightforward solutions.

A more powerful objection to my suggestion is that, given political reality and public choice, a capital structure neutral tax regime will never be enacted. Thus, the best one is likely to observe is a movement away from current inefficiencies toward future inefficiencies or, more bluntly, the replacement of one set of arbitrary and irrational lines with another. Whenever only a partial solution to an existing problem is possible, the relevant policy question must be whether there is a net gain from implementing that partial solution, taking into account that there will be both gains and losses.(10) I do not dwell on such tradeoffs because the goal of this Article is merely to provide a theoretical characterization of capital structure neutral tax regimes, not an ordering of second best non-capital structure neutral tax regimes.

Part I demonstrates how corporate capital structure can affect tax collections. It defines as "capital structure neutral" a tax regime in which corporate capital structure does not affect tax collections. Part II argues that a capital structure neutral tax regime is desirable. Part III shows that integrated corporate tax regimes are not necessarily capital structure neutral. Part IV gives some examples of possible capital structure neutral tax regimes. Part V develops a theoretical deconstruction of any corporate tax regime. This deconstruction forms the basis for the general description of all capital structure neutral tax regimes. Part VI is the proof that any capital structure neutral tax regime must have a certain form. Part VII describes some implications and, in particular, demonstrates that in a world with multiple interest holder tax rates, a corporate deduction for interest expense is incompatible with capital structure neutrality. Part VIII is a brief conclusion.

PART I -- AN EXAMPLE AND A DEFINITION

Assume that the world is governed by certainty: cash flows and changes in the fair market value of assets are all certain. Reality, of course, differs from this assumption. Nonetheless, the assumption is useful because it will make the examples that follow tractable. Moreover, the points illustrated hold under uncertainty as well. Any tax regime that cannot pass muster under the assumption of certainty does not have a chance in the more complex world of uncertainty.

I focus on a single corporation, X. Economically, X is owned in proportions of equal value by two interest holders, A and B. However, the legal form which A's and B's ownership interests will take is a variable to be determined.

I also focus on only a single period. This can be justified in part in that the collection of income taxes has historically been periodic. But more, it is a nod to the fact that once a tax code permits the tax on economic income to be deferred for even one year, it will generally be powerless to prevent subsequent replications of the deferral. The end result is that the tax, which is ultimately paid in the distant future, will have an arbitrarily low net present value. Thus, for modeling purposes, the tax rate in all periods but the present is assumed to be zero.

The tax rates imposed on X, A, and B are variables. They will be set so as to allow the Fisc to collect $50 of tax directly or indirectly from the X business in a base case. The only arbitrary constraint I impose is that the tax rate on income taxed to A will be four times the tax rate on income taxed to B. This allows A to serve as a proxy for high-income taxpayers, and B as a proxy for low-income or tax-exempt taxpayers.

Finally, I assume that each of A and B invests $1000 in the X business and that, in the year in question, X generates cash flow of $200 and what I shall call "capital-structure-independent" taxable income of $100. In addition, during the year, the X assets actually increase in value by $200, rather than depreciating by $100 as implied by the difference between cash flow and taxable income.

Base Case: Common Equity with Current Payout

Assume X intends to distribute annually all net taxable income (that is, taxable income remaining after payment of taxes). Under current tax principles, X would be taxed on $100 of corporate taxable income. The Fisc would like to tax a certain amount of income "at the source," and so imposes a 33.33% tax rate on corporate income. (This rate is arbitrarily chosen (the first variable to be fixed); a different rate will simply require different shareholder rates.) Thus, the Fisc collects $33.33 of tax. If X indeed distributes the $66.67 of net income -- $33.33 each to A and B -- the Fisc will need to impose additional shareholder-level taxes on A and B at rates of 40% and 10%, respectively, to achieve its stated aggregate tax collection objective of $50.

First Variation: Deferral

One way that A and B can keep at least a part of the Fisc's intended take is simply to have X retain its after-tax income. That saves $16.67 of tax. While this is particularly effective under the instant facts -- where future tax rates drop to zero -- it is in fact an opportunity whenever the effective tax rate on implicit reinvestment in a business, by having such business retain its earnings, is lower than the effective tax rate of explicit reinvestment in the business (where interest holders receive distributions and must affirmatively reinvest them -- after paying tax, of course).(11)

Second Variation: Base Erosion

A and B are no fools. Thus, they each structure their ownership interests in X as partially debt -- say $500 of debt accruing interest at a 10% rate. This generates an aggregate $100 of interest deductions for X, which zeroes out X's income and, of course, its payment to the Fisc. Since X has no taxable income, it...