I'm here today first to continue the discussion that Larry Summers began yesterday in his speech on corporate tax shelters by going into greater detail; and then I would like to discuss our project reviewing the treatment of income derived from activities of U.S.- controlled foreign subsidiaries.

I. Corporate Tax Shelters

Deputy Secretary Summers yesterday laid out for you our concerns about corporate tax shelters. He alluded to the administrative actions we have taken recently against particular abusive transactions for example, against step-down preferred by notice and then by regulation, against lease-in, lease-out transactions by revenue ruling, Notice 98-5, dealing with foreign tax credit abuses and to the passage at our instance of a number of specific provisions over the same time period, for instance with COLI (a job waiting to be finished), and with liquidating REITs. This year we have proposed both general legislation, aimed at deterring all parties from entering into tax avoidance transactions, and specific amendments to deal with particular abuses that we have been able to identify.

We at Treasury appreciate that numerous eminent tax practitioners and representatives of Fortune 500 companies have expressed to us their support for taking action against corporate tax shelters, and affirmed our view that there is a serious problem that needs to be addressed. We must staunch the revenue losses, but more importantly we must preserve the system's integrity in which we all have a stake. A view that well-advised corporations can and do avoid their legal tax liabilities by engaging in transactions unavailable to most other taxpayers may lead to a perception of unfairness and, if unabated, may lead to a decrease in voluntary compliance.

We think your support is critical, and that the corporate community will benefit from changes to the system that provide for a more level playing field, between those who are consumers of tax shelter products and those who are not, changes that will help to steer resources away from non-economic use both in the private sector and in the Government.

To date, most attacks on corporate tax shelters have been targeted at specific transactions and have occurred on an ad hoc, after-the-fact basis through legislative proposals, administrative guidance, and litigation. Yet we are hearing that these significant efforts only scratch the surface. It is not possible to identify and address specifically all current and future sheltering transactions. Second, legislating on a piecemeal basis not only complicates the Code but encourages the promoters to ignore and downplay the viability of common law tax doctrines such as sham transaction, business purpose, economic substance and substance over form. And of course, using a transactional legislation approach to corporate tax shelters emboldens some promoters and participants to rush shelter products to market on the belief that in most cases, any fix will be applied on a prospective basis.

Corporate tax shelters are designed to manufacture tax benefits that can be used to offset unrelated income of the taxpayer or to create tax-favored or tax-exempt economic income. Most corporate tax shelters rely on one or more discontinuities in the tax law, or exploit a provision in the Code or Treasury regulations by a literal reading in a manner otherwise not intended by Congress or the Treasury Department. In doing so it appears that they have forgotten what was basic truth in my years of practice, as articulated by Learned Hand 65 years ago in Gregory:

It is quite true . . . that as the articulation of a statute increases, the room for interpretation must contract; but the meaning of a sentence may be more than that of the separate words, as a melody is more than the notes, and no degree of particularity can ever obviate recourse to the setting in which all appear, and which all collectively create.

Corporate tax shelters may take several forms, and any definition must be carefully constructed not to be over inclusive or to fall short of the mark. It must not include benefits which Congress intended. We can, however, identify certain common characteristics that corporate tax shelters typically share. For example, through hedges, circular cash flows, defeasements, or other devices, corporate participants in a shelter often are insulated from any significant risk of economic loss or opportunity for economic gain with respect to the sheltering transaction. Thus, corporate tax shelters are transactions without genuine economic substance. Frequently, the financial accounting treatment of a shelter is significantly more favorable than the corresponding tax treatment, such as when the shelter produces a tax "loss" that is not reflected as a book loss.

We see corporate tax shelter schemes that are marketed as an off the rack product sold by their designers or promoters to multiple corporate taxpayers and that involve property or structures created that are wholly unrelated to the corporate participant's core business. These features distinguish corporate tax shelters from traditional tax planning that set the business objective first and only then the arrangement most favorable taxwise under the law.

Many corporate tax shelters involve arrangements between corporate taxpayers and persons not subject to U.S. tax such that these tax indifferent parties absorb the taxable income from the transaction, leaving tax losses to be allocated to the taxpaying corporation. The tax indifferent parties in effect "rent" their tax exempt status in return for an accommodation fee or an above-market return on investment. Tax indifferent parties include foreign persons, tax- exempt organizations, Native American tribal organizations, and taxpayers with loss or credit carryforwards.

