The United States taxes both corporations and shareholders on corporate profits. In principle, the United States could rely on only one of these taxes, as many commentators have suggested. Although choosing to tax the corporation or its owners may seem like taking money from one pocket or the other, this Essay emphasizes a key dif­ference: These taxes prompt different planning. Relying on one or the other mitigates some distortions and leaks, while exacerbating others. As a result, choosing which to impose is like navigating between Scylla and Charybdis.

In response, this Essay recommends using both taxes for three reasons. First, if one tax is avoided, the other still raises some revenue. Second, if the goal is to deter a planning strategy, cutting the rate to zero is an overreaction. If the rate is low enough, paying a tax is cheaper than avoiding it, since tax planning is not free. Third, if one tax is cut instead of repealed, the other can be correspondingly lower.

Even so, using two taxes poses challenges as well. First, although the taxes are supposed to backstop each other, they cannot do so when a planning strategy avoids both. Second, using two taxes is likely to increase administrative costs. Third, coordinating the taxes to produce the right combined rate—ideally the rate for noncorporate businesses—is not easy.

This Essay also canvasses reforms to shore up both taxes. While the focus is on incremental reform, this Essay’s central recommendation extends to more ambitious reforms as well. These reforms also benefit from using two taxes, instead of one.


    1. Corporate or Shareholder Taxes: Why It Matters
      1. Current Law: Different Tax Burdens and a Different Number of Taxes
      2. Normal Equivalence of Corporate and Shareholder Taxes
      3. Why Nominally Equivalent Taxes Are Different: Component-Rate Strategies
      4. Allocating the Tax Burden Between Corporations and Shareholders
    2. High Corporate Taxes: Planning Strategies and Other Challenges
      1. Shifting Income Abroad
      2. Inversions
      3. Debt
      4. Corporate Lock-In
      5. Comparative Advantage of Corporations over Shareholders in Tax Planning
      6.  Incidence
      7. Tailoring the Tax Burden to Individual Circumstances
    3. High Shareholder Taxes: Planning Strategies and Other Challenges
      1. Shareholder Lock-In
      2. Trapped Earnings
      3. Salary Shifting
      4. Tax-Exempt Shareholders
      5. Foreigner Shareholders
      6. Administrative Costs
      7. Tax Expenditures
    1. Value of Keeping Both Taxes
      1. Built-In Redundancy
      2. Repeal Is an Overreaction
      3. Repeal of One Tax Puts More Pressure on the Other
    2. Challenges in Keeping Both Taxes
      1. Combined-Rate Decisions: Irrelevance of Rate Allocation
      2. Imperfect or Unstable Coordination
      3. Administrative Costs
    1. Three Interrelated Choices for Incremental Reform
      1. The Combined Rate on Corporate Profits
      2. The Balance Between Corporate and Shareholder Rates
      3. Adjustments for Targeted Reforms
    2. Political Constraints on Reform
    3. Incremental Reforms to Ease Distortions from a High Shareholder Rate and a Low Corporate Rate
      1. Tax-Free Appreciation: Charitable Contributions and Stepped-Up Basis
      2. Salary Shifting and Trapped Earnings in Private Firms
      3. Trapped Earnings in Public Firms
      4. Tax-Exempt Shareholders
      5. Foreigner Shareholders
    4. Incremental Reforms to Ease Distortions from a High Corporate Rate
      1. Interest
      2. Income Shifting
      3. Residence Shifting
    1. Comprehensive Business Income Tax
    2. Dual BEIT
    3. Mark-to-Market Accounting or Interest Charges
    4. Imputation
    5. Cash-Flow Corporate Taxation



So we sailed on through the narrow straits, crying aloud for fear of Scylla on the one hand while divine Charybdis sucked the sea in terribly on the other. 1 12 Homer, The Odyssey (A.S. Kline trans., Poetry in Translation 2004) (n.d.) (ebook).

The United States taxes corporate profits twice. Corporations pay one tax, and shareholders pay another when they receive dividends or sell stock. Many commentators have criticized this “double tax,” pro­posing instead to use only one of these taxes. 2 See infra notes 3–6 and accompanying text. For instance, some want to replace the corporate tax with a mark-to-market tax on shareholders. 3 The “realization” rule taxes shareholders only when they sell shares; in contrast, a “mark-to-market” rule taxes shareholders every year on changes in their shares’ value, even if they do not sell. See Eric Toder & Alan D. Viard, Tax Policy Ctr., Major Surgery Needed: A Call for Structural Reform of the U.S. Corporate Income Tax 26–27 (2014) [hereinafter Toder & Viard, Major Surgery Needed],‌publications/major-surgery-needed-call-structural-reform-us-corporate-income-tax/full [] (proposing to replace the corporate tax with a mark-to-market shareholder tax); Eric Toder & Alan D. Viard, Am. Enter. Inst. & Tax Policy Ctr., A Proposal to Reform the Taxation of Corporate Income 2 (2016) [hereinafter Toder & Viard, A Proposal], [] (revising the proposal so corporate tax is retained but at only a 15% rate); see also Joseph M. Dodge, A Combined Mark-to-Market and Pass-Through Corporate–Shareholder Integration Proposal, 50 Tax L. Rev. 265, 266–68 (1995) (proposing to repeal the corporate tax and to replace it with a mark-to-market tax on shareholders of public corporations and a pass-through regime for privately held corporations); Deborah H. Schenk, Complete Integration in a Partial Integration World, 47 Tax L. Rev. 697, 697 (1992) (“I propose that an expanded subchapter S applicable to most nonpublicly held corporations be combined with an accrual tax on publicly held corporations to produce a workable complete integration system.”); David J. Shakow, Taxation Without Realization: A Proposal for Accrual Taxation, 134 U. Pa. L. Rev. 1111, 1134–37 (1986) (proposing mark-to-market taxation for publicly traded stock and repeal of the corporate tax); Victor Thuronyi, The Taxation of Corporate Income—A Proposal for Reform, 2 Am. J. Tax Pol’y 109, 110 (1983) (“The basic thrust of the proposal is to repeal the corporate income tax on publicly held corporations and to tax their shareholders on the annual increase in value of their shares.”).
Professor Michael Knoll would also tax corporate income based on changes in the value of corporate securities, but he would collect this tax from corporations. Michael S. Knoll, An Accretion Corporate Income Tax, 49 Stan. L. Rev. 1, 2–4 (1996). Professor Knoll assumes there will be a separate entity tax, in addition to the personal income tax, but does not defend the use of a second tax. Id. at 5 n.18 (“Of course, this article presumes that there is an entity-level corporate tax. I defend neither the double tax on corporate income nor the income tax . . . .”). Professor Joseph Bankman has also proposed taxing corporations on annual changes in their market value. See Joseph Bankman, A Market-Value Based Corporate Income Tax, 68 Tax Notes 1347, 1348 (1995). Like Professor Knoll, Professor Bankman assumes there will be a separate entity-level tax, but he observes that it “could be easily adapted to an integrated tax system. In that event, the [market value tax] would replace both the current entity-level corporate tax and the shareholder-level tax on corporate distributions.” Id. at 1349.
Other proposals to repeal the corporate tax emerged in the 2016 presidential campaign. 4 For example, former Governor of Louisiana Bobby Jindal proposed to eliminate the corporate tax. Alex Swoyer, Bobby Jindal Unveils Tax Plan, Would Eliminate Corporate Tax, Breitbart (Oct. 7, 2015), []. Senator Rand Paul proposed to replace it with a value-added tax. Rand Paul, Blow Up the Tax Code and Start Over, Wall St. J. (June 17, 2015, 7:09 pm), (on file with the Columbia Law Review) (advocating for a “low, broad-based tax of 14.5% on individuals and businesses”); see also John H. Cochrane, Here’s What Genuine Tax Reform Looks Like, Wall St. J. (Dec. 22, 2015, 7:00 pm), (on file with the Columbia Law Review) (calling for a repeal of the corporate tax). The mirror image is to keep the corporate tax and repeal the shareholder tax. For example, the “comprehensive busi­ness income tax” (CBIT), considered in an influential 1992 U.S. Treasury report, 5 U.S. Dep’t of the Treasury, Integration of the Individual and Corporate Tax Systems, Taxing Business Income Once (1992) [hereinafter Treasury, Integration Study], reprinted in Michael J. Graetz & Alvin C. Warren, Jr., Integration of the U.S. Corporate and Individual Income Taxes 35, 119 (1998) [hereinafter Graetz & Warren, U.S. Corporate and Individual Income Taxes] (analyzing CBIT prototype). In the 2016 campaign, Senator Marco Rubio made a proposal that was somewhat like CBIT: It would repeal investor-level taxes and tax corporations only at the entity level. See Elaine Maag, Roberton Williams, Jeff Rohaly & Jim Nunns, Tax Policy Ctr.: Urban Inst. & Brookings Inst., An Analysis of Marco Rubio’s Tax Plan 1 (Feb. 11, 2016), http://www.‌ [] (analyzing Rubio’s suggested conversion of the federal income tax into a consumption tax by exempting most savings from both personal and business income taxes). would eliminate the dividend tax and possibly also the capital gains tax. 6 In analyzing CBIT, the Treasury did not take a firm position about whether to tax capital gains, noting that “the fundamental problem of capital gains taxation in CBIT is similar to that encountered in other integration prototypes and either resolution (to tax or to exempt capital gains) will be controversial.” Treasury, Integration Study, supra note 5, at 152. The Treasury observed that firms could eliminate capital gains tax on retained earnings with a dividend reinvestment plan. Id. (“If capital gains are taxed under CBIT, corporations might implement a dividend-reinvestment plan . . . to reduce the incidence of double taxation on retained earnings.”).

Although choosing to tax the corporation or its investors may seem like taking money from one pocket or the other, this Essay emphasizes a key difference: Corporate and shareholder taxes prompt different tax planning. Relying on one or the other mitigates some distortions and leaks, while exacerbating others.

At one end of the spectrum, what if Congress repeals the corporate tax and relies only on a (higher) shareholder tax? The good news is that this step eliminates familiar distortions from the corporate tax, such as a firm’s incentive to earn income abroad and to change its tax residence. 7 See infra section I.B.1–.2. But there is bad news as well. Because the shareholder tax is higher, the distortions it causes are more severe. For instance, shareholders have a stronger tax motivation to keep appreciated stock. In addition, tax-exempt and foreign shareholders no longer pay tax (indirectly) through the corporate tax. 8 See infra section I.C.4–.5.

As a result, relying only on the shareholder tax is problematic. To deal with these concerns, Congress can move to the other end of the spectrum, repealing the shareholder tax and relying solely on a (higher) corporate tax. But although this step eliminates distortions from the shareholder tax (such as the incentive to hold onto appreciated stock), it exacerbates distortions from the corporate tax (such as the incentive to shift earnings). 9 See infra section I.B. Since there are distortions either way, choosing which tax to use is like navigating between Scylla and Charybdis.

In response, this Essay recommends using both corporate and shareholder taxes. The two rates should be coordinated so that they aggregate to the combined rate Congress wants, which ideally would be the same as the rate on pass-through businesses (or, at least, close to it). 10 In the United States, “pass-through businesses,” which include partnerships, limited liability companies, and S-corporations, are subjected to only one level of tax. In contrast, “C-corporations” are subjected to two taxes. See infra section I.A.1. The main goal of this Essay, then, is to defend the use of both taxes—in essence, to show “how I learned to love the double tax” 11 Professor Herwig Schlunk uses a similar phrase in offering a very different justification for two taxes: One (the corporate tax) can fund services to businesses, and the other (the shareholder tax) can fund services to individuals. See Herwig J. Schlunk, How I Learned to Stop Worrying and Love Double Taxation, 79 Notre Dame L. Rev. 127, 171–77 (2003). —and to cali­brate the balance between them. At first blush, this argument seems counterintuitive. If both taxes are distortive, isn’t it better to get rid of at least one? It seems odd to contend that “each tax is so flawed that we really need them both.” But in fact, this is the case.

Using both taxes has three important advantages. First, if one is avoided, the other still raises some revenue. Second, if the goal is to deter a planning strategy, cutting the rate to zero is an overreaction. When the rate is low enough, paying a tax is cheaper than avoiding it, since tax planning is not free. For example, a firm’s incentive to shift income abroad can be eliminated not only by repealing the corporate tax, as noted above, but also by cutting the rate to 10% or 15% (so it is competitive with the corporate rate in other countries). Third, if one tax is cut instead of repealed, the other can be correspondingly lower, and thus induces less planning. Proposals to use only one tax on business profits typically neglect these advantages of using two. 12 Professor David Gamage has also emphasized the advantages of using multiple taxes, instead of a single instrument. David Gamage, The Case for Taxing (All of) Labor Income, Consumption, Capital Income, and Wealth, 68 Tax L. Rev. 355, 357 (2014). But he does not focus on corporate tax reform. As this Essay shows, this is a setting where this argument has particular resonance. Instead, Professor Gamage recommends separate taxes on labor income, consumption, capital income, and wealth. Id. at 358.

Even so, using two taxes poses challenges as well. First, although the taxes are supposed to backstop each other, they cannot do so when a planning strategy avoids both. 13 See infra section II.B.1. Second, ensuring that two taxes agg­regate to a particular rate is not easy. 14 See infra section II.B.2. This coordination is fairly blunt unless we shore up both taxes, so they are hard to avoid, and also use a sophisticated mechanism like an imputation system, which gives share­holders a credit for tax already paid by the corporation. 15 See infra section IV.D. Third, using two taxes can increase administrative costs. 16 See infra section II.B.3.

Once Congress chooses the combined rate on corporate profits, how should it allocate this burden between corporate and shareholder taxes? Since lower rates discourage planning, the more distortive tax should be lower. The corporate tax is probably more distortive, so it should be cut significantly. 17 See infra section III.A.2. The shareholder tax should be increased to make up the difference (or at least some of it).

How much of the tax burden should be shifted from the corporation to shareholders? Since the answer depends on how distortive each tax is, the analysis changes when reforms ease distortions in one tax or the other. For example, if a reform makes the shareholder tax harder to avoid, even a high shareholder rate does not motivate taxpayers to avoid the tax (since the reform keeps them from doing so). Once the share­holder tax becomes more reliable in this way, the corporate tax does not need to collect as much revenue, so its rate can be cut. In other words, shoring up the shareholder tax allows more of the tax burden to be shifted from corporations to shareholders. At the same time, the mirror image is also true: If a reform makes the corporate tax harder to avoid, this tax collects more revenue with fewer distortions, so the shareholder tax can be cut. Therefore, both taxes should be strengthened with incre­mental reforms. As each becomes more reliable—and thus collects more revenue with fewer distortions—the other does not have to collect as much revenue. This allows the other tax to use a lower rate, so it becomes less distortive. The bottom line, then, is that reforms ease the tradeoff between collecting tax from corporations, on the one hand, and shareholders, on the other.

In theory, if reforms could eliminate all distortions from one tax—so it could collect as much revenue as is needed without changing taxpayer behavior—the other tax would no longer be necessary. Yet as a practical matter, even fundamental reforms cannot eliminate every distortion; as a result, this Essay shows that these more ambitious reforms also benefit from using two taxes, instead of one. 18 See infra Part IV. For instance, repealing the corporate tax and replacing it with a mark-to-market tax on shareholders offers the significant advantage of eliminating the incentive to shift income. But this reform creates a new distortion: Stock in a public firm is taxed less favorably than investments that do not have to be marked to market. Even so, using two taxes can mitigate this distortion. If a modest corporate tax is retained, the shareholder rate can be lower, making stock more competitive with other investments.

Before proceeding to the analysis, six clarifications are in order. First, this Essay focuses on publicly traded businesses, since private firms usually do not pay corporate tax under current law and use other planning strategies that are mostly beyond this Essay’s scope. 19 For example, high corporate rates can induce controlling shareholders to spend more on corporate expenses with a consumption element, such as expensive health plans, business travel, or company cars. See, e.g., Scott Lynch, 26 Small Business Experts Reveal Their Top Tax Tips for Small Business Owners to Get Bigger Tax Breaks, Direct Capital: Bus. Insights (Dec. 17, 2014), []. In addition, as discussed below, diff­erences between the corporate and personal rates can influence how much cash controlling shareholders leave in the firm, as well as their preference for salary or dividends. See infra sections I.C.2–.3. Second, this Essay generally assumes that income is the tax base, although some consumption-tax alternatives are discussed briefly as well. 20 This assumption does not flow from a conviction that income is a better base, since there are reasons to prefer consumption. See generally Joseph Bankman & David A. Weisbach, The Superiority of an Ideal Consumption Tax over an Ideal Income Tax, 58 Stan. L. Rev. 1413, 1414 (2006). Rather, this Essay focuses on income based on the assumption that corporate tax reform is likely to proceed incrementally. Third, the goal is to maximize national welfare, instead of global welfare. 21 For a discussion of the relative merits of maximizing global or national welfare in designing international tax rules, see Daniel N. Shaviro, Fixing U.S. International Taxation 15 (2014). Fourth, this Essay does not focus on distribution, since the alternatives it compares—corporate and shareholder taxes—both tax capital income, and thus reach high-income taxpayers. Even so, distributional effects can still vary if the incidence of these taxes is different. 22 The concept of tax incidence describes who bears the economic burden of a tax. In some cases, those who pay a tax—in the sense that they write the check—are able to shift this cost to others and thus do not really bear it. For a discussion of this concept, see infra section I.B.6. Fifth, this Essay focuses on where rates and rules should end up. How the transition should be managed—including whether changes apply only to new firms or new equity (e.g., to avoid windfalls), as well as whether changes should be phased in gradually—is beyond this Essay’s scope. Finally, the focus here is on policy instead of politics, although political constraints are discus­sed briefly. 23 See infra section III.B.

This Essay proceeds in four Parts. Part I explains why nominally equivalent corporate and shareholder taxes induce different planning, and it surveys strategies used to avoid a high corporate tax, on the one hand, and a high shareholder tax, on the other. In response to these dueling distortions, Part II recommends using both taxes. Part III considers what the division of labor should be between these two taxes, as well as how to strengthen each with targeted reforms. Part IV extends the analysis to more comprehensive reforms, showing that they also benefit from using two taxes, instead of one.

I. Dueling Distortions: Some Arise Only in Taxing Corporations, While Others Arise Only in Taxing Shareholders

Should tax on businesses be collected from the business itself, its investors, or both? This choice would be unimportant—like taking money from one pocket or the other—if corporate and shareholder taxes were interchangeable. But in fact, these taxes measure income differently. Some planning strategies avoid one tax but not the other. As a result, the choice to tax corporations or shareholders influences which distortions arise, as well as how much revenue is raised.

In principle, there are two ways to deter these planning strategies. One is to tighten up the relevant rules, so the strategies would no longer be effective at reducing the tax bill. Parts III and IV offer examples of how to reform the relevant rules. But unfortunately, these reforms face familiar administrative and political barriers, which have impeded reform so far and may well continue to do so.

Alternatively, a second way to deter planning strategies is to cut (or even repeal) the relevant tax. After all, avoiding a 15% tax offers less tax savings—and thus justifies less effort and expense—than avoiding a 40% tax.

Yet even though cutting one tax eases the distortions it causes, this progress comes at a cost: If Congress makes up the revenue by increasing the other tax, the distortions from this tax become more severe. In other words, changing the mix of rates solves one set of problems but exa­cerbates another. For instance, a low (or 0%) corporate rate reduces the appeal of strategies to avoid the corporate tax, which are described in section I.B. Yet if the cut in the corporate tax is funded with an increase in the shareholder tax—so, for instance, the corporate tax is 0% and the shareholder tax is 40%—taxpayers would have a correspondingly stronger incentive to avoid shareholder taxes. A key point, moreover, is that they would use different strategies to do so. As a result, the system would confront another set of distortions and challenges, which are canvassed in section I.C.

Because of these dueling distortions, collecting tax from either corporations or shareholders is quite difficult, especially when the rate is high. As a result, relying on only one of these taxes is problematic. Part II shows why it is better to use both taxes, instead of only one. But first, this Part surveys the challenges with each tax to show why it matters whether tax is collected from corporations or shareholders.

A. Corporate or Shareholder Taxes: Why It Matters

The current rules for taxing businesses cause familiar distortions. Corporations are not taxed the same as pass-through entities, and debt is taxed differently than equity. 24 See Alvin C. Warren, Jr., Am. Law Inst., Integration of Individual and Corporate Income Taxes (1993) [hereinafter Warren, ALI Integration Report], reprinted in Graetz & Warren, U.S. Corporate and Individual Income Taxes, supra note 5, at 618–22 (noting that differences in the treatment of corporations and pass-through entities, as well as differences in the treatment of corporate debt and equity, breed familiar distortions). Not only are the rates different, but the number of taxes is also not the same. In some cases, only the investor is taxed; in others, the entity also pays tax. At first blush, who is billed for a tax seems much less important than the size of this bill. But in fact, the choice to tax corporations or shareholders matters a great deal, since these taxes define income differently. As a result, different planning strategies are used, so the distortions and revenue are not the same.

1. Current Law: Different Tax Burdens and a Different Number of Taxes. — Under current law, two familiar variables influence how large a business’s tax burden is, as well as whether the business or its investors write the check. The first variable is whether the business is taxed as a corporation or a pass-through entity. Businesses taxed as corporations are called “C-corporations.” They are subject to a heavier tax burden, and tax is collected from both the firm and its owners. In the top bracket under current law, the firm pays a 35% corporate tax, 25 I.R.C. § 11(b)(1)(D) (2012) (establishing a 35% tax rate for income above $10 million). and shareholders pay a 23.8% tax when they receive a dividend or sell stock. 26 The stated rate is 20%. Id. § 1(h)(1)(D). There also is a 3.8% surtax on net investment income. Id. § 1411(a)(1).

