Introduction

The federal government loses both individual and corporate income tax revenue from the shifting of profits and income into low-tax countries. The revenue losses from this tax avoidance and evasion are difficult to estimate, but some have suggested that the annual cost of offshore tax abuses may be over $100 billion per year.¹ International tax avoidance can arise from wealthy individual investors and from large multinational corporations; it can reflect both legal and illegal actions.

Tax avoidance is sometimes used to refer to a legal reduction in taxes, whereas evasion refers to tax reductions that are illegal. Both types are discussed in this report, although the dividing line is not entirely clear. A multinational firm that constructs a factory in a low-tax jurisdiction rather than in the United States to take advantage of low foreign corporate tax rates is engaged in avoidance, whereas a U.S. citizen who sets up a secret bank account in the Caribbean and does not report the interest income is engaged in evasion. There are, however, many activities, particularly by corporations, that are often referred to as avoidance but could be classified as evasion. One example is transfer pricing, where firms charge low prices for sales to low-tax affiliates but pay high prices for purchases from them. If these prices, which are supposed to be at arms-length, are set at an artificial level, then this activity might be viewed by some as evasion, even if such pricing is not overturned in court because evidence to establish pricing is not available.

Most of the international tax reduction of individuals reflects evasion, and this amount has been estimated at around $40 billion.² This evasion has occurred in part because the United States does not withhold tax on many types of passive income (such as interest) paid to foreign entities; if U.S. individuals can channel their investments through a foreign entity and do not report the holdings of these assets on their tax returns, they evade a tax that they are legally required to pay. In addition, individuals investing in foreign assets may not report income from these assets. In 2010, Congress enacted the Foreign Account Tax Compliance Act (FATCA),³ which has recently become effective and requires foreign financial institutions to report information on asset holders or be subject to a 30% withholding rate.

Its consequences for evasion have yet to be determined.

Corporate tax reductions arising from profit shifting also have been estimated, although most of those estimates were based on data prior to the major change in the U.S. international regime as part of the Tax Cuts and Jobs Act (TCJA; P.L. 115-97). That change shifted from a system where dividends paid by foreign subsidiaries to U.S. parents were taxed but income retained abroad was not, to a minimum tax on foreign income. Estimates of the revenue losses from corporate profit shifting varied substantially, ranging from about $50 billion to more than $100 billion.⁴ This activity appears to have increased substantially in recent years. Only one estimate of the revenue loss from profit shifting after the TCJA was enacted has been found, indicating a loss of $77 billion.⁵

In addition to differentiating between individual and corporate activities, and evasion and avoidance, there are also variations in the features used to characterize tax havens. Some restrictive definitions would limit tax havens to those countries that, in addition to having low or nonexistent tax rates on some types of income, also have such other characteristics as the lack of transparency, bank secrecy, and the lack of information sharing, and requiring little or no economic activity for an entity to obtain legal status. A definition incorporating compounding factors such as these was used by the Organization for Economic Development and Cooperation (OECD) in their 2000 tax shelter initiative. Others, particularly economists, might characterize as a tax haven any low-tax country with a goal of attracting capital, or simply any country that has low or nonexistent taxes. This report addresses tax havens in their broader sense as well as in their narrower sense.

Although international tax avoidance can be differentiated by whether it is associated with individuals or corporations, whether it is illegal evasion or legal avoidance, and whether it arises in a tax haven narrowly defined or broadly defined, it can also be characterized by what measures might be taken to reduce this loss. In general, revenue losses from individual taxes are more likely to be associated with evasion and more likely to be associated with narrowly defined tax havens, while corporate tax avoidance occurs in both narrowly and broadly defined tax havens and can arise from either legal avoidance or illegal evasion. Evasion is often a problem of lack of information, and remedies may include resources for enforcement, along with incentives and sanctions designed to increase information sharing, and possibly a move towards greater withholding. Avoidance may be more likely to be remedied with changes in the tax code.