Taxpayers entering into corporate tax shelter transactions often view such transactions as risky because the expected tax benefits may be successfully challenged. To protect against such risk, purchasers of corporate tax shelters often require the seller or a counterparty to enter into a tax benefit protection arrangement. Thus, corporate tax shelters are often associated with contingent or refundable fees, unwind clauses, or insured results.

Our proposals target transactions that make use of these common indicia in an effort to deter tax shelter transactions. This, we hope, will provide an ex ante solution to the corporate tax shelter problem.

Consistent with the shared responsibility of the users, the abettors and expediters, the proposals don't just apply to the taxpayer, but are designed so that all the players bear the risk if the transaction fails unreasonably to comport with established principles.

Under our proposed definition a corporate tax shelter includes transactions for which:

• the reasonably expected pre-tax profit (on a present value basis) is insignificant relative to the reasonably expected net tax benefits, or
  

• the result would improperly eliminate or significantly reduce tax on economic income.

Our basic definition excludes benefits intended by Congress. Thus, for example, low- income housing partnerships whose viability depends upon tax benefits are not within our definition, nor are traditional financial leases that meet the contours of established rulings and case law.

Critics of our proposals have suggested that our definitions are too broad or may create too much uncertainty and thus may penalize otherwise legitimate transactions. We have announced, and repeat here, that we will work with Congress and the corporate community to refine our definition in a manner that will protect from penalty any legitimate, normal-course-of- business transactions. We also invite descriptions of specific cases that require clarification. We have attempted a definition of corporate tax shelter that is narrower and therefore less uncertain than other definitions and formulations used in the Code. Some examples of imprecise, but well understood formulae, already in the law are:

• section 482, which grants authority to reallocate income, deductions etc., between organizations if necessary to prevent evasion of tax or clearly to reflect income;
  

• section 446, which prescribes a change of method of accounting if necessary to clearly reflect income; and
  

• sections 269 and 357, to pick at random two sections that contain as a test, a purpose of tax avoidance or evasion.

Also our definition builds on a firm foundation in court decisions articulated in cases such as ACM and Sheldon.

Thus we strike no new ground in defining the nature of tax shelters. Taxpayers and practitioners have lived with the concepts our definitions embody as they have been enunciated by the courts since the 1920's. Whatever uncertainty is inherent in the law today has been well tolerated. This is really no more than a debate on rules vs. standards. Bright-line/safe-harbor tests, although appropriate in some circumstances, encourage aggressive positions and playing the examination lottery. As Professor James Eustice wrote in 1976, "I personally have viewed some transactions that seem to me to fly only by principles of levitation. ... excessive concentration on technical matters to the exclusion of the broader issues has obviously raised the level of complexity throughout the entire tax system." Standards, in contrast, require the application of common olfactory sense. Moreover some uncertainty may be useful in discouraging taxpayers from venturing too close to the edge, and thereby going over the edge, of established principles.

Our first proposal would modify the substantial understatement penalty for corporate tax shelters. The current 20-percent substantial understatement penalty imposed on corporate tax shelter items can be avoided if the corporate taxpayer had reasonable cause for the tax treatment of the item and good faith. The reasonable cause exception would be deleted and we propose to increase the substantial understatement penalty on corporate tax shelter items to 40 percent, unless the taxpayer:

• discloses to the National Office of the Internal Revenue Service within 30 days of the closing of the tax shelter transaction appropriate documents describing the transaction; and
  

• files a statement with its tax return verifying that such disclosure has been made; and
  

• provides adequate disclosure on its tax returns as to the book/tax differences resulting from the corporate tax shelter item for the taxable years in which the tax shelter transaction applies.

Despite today's heightened substantial understatement penalty for corporate tax shelters, there continues to be a significant number of abusive tax shelter transactions involving corporate taxpayers. The more narrow reasonable cause exception for corporate tax shelters does not appear to adequately deter such transactions. Moreover we are hard pressed to determine a reasonable cause for entering into a corporate tax shelter.

Some have suggested that advance disclosure to the IRS should be sufficient to avoid the penalty and have asked us to consider the establishment of an advance ruling procedure. Under such a procedure, if a transaction is fully disclosed to the IRS in advance, it would be made possible to obtain an expedited ruling from the Service on the tax shelter penalty question without determining the underlying substantive liability questions. Others have suggested a coordinated review process of corporate tax shelters. This could be facilitated by the in-process reorganization of the IRS. We are considering these suggestions. Certainly many of the transactions we seek to eliminate cannot tolerate sunlight.