In contrast, pass-through entities are subject to a lower tax burden and are taxed only once. The pass-through entity itself pays no tax. Instead, its owners pay the only tax, using the “personal” rate that applies to salaries, which currently is 39.6% in the top bracket. 27 For example, assume a limited liability company (LLC) that is wholly owned by an individual earns $100. The LLC pays no tax, and the owner pays $39.60 of tax. See id. § 1 (establishing a 39.6% rate for ordinary income in the top bracket). As a result, the combined rate for C-corporations (50.5%) 28 For example, if a corporation that is wholly owned by one shareholder earns $100, the corporation pays $35 of corporate tax. If the corporation pays a $65 dividend, its shareholder pays a 20% capital gains tax of $13 and a 3.8% surtax on net investment income of $2.47 (assuming the shareholder is in the top bracket). See id. § 1411. As a result, $49.53 is left after $50.47 is paid in taxes. is higher than the rate for pass-through entities (39.6%).

A second variable also affects a business’s tax treatment: whether the firm is capitalized with debt or equity. Under current law, the extra tax burden on C-corporations, described above, arises only for equity and not for debt. When a C-corporation makes interest payments, the rev­enue funding these payments is taxed only once; the corporation is not taxed, since it can deduct interest payments. 29 Id. § 163(a) (authorizing a deduction for an interest expense). Instead, only investors are taxed (i.e., on the interest payments they receive). 30 Id. § 61(a)(4) (including interest income in gross income). These lenders are taxed at their personal rate, which is 39.6% in the top bracket. 31 For example, assume a corporation borrows $2,000 from a lender, paying 5% interest of $100 per year. If the corporation uses this capital to generate $100 of revenue, the corporation can deduct the $100 of interest from $100 of revenue and therefore has no taxable income (and nothing to distribute as a dividend to shareholders). The lender has $100 of interest income and pays $39.60 of tax (assuming the lender is in the top bracket). So under current law, corporate equity bears a higher tax burden (50.5%) than corporate debt (39.6%).

Notably, the same tax burden (nominally) applies to both pass-through equity and corporate debt (39.6%), and each is subject to only one tax. This tax is collected from owners in the case of pass-through entities and from lenders in the case of debt owed by C-corporations.

2. Nominal Equivalence of Corporate and Shareholder Taxes. — Of the two differences highlighted above—the rate and the number of taxes—the first initially seems more important. The difference between the 50.5% and 39.6% rates obviously matters. But the choice to collect tax from the firm or its shareholders appears to be less significant, since these taxpayers are related. After all, if a taxpayer has two bank accounts, does it matter if tax is collected from one or the other? Similarly, why should it matter if tax is collected from the owner’s bank account, instead of the business’s bank account?

If this initial impression is correct—so rates are important, but the choice to tax the business or its owners is not—the right solution is to conform the rates. This step, by itself, should eliminate the relevant distortions.

For example, assume Congress wants to tax all business profits at 40% and also uses 40% as the personal rate. (This Essay uses 40% as a recurring example, but this round number is merely illustrative and is not intended as a recommendation.) 32 For a discussion of factors relevant to the choice of the combined rate, see infra section III.A.1. In principle, a 40% combined rate can be implemented with a 40% corporate rate and no shareholder tax, 33 A total of $40 of corporate tax is imposed on $100 of earnings. The remaining $60 can be distributed tax-free to the investor. a 40% shareholder tax and no corporate tax, 34 If the corporation earns $100, it pays no entity-level tax. The shareholder pays a $40 tax, leaving $60 after taxes. or a 22.5% rate for both. 35 The right percentage is 22.5%, instead of 20%, since one tax is imposed on what is left over after the other has been paid. For instance, if $100 of corporate earnings is taxed at 22.5%, and the remaining $77.50 is distributed, the shareholder pays tax on the $77.50 and not on the original $100. Therefore, the shareholder has $60 after paying a 22.5% tax of $17.50. Indeed, any of the pairs of rates in the following table seem to raise the same revenue and affect taxpayers in similar ways: 36 More generally, the pass-through tax (P) equals the combined corporate (C) and shareholder tax (S) when: (1 – P) = (1 – C) (1 – S).

Table 1: Nominally Equivalent Pairs of Corporate and Shareholder Tax Rates: Combined Tax of 40%

Personal Rate Corporate Rate Shareholder Rate Corporation’s After-Tax Profit on $100 of Earnings Dividend Tax if Corporation Distributes After-Tax Profit Shareholder’s After-Tax Profit
40.00% 35.00% 7.80% $65.00 $5.00 $60.00
40.00% 29.50% 15.00% $70.50 $10.50 $60.00
40.00% 27.00% 18.00% $73.00 $13.00 $60.00
40.00% 25.00% 20.00% $75.00 $15.00 $60.00
40.00% 20.00% 25.00% $80.00 $20.00 $60.00
40.00% 18.00% 27.00% $82.00 $22.00 $60.00
40.00% 15.00% 29.50% $85.00 $25.00 $60.00
40.00% 7.80% 35.00% $92.20 $32.20 $60.00


Since these pairs of rates all yield the same combined 40% rate, they all initially seem equivalent, and thus seem equally able to eliminate the inconsistencies described above. All of these pairs seem to align the 40% personal rate with a 40% combined tax on corporate equity. 37 Cf. George K. Yin, Achieving Integration Through Double Taxation, 56 Tax Notes 1365, 1367 (1992) (proposing an approach that would ask “whether the total income tax burden of a particular investor on corporate­-source earnings, taking into account both the shareholder and corporate taxes, can be equated with the burden of that investor on the same amount of noncorporate-source earnings”). In so do­ing, each rate pair appears to conform the tax on corporate equity with the tax on pass-through equity, as well as the tax on debt.

3. Why Nominally Equivalent Taxes Are Different: Component-Rate Strategies. — But in fact, conforming rates is not enough. Although a 40% corporate rate might seem the same as a 40% shareholder rate, this apparent equivalence is misleading. Even when rates are the same, the tax treatment can still be different, since rates alone do not determine the tax burden. Rather, the definition of income is another key factor, and corporate and shareholder taxes use different definitions under current law. As a result, these taxes have different gaps, so a 40% cor­porate tax does not have the same effects or collect the same revenue as a 40% shareholder tax. In principle, Congress could conform these def­initions of income, but this would require substantial changes in the relevant rules.

Indeed, there are five well-understood differences in these regimes. First, the corporate tax measures a firm’s earnings, while the shareholder tax measures a shareholder’s dividends and capital gains. In the long run, these metrics may well yield the same amount of income, since stock prices—and thus an investor’s capital gains—are supposed to be a forecast of the firm’s future earnings. 38 See Knoll, supra note 3, at 2–3 (“In an efficient market, . . . the change in the aggregate market value of the corporation’s shares will equal the change in the corporation’s expected future returns, or its periodic income.”). But in the short run, these two measures can diverge.

Second, and relatedly, corporate and shareholder taxes do not come due at the same time, since some steps trigger one tax but not the other. For example, assume a shareholder buys stock in a corporation, which then rises in value because the firm buys an asset that appreciates. Sale of the asset triggers corporate tax, while sale of the stock triggers share­holder tax. Taking one of these steps triggers only one tax, not the other. 39 Professor David Weisbach has described this as the “dual ownership” problem and has argued that it is a source of “irreducible complexity” in the corporate tax. David A. Weisbach, The Irreducible Complexity of Firm-Level Income Taxes: Theory and Doctrine in the Corporate Tax, 60 Tax L. Rev. 215, 215 (2007).

Third, these taxes have different geographic boundaries. As a result, the shareholder tax sometimes applies when the corporate tax does not, and vice versa. For instance, assume a foreign corporation earns money abroad and distributes it to shareholders who are U.S. citizens. The corporation does not pay tax on this profit (since foreign corporations pay tax only on income earned in the United States), 40 I.R.C. §§ 881–882 (2012) (taxing foreign corporations on income from U.S. sources). but U.S. share­holders do pay tax on these dividends (since dividends of U.S. share­holders are taxed, regardless of where the firm earned the profits). 41 Id. § 61(a)(7); see also Reuven S. Avi-Yonah, The Structure of International Taxation: A Proposal for Simplification, 74 Tex. L. Rev. 1301, 1317 (1996) (“Domestic taxpayers are taxed by the United States because of their personal connection to the United States, that is, on the basis of residence; . . . [f]oreign taxpayers are taxed . . . on the basis of their territorial connection to the United States, that is, on the basis of source.”). In other words, this profit is taxed at the shareholder level, but not the corporate level. 42 The opposite can also be true. For instance, assume a foreign corporation earns money in the United States, and this profit increases the value of stock held by foreign shareholders. The firm has to pay U.S. corporate tax (since the firm earned the income in the United States), see I.R.C. §§ 881–882, but foreign shareholders do not owe any U.S. tax (since foreigners do not have to pay U.S. capital gains tax). Id. §§ 881, 871(a)(2) (taxing foreigners’ capital gains only if they satisfy the residence requirement). So in this case, the U.S. government taxes the profit at the corporate level, but not at the shareholder level.

A fourth difference arises when Congress pursues goals other than raising revenue. For some tax expenditures, 43 A “tax expenditure” is a tax rule that advances policy goals other than raising revenue. For instance, the deduction for mortgage interest promotes home ownership. See generally David M. Schizer, Limiting Tax Expenditures, 68 Tax L. Rev. 275, 284–85 (2015) (noting that tax expenditures fund a wide range of activities, such as charity, clean energy, and education). Congress uses the shareholder tax, but not the corporate tax. For other tax expenditures, the opposite is true. For example, Congress enacted the production tax credit to promote green energy. 44 I.R.C. § 45(a). If a corporation produces electricity with wind turbines, this credit reduces its corporate tax bill. But if the firm distributes profits from wind energy to shareholders, the dividend tax still applies. 45 See generally Michael J. Graetz & Alvin C. Warren, Jr., Integration of Corporate and Shareholder Taxes, 69 Nat’l Tax J. (forthcoming 2016) [hereinafter Graetz & Warren, Integration of Corporate and Shareholder Taxes] (manuscript at 14),‌abstract=2780490 (on file with the Columbia Law Review) (noting that the current law avoids passing through corporate tax preferences to shareholders).

Finally, another difference between corporate and shareholder taxes is that taxpayers face different nontax costs (or “frictions”) in avoiding them. For example, to reduce their corporate tax bills, firms can draw on sophisticated legal and financial expertise. In contrast, shareholders who want to avoid the shareholder tax may not have access to the same expertise.

Because corporate and shareholder taxes do not have the same gaps, some planning strategies reduce one tax, but not the other. To describe strategies that avoid only one of these taxes, this Essay uses the term “component-rate” strategies. For instance, holding on to appreciated stock defers only the shareholder tax. As a result, this strategy affects only a component of the overall tax on corporate profits (i.e., the shareholder tax, but not the corporate tax).

In contrast, “combined-rate” strategies avoid both corporate and shareholder taxes. For example, organizing a venture as a nonprofit avoids both taxes. 46 See I.R.C. § 501(a) (exempting certain organizations from tax). The same is true of organizing a business as a pass-through entity; the personal rate applies, instead of the corporate or shareholder rate.

4. Allocating the Tax Burden Between Corporations and Shareholders. — To summarize, taxpayers respond differently to corporate and share­holder taxes. So even when these taxes are coordinated to yield a particular combined rate, taxpayers use different planning strategies, depending on whether tax is collected from corporations or share­holders. Yet although the allocation between these taxes is very imp­ortant for some strategies, it is less important for others. Specifically, this allocation matters more for component-rate strategies (which avoid only one tax) than for combined-rate strategies (which avoid both).

a. Effect of the Allocation on Component-Rate Strategies. — For comp­onent-rate strategies, the mix of corporate and shareholder taxes is important for two reasons. First, it influences how much revenue the government loses from these strategies. For instance, holding stock until death avoids the shareholder tax but not the corporate tax. 47 Id. § 1014. Therefore, revenue falls to zero if the shareholder tax is 40% (and the corporate tax is 0%), but it is unaffected if the shareholder rate is 0% (and the corporate rate is 40%).

Second, this allocation also affects whether component-rate strat­egies are cost effective for taxpayers. After all, tax planning is not free. Taxpayers have to pay advisors and modify their behavior. These nontax costs are justified only when they are less than the tax savings. The break-even point varies among taxpayers, since a strategy can be easier for some than others. But in general, some planning is cost effective for avoiding a high tax, but not a low tax.

For instance, if the corporate rate is 40% and the shareholder rate is 0%, taxpayers abandon even the cheapest ways to avoid shareholder tax. But if the shareholder rate is increased to 22.5%, some strategies become cost effective. 48 For instance, hedging appreciated stock with derivatives is more plausible to avoid a 30% tax than a 10% tax. For a discussion of hedging stock, see David M. Schizer, Frictions as a Constraint on Tax Planning, 101 Colum. L. Rev. 1312, 1339–59 (2001) [hereinafter Schizer, Frictions]. If the rate is increased further, additional strategies become appealing. 49 Once the shareholder rate is high enough, most strategies are already in use, so further increases induce relatively little additional planning. I am grateful to Professor Tom Brennan for this observation. At the same time, if the corporate rate is cut—to keep the combined rate at 40%—some ways to avoid the corporate tax cease to be cost effective. In other words, the balance between share­holder and corporate taxes influences the viability of strategies targeting one tax or the other.

b. Effect of the Allocation on Combined-Rate Strategies. — Yet although this allocation between shareholder and corporate taxes is quite important for a component-rate strategy (which avoids one tax), it is less important for a combined-rate strategy (which avoids both). In deciding whether to use combined-rate strategies, taxpayers still compare the strat­egies’ costs and benefits (as taxpayers do with component-rate strat­egies), but the payoff from combined-rate strategies is different: avoiding both taxes, instead of only one.

For instance, in deciding whether to use a pass-through entity in­stead of a corporation, taxpayers focus on the aggregate of the corporate and shareholder rates. If the personal rate is 40%, a pass-through entity offers no tax advantage when the corporate rate is 40% and the shareholder rate is 0%. The same is true if the corporate rate is 0% and the shareholder rate is 40%. The key is that the corporate and share­holder rates aggregate to 40%, but the balance between them is less important.

Yet it is an oversimplification to say that the allocation between corporate and shareholder rates is entirely irrelevant to combined-rate strategies. Rather, the balance between these taxes can have second-order effects, as long as a key condition is satisfied: One of the component taxes—that is, either the corporate tax or the shareholder tax—has to be easier to avoid than the other. When this is the case—and Congress allocates more of the combined burden to the easily avoided tax—less combined tax is collected.

For instance, assume the corporate tax is easy to avoid, but the share­holder and personal taxes are not. If the (easy-to-avoid) corporate tax is 40% and the (hard-to-avoid) shareholder tax is 0%, the true burden is less than 40% (since taxpayers will avoid some of this 40% tax). As a res­ult, corporations have an advantage over pass-through entities. But this advantage is eliminated—so the true burden is closer to 40%—if the (easy-to-avoid) corporate rate is 0% and the (hard-to-avoid) shareholder rate is 40%. 50 In principle, the allocation between corporate and shareholder taxes also matters if these taxes pursue different goals. For instance, Professor Reuven Avi-Yonah has argued that the corporate tax limits corporate power by reducing the resources under corporate control. See Reuven S. Avi-Yonah, Corporations, Society, and the State: A Defense of the Corporate Tax, 90 Va. L. Rev. 1193, 1244–46 (2004). Although shareholder taxes do not drain corporate resources in the same way, this Essay does not focus on this difference, since the corporate tax is a blunt way to limit corporate power; unlike antitrust, it is not limited to firms with market power. Another potential difference is that these taxes might fund different government services. According to Professor Schlunk, the corporate tax should fund services to businesses, while the shareholder tax should fund services to individuals. See Schlunk, supra note 11, at 138–39. Yet this distinction is not easy to draw. Is thwarting a terrorist attack on an Apple store a service to the business or the people who work and shop there? In any event, distinguishing services provided to businesses and individuals—especially in the cross-border context, which Professor Schlunk analyzes—can be accomplished in other ways (e.g., different source rules for active and passive income, instead of two taxes).

The bottom line, then, is that the allocation of tax between corporations and shareholders influences the cost effectiveness of various planning strategies. Changing the mix of rates can deter some strategies, while encouraging others.

B. High Corporate Taxes: Planning Strategies and Other Challenges

As the last section showed, even when corporate and shareholder taxes are coordinated to produce a specific combined rate (e.g., 40%), the allocation between these taxes still matters. It affects how taxpayers respond, as well as how much revenue is raised. To show why the allo­cation between corporate and shareholder taxes is so important, this sec­tion canvasses distortions and challenges that arise with a high corp­orate tax, but not a high shareholder tax. For instance, if the corporate tax is 40% and the shareholder tax is 0%, how would taxpayers respond? This section surveys a range of planning strategies to avoid this high corporate tax.

1. Shifting Income Abroad. — Perhaps the most important problem that arises with a high corporate rate—but not a high shareholder rate—is the incentive to shift income to other jurisdictions. Earning income abroad—or, at least, treating it as earned abroad—reduces the corporate tax burden of both foreign and U.S. firms, albeit for somewhat different reasons. Foreign corporations generally pay U.S. corporate tax only on U.S. earnings. 51 I.R.C. §§ 881–882. Likewise, U.S. firms generally pay current U.S. tax only on U.S. earnings and can defer U.S. tax on foreign earnings. 52 When a foreign subsidiary earns income abroad in an active business, no U.S. tax is due until the subsidiary pays a dividend to its U.S. parent. Alvin C. Warren, Jr., Income of Foreign Subsidiaries: A Review of the Basic Analytics, 145 Tax Notes 321, 322–24 (2014) (explaining that foreign earnings in a low-tax jurisdiction can compound at the low-tax jurisdiction’s tax rate). As a result, the corporate tax con­sequences change when income is sourced abroad. This incentive to shift income is especially strong because the U.S. corporate rate (35%) is the highest in the Organisation for Economic Cooperation and Develop­ment (OECD). 53 Table II.1 Corporate Income Tax Rate, Org. for Econ. Co-operation & Dev., [] (last visited Aug. 3, 2016) (listing the corporate tax rates in OECD countries). In contrast, Ireland’s rate is 12.5%, and the United Kingdom’s rate is 20%. 54 Id. Excluding the United States, the nations in the OECD have an average statutory rate (including state and provincial taxes) of 24.8%, compared with a rate of 39.1% in the United States. Martin Sullivan, The Truth About Corporate Tax Rates, Forbes (Mar. 25, 2015, 8:50 am), (on file with the Columbia Law Review).

In motivating firms to shift their income abroad, high corporate taxes induce two types of distortions. First, firms move assets and jobs abroad, even if they would rather operate in the United States. 55 Commentators describe this tax distortion as a violation of capital-export neutrality. See Terrence R. Chorvat, Taxing International Corporate Income Efficiently, 53 Tax L. Rev. 225, 227 (2000) (defining capital-export neutrality). As the gap in rates widens, more investment flows to low-tax jurisdictions, 56 See Ruud A. De Mooij & Sjef Ederveen, Taxation and Foreign Direct Investment: A Synthesis of Empirical Research, 10 Int’l Tax & Pub. Fin. 673, 690 (2003) (finding that a 1% reduction in the host-country tax rate yields a median increase in foreign investment of 3.3%); see also Tim Dowd, Paul Landefeld & Anne Moore, Joint Comm. on Taxation, Profit Shifting of U.S. Multinationals 4 (2016), (on file with the Columbia Law Review) (explaining that rate cuts in low-tax jurisdictions induce even more pronounced effects than rate cuts in high-tax jurisdictions). shifting assets to owners with less expertise or fewer synergies with other assets. 57 This distortion is known as a violation of “capital ownership neutrality.” See Mihir A. Desai & James R. Hines, Jr., Evaluating International Tax Reform, 56 Nat’l Tax J. 487, 494 (2003) (outlining how “capital ownership neutrality” may be satisfied through tax exemptions). Since real activity is harder to shift in some sectors than in others, tax distorts capital allocation across sectors. 58 See Erik Cederwall, Reconciling the Profit-Shifting Debate, 150 Tax Notes 713, 714 (2016) (“There is clear evidence that significant heterogeneity exists among [multinational enterprises] regarding their capacity and desire to shift profits, and the costs and benefits of profit shifting vary substantially.”). Second, income can be shifted not only with changes in real activity but also with familiar planning strategies. 59 For example, even if profitable innovations are developed in Silicon Valley, a subsidiary in a low-tax jurisdiction (a “low-tax sub”) can hold the patents, shifting income when a U.S. affiliate pays (deductible) royalties to the low-tax sub. With the right planning, the United States does not tax this income until it is paid as a dividend to the U.S. parent. See generally Michael J. Graetz & Rachael Doud, Technological Innovation, International Competition, and the Challenges of International Income Taxation, 113 Colum. L. Rev. 347, 394 (2013) (“Income from IP is much more mobile than the highly skilled workers and entrepreneurs who create it.”). Increasing the tax differential between affiliates by 1% is thought to increase the low-tax affiliate’s patent applications by 3.5%. See Tom Karkinsky & Nadine Riedel, Corporate Taxation and the Choice of Patent Location Within Multinational Firms, 88 J. Int’l Econ. 176, 182 (2012) (finding “the corporate tax rate exerts a significantly negative impact on the affiliates’ number of patent applications”); id. at 177 (“Quantitatively, the coefficient estimates are sizable, implying semi-elasticities of around −3.5.”). Likewise, a 1% reduction in this differential is estimated to increase the high-tax affiliate’s intellectual property holdings by 2.2%. Matthias Dischinger & Nadine Riedel, Corporate Taxes and the Location of Intangible Assets Within Multinational Firms, 95 J. Pub. Econ. 691, 697–98 (2011).