Prior to the TCJA, several legislative proposals had been advanced that address international tax issues. President Obama proposed several international corporate tax revisions which relate to multinational corporations, including profit shifting, as well as individual tax evasion. Some of the provisions relating to multinationals had earlier been included in a bill introduced in the 110^th Congress by Chairman Rangel of the Ways and Means Committee (H.R. 3970). Major revisions to corporate international tax rules were also included in S. 3018, a general tax reform act introduced by Senators Wyden and Gregg in the 111^th Congress, and a similar bill, S. 727, introduced by Senators Wyden and Coats in the 112^th Congress.⁶ This bill had provisions to tax foreign source income currently, which could have limited the benefits from corporate profit shifting. In the 113^th Congress, H.R. 694 (Representative Schakowsky) and S. 250 (Senator Sanders), also would have eliminated deferral. Former Ways and Means Chairman Dave Camp proposed a lower corporate rate combined with a move to a territorial tax system (which would exempt foreign source income). His bill, H.R. 1 (a general tax reform bill), was introduced in the 113^th Congress. Because a territorial tax could increase the scope for profit shifting, the proposal contained detailed provisions to address these issues. A territorial tax proposal with anti-abuse provisions had also been introduced by Senator Enzi (S. 2091, 112^th Congress).⁷

The Senate Permanent Subcommittee on Investigations had been engaged in international tax investigations since 2001, holding hearings and proposing legislation.⁸ In the 111^th Congress, the Stop Tax Haven Abuse Act, S. 506, was introduced by the chairman of that committee, Senator Levin, with a companion bill, H.R. 1265, introduced by Representative Doggett. The Senate Finance Committee also had circulated draft proposals addressing individual tax evasion issues. A number of these anti-evasion provisions (including provisions in President Obama's earlier budget outlines) were adopted in the Hiring Incentives to Restore Employment (HIRE) Act, P.L. 111-147. Subsequently, revised versions of the Stop Tax Haven Abuse Act were introduced in the 117^th Congress: S. 725 (Senator Whitehouse) and H.R. 1786 (Representative Doggett). The Permanent Subcommittee also released a study of profit shifting by multinationals in preparation for a hearing on September 20, 2012.⁹

The TCJA provided for major changes in the international tax regime along with reducing the corporate tax rate from 35% to 21%. Prior law taxed income of foreign subsidiaries only when dividends were paid to the U.S. parent, thus that income retained abroad was not taxed. The new system exempted dividends but imposed a lower tax rate on global low-taxed intangible income (GILTI). GILTI allowed a deemed return for tangible income. It also provided for a deduction for U.S. foreign derived intangible income (FDII) to make holding intangibles in the United States subject to close to the same tax treatment as holding them abroad. The system provides additional incentives for profit shifting because foreign tax credits, while limited to tax on U.S. source income, are limited on an overall basis. In that case, taxes in excess of U.S. tax due in high-tax countries can be used to offset U.S. tax due in low-tax countries.

This report first reviews what countries might be considered tax havens, including a discussion of the OECD initiatives and lists. The next two sections discuss, in turn, the corporate profit-shifting mechanisms and evidence on the existence and magnitude of profit-shifting activity. The following two sections provide the same analysis for individual tax evasion. The report concludes with overviews of alternative policy options.

Where Are the Tax Havens?

There is no precise definition of a tax haven. The OECD initially defined the following features of tax havens: no or low taxes, lack of effective exchange of information, lack of transparency, and no requirement of substantial activity.¹⁰ Other lists have been developed in legislative proposals and by researchers. In addition, a number of other jurisdictions have been identified as having tax haven characteristics.

Formal Lists of Tax Havens

The OECD created an initial list of tax havens in 2000. A similar list was used in S. 396, introduced in the 110^th Congress, which would have treated firms incorporated in certain tax havens as domestic companies; the only difference between this list and the OECD list was the exclusion of the U.S. Virgin Islands from the list in S. 396. Legislation introduced in the 111^th Congress to address tax haven abuse (S. 506, H.R. 1265) used a different list taken from Internal Revenue Service (IRS) court filings but had many countries in common. The definition by the OECD excluded low-tax jurisdictions, some of which are OECD members that were thought by many to be tax havens, such as Ireland and Switzerland. These countries were included in an important study of tax havens by Hines and Rice.¹¹ The Government Accountability Office (GAO) also provided a list.¹²

Table 1 lists the countries that appear on various lists, arranged by geographic location. These tax havens tend to be concentrated in certain areas, including the Caribbean and West Indies and Europe, locations close to large developed countries. There are 50 altogether.

Developments in the OECD Tax Haven List

The OECD list, the most prominent list, has changed over time. Nine of the countries in Table 1 did not appear on the earliest OECD list. These countries not appearing on the original list tend to be more developed larger countries and include some that are members of the OECD (e.g., Switzerland and Luxembourg).