The next general proposal would also apply at the taxpayer level and would disallow a deduction, credit, exclusion or other allowance obtained in a tax avoidance transaction. Section 269 currently disallows deductions in transactions with a principal purpose of tax avoidance. The provision, as currently applied, has proven insufficient to address many corporate tax shelter schemes, since the courts have allowed taxpayers with some ingenuity to contrive some semblance of business purpose.

A third proposal would affect the promoters and expediters. It would deny deductions for tax advice in implementing a corporate tax shelter and it would impose a 25 percent excise tax on fees received in connection with the purchase and implementation of corporate tax shelters (including fees related to the underwriting). It would, of course, not apply to advice that a transaction would not stand scrutiny, but only to fees in implementation of a shelter.

Many reputable practitioners have advised us that to control the proliferation of tax shelters, the promoter needs to suffer the consequences, as well as the taxpayer. As the ABA put it in its recent testimony: "all essential parties to a tax-driven transaction should have an incentive to make certain that the transaction is within the law."

Another proposal aimed at discouraging overly aggressive promotion would impose on the corporate purchaser of a corporate tax shelter an excise tax of 25 percent on the maximum payment under a tax benefit protection arrangement at the time the arrangement is entered into.

A tax benefit protection arrangement would include a rescission clause, guarantee of tax benefits arrangement or any other arrangement that has the same economic effect (e.g., insurance purchased with respect to the transaction).

If the taxpayer is unwilling to bear the risk of the legitimacy of its transaction, tax benefit protection arrangements permit even the timid to go forward on a "nothing ventured, nothing gained" theory.

We have been discussing provisions affecting taxpayers, their advisors, and tax shelter promoters. A measure of responsibility also lies with those who peddle their tax-advantaged status to aid and abet tax avoidance. We propose that income allocable to a tax indifferent party with respect to a corporate tax shelter would be taxable to such party. This will help ensure that all parties to the transaction will monitor its propriety. The tax on the income allocable to the tax indifferent party would be determined without regard to any statutory, regulatory, or treaty exclusion or exception. All other participants of the corporate tax shelter (i.e., any participant other than the tax indifferent party in question) would be jointly and severally liable for the tax.

• Many corporate tax shelters involve a mismatch or separation of income or gains from losses or deductions. In these transactions, the tax indifferent party absorbs the income or gain generated in the transaction, leaving the corresponding loss or deductions for the taxable corporate participants. Tax indifferent parties often agree to engage in such transaction in exchange for an enhanced return on investment or for an accommodation fee.
  

• The freedom of taxation of tax indifferent parties should not be a commodity to sell to parties engaged in avoidance through tax sheltering.

The next proposal of general applicability would preclude corporations from taking any position (on any return or refund claim) that the Federal income tax treatment of a transaction is different from that dictated by its form if a tax indifferent party has a direct or indirect interest in such transaction.

Exceptions to this rule would apply in cases of tax return disclosure and otherwise as appropriately provided in regulations.

Taxpayers have control over the form of their transactions. It is thus appropriate to impose restrictions on the taxpayer's ability to argue against the form it has chosen in order to whipsaw the fisc.

In addition to the generic proposals I have just summarized, our budget also addresses a number of specific tax-engineered transactions that require substantive fixes.

II. U.S.-Controlled Foreign Subsidiaries

Frequently related to the tax shelter discussion, but of course, involving much broader issues is our review of taxation of foreign source income. As many of you know, this summer we hope to release for comment our study of Subpart F issues, along with the proposed regulations under Notice 98-35 dealing with hybrid entities.

We are conducting this study in a comprehensive manner, relying on evidence and thorough analysis and without a preconceived agenda. We shall neither be bound to, nor ignore, the policies that have guided us for the past three decades. Upon completion, we expect to set forth whatever conclusions a tabula rasa review of the evidence leads us to. We intend to release our study for public comment as the basis for an organized discussion of views on the issues. Of course, taxpayers will also have the opportunity for public comment on the regulations that we indicated in the Notice would be proposed. No formal decisions will be made before review of such comment or before Congress has had adequate opportunity to review the matters involved.

There are a number of principles that seem to have long guided policymakers in determining the appropriate taxation of international income and that will have to be considered in the study. Although there are differences over the application and relative weight to be given to them, these policy goals in the international tax area have been generally recognized.

• Meet the revenue needs determined by Congress in a fair manner
  

• Minimize compliance and administrative burdens
  

• Minimize distortion by, and maintain neutrality of, tax considerations in making of investment decisions
  

• Take due account of the competitive needs of U.S. multinational business, and
  

• Conform with international norms, to the extent possible.