As a result, a great deal of income is shifted abroad, 60 See, e.g., Jane G. Gravelle, Cong. Research Serv., R40623, Tax Havens: International Tax Avoidance and Evasion 18 (2015)‌R40623.pdf [] (noting that profits reported in 2010 by U.S. affiliates in Bermuda and the Cayman Islands represent 1,614% and 2,065% of these nations’ GDPs and explaining that some of this income is shifted from the United States, while some presumably is shifted from high-tax foreign jurisdictions); see also Kimberly A. Clausing, The Revenue Effects of Multinational Firm Income Shifting, 130 Tax Notes 1580, 1581–82 figs.1 & 2 (2011) (documenting the gap between where U.S. multinationals report income (e.g., Netherlands, Luxembourg, Ireland, Bermuda, etc.) and where they have employees (e.g., U.K., Canada, Mexico, China)); Harry Grubert, Foreign Taxes and the Growing Share of U.S. Multinational Company Income Abroad: Profits, Not Sales, Are Being Globalized, 65 Nat’l Tax J. 247, 257–58 (2012) (noting that many profitable U.S. multinationals have a high percentage of sales in the United States, but a much lower percentage of their taxable income is earned there). Compared with shifting income from the United States, foreign-to-foreign shifting has somewhat different normative implications, since it arguably increases the (residual) tax collected by the United States. See Dhammika Dharmapala, What Do We Know About Base Erosion and Profit Shifting? A Review of the Empirical Literature, 35 Fiscal Stud. 421, 423 (2014) (noting that a 10% differential in tax rates between affiliates induces an 8% increase in the low-tax affiliate’s reported income). costing the United States approximately $100 billion of annual revenue 61 Kimberly A. Clausing, The Effect of Profit Shifting on the Corporate Tax Base in the United States and Beyond, 69 Nat’l Tax J. (forthcoming Dec. 2016) (manuscript at 2), (on file with the Columbia Law Review) (estimating revenue loss between $77 and $111 billion). and distort­ing behavior in a variety of ways. 62 Gabriel Zucman, Taxing Across Borders: Tracking Personal Wealth and Corporate Profits, 28 J. Econ. Persp. 121, 130 (2014) (“I conclude that profit-shifting to low-tax jurisdictions reduces the tax bill of US-owned companies by about 20 percent.”). Managers and advisors invest significant time in shifting income. 63 Cf. Dhammika Dharmapala, The Economics of Corporate and Business Tax Reform 7 (Univ. of Chi. Law Sch., Coase-Sandor Inst. for Law & Econ., Working Paper No. 757, 2016) [hereinafter Dharmapala, The Economics of Corporate and Business Tax Reform], (on file with the Columbia Law Review) (noting the deadweight loss from the commitment of talent and effort to tax planning, instead of to a more socially valuable occupation). In addition, U.S. multinationals keep over $2 trillion of foreign earnings offshore. 64 Fortune 500 Companies Hold a Record $2.4 Trillion Offshore, Citizens for Tax Justice (Mar. 4, 2016, 9:00 am),‌companies_hold_a_record_24_trillion_offshore.php#.V4jcN_krIdV [‌TWG4-KSKS] (noting that earnings have been deemed “permanently reinvested”). To invest it in the United States, they have to pay a 35% repatriation tax or use costly plan­ning to bring it back another way. 65 See Harry Grubert & Rosanne Altshuler, Fixing the System: An Analysis of Alternative Proposals for the Reform of International Tax, 66 Nat’l Tax J. 671, 683–85 (2013) [hereinafter Grubert & Altshuler, Fixing the System] (estimating the planning costs of bringing offshore earnings back to the United States to be 7%, based on analysis of the unexpectedly large response from multinationals when Congress temporarily reduced the repatriation tax to 3.6% in 2005); Melissa Redmiles, IRS, The One-Time Received Dividend Deduction, Stat. Income Bull., Spring 2008, at 102, 102–14 (showing that the stated rate was 5.25%, but a partial foreign tax credit was allowed, so the estimated average effective rate was 3.6%); Stephen E. Shay, The Truthiness of “Lockout”: A Review of What We Know, 146 Tax Notes 1393, 1395 (2015) (noting that firms generally can invest offshore earnings in stock and debt of other firms, but not in their own businesses).

Yet although high corporate taxes encourage firms to shift income abroad, high shareholder taxes do not. They are immune to this dist­ortion because, as noted above, the geographic boundary of shareholder taxes is based on where the shareholder lives, not where the firm earns its income. 66 See supra section I.A.3. As a result, shareholders cannot avoid this tax by sourcing income abroad. For example, if a U.S. citizen receives a dividend from a foreign corporation, U.S. shareholder tax is due for this dividend, reg­ardless of whether the firm earned this profit abroad or in the United States.

This difference between corporate and shareholder taxes has be­come much more important in recent years. It did not loom as large, for instance, when the Treasury and the American Law Institute (ALI) published studies on corporate integration two decades ago. 67 For example, in discussing international tax issues, the Treasury and ALI do not focus on income shifting. See Treasury, Integration Study, supra note 5, at 183–98; Warren, ALI Integration Report, supra note 24, at 735–63. In addition to these reports, the Treasury also addressed integration in a report published in December 1992. Dep’t of the Treasury, A Recommendation for Integration of the Individual and Corporate Tax Systems (Dec. 1992) [hereinafter Treasury, Recommendation for Integration], []. At the time, the economy was less global, foreign rates were higher, and some tax rules used in income shifting, such as the “check-the-box” rules, were not yet on the books. 68 The “check-the-box” rules allow some types of entities to choose to be taxed either as corporations or pass-through entities. Joshua A. Kaplan, Check-the-Box Elections: Relevance in the International Context, Taxes Without Borders (July 3, 2014), []. As one commentator observed, “[o]ne of the most powerful tools in cross-border tax planning is the ability to make a ‘check-the-box’ election.” Id. Perhaps for these reasons, the Treasury and ALI did not focus on the issue. Indeed, the Treasury proposed to rely on the corporate tax and to repeal the shareholder tax on dividends (and, to an extent, the capital gains tax as well). 69 See Treasury, Recommendation for Integration, supra note 67, at 2–3 (recommending a “dividend exclusion” prototype, which would tax profits at the corporate level but impose no shareholder tax on dividends). The proposal nominally preserved the capital gains tax, but scaled it back by allowing firms to adopt a “dividend reinvestment plan,” or “DRIP.” Without a DRIP, dividends would be taxed more favorably than retained earnings. Shareholders could receive a dividend tax-free, and if they used it to buy more shares, the shareholders would have basis in these new shares. The key to this increase in basis—and thus to a lower capital gains tax bill—was to receive and reinvest a dividend. In contrast, if the firm retained earnings, shareholders could not use these earnings to buy more shares and thus would not have additional basis. To eliminate this difference between dividends and retained earnings, a DRIP would allow shareholders to use retained earnings to increase the basis in their stock. In effect, the capital gains tax on retained earnings would be eliminated. See id. at 2 (“The effect will be to reduce the capital gains (or increase the capital losses) realized when shareholders sell their stock by an amount equal to the corporation’s retained previously-taxed earnings.”). Although more plausible in 1992, this proposal is untenable today, as one of its principal architects has emphasized. 70 See Does the Tax System Support Economic Efficiency, Job Creation, and Broad-Based Economic Growth?: Hearing Before the S. Comm. on Fin., 112th Cong. 80 (2011) (statement of Michael J. Graetz, Professor of Law, Columbia Law School) [hereinafter Graetz Testimony] (“I will not insist . . . we were right when the Treasury report was issued, but even if we were right then, that policy is now wrong. It is far easier and, I believe now better tax policy, to collect income taxes from individual citizens and resident shareholders than from multinational[s] . . . .”).

2. Inversions. — In addition to encouraging firms to shift income abroad, high corporate taxes induce another familiar distortion: Some firms respond by becoming a foreign firm, instead of a U.S. firm. Chang­ing tax residence allows a firm to pay U.S. tax only on U.S. income, rather than on worldwide income. 71 Although U.S. firms are taxed on worldwide income, they can defer tax on their active foreign income by earning it through a subsidiary and keeping these earnings offshore. Edward D. Kleinbard, Stateless Income, 11 Fla. Tax Rev. 699, 713–14 (2011) (“In practice the U.S. tax rules do not operate, as many presentations suggest, as a ‘worldwide’ system of taxation, but rather as an ersatz variant on territorial systems, with hidden benefits and costs when compared to standard territorial regimes.”).

Changing tax residence is not especially difficult for corporations. They have to reincorporate abroad or combine with a foreign firm. Yet their headquarters can remain in the United States, along with factories, researchers, and customers. They also can still be listed on a U.S. exchange. 72 For a discussion of inversions, see generally Eric L. Talley, Corporate Inversions and the Unbundling of Regulatory Competition, 101 Va. L. Rev. 1649, 1650 (2015) (“In a typical inversion, a U.S. multinational corporation (‘MNC’) merges with a foreign company. The entity that ultimately emerges from this transactional cocoon is invariably incorporated abroad, yet typically remains listed in U.S. securities markets under the erstwhile domestic issuer’s name.”). Admittedly, although changing tax residence for a cor­poration is not hard, it is not entirely without cost. The firm has to pay an exit tax 73 I.R.C. § 367 (2012). and also may alienate some customers and investors. 74 For instance, in abandoning its plans to invert, Walgreens responded in part to consumer pressure. See Michael J. de la Merced & Alexandra Stevenson, Walgreens Said to Consider an [sic] Tax-Inversion-Free Merger with Alliance Boots, N.Y. Times: Dealbook (Aug. 5, 2014, 3:53 pm), (on file with the Columbia Law Review). Different corporate law rules apply, so the cost of capital could increase if investor protections are weakened. But these costs are unlikely to deter self-interested managers, since they may benefit personally from having more discretion to pursue self-interested goals. Indeed, they can even blame the tax law for “forcing” them to weaken investor protections. 75 See David M. Schizer, Tax and Corporate Governance: The Influence of Tax on Managerial Agency Costs, in Oxford Handbook of Corporate Law and Governance 13 (Jeffrey Gordon & Wolf-Georg Ringe eds., 2015) (ebook) [hereinafter Schizer, Tax and Governance] (“[U.S.] tax rules offer a reason (or excuse) to incorporate in jurisdictions with weaker legal constraints on managerial agency costs.”). This erosion of shareholder protections should not be overstated, moreover, since protections can be included in the firm’s charter, and some also are guaranteed by federal securities law (which still applies to foreign firms, as long as they are listed in the United States). See Talley, supra note 72, at 1693–94.

So although switching tax residence requires some changes in be­havior—and thus involves distortions—these costs often are less than the tax savings, at least when the U.S. corporate rate is high. In response, Congress and the Treasury have tried to block this planning strategy with section 7874, 76 I.R.C. § 7874. Section 7874 denies the tax benefits of an inversion—treating the inverted firm as still a U.S. firm for some purposes—if the inversion fails the statutory test, which focuses on where the inverted firm’s business assets are located and how much of the inverted firm is owned by shareholders who were owners of the pre-inversion domestic corporation. Id. as well as with repeated rounds of regulatory action. 77 See, e.g., I.R.S. Notice 2015-79, 2015-49 I.R.B. 775 (announcing the intention to issue additional regulations targeting inversion structures and post-inversion planning techniques); I.R.S. Notice 2014-52, 2014-42 I.R.B. 712 (announcing the intention to issue regulations targeting inversions); see also T.D. 9761, 2016-20 I.R.B. 746 (issuing temporary regulations to implement proposals described in Notices 2014-52 and 2015-79 and proposing regulations to recharacterize certain intercompany debt as equity). But although these steps raise the cost of this planning strategy, they have not stopped it entirely. 78 In announcing new regulations in April of 2016, Treasury Secretary Jacob Lew said: “[W]e know companies will continue to seek new and creative ways to relocate their tax residence to avoid paying taxes here at home.” Press Release, U.S. Dep’t of the Treasury, Treasury Announces Additional Action to Curb Inversions, Address Earnings Stripping (Apr. 4, 2016),‌jl0405.aspx []. Secretary Lew added that regulations alone will not be enough and “urge[d] Congress to move forward with anti-inversion legislation this year.” Id.  For instance, firms can still combine with foreign firms that are sufficiently large. 79 U.S. shareholders should own less than 80% (or, for some purposes, 60%) of the combined firm. I.R.C. § 7874. Firms also can incorporate abroad ini­tially when they first go into business.

While high corporate taxes induce these distortions, high share­holder taxes are less likely to do so. Admittedly, shareholders enjoy tax advantages in changing tax residence, which roughly parallel those for corporations. Although shareholders still owe U.S. shareholder tax on dividends from U.S. firms, they no longer owe shareholder tax on dividends from foreign firms or, for that matter, on capital gains. Yet although changing residence can offer tax savings to both firms and shareholders, this step is much costlier for shareholders. In addition to owing an “exit” tax, 80 The tax applies to built-in gains above $600,000 for taxpayers with $2 million in assets or average income of at least $160,000 over the past five years. Id. § 877(a) (requir­ing taxpayers with income or assets above specified levels to pay expatriation tax); id. § 877A(a)(3) (providing $600,000 exclusion, which is adjusted for inflation); Expatriation Tax, Internal Revenue Serv. (May 13, 2016),‌international-taxpayers/expatriation-tax [] (noting that income must be above $160,000 in 2015 in order for the tax to apply). shareholders have to renounce U.S. citizenship, leave the United States, and forgo the right to come back. 81 During the ten-year period after they expatriate, former citizens also cannot spend more than thirty days per year in the United States, or they are taxed as U.S. citizens on their worldwide income. I.R.C. § 877(g). In theory, former citizens can even be barred from short visits, since Congress has authorized the Justice Department to deny reentry to those who expatriate for tax reasons. See Mark Nestmann, “Homelanders” to U.S. Expatriates: Don’t Come Back . . . Ever, Nestmann Grp. (June 23, 2013), http://‌ []. These are major life changes, which are much more daunting than the relatively minor implications of changing a corporation’s residence. As a result, changing tax residence is a distortion that is more likely with high corporate taxes than with high shareholder taxes.

3. Debt. — Along with encouraging firms to shift income and change their tax residence, high corporate taxes prompt another familiar planning strategy: Firms issue more debt. Since interest is deductible, corporations pay no tax on revenue that funds interest payments. Only lenders pay the tax, which they avoid if they are tax exempt or foreign. 82 See infra sections I.C.4–.5. When debt finances favorably taxed assets (e.g., those that are eligible for accelerated depreciation), the effective tax rate can be negative. 83 See Cong. Budget Office, Taxing Capital Income: Effective Rates and Approaches to Reform 8 (Oct. 2005), [] (finding that debt-financed corporate investments have an effective tax rate of -6.4%). Intercompany debt is also a familiar way to shift income: When high-tax affiliates borrow from low-tax affiliates, interest payments can shift income to the low-tax jurisdiction. See, e.g., Stephen E. Shay, J. Clifton Fleming, Jr. & Robert J. Peroni, Treasury’s Unfinished Work on Corporate Expatriations, 150 Tax Notes 933, 933–34 (2016) (discussing regulatory responses to impede earnings stripping by inverted companies).

In addition to reducing revenue, the tax preference for debt causes familiar taxpayer responses. On the positive side of the ledger, debt imposes discipline on managers. If a firm has no debt, managers can choose what to do with the firm’s cash, so they are freer to use it in self-interested ways. But once the firm has debt, managers have less discretion, since they have to pay interest. To avoid this constraint, self-interested managers may shy away from issuing debt, so the tax preference for debt plays a useful role in countering this impulse. 84 See Schizer, Tax and Governance, supra note 75, at 9.

Yet on the negative side of the ledger, debt increases the risk of bankruptcy, which is a costly process. As a result, a tax preference for debt is likely to increase bankruptcy costs. 85 The consensus estimate is that cutting the corporate rate by 10% would reduce the debt-to-asset ratio by 2.8%. See Ruud A. De Mooij, Tax Biases to Debt Finance: Assessing the Problem, Finding Solutions, 33 Fiscal Stud. 489, 495 (2012). The bankruptcy costs associated with tax-induced leverage are estimated to be approximately 1% of corporate tax revenue. See Roger H. Gordon, Taxation and the Corporate Use of Debt: Implications for Tax Policy, 63 Nat’l Tax J. 151, 152–53 (2010). Alternatively, it can induce costly financial innovation, as hybrid securities are developed that offer a tax deduction but do not increase bankruptcy risk as much. 86 See Schizer, Tax and Governance, supra note 75, at 9. When firms respond by taking on more insolvency risk, the tax preference for debt can cause another distortion as well: Managers have the incentive to take unwise risks when firms are on the brink of insolvency. Even if a risky initiative is unlikely to pay off, managers “swing for the fences” because they personally share in gains (by avoiding the reputational cost of bank­ruptcy) but can shift losses to creditors (by reducing their recovery in bankruptcy). 87 See id.

4. Corporate Lock-In. — The high corporate tax distorts managerial behavior in another familiar way: Managers hesitate to sell appreciated assets, since selling them triggers corporate tax on the appreciation. 88 Under the realization rule, tax on an appreciated asset is not due until it is sold. See David M. Schizer, Realization as Subsidy, 73 N.Y.U. L. Rev. 1549, 1551 (1998) [hereinafter Schizer, Realization as Subsidy]. If the tax is high enough, managers keep unwanted business units, intel­lectual property, and passive investments. Although managers are sup­posed to allocate scarce capital to the most promising opportunities, this “lock-in” effect impedes this process by favoring appreciated assets that the firm already owns.

In some cases, managers can use planning strategies to avoid lock-in. In effect, they dispose of appreciated assets tax-free, for instance, with spinoffs or derivative transactions. 89 See Schizer, Frictions, supra note 48, at 1345–47 (describing how derivatives are used to hedge appreciated stock). When these strategies are available, lock-in no longer impedes the efficient allocation of a firm’s assets. Yet these strategies cause other distortions instead. In addition to paying advisory fees, the firm has to comply with a host of technical require­ments, which can affect the structure and economic return. 90 See id. at 1345–59 (describing frictions that impede hedging or make it costlier).

While high corporate taxes encourage firms to hold onto appre­ciated assets, high shareholder taxes do not; instead, high shareholder taxes induce a different form of lock-in by encouraging shareholders to keep appreciated stock. These dueling lock-in effects are a quintessential example of why shifting tax from corporations to share­holders alleviates some distortions, while exacerbating others. For example, suppose a cor­poration has appreciated assets while the firm’s shareholders have appreciated stock. Both the corporation and its shareholders have a tax incentive not to sell. Cutting the corporate rate mitigates the cor­poration’s lock-in. But if this cut is funded by increasing the shareholder rate, shareholder lock-in is exacerbated.

5. Comparative Advantage of Corporations over Shareholders in Tax Planning. — Another challenge in collecting tax from corporations, inst­ead of from shareholders, is that corporations enjoy greater economies of scale in tax planning than do shareholders. When firms incur advisory fees and other structuring costs, all shareholders share in these costs. In contrast, individual shareholders cannot pool costs in the same way. 91 Of course, shareholders have advisors that may be able to spread costs among cli­ents, such as investment bankers, accountants, and lawyers. But these advisors probably are better positioned to spread the costs of developing strategies than of implementing them.

Admittedly, this difference should not be overstated, since corp­orations do not always use their comparative advantage. Managers can be wary of tax planning, since they bear reputational costs when strategies fail but share tax savings with shareholders when strategies succeed. Even so, there are familiar ways to mitigate these agency costs, including legal opinions to protect managers’ reputations, as well as bonuses for lowering the firm’s effective tax rate. 92 See, e.g., Daniel Shaviro, Disclosure and Civil Penalty Rules in the U.S. Legal Response to Corporate Tax Shelters, in 3 MPI Studies on Intellectual Property, Competition and Tax Law: Tax and Corporate Governance 229, 242 (Wolfgang Schön ed., 2008) (“[A]voiding penalties may depend . . . on whether the taxpayer can demonstrate reliance on a sufficiently sanguine opinion of counsel. Tax lawyers therefore have at least a limited power to grant penalty protection.”).

When managers are motivated to engage in tax planning, they focus on corporate taxes instead of on shareholder taxes. Corporate taxes are more likely to affect standard metrics of their performance, such as reported earnings. 93 See, e.g., Edward D. Kleinbard, Rehabilitating the Business Income Tax 56 (Brookings Inst., Hamilton Project, Discussion Paper 2007-09, 2007), http://www.‌ [‌8H2E-BPVT] (positing that for agency-cost reasons “investor tax consequences will not materially affect managers’ business decision making”). Of course, if accounting for shareholder taxes reduces a firm’s cost of capital, this savings could influence metrics that matter to managers. Managers also have less control over shareholder taxes, which often turn on shareholder choices (e.g., about when to sell) and vary with each shareholder’s tax bracket. 94 Even so, when deciding whether to pay dividends or repurchase shares, managers do have influence on shareholder taxes, and there is some evidence that they consider them. See, e.g., Jennifer Blouin, Jana Raedy & Douglas Shackelford, Dividends, Share Repurchases, and Tax Clienteles: Evidence from the 2003 Reductions in Shareholder Taxes 32 (Nat’l Bureau of Econ. Research, Working Paper No. 16129, 2010), http://‌ (on file with the Columbia Law Review) (showing firms repurchased fewer shares after the rates for dividends and capital gains were aligned in 2003).

6. Incidence. — A further challenge with relying on the corporate tax is the uncertainty about who actually pays it. Obviously, the corporation writes the check. But does this payment reduce the invest­ment return of shareholders, lower the salary of employees, or raise the price paid by consumers? If the goal of taxing “rich” corporations is to tax high-income investors—as voters often assume—this goal is not achieved if the burden actually falls on low-income workers.

A key reason why corporate taxes may burden workers is income shifting. If firms respond by shifting real activity overseas, there is less capital in the United States, reducing the productivity and wages of U.S. workers. 95 Since high corporate rates are more likely to trigger shifting than low rates, a low corporate tax presumably is more likely to burden investors than a high tax. Yet this effect will not necessarily arise if firms shift income with planning, instead of moving real activity. 96 Kimberly A. Clausing, Who Pays the Corporate Tax in a Global Economy?, 66 Nat’l Tax J. 151, 171 (2013) (“If firms can respond to tax differences among countries through financial or organizational decisions, this will lower the tax sensitivity of real activity, thus reducing adverse effects on labor associated from tax-induced reductions in the capital stock.”). Shifting income also is not a viable strategy—so incidence is not likely to be on workers—when U.S. firms earn rents, are not capital intensive, or cannot easily rely on foreign production. Therefore, it is unclear how much of the corporate tax is borne by investors, as opposed to workers.

Does the same issue arise with a shareholder tax? Although gen­eralizations are hard, there is a reason why shifting the corporate tax to workers could be easier than shifting the shareholder tax: If workers refuse to take a pay cut, managers can avoid this expense by moving the business offshore, which is costly to workers. “If you will not absorb this cost for us,” managers can say, “we will move our operations out of the United States, and you will lose your jobs.” But unlike the corporate tax, the shareholder tax cannot be avoided by moving the business offshore. As noted above, U.S. shareholders are still taxed on dividends and capital gains, even if the underlying profit is earned abroad. 97 See I.R.C. § 61(a)(7) (2012); see also supra section I.B.1. Since this tax cannot be avoided another way—which is even more costly to workers—employees are under less pressure to accept lower pay. For this reason, the incidence of corporate and shareholder taxes could be different. Yet even so, it is hard to know whether this is actually the case.

7. Tailoring the Tax Burden to Individual Circumstances. — Finally, another limitation of corporate taxes is that, unlike shareholder taxes, they use a rate that does not account for the other income of share­holders (or others who bear the tax’s economic burden). 98 See supra section I.B.6. The brackets of a corporate tax are based on the income of the firm, not the income of shareholders (or other stakeholders). 99 I.R.C. § 11. So if a corporation earns enough to be in the top bracket, the same (top) corporate rate applies (indirectly) to all shareholders (and other stakeholders). Retirees with low incomes are treated the same as the CEO. 100 Although individual tailoring is not feasible if Congress relies solely on the corporate tax, it is feasible if Congress relies only partially on it. Congress can impose a flat tax at the corporate level, while also collecting a shareholder tax that varies with the shareholder’s income. Professor George Yin proposed this approach, describing the shareholder tax as a “surtax.” See George K. Yin, Corporate Tax Integration and the Search for the Pragmatic Ideal, 47 Tax L. Rev. 431, 481–83 (1992). For instance, the corporate tax can be a flat 22.5% tax, while the shareholder tax can vary between 0% and 22.5%, depending on the shareholder’s income.

This blunt approach creates familiar efficiency and distributional concerns. Corporations are treated differently than pass-through busi­nesses, whose owners are taxed at their personal rate, which accounts for other income. 101 I.R.C. § 701 (“A partnership as such shall not be subject to the income tax imposed by this chapter. Persons carrying on business as partners shall be liable for in­come tax only in their separate or individual capacities.”). This discontinuity can distort behavior. 102 These issues featured prominently in the ALI’s recommendation to use an imputation system, which applies the shareholder’s individual rate to dividend income. See Warren, ALI Integration Report, supra note 24, at 618–22, 631–35. In addition, using a rate that fails to account for other income is a flawed way to pur­sue distributional goals.

C. High Shareholder Taxes: Planning Strategies and Other Challenges

To show why the allocation between corporate and shareholder taxes is important—even when rates are coordinated to produce a specific combined rate (e.g., 40%)—the last section showed that a high corporate tax motivates a range of planning strategies. 103 See supra section I.B. For instance, if the corporate tax is 40% and the shareholder tax is 0%, corporations are likely to respond by shifting income abroad, changing their tax resi­dence, issuing more debt, and delaying the sale of appreciated assets.

In response, these planning strategies can be preempted by cutting (or even repealing) the corporate tax. Since taxpayers are less motivated to avoid a low tax, they would have less reason to invest in these strategies or to change their behavior in other ways. But unfortunately, this pro­gress is not free. If the cut in the corporate tax is funded with an increase in the shareholder tax—so the corporate tax is 0% and the shareholder tax is 40%—taxpayers have a correspondingly stronger incentive to avoid shareholder taxes. Instead of using the strategies discussed above, they would use other strategies to avoid the shareholder tax. This section can­vasses planning strategies that are appealing when shareholder rates are high and corporate rates are low. The bottom line is that changing the mix of rates eases one set of distortions but exacerbates another.

1. Shareholder Lock-In. — A familiar challenge with relying on shareholder taxes, instead of on corporate taxes, is that shareholders control the timing of an important component: the tax on share appreciation. Unlike corporate income, which is taxed every year, share appreciation is taxed only when stock is sold. 104 See, e.g., Tim Dowd, Robert McClelland & Athiphat Muthitacharoen, New Evidence on the Tax Elasticity of Capital Gains, 68 Nat’l Tax J. 511, 513–14 (2015) [hereinafter Dowd et al., New Evidence on Tax Elasticity] (showing the significant effect of capital gains tax on realizations). By discouraging sales, a capital gains tax “locks in” shareholders, distorting portfolio choices and encouraging costly hedging transactions. Not only can this tax be delayed, but some­times it can also be avoided entirely: Shareholders are never taxed on capital gain if they hold appreciated stock until they die 105 For instance, if a taxpayer buys stock for $1 per share and dies when it is valued at $100 per share, the taxpayer’s heir’s basis is $100, not $1. See I.R.C. § 1014 (providing that the basis of property acquired from a decedent is “the fair market value of the property at the date of the decedent’s death”). The reason is that heirs can “step up” the basis of inherited property to its fair market value when they inherit it. If the heir immediately sells the shares for $100, there is no taxable gain. In other words, no tax is ever collected on the $99 of appreciation that occurred during the taxpayer’s lifetime. or contribute it to charity. 106 For example, if a taxpayer buys stock for $1 per share, and contributes it to charity when it is worth $100 per share, this contribution is not taxed as a sale. Since the charity is tax exempt, it can sell it immediately for no gain. At the same time, the taxpayer can still deduct the stock’s fair market value ($100 per share), instead of the stock’s basis ($1 per share), if the taxpayer holds it for more than a year. See id. § 170. As a result, the $99 of appreciation is never taxed. Obviously, these strategies become more appealing when the shareholder rate increases. 107 Empirical estimates suggest that taxpayers are quite sensitive to tax rates in deciding how much gain to realize. See, e.g., Dowd et al., New Evidence on Tax Elasticity, supra note 104, at 527 (estimating elasticity of persistent tax changes to be about -0.72, with transitory elasticity above 1). So if the shareholder tax is raised in order to fund a cut in the corporate tax, more shareholders will use these strategies to delay or avoid the capital gains tax.

2. Trapped Earnings. — Like the tax on share appreciation, the tax on dividends can also be delayed. This tax is not due until a dividend is actually paid. 108 I.R.C. § 301(c) (including dividend in income). At first blush, delaying the dividend tax may seem advantageous, but delaying this tax actually does not reduce its present value (as long as the rate does not change). 109 See Daniel N. Shaviro, Decoding the U.S. Corporate Tax 37 (2009) [hereinafter Shaviro, Decoding the U.S. Corporate Tax] (“Ordinarily in income tax planning, deferral . . . is a good thing . . . . In the corporate distributions setting, . . . things are more complicated because the amount that ultimately will have to be distributed . . . is growing by reason of the positive return to the funds the company is keeping on hand in the interim.”).

Even so, there is a different tax reason to delay a dividend, which is not prompted by a high shareholder rate, but by a low corporate rate. When the corporate rate is low, firms have a comparative advantage over shareholders in investing money. 110 See Daniel I. Halperin, Mitigating the Potential Inequity of Reducing Corporate Rates, 126 Tax Notes 641, 648 (2010) [hereinafter Halperin, Mitigating Inequity] (“[I]f distributions are taxed and the combined rate is equal to the individual rate, there is no advantage from the deferral of the tax on the initial corporate earnings. The advantage is the higher return on reinvested earnings.”). For instance, assume the shareholder rate is 40%, the corporate rate is 0%, and the personal rate is 40% (i.e., on pass-through businesses, rents, interest, etc.). Because the corporate rate is 0%, investments grow tax-free inside a corporation. But if the firm distributes earnings, and shareholders do not reinvest them in a corp­oration, the shareholders’ investments generally are taxed at the 40% personal rate. 111 For instance, periodic income from pass-through businesses, loans, and rental property would generally be taxed at the personal rate. See I.R.C. § 61. However, capital gain from selling investments would generally be taxed at the capital gains rate. See id. § 1(h). So if the pretax return is the same, the investment grows faster inside a corporation. As a result, corporations function like a tax-advantaged account when the corporate rate is lower than the personal rate. 112 For instance, assume a firm can either distribute $100 of profit to shareholders or retain it for three years and then distribute it. Assume the annual pretax return during those three years is 10% either way, and that the personal rate is 40%, the corporate rate is 0%, and the shareholder rate is 40%. If the firm keeps these earnings for three years, it can earn a 10% after-tax return (since the 0% corporate rate applies). But if the firm distributes the earnings immediately, the shareholder can earn only a 6% return (since the 40% personal rate applies). At the end of three years, the shareholder has $79.86 if the corporation retains these earnings, compared with $71.46 if the corporation had dist­ributed them immediately. This tax advantage of corporations was widely used when corpo­rate rates were significantly lower than personal rates. 113 See Halperin, Mitigating Inequity, supra note 110, at 642. Although this planning receded when this gap narrowed in 1981, it would revive if the gap were to widen. 114 Cf. id. (“[I]f the corporate rate is lower than the top individual rate—a dynamic that has not been in place since 1986—closely held businesses might again choose to be subject to the corporate tax to shelter high-income individuals from high individual rates.”). Indeed, if Congress wants to use the same rate for noncorporate and corporate businesses—so the corporate and shareholder rates aggregate to the personal rate—the corporate rate has to be lower than the personal rate, as long as at least some tax is collected from shareholders (and the timing of these taxes varies). Specifically, if (1–P) = (1–C) (1–S), and S>0, C has to be less than P.

Unfortunately, this tax preference for retained earnings is likely to exacerbate agency costs. Retained earnings are problematic because managers have unfettered discretion to spend them. In contrast, if profits are distributed to shareholders, managers no longer have capital on hand. To pursue new initiatives, they have to persuade investors to write a check. The need to make a case disciplines managers, discouraging them from proposing self-interested uses of capital. 115 See Schizer, Tax and Governance, supra note 75, at 10. Since leaving cash in the corporation undermines this discipline, tax rules that encourage firms to keep cash breed agency costs. 116 Conversely, when corporate rates are higher than personal rates, there is a tax incentive to distribute earnings, which could keep firms from pursuing valuable projects (e.g., if outside investors will not fund them because of asymmetric information). This is a problem not only with favoring retained earnings, but also with collecting tax from shareholders instead of corporations: If the check comes from shareholders, the corporation keeps more of its cash, leaving managers in control of more resources.

Shareholder taxes can also affect corporate cash flow in another way. Taxing dividends creates a timing mismatch for shareholders whose stock has not appreciated: Current income from the dividend is offset by future capital loss when they sell stock. 117 For example, assume a shareholder buys stock for $100 and immediately receives a $5 dividend, which reduces the stock price to $95. The shareholder has five dollars of taxable dividend income, but no economic profit. As a result, this current income is offset by five dollars of future capital loss, which is not deductible until the stock is sold. One way to avoid this mis­match, emphasized by Professor David Weisbach, is to accelerate divi­dends, paying them before stock is sold to a new owner with shares that have not appreciated. 118 David Weisbach, Capital Gains Taxation and Corporate Investment 16 (Univ. of Chi. Law Sch., Coase-Sandor Inst. for Law & Econ., Working Paper No. 740, 2016), (on file with the Columbia Law Review). Conversely, another solution is to delay divi­dends until shares appreciate. Still another (partial) fix is to cut divi­dends, so the magnitude of this mismatch is reduced. 119 The scope of the mismatch increases when dividends are large, and also when shareholder taxes are high; either one increases the size of the initial income inclusion, which is offset by a deferred capital loss.

3. Salary Shifting. — When the shareholder rate is high—or, to be precise, when the corporate rate is low—another planning strategy becomes appealing: Owners who are also employees pay themselves below-market salaries. This strategy is appealing when the personal rate (which governs salaries) is higher than the corporate rate. To implement this strategy, employee-owners leave money inside the corporation, instead of taking it out as salary. There is not much economic sig­nificance to leaving money in the corporation, since the employee-owner owns and controls it. But the tax treatment changes: (High-taxed) salary is transformed into (low-taxed) corporate earnings. 120 For example, assume that an entrepreneur is the sole owner and employee of a corporation, which earns $500,000, and assume also that the personal rate is 40%, the corporate rate is 20%, and the shareholder rate is 25%. If the entrepreneur takes all the corporation’s $500,000 of revenue as salary, the entrepreneur pays a 40% tax (of $200,000). Since the corporation deducts this salary payment, it has no taxable income. As a result, all the business’s economic return is taxed as salary (at the higher personal rate), and none is taxed as corporate profit (at the lower corporate rate). But this changes if the entrepreneur takes only a $100,000 salary (taxable at 40%), leaving the firm with a $400,000 profit (taxable at 20%). Instead of a current tax of $200,000, the government receives a current tax of only $120,000 (i.e., $40,000 on the salary and $80,000 on the corporate profit). In other words, the entrepreneur turns $400,000 of (high-taxed) salary into (low-taxed) corporate profit. Obviously, an additional 25% shareholder tax would be imposed if the firm distributes its profit to the owner.

This “salary shifting” loses its appeal if the corporate rate is increased to match the personal rate. But unfortunately, raising the cor­porate rate exacerbates income and residence shifting, as well as the other problems discussed in section I.B. In theory, to avoid all of these distortions, the U.S. corporate rate has to equal both the U.S. personal rate (to deter salary shifting) and foreign corporate rates (to deter income and residence shifting). But this solution is mathematically impossible under current law, since the U.S. personal rate is significantly higher than foreign corporate rates. So once again, we have to pick our poison. 121 See Dharmapala, The Economics of Corporate and Business Tax Reform, supra note 63, at 14 (“[T]he corporate tax rate is a single instrument that affects two very different kinds of behavior—the incorporation decisions of (typically small) business ent­ities or individuals, and the location and investment decisions of large MNCs.”).

4. Tax-Exempt Shareholders. — Another familiar problem with shareholder taxes is that, unlike corporate taxes, they do not reach tax-exempt shareholders. 122 I.R.C. § 501(a) (2012). For instance, assume that tax-exempt share­holders own a for-profit corporation, which earns $100 million. If the corporate tax is 40% and the shareholder tax is 0%, $40 million of tax is collected. But if these rates are flipped—so the corporate rate is 0% and the shareholder rate is 40%—no tax is collected.

This gap in investor-level taxes is already an issue for debt. Under current law, tax-exempt lenders pay no tax on interest, but corporations can still deduct it. 123 See id. § 163(a) (authorizing deduction for interest); id. § 501(a) (exempting income of nonprofits). As a result, the United States collects no tax on revenue servicing this debt. This hole in the tax base would grow if equity were also taxed only (or primarily) at the investor level. Significant rev­enue is at stake, since less than half of U.S. equities are held by taxable U.S. shareholders; tax-exempt and foreign shareholders own the rest. 124 See Jane G. Gravelle & Donald J. Marples, Cong. Research Serv., R44242, The Effect of Base-Broadening Measures on Labor Supply and Investment: Considerations for Tax Reform 27 (Oct. 22, 2015), [http://‌] (estimating that 25% of dividends are paid to taxable accounts); Steven M. Rosenthal & Lydia S. Austin, The Dwindling Taxable Share of U.S. Corporate Stock, 151 Tax Notes 923, 923 (2016) (estimating that household accounts held only 24.2% of U.S. equities in 2015); Alan J. Auerbach, Who Bears the Corporate Tax? A Review of What We Know 5 (Nat’l Bureau of Econ. Research, Working Paper No. 11686, 2005), [] (estimating that less than half of corporate equities are owned by households). This gap in shareholder taxes complicates any effort to fund a corporate tax cut with them.

Obviously, the treatment of tax exempts does not have to be binary. Instead of paying a full tax or no tax at all, they could pay an inter­mediate amount, as with the (indirect) corporate tax under current law. How much they pay should depend, at least in part, on the reasons for exempting them. Since the rationales vary with the type of tax exempt, the tax could vary too. For example, individual retirement accounts (IRAs) encourage savings, especially for low-income people. So do pen­sion funds, although distribution features less prominently (since there are no income limits).

By contrast, exempting investment income of universities, religious organizations, and cultural institutions is supposed to promote their charitable missions. But another tax preference—the charitable deduc­tion—also advances this goal and is arguably more effective. The deduc­tion rewards charities for attracting support from donors; government money flows only if donors are willing to contribute their own money and thus have “skin in the game.” 125 By offering a deduction, the government allows donors to use charitable contributions to reduce their tax bills. See I.R.C. § 170. For instance, if a donor’s personal tax rate is 40%, and the donor contributes $1 million, this contribution can reduce her tax bill by $400,000 (by reducing her taxable income by $1 million and thus reducing her taxes by $400,000). With this $1 million contribution, then, the donor is putting in $600,000 of her own money and attracting a $400,000 matching contribution from the government (in the form of a reduced tax bill). Therefore, the government’s $400,000 of support becomes available only if the donor is willing to put in $600,000 of her own money. In contrast, exempting investment income rewards charities for running a surplus and investing it successfully, which is a less reliable signal of social value. 126 For a discussion, see David M. Schizer, Subsidizing Charity: Donations, Investment Income, and the Governance of Nonprofits 21 (Oct. 14, 2016) (unpublished manuscript) (on file with the Columbia Law Review). In fact, non­profits with investment income are under less pressure to raise new money. They have less need to demonstrate continuing value to donors, especially when investment income is tax-free and thus covers more of their budgets. As a result, the deduction arguably promotes positive exter­nalities more effectively than the exemption does. Yet even if these rules are equally effective, charity can still be subsidized without the exemption, as long as the deduction is available. A cut in one can be offset with enhancements to the other.

5. Foreigner Shareholders. — Like tax exempts, foreigners are taxed indirectly by the corporate tax, but (largely) escape shareholder taxes. 127 The assumption here is that these foreigners do not live in the United States; if they do, they are taxed like U.S. citizens. See, e.g., I.R.C. § 871(a)(2) (stipulating that the capital gains of foreigners be taxed if they satisfy the residency test). So if shareholder taxes are increased to fund a cut in corporate taxes, for­eign shareholders would mostly be exempt from this increased share­holder tax. Under current law, foreigners pay no tax on capital gains. 128 See id. Although foreigners’ dividends are subject to withholding, treaties often reduce the rate of the withholding tax that applies to these dividends (e.g., from 30% to 15%). 129 See id. Foreigners used to avoid even this reduced withholding by investing in derivatives. Although a reform, section 871(m), was supposed to plug this gap, questions remain about its effectiveness. Id. § 871(m); see also Thomas J. Brennan & Robert L. McDonald, The Problematic Delta Test for Dividend Equivalents, 146 Tax Notes 525, 529 (2015) (discussing various ways that taxpayers could avoid withholding tax by manipulating the delta test). For a discussion of these regulations, see Tax Section, N.Y. State Bar Ass’n, Report No. 1340, Report on Regulations Under Section 871(m) (Mar. 28, 2016), []. So if foreigners own a U.S. corporation that earns $100 million, $40 million of tax is collected with a 40% corporate tax and a 0% shareholder tax. But no tax is collected if these rates are reversed, as long as shareholders earn capital gains instead of dividends. Like tax exempts, foreigners already avoid both corporate and investor taxes in lending to corporations. This gap would grow if equity were also taxed only (or primarily) at the investor level.

A key question, then, is how much tax foreigners should pay. Taxing them enhances national welfare in three ways. First, it keeps resources in the U.S. economy. 130 For instance, if foreign owners of a U.S. corporation would otherwise receive $100 of corporate profits, a 40% tax on the corporation keeps $40 in the United States. Second, revenue targets can be met with a lower (and less distortive) tax on U.S. firms and shareholders. Third, taxing foreign owners of corporate businesses matches the treatment of non­corporate businesses, whose foreign owners already pay tax on U.S. pro­fits. (The politics are tempting as well, since foreign investors do not vote in the United States.)