The first three goals apply to income taxation in general.

Raising Revenue Fairly. The first goal is of primary importance. The credibility of our tax system depends upon the perception that revenue is being raised fairly and that the intended tax base is protected from avoidance.

Application of the concept of fairness, however, inevitably produces disagreement. For example, what is a fair tax burden on foreign business income compared to the tax burden on business income from domestic investment? And should there be a lower rate of tax imposed on income from business activities than the rate of tax imposed on income from labor? These fairness questions must be answered with significant popular satisfaction of some significant majority.

Minimizing Compliance and Administrative Burdens. The second goal, minimizing compliance and administrative burdens and avoiding complexity, receives universal acceptance, or at least lip service. The trouble is that simplification frequently comes with a cost.

For example, we might be able to drastically simplify subpart F, by adopting an alternative rule which eliminates the need to distinguish passive business income from active business income and mobile income from non-mobile income. To accomplish this, subpart F could be limited to situations where the effective rate of foreign tax on foreign business income is below a threshold, and would not apply if the income is subject to foreign tax at or in excess of the threshold. We recognize that the attractiveness of this relative simplification would largely depend upon the threshold rate.

Neutrality and Competitiveness. The goals of neutrality and competitiveness call for particular attention, because much of the debate about subpart F has revolved around these objectives. The objectives are sometimes in conflict, for example if reducing undue burdens on competitiveness requires reducing tax on foreign income. To what extent can this be accomplished without distorting investment decisions by favoring foreign investment over domestic investment? At the recent hearing before the Senate Finance Committee on taxation of foreign source income, a number of Members bridled at suggestions that moving corporate situs offshore to minimize involvement with U.S. taxation was a tolerable option. And arguments that investment location decisions are determined by factors other than taxation are at first blush at least inconsistent with suggestions that U.S. taxes are strangling the competitive position of our companies.

Thus we note that, "competitiveness" means different things to different people. To quote a 1991 Joint Committee on Taxation Report, "although the term competitiveness is used frequently, it does not have a consistent definition."

Conforming to International Norms. Similarly, the goal of conforming with international norms is one that can mean different things to different people. As traditionally conceived, conformity requires, not rigid conformity of rates and base among countries, but that we adopt policies, such as a foreign tax credit, that have historically been adopted by developed countries to avoid double taxation. Some, however, would interpret conformity as requiring that we adopt policies that would facilitate world-wide escape from taxation. We do not accept the lowest common denominator as setting the standard in the areas of bribery, environmental regulation or fair labor laws, and we should not accept it in the tax area either.

Before I close, I'd like to address a topic that is of great concern to us in the Government and, from what we are learning, is of growing concern to you: Disclosure of Advance Pricing Agreements.

As most of you know, the IRS recently took the position in litigation that, although APAs contain return information that is protected from disclosure under 6103, they are also "written determinations" subject to disclosure in redacted form under 6110. This position has raised the concerns of some that their confidential information would be disclosed by the IRS.

Although this matter is still in litigation and, therefore, my comments must be limited, let me say a few things about it. First, we at the Treasury, as well as those at the IRS, from Commissioner Rossotti and Chief Counsel Brown on down, care deeply about the APA program and its success, and all of the decisions in this matter are being made with due regard to the viability of the program. Second, we at Treasury have confidence that, with respect to any given APA, if we put the taxpayer and the IRS in a room, an appropriately redacted APA could be agreed that would satisfy the public's reasonable need to know. Third, for purposes of analyzing the disclosure issue, we believe that there may be a distinction to be made between APAs themselves and background files. It appears that the background files contemplated for disclosure under 6110, for example the background files relating to private letter rulings are far more limited in factual disclosure by the taxpayer from the background files relating to APAs. There is some question about whether the current law adequately recognizes such a distinction and, if it does not, legislation to do so may be necessary. And although the Treasury has not formulated a position on any such legislation, those who may be thinking that broader legislation is appropriate, such as legislation to prohibit any disclosure of any contents even of the APA itself, may not be acting in the best long-term interests of the program and may need to refine their analysis.

III. Conclusion

Focusing on corporate tax shelters and on the appropriateness of the current-law rules under Subpart F are major undertakings for the Office of Tax Policy, the Treasury, and the Administration. In important ways, each is a part of a more general set of concerns, not confined to tax shelters per se alone nor to principles of international taxation, but lying in the general ground of the effects of the globalization of the economy on our system of corporate taxation. Your participation, criticism, and suggestions are all critical to the success of these projects. We look forward to working with you in the coming months.