Even so, taxing foreign investors has two offsetting costs. First, the United States might attract less foreign investment. Yet a zero rate seems like an overreaction, even though the United States exempts interest for this reason. After all, investing in the United States is appealing, 131 See, e.g., Dep’t of Commerce & The President’s Council of Econ. Advisors, Foreign Direct Investment in the United States 3 (Oct. 2013), http://www.whitehouse.‌gov/sites/default/files/2013fdi_report_-_final_for_web.pdf [] (“With the world’s largest consumer market, skilled and productive workers, a highly innovative environment, appropriate legal protections, a predictable regulatory environment, and a growing energy sector, the United States offers an attractive investment climate for firms across the globe.”). so for­eigners presumably will pay some tax to do so. Second, if the United States raises taxes on foreigners, other jurisdictions might reciprocate with higher taxes on U.S. investors. If Americans claim more foreign tax credits, the United States might not raise (net) revenue.

So far, the focus has been on revenue losses when a corporate tax cut is funded with higher shareholder taxes: Less revenue is collected from foreign and tax-exempt shareholders of U.S. corporations. How­ever, these rate changes offer offsetting revenue gains as well: More is collected when U.S. shareholders invest in foreign firms that earn money abroad. 132 See Harry Grubert & Rosanne Altshuler, Shifting the Burden of Taxation from the Corporate to the Personal Level and Getting the Corporate Tax Rate Down to 15 Percent, at 8 (Rutgers Univ. Dep’t of Econ., Departmental Working Paper No. 2015-06, 2016) [hereinafter Grubert & Altshuler, Shifting the Burden],
2802109 (on file with the Columbia Law Review) (“[F]or proposals that include an increased tax on capital gains and dividends it would also be necessary to add U.S. resident holdings of foreign shares.”).
The corporate tax does not reach foreign profits of foreign firms, but the shareholder tax does. Therefore, cutting the corporate rate and raising the shareholder rate collects more tax on these profits.

6. Administrative Costs. — In addition to the revenue losses from exempt and foreign shareholders, another challenge in taxing share­holders, instead of corporations, is the loss of economies of scale in compliance and collection. Corporations can spread the cost of advice, record-keeping, and other infrastructure across all shareholders. For the government, monitoring the firm is more cost effective than policing (and suing) owners of small blocks of stock. 133 See Treasury, Integration Study, supra note 5, at 102 (noting that “tax is most likely to be collected if paid at the corporate level”). Even so, these admin­istrative advantages have become less compelling over time. Superior expertise has enabled corporations not only to comply with rules, but also to avoid them, as emphasized above. 134 See supra section I.B.5. At the same time, third-party reporting of dividends and basis has helped the government monitor shareholder compliance more effectively. 135 Securities brokers have been required to report the cost basis of securities since 2011. Third Party Reporting Information Center, Internal Revenue Serv., http://‌ [‌AB8E-X54A] (last updated Mar. 17, 2016). Studies show that third-party reporting increases compliance. See, e.g., James Alm, John A. Deskins & Michael McKee, Third-Party Income Reporting and Income Tax Compliance 3–4 (Andrew Young Sch. of Policy Studies, Working Paper No. 06-35, 2006), (on file with the Columbia Law Review) (using experimental methods to show that third-party reporting increases compliance).

7. Tax Expenditures. — Finally, another difference between cor­porate and shareholder taxes is that shareholder taxes have less influence on managers. In theory, since shareholder taxes affect a share­holder’s return, a manager should try to minimize them. But tax-exempt and foreign investors do not want managers to devote resources to these efforts—since these shareholders would not benefit—and shareholder taxes also have a less immediate impact on reported earnings. 136 Halperin, Mitigating Inequity, supra note 110, at 645 (“[S]ince only the corporate tax affects reported corporate earnings and earnings per share, the shareholder’s tax burden on distributions may not affect corporate decision-making, except to the extent it raises the cost of capital.”). Since managers do not focus on shareholder taxes, tax expenditures have less influence on corporate behavior when included in shareholder taxes, instead of in corporate taxes. Of course, this is a disadvantage only if the relevant tax expenditures are good policy. If they are not, reducing their impact is actually an advantage. 137 Even then, a further issue in relying less on the corporate tax—so the gap between it and the personal rate widens—is that the same tax expenditure becomes more valuable for pass-through businesses (in avoiding the higher personal rate) than for corporate businesses (in avoiding the lower corporate rate). See Toder & Viard, A Proposal, supra note 3, at 19 (explaining that tax deductions are more valuable for flow-through businesses). A solution is to use credits, instead of deductions, so corporations’ generosity does not depend on the tax rate. See id. (“The corporate tax rate reduction would not directly reduce the tax savings from claiming a credit, which depends on the credit rate rather than marginal tax rates.”).

II. Balancing Component-Rate Distortions: Advantages of Using Two Taxes

Part I laid out key challenges in taxing corporations, on the one hand, and shareholders, on the other. Cutting one rate and raising the other fixes some problems while complicating others. What is the right response to these dueling distortions? Obviously, one answer is to fix the relevant rules, so they cannot be gamed. In theory, if all the holes in each tax were plugged, taxpayers would no longer be able to avoid either one. Each tax would measure income in the same (perfectly accurate) way. But although the rules can certainly be improved, and these improve­ments should be an important priority, it is too optimistic to expect all these planning strategies to be foreclosed. The administrative and pol­itical challenges are too formidable. Instead, it is more realistic to expect at least some gaps in each tax to endure.

So if these dueling distortions are likely to be a perennial challenge, what is the right response? Since each tax has significant gaps, one option is to get rid of at least one of these porous taxes, and to do the best we can with the other. But as this Part explains, the opposite is act­ually a better strategy. Instead of relying on one tax or the other, this Essay’s central recommendation is to use both. Although it may seem counterintuitive, the fact that each tax is flawed is all the more reason why both are needed.

A. Value of Keeping Both Taxes

1. Built-In Redundancy. — There are three reasons to rely on both corporate and shareholder taxes. First, the government gets two bites at the apple. If one tax is not collected, the other still is. For instance, even if corporations avoid the corporate tax by shifting income abroad, share­holders are still taxed on dividends and capital gains. To make a system more reliable, engineers often add a backup. 138 John Downer, When Failure Is an Option: Redundancy, Reliability and Regulation in Complex Technical Systems 2–3 (London Sch. of Econ. & Political Sci., Ctr. for Analysis of Risk & Regulation, Discussion Paper No. 53, 2009),‌accounting/CARR/pdf/dps/disspaper53.pdf []. This built-in redundancy is useful not only for planning strategies already in use, but also for ones that have not been developed yet. Just as the Treasury’s integration study twenty years ago did not worry about income shifting and inversions, 139 See Graetz Testimony, supra note 70, at 7 (insisting the Treasury’s recommendation to tax business profits only at the entity level was no longer sound policy, as the internationalization of the economy made taxing at the individual level a more plausible option); see also Michael J. Graetz & Alvin C. Warren, Unlocking Business Tax Reform, 145 Tax Notes 707, 707 (2014) [hereinafter Graetz & Warren, Unlocking Business Tax Reform] (“Whether reducing taxes at the shareholder rather than the corporate level was appropriate in the 1990s, it is no longer a sensible component of business and investment taxation.”). it is not possible to predict what leaks and distortions will arise two decades from now.

Admittedly, built-in redundancy is effective only when two systems have different vulnerabilities. If the two taxes measure income in precisely the same way, there is no advantage in using both taxes instead of only one. Rather, for one tax to work when the other does not, their limitations have to be somewhat uncorrelated. Corporate and share­holder taxes satisfy this condition. As this Essay has emphasized, these taxes define income somewhat differently and thus are avoided with dif­ferent planning strategies. Admittedly, these disparities are conceptually unsatisfying, since it is not clear why two regimes that measure business profits should yield different answers. But ironically, these disparities facilitate built-in redundancy, which is a meaningful silver lining.

2. Repeal Is an Overreaction. — A second reason to use both taxes is that, if the goal is to cure distortions from a tax, repeal is too drastic a remedy. Instead, cutting the rate is often sufficient. As long as the rate is low enough, paying a tax is cheaper than avoiding it. After all, tax plan­ning is not free. Taxpayers have to hire advisors, incur transaction costs, and change business decisions in ways that otherwise are unappealing. These costs are justified only if they are less than the tax.

Obviously, the magnitude of these costs varies for different strat­egies. For instance, moving factories abroad might be cost effective when the U.S. corporate rate is 10% higher than in other jurisdictions, but not when it is only 7% higher. Likewise, the threshold for moving patents offshore could be a 7% gap, while the threshold for intercompany debt could be 5%. If this differential narrows to 3%, shifting income may no longer be worth the trouble.

When tax planning has this sort of cost curve, cutting the rate winnows out costlier strategies. As the rate is reduced, fewer strategies remain cost effective. If even the cheapest strategies involve significant costs, rate cuts can meaningfully ease distortions—indeed, almost as eff­ectively as repealing the tax—while still collecting some revenue.

3. Repeal of One Tax Puts More Pressure on the Other. — Third, this revenue is another advantage of using both shareholder and corporate taxes. If one of these taxes is eliminated—instead of merely cut—the surviving tax has to be even higher (assuming the two taxes, in comb­ination, have to meet a revenue target). For instance, assume the goal is to mitigate distortions from both corporate and shareholder lock-in. The severity of corporate lock-in depends on the corporate rate (C), while the magnitude of shareholder lock-in depends on the shareholder rate (S). If tax must be collected either from corporations or shareholders, there is a tradeoff: To ease shareholder lock-in (by cutting S), we have to exac­erbate corporate lock-in (by raising C), and vice versa. Cutting one tax more than necessary puts more pressure on the other. 140 See, e.g., Henrik Jacobsen Kleven & Joel Slemrod, A Characteristics Approach to Optimal Taxation and Tax-Driven Product Innovation 25 (Sept. 2009) (unpublished manu­script), [] (“[T]ypically, it is not optimal to eliminate distortions completely: it is better to have several small distortions than to have large distortions somewhere and none elsewhere.”). If a 0% rate is used for one, when a 10% rate would be just as effective at eliminating the relevant distortion, the other rate has to be correspondingly higher, and thus more distortive. 141 The assumption here is that raising the rate increases the appeal of tax planning and thus induces taxpayers to use increasingly costly strategies. Admittedly, though, this acceleration should continue only up to a certain point. When the rate is high enough, all plausible strategies are already in use.

B. Challenges in Keeping Both Taxes

1. Combined-Rate Decisions: Irrelevance of Rate Allocation. — To be clear, although using two taxes can ease distortions from component-rate planning (such as corporate and shareholder lock-in), it cannot do so for combined-rate planning. For instance, only the combined rate matters when taxpayers decide to use a corporation instead of a partnership or, for that matter, to save money instead of spending it. For these com­bined-rate choices, whether tax is collected from corporations or share­holders usually is irrelevant; it matters only in changing the combined rate (e.g., by altering the present value of what is actually collected) or in shifting the incidence of the tax, as noted above. 142 See supra section I.A.4.b (discussing the effect of the allocation between cor­porate and shareholder rates on combined-rate strategies); supra section I.B.6 (discussing incidence).

2. Imperfect or Unstable Coordination. — Moreover, even for com­ponent-rate planning, using two taxes is not a panacea. Although relying on both taxes has distinct advantages, a key challenge is that it is not easy to coordinate corporate and shareholder rates so they always aggregate to a particular combined rate. After all, an important rationale for using two taxes is that, if one is avoided, the other is still collected. When this happens, only a portion of the combined tax is collected, rather than the entire amount. This means the combined rate does not apply uniformly to all economic activity. Of course, these disparities are likely to be even greater if the system relies on only one tax, which sometimes can be avoided entirely. But even so, using two taxes is not a complete remedy. It still is necessary to shore up both taxes, so they are actually collected. In addition, the challenge is not just to coordinate rates now, but also over time. This is particularly difficult, since Congress constantly tinkers with tax rates. 143 Cf. Schizer, Realization as Subsidy, supra note 88, at 1579–82 (discussing fluct­uations in the capital gains tax rate over time).

Ideally, rates would be coordinated so one adjusts automatically when the other is avoided or changed. Current law does not supply this sort of coordination, but there are various ways to do so. For example, a reduced shareholder rate for dividends could be available only if the firm has paid tax on the distributed profits. Alternatively, dividends could be deductible only if the firm withholds shareholder tax on the distribution. Likewise, shareholders could reduce their tax bills by claiming a credit for tax the corporation already has paid; under this “imputation” system, shareholders would backstop any tax the corporation was supposed to pay. By cutting its tax bill, a corporation would reduce the shareholder’s credit and thus increase the shareholder’s tax liability. 144 See Graetz & Warren, Unlocking Business Tax Reform, supra note 139, at 708–09 (describing shareholder credit integration and arguing that it would be preferable to current law); Graetz & Warren, Integration of Corporate and Shareholder Taxes, supra note 45 (manuscript at 8–9) (discussing shareholder credit integration).

Instead of an automatic adjustment when a taxpayer avoids one of the taxes, a less ambitious approach is an adjustment when Congress changes the rate schedule. For instance, Congress could impose a stat­utory relationship between corporate, shareholder, and personal rates, so that a change in one automatically triggers adjustments in the others. Obviously, Congress would be free to repeal this formula (or to override it explicitly) when changing rates. Yet if it does not do so, the desired balance is maintained as rates change. 145 I am grateful to Professor Brennan for this suggestion.

3. Administrative Costs. — Another challenge with using two taxes is that administrative costs are higher. 146 See, e.g., R. Glenn Hubbard, Corporate Tax Integration: A View from the Treasury Department, 7 J. Econ. Persp. 115, 116 (1993) (noting the current system of cor­porate taxation imposes significant administrative costs). The government has to administer both taxes, and taxpayers have to comply with them. With one regime, administrative costs are likely to be lower. 147 For instance, in their 2014 proposal to repeal the corporate tax and replace it with a mark-to-market tax on shareholders, Professors Eric Toder and Alan Viard empha­sized the advantage of eliminating the costs of administering the corporate tax. Toder & Viard, Major Surgery Needed, supra note 3, at 26 (“Depreciation schedules, amortization rules, inventory accounting, uniform capitalization, and a host of other complexities would not need to be considered with respect to corporations.”). This is not a trivial advan­tage, since these regimes are complex. The corporate tax, in particular, is extraordinarily intricate.

Even so, the administrative savings from repeal may not fully materialize for two reasons. First, if only one tax is used, it will have to be shored up, 148 See supra sections I.B–.C (surveying gaps in corporate and shareholder taxes). and these enhancements are likely to increase compliance and enforcement costs. Second, if taxpayers believe the repealed regime may someday be reinstated, their compliance costs will not go down as much. For instance, if the corporate tax is repealed, would corporations be confident enough in the permanence of this reform to stop tracking their tax basis in assets? In any event, even if a second tax adds admin­istrative costs, this is the price of built-in redundancy and the other advantages of two taxes.

III. Implications for Incremental Reforms

As Part II emphasized, the fact that corporate and shareholder taxes have different gaps allows them to backstop each other. 149 See supra section II.A.1. One tax can still raise some revenue, even when the other is avoided. 150 See supra section II.A.1. In addition, using two taxes allows each one to have a lower rate. 151 See supra section II.A.2. As a result, plan­ning strategies that avoid only one tax become less appealing. 152 See supra section II.A.2.

In theory, if all the gaps in one tax are eliminated, the other tax is no longer needed as a backstop. But as a practical matter, even the most ambitious reforms are unlikely to plug all the gaps in one tax (without creating new ones). Part IV argues that even fundamental reforms ben­efit from using both taxes, instead of only one. 153 See infra Part IV.

The same is true of more modest reforms that preserve the broad outlines of current law. This Part focuses on modest reforms, which, as this Essay has emphasized, should preserve both taxes. 154 Admittedly, this sort of targeted reform is still valuable even if corporate profits are subject to only one tax. Yet the use of both taxes affects the need for these reforms in competing ways. On the one hand, more reforms are needed, since Congress has to keep both taxes in working order. On the other hand, each individual reform becomes less impor­tant; since each tax has a lower rate, planning strategies to avoid only one tax become less distortive.

While the focus here is on policy, politics obviously also plays a role. So in addition to highlighting important policy choices and offering illustrative examples of targeted reforms, this Part also briefly surveys pol­itical dynamics that are likely to influence business tax reform.

A. Three Interrelated Choices for Incremental Reform

After deciding to keep both the corporate and shareholder tax, Congress still faces three interrelated choices, which this section con­siders in turn. First, what should the total tax burden be on corporate profits? Second, how should this combined burden be divided between corporations and shareholders? This allocation should depend on the severity of the component-rate distortions from each tax: The more dis­tortive tax should have a lower rate. Third, what targeted reforms should be used to shore up each tax? Notably, these reforms can influence the second decision—that is, the balance between the shareholder and cor­porate rates. While the less distortive tax should have the higher rate, the fact that one is less distortive now does not mean it will always be less distortive, since targeted reforms can make a tax less distortive. If one tax is shored up more than the other, its share of the overall tax burden should increase.

1. The Combined Rate on Corporate Profits. — In any event, in using one tax or two, Congress must decide how much to tax corporate profits. In making this choice, Congress should consider a range of familiar issues, including empirical estimates of elasticity, incidence, and distri­butional impact, as well as the nation’s revenue needs and social-welfare func­tion. 155 These issues are the nuts and bolts of public finance. For an introduction, see Louis Kaplow, The Theory of Taxation and Public Economics (2008); Harvey S. Rosen & Ted Gayer, Public Finance (10th ed. 2014). Since these issues are beyond this Essay’s scope, this Essay does not recommend a rate and offers 40% only as an illustrative example.

Even so, in focusing on planning, this Essay’s analysis has four implications for rates. First, the rates on corporate and noncorporate businesses should be comparable, since disparities breed familiar distor­tions. 156 These distortions are well documented in the literature on corporate integration. While a premium can be charged for access to public markets, the prevalence of private-equity financing suggests that this premium should be modest. See, e.g., Michael Cooper et al., Business in the United States: Who Owns It and How Much Tax Do They Pay? 2 (Nat’l Bureau of Econ. Research, Working Paper No. 21,651, 2015),‌abstract=2679689 (on file with the Columbia Law Review) (noting that pass-through entities earned 54.2% of business income in the United States in 2011, compared with 20.7% in 1980). Second, corporate and shareholder taxes are inefficient revenue sources. To ease the pressure on these porous regimes, rates should be low. The revenue cost of low rates is less than it may seem, since effective rates are much lower than stated rates under current law. 157 For example, the Government Accountability Office recently estimated that the average effective tax rate on corporations is only 12.6%, which is approximately one-third of the stated rate. U.S. Gov’t Accountability Office, GAO-13-520, Corporate Income Tax: Effective Tax Rates Can Differ Significantly from the Statutory Rate 14 (2013), http://‌ [] (“For tax year 2010, profitable Schedule M-3 filers actually paid U.S. federal income taxes amount­ing to 12.6 percent of the worldwide income that they reported in their financial state­ments (for those entities included in their tax returns).”). Therefore, lower stated rates can generate the same revenue, as long as they are actually collected. Third, high rates advance distri­butional goals only if the incidence of these taxes is on high-income investors, instead of on low- and middle-income employees. 158 See supra section I.B.6 (explaining how corporate taxes may burden low-income workers). Even then, these inefficient taxes may not pursue distributional goals as effectively as other policy instruments, such as progressive wage taxes. Fourth, Con­gress should be wary of taxing private businesses at a (much) lower rate than the personal rate. Otherwise, employee owners will convert salary to business profits, treat personal consumption as business expenses, and the like. 159 See supra section I.C.3 (discussing salary-shifting strategies).

2. The Balance Between Corporate and Shareholder Rates. — Once Congress picks the combined rate for corporate profits, it must allocate this burden between corporate and shareholder taxes. For each one, a lower rate would induce fewer distortions. But since these taxes have to aggregate to the combined-rate target (e.g., of 40%), cutting one means increasing the other. If these taxes are equally distortive, the goal should be for both taxes to be as low as possible, so the rates should be the same. Each should be 22.5%, which is the minimum rate for both taxes that aggregates to 40%. 160 See supra note 35 (reaching this conclusion). If one is lower than 22.5%, the other has to be higher.

Yet these rates should not necessarily be the same; rather, if one tax is more distortive than the other, its rate should be lower in order to mitigate its (more severe) distortions. If one is slightly more distortive, it should have a modestly lower rate. But if one is significantly more dist­ortive, it should have a much lower rate. If the corporate tax induces more planning—as seems likely, although this empirical question cannot be resolved here—its rate should be lower. 161 See Graetz & Warren, Integration of Corporate and Shareholder Taxes, supra note 45 (manuscript at 5) (“Given the ability of multinational corporations . . . to shift items of income and deduction among countries to obtain tax advantages, and even to change the residence of the parent company, it is the corporate, not the shareholder, rate that needs to be reduced today.”). Obviously, this is the oppo­site of current law (35% for corporations and 23.8% for shareholders). 162 See supra section I.A.1. In any event, instead of 22.5% for both rates, a better split would be 20% for corporations and 25% for shareholders or, for that matter, 15% for corporations and 29.5% for shareholders.

3. Adjustments for Targeted Reforms. — Raising the shareholder rate obviously increases the pressure on this tax. In response, targeted reforms should shore up some of its familiar vulnerabilities. With these changes, Congress could increase the shareholder rate even more.

Likewise, Congress could cut the corporate tax a bit less if targeted reforms plug some of its leaks. More generally, the appropriate balance of rates can shift if targeted reforms address some of the distortions dis­cussed above. These reforms are needed, moreover, to enable the com­bined rate that is actually collected to come closer to the combined rate on the books.

B. Political Constraints on Reform

While decisions about rates and reforms should be based on rigorous policy analysis, politics inevitably plays a role as well. So before the next two sections offer examples of incremental reforms, this section provides a few observations about political dynamics that are likely both to constrain and shape business tax reform.

Fortunately, both sides of the aisle recognize the need to reduce the corporate rate and reform the rules for multinationals. Nevertheless, support for reform in the abstract does not translate into a consensus for specific proposals, which are vigorously debated.

Reaching a consensus is difficult for four reasons. First, the business community is divided. To fund a corporate rate cut, firms want to repeal tax breaks for others, instead of tax breaks for themselves. For instance, limits on income shifting are more appealing to sectors that cannot shift income anyway, such as retail and real estate. 163 In contrast, income shifting is a mainstay of tax planning for tech companies. Richard Rubin, U.S. Companies Are Stashing $2.1 Trillion Overseas to Avoid Taxes, Bloomberg (Mar. 4, 2015, 5:00 am), [] (“‘Computing and IT companies especially have a lot of flexibility in where they declare their profits.’” (quoting Joseph Kennedy, senior fellow at the Information Technology and Innovation Foundation)). Similarly, cutting the corporate rate does not benefit pass-through businesses, so they do not want it funded with increases in their taxes, such as repeal of accelerated depreciation. Self-interest also colors the perspective of managers. For instance, since they benefit from retaining earnings, they may resist reforms that encourage distributions.

Second, statutory rates for corporations and shareholders are politically salient. As a result, some Democrats are reluctant to support tax cuts for “rich corporations,” even though the tax burdens workers as well as investors. 164 For example, Senator Elizabeth Warren has called the corporate tax a “rigged game” and has said that “the problem with our corporate tax code” is “not that taxes are far too high for giant corporations, as the lobbyists claim. No, the problem is that the revenue generated from corporate taxes is far too low.” Dave Johnson, Must Watch—Warren’s Warning About the Corporate Tax Giveaway, Campaign for Am.’s Future (Nov. 19, 2015), []. Meanwhile, Republicans face a parallel political con­straint. Although they want to cut the corporate rate, they are reluctant to pay for it with higher shareholder rates. 165 As one commentator put it, “The Republicans running for president all agree on one thing, which is significant since they tend to agree on so little: Cut the capital gains tax top rate.” Ryan Ellis, Every GOP Presidential Candidate Wants to Cut the Capital Gains Tax, Forbes (Jan. 27, 2016, 9:30 am),‌27/capital-gains-tax-rate-cuts-common-to-all-gop-presidential-contenders/#6ff6ad485c7f (on file with the Columbia Law Review).

Third, although rates are politically salient, gaps and planning strategies are not. The average voter does not have the esoteric know­ledge to understand these gaps or to monitor efforts to plug them. Therefore, organized interest groups wield disproportionate influence, which they use to block or water down these reforms. 166 See David M. Schizer, Fiscal Policy in an Era of Austerity, 35 Harv. J.L. & Pub. Pol’y 453, 474–77 (2012) (noting that “[t]he main problem with cutting the corporate tax rate is political” and describing the influence of interest groups on fiscal policy). The result is stated rates that are much higher than effective rates, a combination that has familiar political advantages and policy disadvantages.

Finally, a political downside of cutting the corporate rate is that this tax is invisible to many who bear it. 167 See Shaviro, Decoding the U.S. Corporate Tax, supra note 109, at 12–13 (“And even insofar as voters understand that a corporate tax necessarily hits some set of people, the ambiguity of exactly whom it is taxing may be politically advantageous.”). Shareholders are more likely to focus on taxes they pay than on taxes corporations pay for them. Unso­phisticated investors ignore the corporate tax because they do not write the check, while sophisticated investors know that some of this burden is shifted to workers. Meanwhile, workers are even less aware of the cor­porate tax’s impact on them. As a result, many beneficiaries of this tax cut would not realize they are benefitting and thus would show less grat­itude to politicians for supporting it.

Even so, the political payoff would still be substantial if reforms improve the anemic U.S. growth rate. Rising wages, lower unemploy­ment, and booming real estate and stock markets generate good will (and greater job security) for elected officials. To reap these policy and political benefits, political leaders may seek a “grand bargain,” funding a corporate tax cut with base-broadening reforms and higher shareholder rates. But admittedly, the political case for this package is weaker than the policy case, so its prospects are uncertain.

C. Incremental Reforms to Ease Distortions from a High Shareholder Rate and a Low Corporate Rate

Since the corporate tax is probably more distortive than the shareholder tax, Congress should cut the corporate rate and make up the revenue by increasing the shareholder rate, as noted above. But how large should this shift be? If the combined rate is 40%, should the cor­porate and shareholder rates be 20% and 25%? Or 18% and 25%? Or 15% and 29.5%? Or something else?

As emphasized above, two constraints complicate this shift from corporate to shareholder taxes. First, the shareholder tax also has gaps, which are more likely to be exploited as the shareholder rate increases. Second, a low corporate rate—and, in particular, one that is lower than the personal rate—causes distortions of its own. In response, Congress should consider reforms to ease these distortions. This section gives illus­trative examples of how to do so. Of course, even without this shift from corporate to shareholder taxes, these reforms should be considered. Yet this shift in rates lends greater urgency to these reforms.

In offering examples of incremental reforms, this Essay’s goal is not to offer a comprehensive analysis of any specific reform, advocate for a particular reform package, or argue that these reforms are politically plausible. Although the reforms in this section largely preserve current law, 168 This analysis uses the same methodology as the Treasury’s December 1992 integration report, which favored reforms that “retain current law” and “rely on established principles and rules.” Treasury, Recommendation for Integration, supra note 67, at 6. some are more politically plausible than others. Rather, these examples are supposed to provide a more concrete sense of how the shareholder tax can be shored up and also to emphasize the connection between reforms and rates. This sort of reform allows Congress to raise the shareholder rate—and cut the corporate rate—even more.

1. Tax-Free Appreciation: Charitable Contributions and Stepped-Up Basis. — One reform strategy is to block two familiar ways of avoiding the shareholder tax: contributing appreciated stock to charity and holding it until the shareholder dies. Under current law, these steps permanently shelter share appreciation from income tax. 169 See supra section I.C.1. To plug this gap, one option is to tax the holder of the appreciated stock. To do so, a reform could treat bequests and charitable contributions as taxable sales. Another alt­ernative is to limit the charitable deduction to the taxpayer’s basis 170 For instance, if a taxpayer buys stock for $1 per share and contributes it to charity when it is worth $100, the deduction could be $1 per share instead of $100. As a result, the $99 of appreciation is taxed indirectly (in the form of a reduced deduction). While this reform reduces the subsidy for contributions of stock, it aligns their treatment with con­tributions of cash. There is no reason for them to be different. If the government wishes to provide a more generous subsidy, there are other ways to do so more uniformly, such as increasing the subsidy rate. and to require heirs to use the donor’s “carryover” basis. 171 A more modest approach is to disallow the basis step up only for retained earnings. See Halperin, Mitigating Inequity, supra note 110, at 654; Warren, ALI Integration Report, supra note 24, at 705–06. If Congress has the political will to revoke these tax benefits—and, frankly, it is not clear that Congress does—these reforms would make the shareholder tax a more reliable source of revenue, so steeper cuts in the corporate tax would be feasible.

2. Salary Shifting and Trapped Earnings in Private Firms. — Two other familiar planning strategies complicate any effort to make steep cuts in the corporate rate. As noted above, these strategies become appealing when corporate rates are lower than personal rates (which apply to salaries and noncorporate businesses). First, entrepreneurs take below-market salaries, transforming high-taxed salary into low-taxed co­rporate income. 172 See supra section I.C.3. Second, earnings are retained, compounding at a higher after-tax rate than if they were distributed. 173 See supra section I.C.2.

At first blush, these planning strategies seem to create a difficult tradeoff for Congress in setting corporate rates: Although low corporate rates exacerbate these distortions, they ease other distortions—for instance, by reducing the incentive to shift income 174 See supra section I.B.1. or reincorporate abroad. 175 See supra section I.B.2.

But this tradeoff in using a low corporate rate is not as stark as it first seems, since these planning strategies appeal to different types of firms: One set of strategies is used by multinationals, while the other is popular with family firms. Specifically, income and residence shifting are plausi­ble for multinationals, but not for small family businesses. 176 The fixed costs of this activity arguably are not justified for businesses with modest income. Special tax rules, such as the controlled foreign corporation rules, are more likely to apply to family-owned businesses. In contrast, salary shifting makes sense for family businesses, but not for multi­nationals. After all, for salary shifting to be a viable strategy, the employee has to own the business. Otherwise, she is sharing her salary with other owners, something the CEO of a multinational obviously would not do. Likewise, using a corporation as a tax-advantaged savings vehicle is also especially plausible for a family business, since the corporation can fun­ction like a family investment account.

Since a low corporate rate deters planning by multinationals, but encourages planning by family businesses, a solution is for Congress to offer the low rate only to multinationals. There are three ways to do this. Since multinationals are almost always public companies, one option is for the C-corporation form—and, therefore, the reduced rate—to be available only to public firms. Private businesses would have to be pass-through entities (e.g., LLCs, partnerships, or S-corporations) and thus would not have access to the reduced corporate rate.

Second, if the C-corporation form were still available to private firms, the rate structure should be adjusted so a higher marginal rate applies to family businesses (to deter salary shifting) than to multi­nationals (to deter income shifting). To accomplish this, the top personal rate (39.6%) should apply to the first $30 million of income—or another threshold approximating the earnings of successful family businesses—and a lower rate (e.g., 15% or 20%) should apply above that level. 177 Cf. Dharmapala, The Economics of Corporate and Business Tax Reform, supra note 63, at 14 (“In an admittedly rough way, this would subject the income of smaller corporations to a higher rate (which can be set to ensure neutrality with regard to organizational form) while giving larger corporations the benefit of a lower marginal rate.”). Admittedly, this income would be overtaxed if the regular shareholder rate applied to dividends and capital gains, but taxpayers could avoid this issue by using pass-through entities instead. Another approach, proposed by Professor Daniel Halperin, is to apply the lower rate to active business income, while applying the higher rate to passive income. See Halperin, Mitigating Inequity, supra note 110, at 652 (“The reduced rate also should not apply to investment earnings of an active business corporation that retains profits beyond the needs of the business.”).

Third, a special tax could be imposed when employees own more than a minimum percentage of a firm. 178 This threshold could be defined as a share of both vote and value, and it would use attribution rules (e.g., for family members, controlled corporations, options, etc.). The profits of these firms could be divided into the return on capital (taxed at the lower corporate rate) and disguised labor income (taxed at the higher personal rate). As Professor Edward Kleinbard has proposed, these returns could be distinguished by applying an assumed rate of return to capital invested in the firm (e.g., measured by tax basis). 179 See Edward D. Kleinbard, An American Dual Income Tax: Nordic Precedents 5 Nw. J.L. & Soc. Pol’y 41, 57–58 (2010) (describing model in which business income is split into “labor and capital components” in order to calculate the taxable income). Capital invested could be measured with either the firm’s basis in assets or the investors’ basis in their investments in the firm. A choice is needed about whether to focus on the firm’s net or gross assets at the firm level, and how to treat debt at the investor level. Id. An extra tax could apply to profits above this level. 180 This sort of rule has been used in Nordic countries, which have imposed this tax at the entity level. See id. The bottom line, then, is that a corporate-rate cut implemented in one of these three ways can mitigate distortions from a high corporate rate (e.g., income shifting) without exacerbating distortions from a low corporate rate (e.g., salary shifting).

3. Trapped Earnings in Public Firms. — Yet cutting the corporate rate can create distortions not only in private firms, but also in public ones. Specifically, a low rate can discourage public firms from dist­ributing earnings, as noted above. 181 See supra section I.C.2 (discussing practice of retaining dividends). This is undesirable because the managers of public firms already have familiar agency-cost reasons to hoard cash. By not distributing this cash, managers maintain control over it and can use it in self-interested ways (e.g., for perquisites or pet projects). 182 See Schizer, Tax and Governance, supra note 75, at 10.

Yet targeted reforms can counter this tax incentive to retain earnings so that the corporate rate can be cut without “trapping” earnings in this way. For example, the accumulated earnings tax under current law is supposed to stop retained earnings “beyond the reasonable needs of the business.” 183 I.R.C. § 533 (2012). But this test sets a low bar, since the IRS and courts lack the expertise to apply it effectively; reinvesting in the business is sufficient, even if this is not the best use of capital. 184 See, e.g., Homer L. Elliott, The Accumulated Earnings Tax and the Reasonable Needs of the Business: A Proposal, 12 Wm. & Mary L. Rev. 34, 40 (1970) (“In practice, widely held public corporations have been exempted from the accumulated earnings tax.”). Firms can also keep a sig­nificant amount of cash on hand. 185 See Bardahl Mfg. Corp. v. Comm’r, 24 T.C.M. (CCH) 1030, 1034, 1044 (1965), 1965 WL 1121 (permitting the firm to keep enough cash for one full operating cycle that accounts for turnover in inven­tory, accounts receivable, and accounts payable).

Another approach, known as a “split rate” system, taxes retained earnings at a higher rate than distributed earnings. 186 See Warren, ALI Integration Report, supra note 24, at 641–42. Compared with the accumulated earnings tax, this alternative is easier to administer: Instead of imposing an extra tax only on excess retained earnings—a category that is hard to define—split-rate systems impose an extra cost on all retained earnings. 187 For instance, if the corporate rate is 30% and firms can deduct one-half of each dividend, the effective rate for dividends (15%) is half the rate for retained earnings (30%). This differential is desirable, since dividends enhance corporate governance (by allowing investors to control this cash), but are undersupplied for agency-cost reasons (since managers usually prefer to control cash). 188 See Schizer, Tax and Governance, supra note 75, at 10. As a result, the tax cut is conditioned on a step we want managers to take, instead of ones we do not want to encourage, such as shifting real activity abroad for tax reasons. 189 More fundamental reforms can also counter the incentive to retain earnings. For instance, if shareholders are required to mark their stock to market, retained earnings are taxed currently, to the extent they increase the stock’s value. Imposing an interest charge on dividends, as a way to compensate for deferral, would have a similar effect.

4. Tax-Exempt Shareholders. — Under current law, the shareholder tax does not reach tax-exempt shareholders, but the corporate tax does so indirectly. In relying more heavily on shareholder taxes, then, Congress has two choices: It can give up on this revenue, or it can begin taxing tax exempts on dividends and capital gains. 190 Tax exempts should also be taxed on equity derivatives, which are close sub­stitutes for corporate stock. Arguably, tax exempts should also pay tax on interest, so their debt and equity investments are taxed the same way. 191 As noted above, corporate revenue that funds interest payments to a tax exempt are never taxed under current law, since the corporation can deduct these payments. See supra section I.C.4. A further question is whether this tax should apply only to corporate debt or to other types of debt as well (to avoid creating a tax preference for the debt of pass-through businesses, governments, etc.). See Graetz & Warren, Integration of Corporate and Shareholder Taxes, supra note 45 (manuscript at 104–05).

In setting the rate for this new shareholder tax on tax exempts, one possibility is to offset the corporate rate cut. 192 See Treasury, Integration Study, supra note 5, at 79 (“This Report recommends, in general, retaining the current level of taxation of corporate equity income allocable to tax-exempt shareholders.”). For instance, if the corporate rate falls from 35% to 20%, this 15% of corporate tax can be replaced with an 18.75% shareholder tax. 193 Assume a nonprofit owns a share of stock that earns $100 per share. With a 35% rate, there would be $65 left over to distribute (tax-free) to the nonprofit. If the corporate tax is cut to 20%, so that $80 is distributed, an 18.75% tax on the nonprofit shareholder would bring the after-tax amount back to $65, since 0.1875 * 80 = 15. The assumption here is that a 15% reduction in the tax would lead to a 15% cut in corporate tax collections. But if the rate cut leads to less avoidance, a more modest tax on nonprofits would cover the difference. The rate could also be lower if it applied to interest income as well, since the base for this tax would be broader. Politically, the easiest way to justify this new tax on tax exempts is to argue that it replaces the tax they were already paying indirectly. 194 Notably, tax exempts are already taxed on equity investments in pass-through businesses, since these returns are usually taxed as unrelated business taxable income. I.R.C. § 513 (2012).

Yet in principle, the rate should not turn on what tax exempts already pay, but on a policy judgment about the right level. This assessment should account for social benefits of tax exempts, which vary for different types of exempt organizations. A further question is whether exempting investment returns is the most effective way to encourage the good work that nonprofits do (e.g., in providing health care, education, social services, etc.). As noted above, the tax exemption is arguably less effective at supporting this good work than the charitable deduction, which rewards nonprofits that attract donations instead of ones that invest a surplus successfully. 195 See supra section I.C.4.

5. Foreigner Shareholders. — Like tax exempts, foreign investors are taxed indirectly by the corporate tax, but they (mostly) avoid the shareholder tax under current law. 196 See supra note 127 and accompanying text. Any effort to rely more heavily on shareholder taxes must either forgo this revenue or change the treat­ment of foreign shareholders.

Changing the rules for foreigners is harder than for tax exempts for two reasons. First, a vast network of tax treaties would have to be renegotiated. 197 See, e.g., U.S. Dep’t of Treasury, United States Model Income Tax Convention, arts. 11, 13 (2016),‌Treaty-US%20Model-2016.pdf []. In principle, the United States could override these treaties with legislation, as long as the intent to do so is clear. 198 E.g., Posadas v. Nat’l City Bank, 296 U.S. 497, 503 (1936); Richard L. Doernberg, Legislative Override of Income Tax Treaties: The Branch Profits Tax and Congressional Arrogation of Authority, 42 Tax Law. 173, 198 (1989). But the United States rarely overrides treaties 199 For instance, when the United States in 1986 enacted section 163(j), the branch profits tax, which was arguably inconsistent with existing treaties, it deferred imple­menting the tax until treaties were renegotiated. Reuven S. Avi-Yonah, Tax Treaty Overrides: A Qualified Defence of U.S. Practice, in Tax Treaties and Domestic Law 65, 75–76 (Guglielmo Maisto ed., 2006). and has done so mainly to target abuses. 200 Examples include the branch profits tax, the earnings stripping rule, the multiparty financing regulations, and the reverse hybrid rule. See id. at 75–78. In a sense, shifting the tax to shareholders is a reaction to abuse—in particular, the gaming of source and other rules by multinationals—but the solution is at some remove from the abuse. Second, to collect this tax, the United States would need infor­mation about the holdings of foreigner shareholders, as well as infrastructure to withhold the tax. Although this infrastructure is already in place for dividends, one would have to be created for capital gains, which is not straightforward.

Assuming a tax on foreign shareholders can be collected, would it enhance national welfare? This revenue would add to the U.S. GDP, while allowing a lower rate to govern U.S. taxpayers. But there are two reasons to tread carefully, as noted above. First, the United States benefits from attracting foreign investors, although a low tax is unlikely to deter them given the unique appeal of the U.S. market. Second, if foreign governments respond by raising their taxes on U.S. investors, U.S. revenue would fall as U.S. taxpayers claim foreign tax credits. Therefore, taxing foreign shareholders enhances U.S. welfare only if the United States can still attract enough foreign investment and raise revenue on a net basis.

D. Incremental Reforms to Ease Distortions from a High Corporate Rate

Since the corporate tax is probably more distortive than the shareholder tax under current law, Congress has good reason to cut the corporate rate and make up the revenue by increasing the shareholder rate, as noted above. But these rate adjustments exacerbate other distor­tions, and the prior section canvassed various ways to mitigate them.

Of course, instead of cutting the corporate tax, a different strategy is to shore up this tax. If the corporate tax is strengthened, there is less need to cut this tax (or, at least, to cut it as much). If this strategy is successful, there also is less pressure to increase the shareholder tax (or, at least, to increase it as much). This section offers illustrative examples of how to diminish the need for these rate adjustments by strengthening the corp­orate tax.

Since there is an extensive literature on reforming the corporate tax, 201 For a recent distillation of key issues, see Dharmapala, The Economics of Corporate and Business Tax Reform, supra note 63. only a few examples are offered here, which preserve the broad contours of current law. Again, the goal is not to advocate for particular reforms. Instead, these examples illustrate how to strengthen the cor­porate tax with incremental reforms, while showing how difficult this task is. In addition, this discussion emphasizes the connection between reforms and rates: Targeted reforms to strengthen the corporate tax can enable the cut in the corporate rate—and the offsetting increase in the shareholder rate—to be more modest.

1. Interest. — While this Essay recommends using two taxes, debt—unlike equity—is taxed only once under current law. Because firms can deduct interest, lenders pay the only tax on debt-financed revenue. 202 See supra note 83 and accompanying text. For example, assume a corporation borrows $100 million to buy a factory. If the factory generates $5 million of revenue, and the corporation has to pay $5 million of interest expense, the corporation has no taxable income from the factory. The $5 million interest deduction shields the factory’s $5 million of revenue from tax. Instead, only the lenders pay tax (on the $5 million of interest payments they receive from the firm). Using only one tax creates three problems. First, there is no built-in redundancy, so no tax is collected when lenders are foreign or tax exempt. Second, if the goal is to discourage tax planning, cutting this tax could be as effective as eliminating it. In a sense, the interest deduction is a form of repeal. Instead of cutting the corporate tax on debt-financed income, it eliminates this tax entirely. In doing so, the interest deduction is not an effective way to combat corporate tax planning; on the contrary, this deduction is actually an essential ingredient of many planning strat­egies for avoiding corporate tax. 203 For instance, intercompany debt often is used to shift income out of the United States. See Michael J. Graetz, A Multilateral Solution for the Income Tax Treatment of Interest Expenses, 62 Bull. for Int’l Tax’n 486, 486 (2008) [hereinafter Graetz, A Multilateral Solution] (noting the “increased government concerns with the role of debt in sophisticated tax avoidance techniques”). Third, since no revenue is collected at the corporate level, the investor tax on interest has to be correspondingly higher and thus more distortive.

In response, some of this tax burden should be shifted to corporations. One way of doing so is to offer only a partial deduction for interest. Like travel and entertainment expenses, a portion of interest would be deductible (e.g., 50%). At the same time, Congress could offer an offsetting reduction in the lender’s tax on interest income. For instance, the same fraction of this income could be excluded (e.g., 50%). If the same deduction and exclusion were also applied to dividends, the tax preference for debt would be eliminated, along with the familiar distortions it causes.

Another alternative is to eliminate the interest deduction entirely, while cutting the corporate tax substantially. Broadening the base in this way can help fund the rate cut. At the same time, the treatment of interest and dividends could be conformed at the investor level. For example, if the target combined rate is 40%, the corporate rate could be 15% or 20% (with no interest deduction), while the rate for dividends and interest could be 29.5% or 25%. 204 Admittedly, a full interest deduction is needed for financial firms. Their business model is to make loans with borrowed money, earning a profit on the spread. To tax only this spread, an interest deduction is needed. For example, if a firm borrows $100 million at 2%, and lends it at 3%, it will have $3 million of revenue and $1 million of profit. If the $3 million of revenue is taxed at 20%, the $600,000 tax is equivalent to a 60% tax on the $1 million profit. To address this issue, banks could have their own tax regime, like securities dealers and insurance companies under current law. If such a regime is developed, there are good reasons to conform the treatment of debt and equity. Eliminating the tax preference for debt—for instance, by offering a cost of capital allowance for equity—reduces the need for burdensome (and potentially ineffective) financial regulation. See Mark Roe & Michael Tröge, Degradation of the Financial System Due to the Structure of Corporate Taxation 7–9 (European Corp. Governance Inst., Working Papers Series in Law Working Paper, No. 317/2016, 2016), (on file with the Columbia Law Review).

2. Income Shifting. — In addition to the debt–equity distinction, another reform priority is to block income shifting. But this problem is excruciatingly difficult to overcome, as long as the United States continues to have a much higher corporate rate than other countries. In fact, the two leading options for international reform would exacerbate this problem, instead of solving it, by offering favorable treatment to foreign earnings: First, a territorial system would tax U.S. multinationals only on U.S. earnings, exempting foreign earnings; second, a minimum tax would tax U.S. multinationals on both their U.S. and foreign earnings, but would apply a reduced rate to foreign earnings. 205 Harry Grubert, a leading economist at the Treasury Department, and Professor Rosanne Altshuler have proposed a minimum tax on all foreign income or just on above-market returns earned abroad. See Grubert & Altshuler, Fixing the System, supra note 65, at 676–78 (providing an analysis of alternative proposals). The Obama Administration has also proposed a minimum tax. See Dep’t of Treasury, General Explanations of the Administration’s Fiscal Year 2016 Revenue Proposals 20 (2015),‌resource-center/tax-policy/Documents/General-Explanations-FY2016.pdf [http://.perma.‌cc/3RZL-CGKX].

In theory, income shifting can be blocked with better source rules, which treat income as really earned in the United States, even if taxpayers say otherwise. A virtue of reforming source rules is that they govern both U.S. and foreign firms, so tougher rules do not put U.S. firms at a comparative disadvantage. 206 See Shaviro, Decoding the U.S. Corporate Tax, supra note 109, at 106 (“[F]or business activity in the U.S., foreign companies are taxed on their U.S. activity in much the same way that U.S. companies would be taxed on such activities . . . .”). But source rules are hard to improve for a familiar reason: Conceptually, the source of income is often unclear. 207 See id. at 107 (“‘[S]ource’ is an idea that verges on having ‘no there there’ . . . .”); Michael P. Devereux, Issues in the Design of Taxes on Corporate Profit, 65 Nat’l Tax J. 709, 725 (2012) (“There is simply no answer to the question: in which country is profit generated? All . . . elements of the company’s activities play a part in generating worldwide profit.”). For instance, profit from a cellphone could be characterized as originating not only from where its technology was invented, but also from where it was manufactured or sold. Since source rules resolve this sort of issue in arbitrary ways, they are easy to manipulate. 208 See Edward D. Kleinbard, The Lessons of Stateless Income, 65 Tax L. Rev. 99, 143 (2011) (arguing that source rules are “largely artificial” and “devoid of any conceptual foundation”). But see Mitchell A. Kane, A Defense of Source Rules in International Taxation, 32 Yale J. on Reg. 311, 314 (2015) (defending source rules as potentially coherent). Likewise, when value is added in different jurisdictions, the allocation of profit among them is malleable. Multinationals include low-tax affiliates in the production process and allocate disproportionate profits to them. 209 Current law accounts for each corporation separately, even if they are commonly owned. Although affiliates are supposed to pay the same “arm’s length” price as an unrelated party, this standard is famously malleable, especially if there are no comparable transactions with third parties. See Reuven S. Avi-Yonah, Kimberly A. Clausing & Michael C. Durst, Allocating Business Profits for Tax Purposes: A Proposal to Adopt a Formulary Profit Split, 9 Fla. Tax Rev. 497, 501 (2009). This internal accounting has little significance—aside from cutting the U.S. tax bill—since the same shareholders own all these affiliates.

In response, a familiar alternative is to ignore this internal accounting. Instead, income can be allocated with a formula based on the location of the firm’s employees, property, and sales. 210 See Rosanne Altshuler & Harry Grubert, Formula Apportionment: Is It Better than the Current System and Are There Better Alternatives?, 63 Nat’l Tax J. 1145, 1145 (2010) (“Under formula apportionment . . . income allocated to a jurisdiction depends on the share of worldwide measurable factors such as capital, payrolls and sales that are located there.”). Yet this sort of formula also can be manipulated in familiar ways, for instance, by moving employees and capital to low-tax jurisdictions. 211 See id. at 1151; James R. Hines, Jr., Income Misattribution Under Formula Apportionment 2 (Nat’l Bureau of Econ. Research, Working Paper No. 15185, 2009), [] (discussing how this formula “creates incentives to modify . . . operations in order to reduce associated tax burdens”). At first blush, sales may seem harder to manipulate, since firms want to sell their products in the United States. 212 See Avi-Yonah et al., supra note 209, at 508–10 (“The key advantage of a sales-based formula is that sales are far less responsive to tax differences across markets . . . .”). But firms can still access the U.S. market without (technically) having any U.S. sales: They can sell to an independent distributor in a low-tax jurisdiction, which then sells in the United States. Even if these avoidance strategies are blocked, the formula still has to be coordinated with other jurisdictions to prevent double taxation. This is a tall order, since the OECD opposes formulary apportionment. 213 See Org. for Econ. Co-operation & Dev., Action Plan on Base Erosion and Profit Shifting 14 (2013), [‌‌3VZB-PEH4] (“[T]here is consensus among governments that . . . a system of formulary apportionment of profits is not a viable way forward; it is also unclear that the behavioural changes companies might adopt in response . . . would lead to investment decisions that are more efficient . . . than under a separate entity approach.”).

Instead of general changes in source rules, narrower reforms can target especially mobile deductions and income, such as interest deduc­tions and intellectual property income. For instance, firms should not be allowed to source interest deductions in the United States—where they offset high-tax income—if the loan generates income in low-tax juris­dictions. A formula can block this strategy, at least to an extent, by allocating interest based on each affiliate’s earnings, assets, and equity. 214 After all, if multinationals borrow in the United States to invest in low-tax jurisdictions, this matching of high-tax deductions with low-tax income subsidizes overseas investment. See Graetz, A Multilateral Solution, supra note 203, at 492–93 (proposing a worldwide formula based on assets or income); Mihir A. Desai & Dhammika Dharmapala, Interest Deductions in a Multijurisdictional World 3–4 (Univ. of Chi. Law Sch., Coase-Sandor Inst. for Law & Econ., Working Paper No. 725, 2015), http://chicagounbound.‌ [http://‌] (analyzing ways in which worldwide debt cap or net financing deduction (NFD) regimes ease ownership distortions and also address concerns about base erosion). In this spirit, the OECD recently released a discussion draft to explore various options. Org. for Econ. Co-operation & Dev., BEPS Action 4: Interest Deductions and Other Financial Payments 2 (2014), []. Likewise, a tech firm’s profits derive largely from intellectual property. Instead of allowing firms to choose where to source these profits, a formula (or other rule) can allocate them based on the location of research or sales. 215 Graetz & Doud, supra note 59, at 426 (proposing to source intellectual property (IP) income based on the location of research or sales). In contrast, current law leaves multinationals considerable discretion to source IP income where they want. For instance, although the sale of IP for a fixed price is sourced in the seller’s resident country, the sale for a contingent price is generally sourced where the IP is used. While this sort of narrow reform can offer some relief, income shifting will continue to be a serious problem as long as the corporate rate in the United States remains uniquely high. Thus, income shifting is a compelling reason to cut the corporate rate, since other reforms are unlikely to solve this problem on their own. 216 Professors Toder and Viard estimate that a reduction in income shifting offsets a meaningful portion of the revenue loss from cutting the corporate rate: If the rate is cut to 15%, corporate tax revenue declines by $212 billion without any change in income shifting, but by only $162 billion once this change is taken into account. Toder & Viard, A Proposal, supra note 3, at 48.

3. Residence Shifting. — In contrast, cutting the corporate rate is not the only effective way to eliminate the appeal of inversions. A territorial system accomplishes this goal by exempting a U.S. firm’s foreign earnings from U.S. tax. Like foreign firms, they pay only foreign tax on foreign operations and can repatriate earnings tax-free. At first blush, a territorial system seems to collect substantially less revenue, but this actually may not be the case, since the effective rate on foreign earnings is already low. 217 The Treasury estimates a loss of $130 billion over ten years from switching to a territorial system, while earlier estimates by the Treasury, the Joint Committee on Taxation, and the Congressional Budget Office have estimated a loss of between $40 billion and $76 billion in that window. President’s Econ. Recovery Advisory Bd., Report on Tax Reform Options: Simplification, Compliance, and Corporate Taxation 90 (2010),‌Report.pdf []. Other commentators estimate that the shift will involve no substantial revenue losses, on the theory that firms are already using deferral and foreign tax credits to shelter earnings. See, e.g., Eric Drabkin, Kenneth Serwin & Laura D’Andrea Tyson, Implications of a Switch to a Territorial Tax System in the United States: A Critical Comparison to the Current System 5 (2013) (unpublished working paper),‌20of%20Territorial%20Tax_Nov2013.pdf []. Even so, an important downside of a territorial system, noted above, is that the incentive to shift income persists and could even grow stronger.

However, if the United States continues to tax worldwide income, inversions will remain a challenge, unless Congress cuts the corporate rate substantially. For example, section 7874 blocks some inversions, while allowing others. 218 See I.R.C. § 7874 (2012) (taxing an inverted firm as a U.S. firm if it does not have sufficient business assets in its new jurisdiction of incorporation and if a sufficiently large percentage of the inverted corporation is owned by shareholders of the predecessor domestic corporation). For a discussion of inversions, see supra section I.B.2. Although the Treasury has introduced a series of regulations to tighten up the rules, a truly comprehensive ban on inversions would be overbroad. For instance, if the rule were “once a U.S. firm, always a U.S. firm,” it could deter mergers with foreign firms that are not tax motivated, while also not stopping new businesses from org­anizing as foreign corporations. 219 Another strategy is to use the location of the headquarters instead of the site of incorporation to define residence, on the theory that managers might be reluctant to move. Yet if they do move, this avoidance strategy becomes even more distortive, especially if there are positive externalities in having the headquarters (and highly skilled workers) in the United States. Given these difficulties, current law does not seek to stop all inversions, but to make them incrementally more costly so that the trickle does not become a flood. 220 Daniel Shaviro, The David R. Tillinghast Lecture: The Rising Tax-Electivity of U.S. Corporate Residence, 64 Tax L. Rev. 377, 413 (2011) (“Rising electivity of the worldwide residence-based U.S. corporate tax is both plausible and, to a modest degree, demonstrable . . . . Yet complete electivity appears to remain far off.”). This task is more daunting when the corporate rate is 35%, instead of 15% or 20%.

IV. Implications for Fundamental Tax Reform

So far, this Essay has focused on incremental reform, but its core argument also applies to more ambitious reforms: Even when funda­mental reforms seem economically comparable, they can induce different planning strategies, which usually are easier to block with two taxes than with one. This Part assesses the vulnerability of five fundamental reforms to the various planning strategies discussed above: first, the comprehensive business income tax; second, the “dual BEIT”; third, mark-to-market accounting or interest charges for shareholders; fourth, imputation; and fifth, corporate cash-flow taxation. While a comp­rehensive analysis of these proposals would fill several volumes, the modest goal here is to show the relevance of this Essay’s analysis to fundamental reforms.

A. Comprehensive Business Income Tax

One option, which the Treasury called the “comprehensive business income tax” in its 1992 integration report, is to replace current corporate and shareholder taxes with a single entity-level tax. 221 See Treasury, Integration Study, supra note 5, at 119. This proposal repeals the shareholder tax on distributions and possibly also on capital gains. 222 The Treasury discusses different ways to treat capital gains, without specifying a particular approach. For instance, to avoid taxing capital gains from retained earnings, the Treasury considers using a dividend reinvestment plan or relying on a purchaser’s built-in capital loss to offset a seller’s capital gain. Id. at 211. Similarly, businesses no longer deduct interest, and creditors do not include it.

This proposal offers four advantages. First, aligning the treatment of debt and equity eases an important distortion under current law. 223 See supra section I.B.3. Second, since debt features prominently in many tax-planning strategies, CBIT defangs these strategies. Third, a high entity-level tax eliminates the incentive to retain earnings or recharacterize salary as corporate income. Fourth, if capital gains are not taxed, shareholder lock-in ceases to be an issue.

Although these advantages are significant, they come at a steep cost. A high entity-level tax breeds familiar distortions, including income and residence shifting and corporate lock-in. 224 See supra section I.B (describing planning strategies if the shareholder rate is zero and all tax is collected from the corporation). CBIT forgoes the advantages of built-in redundancy by collecting tax only once. Moreover, eliminating the shareholder tax, instead of merely cutting it, goes further than necessary to ease shareholder-level distortions.

B. Dual BEIT

An important problem under current law, emphasized above, is that debt and equity are taxed differently. A firm can deduct its cost of capital in issuing debt (since interest is deductible), but not in issuing equity (since dividends are not deductible). In principle, the treatment of debt and equity can be conformed in two different ways: On the one hand, a deduction for the cost of capital can be disallowed for both; on the other hand, this deduction can be allowed for both. The reform discussed in the last section, CBIT, uses the first strategy by repealing the deduction for interest expense. In contrast, another option is to authorize a deduction for equity. Put another way, the current (nonde­ductible) treatment of equity can be extended to debt (as CBIT does), or the current (deductible) treatment of debt can be extended to equity.

This second approach, which offers corporations a deduction for the cost of equity capital, is a prominent feature of the “dual BEIT.” This proposal’s author, Professor Kleinbard, seeks to tax corporate stock with rules based on current law’s treatment of debt. Every year, a firm would deduct (and investors would include) an amount based on its cost of capital, regardless of whether the firm issues debt or equity. 225 For example, if a firm issues $5 billion of stock and $5 billion of debt, pays no current interest or dividends, and has a cost of capital of 6%, it deducts (and investors include) $600 million each year. In effect, dual BEIT extends the original issue discount rules of section 1271 through section 1275 to corporate stock. See Edward D. Kleinbard, We Are Better than This: How Government Should Spend Our Money 400 (2015) (replacing firm-level interest deductions with cost-of-capital allowance, which is allowed for equity as well as for debt). The amount that the firm deducts (and investors include) is based on what the firm’s cost of capital is assumed to be, and does not necessarily equal the amount the firm pays out in a given year.

Yet like current law’s treatment of interest, this approach has an important disadvantage: Since this assumed yield is taxed only at the shareholder level, not at the corporate level, there is no built-in redun­dancy. As a result, when investors are foreign or tax exempt, no tax is collected.

In addition to the investor-level tax, the dual BEIT also has an entity-level tax, but its use of two taxes is not consistent with the approach recommended in this Essay. The difference is that Professor Kleinbard divides a business’s return into two components—normal returns and above-normal returns—and uses a separate tax for each: Investors pay the only tax on normal returns, while the entity pays the only tax on above-normal returns. Therefore, the dual BEIT taxes each component of the return once, and only once. In contrast, this Essay recommends col­lecting some tax at both levels for both types of returns: Normal returns should be taxed at both the entity and investor level, as would above-normal returns. In a sense, Professor Kleinbard is slicing a business’s returns vertically (so each type of return is taxed by different taxes), while this Essay recommends slicing returns horizontally (so each type of return is taxed by both taxes). Therefore, even though the dual BEIT deploys two taxes, the division of labor between them is quite different.

Instead of relying on built-in redundancy and low rates to constrain planning, Professor Kleinbard uses more sophisticated rules to address various distortions. For example, he shows that taxing shareholders based on an assumed yield can mitigate lock-in and reduce an employee-owner’s incentive to recharacterize salary as investment return. While these reforms are promising, they could still be used in a system that taxes the entire return at both levels and thus offers built-in redundancy. 226 For example, investors could be taxed initially on the assumed yield, but a subsequent adjustment could be added at realization, as in the contingent debt regime under current law. See Treas. Reg. § 1.1275-4 (2012). To ensure that an entity-level tax is also collected, a firm could be allowed to deduct only a portion of its cost of capital. For instance, the firm could deduct half of its assumed yield, as well as half of any additional amounts it pays in the subsequent adjustment. Shareholders would include both of these amounts, and the corporate and shareholder rates would be coordinated to yield the appropriate combined tax rate.

C. Mark-to-Market Accounting or Interest Charges

Another option is to use mark-to-market accounting to strengthen the shareholder tax, as Professors Toder and Viard, as well as others, have proposed. 227 See generally Toder & Viard, Major Surgery Needed, supra note 3, at 29–31 (proposing to replace the corporate tax with a mark-to-market tax on shareholders); Toder & Viard, A Proposal, supra note 3, at 1–2 (presenting a modified version of the 2014 proposal which preserves the corporate tax but with a significantly reduced rate); see also Bankman, supra note 3, at 1347–48 (proposing a mark-to-market or “market-value added” on corporations); Dodge, supra note 3, at 293 (outlining a “two-tier integration proposal” that would abolish the corporate income tax and impose a mark-to-market tax); Knoll, supra note 3, at 24–25 (proposing a mark-to-market tax on corporations). Taxing gains annually—even when shareholders do not sell—offers two important advantages. First, it mitigates shareholder lock-in. Delaying a sale no longer defers capital gains tax. Likewise, bequests and contributions of appreciated stock no longer avoid this tax entirely (since taxpayers have to pay tax on appreciation every year).

A second advantage, which is less familiar, is that a mark-to-market tax on shareholders counters a firm’s tax incentive to retain earnings. Without this reform, if the corporate rate is lower than the personal rate, firms are more tax efficient investors than shareholders, since profits inside the firm are taxed at the (lower) corporate tax, while profits outside the firm are taxed at the (higher) personal tax. Eventually, profits inside the firm also are taxed at the shareholder rate, but this tax is deferred under current law. 228 See supra sections I.C.1–.2. In contrast, this deferral is eliminated under mark-to-market accounting, so investments inside the firm are taxed at both the corporate and shareholder rate. 229 Toder & Viard, A Proposal, supra note 3, at 14. As a result, these investments lose their tax advantage over investments outside the corporation. 230 For example, assume the corporate rate is 0%, shareholder and personal rates are 40%, and a 10% pretax return is available both inside and outside the firm. Under realization accounting, a $1 million investment grows faster if earnings are retained for three years than if they are distributed and invested in a bond: The profit is $198,600, instead of $191,016. Specifically, if earnings are retained in the firm (so they can compound tax-free), the investment’s value grows to $1,331,000 after three years. Distribution of the $331,000 profit yields $198,600 after the 40% shareholder tax. In contrast, if the $100,000 of earnings are distributed each year (and thus are subject to a 40% shareholder tax), the shareholder has only $60,000 each year to invest in a bond, and the 10% return on the bond also is taxable. As a result, the $60,000 paid in the first year compounds to $67,416 (i.e., 60,000 x 1.06 x 1.06), and the $60,000 paid in the second year compounds to $63,600 (i.e., 60,000 x 1.06). When added to the $60,000 paid in the third year, these amounts total only $191,016.
But if the stock is marked to market, the after-tax profit shrinks to $191,016—the same as with the immediate distribution—as long as retained earnings cause a dollar-for-dollar increase in the stock price and shareholders sell a portion of their stock each year to fund the mark-to-market tax. In effect, the $1,000,000 investment grows by only 6% each year, to $1,191,016 after three years.

Yet notwithstanding these benefits, a mark-to-market tax poses fam­iliar valuation and liquidity challenges, which are manageable only for publicly traded assets. 231 If annual payments are too onerous, even the tax on publicly traded shares can apply to only a portion of gains each year. See, e.g., Toder & Viard, A Proposal, supra note 3, at 34 (proposing a smoothing mechanism that taxes 20% of gains each year). Since this reform cannot be used for private businesses, collectibles, or real estate, it creates a tax preference for these investments. Investors may shy away from publicly traded assets, and owners of private firms may hesitate to go public. 232 In response, Professors Toder and Viard propose to tax pre-IPO gains at a reduced rate and to allow this tax to be paid over ten years. Id. at 23–25.

To avoid this sort of distortion, the same timing rule has to apply to all investments. Since a mark-to-market rule cannot do so, a different remedy is needed for shareholder lock-in and trapped earnings. As an alternative, economist Harry Grubert and Professor Rosanne Altshuler propose to delay the shareholder tax until stock is sold, as under current law, but to add an interest charge to compensate the government for this delay. 233 Grubert & Altshuler, Shifting the Burden, supra note 132, at 25. Yet their proposal is less accurate than a mark-to-market rule. Instead of meas­uring the return each year, it uses (imperfect) assumptions to allocate gains across the relevant years, creating planning opportunities when these assumptions are inaccurate. 234 For example, if the asset immediately appreciates, taxpayers have the incentive to keep the stock—a form of lock-in—so some of this gain is allocated to later years. This is all the more true when a taxpayer’s bracket changes during these years. 235 See, e.g., Toder & Viard, A Proposal, supra note 3, at 73–74 (“If the tax rate that prevails in the year of realization [is] assumed to apply throughout the holding period, the incentive to realize in low-tax years would actually be stronger with the deferral charge than under a conventional realization-based tax.”).

In addition to creating new distortions, mark-to-market accounting and interest charges share two other limitations. First, unlike corporate taxes, they do not reach tax exempts and (mostly) miss foreign shareholders as well. Second, both proposals are likely to encounter political opposition. Rules that impose interest charges under current law are unpopular, as is taxing gains that could prove temporary. 236 For instance, taxpayers generally have to pay an interest charge on gains from Passive Foreign Investment Companies (PFICs) and thus shy away from investing in them. See David Kuenzi, Thun Fin. Advisors, L.L.C., Why Americans Should Never Own Shares in a Non-US Mutual Fund (PFIC) 2 (2016),‌uploads/2016/09/Why-Americans-Should-Never-Own-Foreign-Mutual-Funds-2016.pdf [] (“The tax treatment of PFICs is extremely punitive compared to the tax treatment of similar investments that are incorporated in the U.S.”). While these proposals may be easier to sell if they applied only to high-net-worth shareholders, as David Miller has proposed, 237 David Miller has also suggested imposing mark-to-market accounting only on taxpayers with sufficiently high incomes on the theory that they have the sophistication and liquidity to be taxed in this way. See David S. Miller, A Progressive System of Mark-to-Market Taxation, 121 Tax Notes 213, 213 (2008). this limit would introduce another boundary for taxpayers to game.

Therefore, in shoring up the shareholder tax with a mark-to-market rule or interest charges, the goal should not be to eliminate the corpor­ate tax—even though this is what many commentators have suggested—but to cut it. 238 For example, in 2014 Professors Toder and Viard proposed to use mark-to-market accounting to eliminate the corporate tax. Toder & Viard, Major Surgery Needed, supra note 3, at 26. In so doing, they followed Victor Thuronyi and David Shakow. See Shakow, supra note 3, at 1134–37; Thuronyi, supra note 3, at 110. In contrast, Grubert and Professor Altshuler proposed to keep the corporate tax, but at a 15% rate. Grubert & Altshuler, Shifting the Burden, supra note 132, at 25. Notably, in a revised version of their proposal in 2016, Professors Toder and Viard opted to keep a 15% corporate tax. See Toder & Viard, A Proposal, supra note 3, at 12. There are two reasons for this. First, if the corporate tax is repealed, tax exempts would escape tax entirely, and foreigners would (mostly) do so as well, unless the treatment of these taxpayers changes. 239 One way to avoid these gaps is to collect this mark-to-market tax from the corporation instead of from shareholders. Professor Halperin has suggested imposing a tax on publicly traded corporations, which would be based on changes in their market value in the relevant tax year. Professor Halperin considers pairing this tax with an imputation system, so shareholders can claim a credit for tax paid by the corporation. As long as this credit is not available to tax exempts and foreigners, tax is still collected (indirectly) from them. See Daniel Halperin, Saving the Income Tax: An Agenda for Research, 24 Ohio N.U. L. Rev. 493, 501, 507, 509 (1998). Second, collecting some tax from corporations relieves pressure on the shareholder tax. The shareholder rate does not have to increase as much, so shareholder distortions are not as severe (e.g., at the boundary between mark-to-market and realization). 240 In proposing a mark-to-market shareholder tax to replace the corporate tax, Professor Joseph Dodge offers a different reason to retain a second tax: “offer[ing] a vehicle to deliver modest tax preferences.” Dodge, supra note 3, at 308. There­fore, even when fundamental reforms strengthen the shareholder tax, two taxes still have advantages over one.

D. Imputation

Still another reform option, an imputation system, adheres to this Essay’s core recommendation by collecting tax from both corporations and shareholders. The corporation pays the first tax, and shareholders pay a second tax at their individual marginal rates. 241 See Warren, ALI Integration Report, supra note 24, at 639–41 (discussing the benefits of imputation systems). Since shareholders can claim a credit for tax already paid by the corporation, the corporate tax functions as a withholding tax. For instance, if the corporate tax is 35%, a shareholder whose marginal rate is 40% pays an additional 5% of tax, while a shareholder whose marginal rate is 20% receives a 15% refund. 242 For instance, if a corporation with earnings of $10 per share pays $3.50 of corporate tax and a dividend of $6.50 per share, its shareholders pay tax on the full $10 of earnings per share—instead of on the $6.50 of cash they receive—while also claiming a $3.50 credit for the corporate tax already paid. Under this approach, shareholders in the 40% bracket owe a total of $4.00 per share so—in addition to the $3.50 credit—they pay $0.50 per share. In contrast, shareholders in the 20% bracket owe only $2.00, and thus receive a $1.50 per share refund.

The combined effect is to tax business profits at a shareholder’s individual marginal rate. Since a shareholder’s bracket depends on her other income, using this rate advances distributional goals. Another advantage is that this rate is also used for pass-through entities, as well as for corporate debt. Therefore, as Professor Alvin Warren has empha­sized, imputation systems conform the rates for corporations and pass-through businesses, as well as for corporate debt and equity. 243 See Warren, ALI Integration Report, supra note 24, at 649–52.

If an imputation system is enacted, will it deter the various planning strategies considered in this Essay? Overall, this reform is fairly effective in discouraging these strategies. A key reason is that an imputation system relies on two points of collection, instead of on one. As a result, the corporate tax backstops the shareholder tax. For example, the cor­porate tax can still reach foreign and tax-exempt shareholders indirectly, as long as they cannot claim a credit for this corporate tax. 244 Notably, European systems used to rely on imputation but stopped using these credits when the European Court of Justice ruled that they had to be equally available to foreign investors from other European-jurisdiction nations. See Graetz & Warren, Unlocking Business Tax Reform, supra note 139, at 709. Likewise, the shareholder tax can still backstop the corporate tax. For instance, income that is shifted abroad can still be taxed (eventually) at the shareholder level (when shareholders receive a dividend or sell stock); if no corporate tax was paid, shareholders cannot claim a credit.

Arguably, corporate and shareholder taxes backstop each other more effectively in imputation systems than under current law, since these taxes are better coordinated. When a corporation cuts its tax bill, it reduces the credit available to shareholders. As a result, income shifting and other corporate planning becomes less attractive, since it (potentially) increases the shareholder tax. 245 See Graetz & Warren, Integration of Corporate and Shareholder Taxes, supra note 45, at 12 (“[A]s in the Australian system, the incentives for corporations to shift their income or their domicile abroad could be reduced.”); see also Richard Vann, Corporate Tax Reform in Australia: Lucky Escape for Lucky Country?, 1 Brit. Tax Rev. 59, 65 (2013) (“The imputation system was thought to put a floor under tax planning by Australian listed companies . . . .”); Dan Amiram, Andrew M. Bauer & Margaret Frank, Tax Avoidance at Public Corporations Driven by Shareholder Demand: Evidence from Changes in Shareholder Dividend Tax Policy 2 (Darden Bus. Sch., Working Paper No. 21111467, 2016), (on file with the Columbia Law Review) (“Conceptually, corporate tax avoidance in an imputation system simply shifts tax payments from the corporation to its shareholders.”). Some studies suggest that corporations engage in less planning under an imputation system (e.g., in Australia) and ramp up planning when imputation is repealed (e.g., in Europe). 246 See, e.g., Catherine Ikin & Alfred Tran, Corporate Tax Strategy in the Australian Dividend Imputation System, 28 Australian Tax F. 523, 532 (2014) (finding that Australian managers whose firms pay “franked” dividends are less likely to defer corporate income in advance of an announced reduction in the corporate rate); Amiram et al., supra note 245, at 32 (“[P]ublic corporations in countries that eliminate an imputation system increase corporate tax avoidance by at least 5.5 percent of pre-tax income . . . .”).

Yet this effect should not be overstated, since corporations still have three reasons to cut their tax bills under imputation systems. First, these efforts still help shareholders who cannot claim a credit for corporate tax and thus do not face higher shareholder taxes when the corporation cuts its tax bill. Since tax-exempt and foreign shareholders are likely to be in this position, they will prefer firms with low effective tax rates. Indeed, activist pension funds and large foreign shareholders will likely lobby firms to ramp up corporate tax planning.

Second, taxable shareholders also have reason to join this chorus. Even though cutting the corporate tax bill reduces imputation credits, these credits are not always needed. Shareholders have other ways to avoid shareholder tax, such as deferring sales or contributing appreciated stock to charity. When these strategies are available, share­holders do not need imputation credits and are happy for the firm to cut its corporate tax bill. After all, even if taxable shareholders do not benefit when the corporation’s effective rate is lower than their (effective) shareholder rate, they do not want the corporation to have a higher effective rate. The corporate tax functions as a withholding tax, and shareholders do not want the firm to withhold tax that they may never owe. For instance, shareholders are taxed currently on dividends but not on retained earnings (as long as they do not sell shares). So shareholders need imputation credits only for dividends and want the firm to minimize the corporate tax on retained earnings. 247 Even a modest corporate-tax bill should be sufficient to provide imputation credits for dividends. Firms that fall short can respond not only by paying more corporate tax but also by cutting dividends. Grubert & Altshuler, Shifting the Burden, supra note 132, at 13–16. More generally, an imputation system preserves familiar gaps in corporate and shareholder taxes. 248 A response to this limitation is to pair imputation with other reforms. See, e.g., Toder & Viard, A Proposal, supra note 3, at 2 (suggesting the adoption of mark-to-market accounting paired with an imputation system). Although it weakens incentives to exploit these gaps in some cases, it maintains them in others.

Third, even when shareholders do not benefit from corporate tax planning, managers still have agency-cost reasons to invest in it: Reporting higher (after-tax) earnings could inflate their equity com­pensation or bonuses. This incentive is especially pronounced in the United States, where equity compensation is pervasive. So even though evidence suggests that imputation systems have discouraged corporate tax planning in Europe and Australia, 249 See Ikin & Tran, supra note 246, at 526 (discussing Australia); Amiram et al., supra note 245, at 19, 45 tbl.1 (noting countries in their sample that eliminated impu­tation were all in Europe: France, Germany, Finland, Italy, and Norway). these jurisdictions use equity compensation far less frequently, so the U.S. experience could turn out to be different.

Therefore, it is too optimistic to expect imputation, on its own, to shut down the various planning strategies in this Essay. But imputation still is a valuable step, especially if paired with a lower corporate rate, as well as other reforms to shore up the corporate and shareholder taxes. A key advantage of imputation is that—like current law and unlike some other proposals—it relies on two taxes, instead of on one. Imputation also has the advantage of coordinating these taxes more effectively than current law.

E. Cash-Flow Corporate Taxation

A final alternative, a “cash-flow corporate tax,” is a type of consumption tax. Its defining feature is that capital investments are deducted currently, instead of capitalized. Therefore, this reform avoids the challenge of calibrating depreciation to reflect economic reality (or encourage investment). As in a 401(k) plan, the deduction of invest­ments spares the “normal” (or market-wide) return on capital from tax. Since this return is not taxed anyway, tax planning is not needed to shield it from tax.

But this reform is still supposed to tax above-market returns (for instance, when intellectual property creates market power) as well as disguised labor income (when owners are paid a below-market salary). Therefore, taxpayers still have an incentive to shelter these returns from tax. Although their interest in tax planning is muted, it does not disappear.

Even so, the available planning opportunities depend on how the tax is structured. 250 For example, Professor Weisbach has described planning opportunities that arise when consumption taxes are “open,” which means one counterparty can claim a deduction even if the other counterparty does not have an inclusion. See, e.g., David A. Weisbach, Ironing Out the Flat Tax, 52 Stan. L. Rev. 599, 616–17 (2000) [hereinafter Weisbach, Ironing Out the Flat Tax]. This is especially true for taxes that are “R-based,” which means they disregard financial transactions. See id. As an example, Professor Weisbach shows that taxpayers can enter into derivative contracts and then settle them by either delivering the underlying property or paying cash. See id. An R-based system would account for the delivery of property but not cash. As a result, taxpayers can deliver property if they have a loss and settle in cash if they have a gain. See id. A key variable is whether the tax covers economic value produced or consumed in the United States. If the tax applies only when value is produced here (an “origin-based” tax), firms have an incentive to shift income overseas. Admittedly, there is no need to shift normal returns, which are exempt anyway, but shifting above-market returns is advantageous. This is a significant limitation of the “X-tax,” 251 The X-tax is an origin-based cash-flow corporate tax, which uses a flat rate and offers a deduction for wages. A separate tax on wages, using progressive rates, is collected from workers. See David F. Bradford, The X Tax in the World Economy 3–5 (Ctr. for European Policy Studies, Working Paper No. 93, 2003),‌workingpapers/93bradford.pdf []. an origin-based tax endorsed by Professor David Bradford, Robert Carroll, and Professor Viard. 252 See Robert Carroll & Alan D. Viard, Progressive Consumption Tax: The X-Tax Revisited 20–39 (2012); Bradford, supra note 251, at 3–4. Although they acknowledge this problem, 253 See Carroll & Viard, supra note 252, at 113 (stating that “[when] the correct market value cannot be objectively determined, as will often be true in cases involving above-normal returns, the origin-based tax offers an incentive to relocate investment abroad and to misstate prices to prevent the proper amount of tax from being collected in the United States”). they rely on current law’s (porous) responses, such as transfer pricing and a tax on repatriated earnings. 254 Id. at 114 (“The Bradford proposal calls for U.S. firms and households dealing with foreign related parties to include all inflows both real and financial, from the related parties and to deduct all outflows, both real and financial.”). For an illustrative example, see id. Another prominent origin-based proposal, Professors Robert Hall and Alvin Rabushka’s flat tax, shares this limitation. 255 See Weisbach, Ironing Out the Flat Tax, supra note 250, at 641–42 (noting the pressure exerted on transfer pricing under Professors Hall and Rabushka’s flat tax).

In contrast, there is no incentive to shift income when a tax applies to goods and services consumed here, regardless of where they are produced (i.e., destination-based taxes). Even if the product is developed abroad, this tax still reaches above-market returns from U.S. sales. 256 Carroll & Viard, supra note 252, at 113 (“[B]order adjustment eliminates transfer-pricing problems between firms.”). This advantage motivates Professor Alan Auerbach to recommend a destination-based cash-flow corporate tax 257 See Alan J. Auerbach, Ctr. for Am. Progress, Hamilton Project, A Modern Corporate Tax 3 (2010) [hereinafter Auerbach, A Modern Corporate Tax], http://‌ []. and Professor Michael Graetz to favor a destination-based value added tax (VAT). 258 See Michael J. Graetz, The Tax Reform Road Not Taken—Yet, 67 Nat’l Tax J. 419, 424 (2014) [hereinafter Graetz, Tax Reform Road Not Taken].

While immunity from income shifting is a notable advantage—and some commentators argue that destination-based consumption taxes solve all problems with the corporate tax under current law 259 Dharmapala, The Economics of Corporate and Business Tax Reform, supra note 63, at 25 (“The [destination-based cash flow tax] would solve virtually all distortions from the corporate tax.”). —these taxes still have familiar limitations. For example, international trade rules may block variations that tax wages separately with progressive rates (such as cash-flow corporate taxes, but not VATs). 260 The problem arises because the rebate that firms collect on exports would be based on the corporate rate but some of the tax being rebated would have been paid by workers at a lower rate. If the rebate exceeds the tax that was paid, it could be characterized as an export subsidy. Stephen E. Shay & Victoria P. Summers, Selected International Aspects of Fundamental Tax Reform Proposals, 51 U. Miami L. Rev. 1029, 1048–50 (1997). Another limitation is that (by design) destination-based taxes do not reach exports. As a result, they collect less revenue from debtor nations like the United States, which have to run trade surpluses in the future to compensate for past trade deficits and thus will have more (untaxed) exports than (taxed) imports. 261 Carroll & Viard, supra note 252, at 110 (explaining that because “today’s trade deficits must eventually be followed by trade surpluses,” a border-adjusted tax would raise less revenue and estimating the difference to be $0.97 trillion). A further concern is that destination-based taxes can trigger a one-time transfer from the United States to its trading partners, which could be quite large. 262 According to Carroll and Professor Viard, this transition effect arises because the tax applies to Americans who fund imports by liquidating foreign assets but not to foreigners who finance exports by liquidating U.S. assets. This imbalance should reduce the real value of Americans’ foreign holdings and increase the real value of foreigners’ U.S. holdings, causing a one-time transfer from Americans to foreigners, which Carroll and Professor Viard call “a gift to the world, rather than a gain for the United States.” Carroll & Viard, supra note 252, at 110–11 (estimating this increases the tax burden on Americans by $7.88 trillion and reduces the tax burden on foreigners by $8.85 trillion in present-value terms).

Given these distortions and leaks, relying solely on one of these new taxes—without a second tax as a backstop—can be problematic. For instance, an origin-based corporate-cash-flow tax would not reach above-normal returns that are shifted abroad, but a second tax (e.g., on a shareholder’s capital gains and dividends) would reach them. Likewise, a destination-based cash-flow tax would not reach exports, but a second (e.g., shareholder) tax would do so. In this spirit, Professor Auerbach pairs his destination-based corporate tax with the current personal income tax, 263 Auerbach, A Modern Corporate Tax, supra note 257, at 13 (noting that if the proposed corporate tax reform is implemented, scaling back favorable treatment of capital gains and dividends should be considered). while Professor Graetz proposes to use three taxes: a destination-based VAT, a corporate income tax (at a much lower rate), and a personal income tax (for high earners only). 264 Graetz, Tax Reform Road Not Taken, supra note 258, at 424.

Admittedly, coordinating two taxes can be a challenge. If some activity is subject to only one of them, the combined rate varies for different activity. For instance, if an origin-based cash-flow tax of 20% is paired with a shareholder tax of 25%, only the shareholder tax reaches above-normal returns that are shifted abroad. As a result, the combined rate for this foreign income (25%) is lower than for U.S.-source income (40%). This sort of inconsistency can be conceptually unsatisfying and is also likely to induce distortions. Yet these problems are even worse if the origin-based cash-flow tax stands alone, without a shareholder tax as a backstop. Since foreign income is not taxed at all, the disparity is 40%, instead of 15%. So even though the second tax is not a complete solution, it is still helpful.


To sum up, then, a central challenge in taxing corporate profits is that corporate and shareholder taxes prompt different tax planning. A cut in one tax, which is funded with an offsetting increase in the other, mitigates some distortions and leaks, while exacerbating others. To deal with these dueling distortions, this Essay recommends keeping both a corporate tax and a shareholder tax. If taxpayers are able to avoid one, the other can still collect some tax. Moreover, if the goal is to deter planning that targets only one of these taxes, repealing this tax goes further than necessary. Cutting the rate can also accomplish this goal, while generating at least some revenue.

The two taxes should be coordinated so that, in combination, they equal (or approximate) the rate on noncorporate businesses. The balance of rates under current law also should shift, so the corporate rate is lower than the shareholder rate. In addition, targeted reforms should be considered to shore up both taxes, so the combined rate that is actually collected comes closer to the one on the books.

Although this incremental reform strategy would improve current law significantly, it still leaves problems unsolved. For instance, shareholder lock-in remains a concern, unless Congress charges interest on tax deferral or adopts mark-to-market accounting for publicly traded stock (and derivatives based on it). Likewise, since the coordination between corporate and shareholder rates is likely to be rough, an imputation system can improve it. These more ambitious reforms are appealing, if the political will can be mustered to enact them. But even with more fundamental reforms, there are significant advantages in rely­ing on two taxes, instead of on one alone.