It is also important to distinguish between OECD's original list and its blacklist. OECD subsequently focused on information exchange and removed countries from a blacklist if they agree to cooperate. OECD initially examined 47 jurisdictions and identified a number as not meeting the criteria for a tax haven; it also initially excluded six countries with advance agreements to share information (Bermuda, the Cayman Islands, Cyprus, Malta, Mauritius, and San Marino). The 2000 OECD blacklist included 35 countries; this list did not include the six countries eliminated due to advance agreement. The OECD had also subsequently determined that three countries should not be included in the list of tax havens (Barbados, the Maldives, and Tonga). Over time, as more tax havens made agreements to share information, the blacklist dwindled until it included only three countries: Andorra, Liechtenstein, and Monaco.

A study of the OECD initiative on global tax coordination by Sharman, also discussed in a book review by Sullivan, argues that the reduction in the OECD list was not because of actual progress towards cooperation so much as due to the withdrawal of U.S. support in 2001, which resulted in the OECD focusing on information on request and not requiring reforms until all parties had signed on.¹³ This analysis suggests that the large countries were not successful in this initiative to rein in on tax havens. A similar analysis by Spencer and Sharman suggests little real progress has been made in reducing tax haven practices.¹⁴

Interest in tax haven actions has increased recently. The scandals surrounding the Swiss bank UBS AG (UBS) and the Liechtenstein Global Trust Group (LGT), which led to legal actions by the United States and other countries, focused greater attention on international tax issues, primarily information reporting and individual evasion.¹⁵ The credit crunch and provision of public funds to banks has also heightened public interest. The tax haven issue was revived with a meeting of the G20 industrialized and developing countries that proposed sanctions, and a number of countries began to indicate commitments to information sharing agreements.¹⁶

The OECD currently has three lists: a white list of countries implementing an agreed-upon standard, a gray list of countries that have committed to such a standard, and a black list of countries that have not committed. On April 7, 2009, the last four countries on the black list, which were countries not included on the original OECD list—Costa Rica, Malaysia, the Philippines, and Uruguay—were moved to the gray list.¹⁷ The gray list includes countries not identified as tax havens but as "other financial centers." According to news reports, Hong Kong and Macau were omitted from the OECD's list because of objections from China, but are mentioned in a footnote as having committed to the standards; they also noted that a "recent flurry of commitments brought 11 jurisdictions, including Austria, Liechtenstein, Luxembourg, Singapore, and Switzerland into the committed category."¹⁸ As of May 18, 2012, only one country (Nauru) appeared on the gray list for tax havens and one (Guatemala) appeared on the gray list for financial centers.¹⁹

Many countries that were listed on the OECD's original blacklist protested because of the negative publicity and many now point to having signed agreements to negotiate tax information exchange agreements (TIEA) and some have negotiated agreements. The identification of tax havens can have legal ramifications if laws and sanctions are contingent on that identification, as is the case of some current proposals in the United States and of potential sanctions by international bodies.

More recently, the OECD has focused attention on its Base Erosion and Profit Shifting (BEPS) initiative. Among the elements of this initiative is the Global Forum on Transparency and Exchange of Information for Tax Purposes, which has begun rating countries on various criteria. As of October 2014, it had under way 105 reviews of countries based on various standards.²⁰ The countries are rated as compliant, largely compliant, partially compliant, or noncompliant. As of 2014, 71 jurisdictions had received a full review, with 42 of those rated as noncompliant. Of 34 countries that had undergone only a phase 1 review, which examines the legal and regulatory framework, 12 were not able to advance to the final (phase 2) review, which looks into the implementation of the regulatory framework in practice. As with the evolution of the OECD list, these evaluations focus on one aspect of the characteristics of tax havens.

The European Union also developed, beginning in 2017, a blacklist and greylist of tax havens. Its focus is on harmful tax practices and excludes EU countries. The countries currently on the blacklist are American Samoa, Fiji, Guam, Palau, Panama, Samoa, Trinidad and Tobago, U.S. Virgin Islands, and Vanuatu.²¹ The EU list has a narrower focus as it does not include countries based on their corporate tax rate but largely on issues such as transparency. The EU blacklist been criticized as being ineffective, and the European Parliament has passed a resolution demanding reform of the blacklist by the end of 2021.²²

Other Jurisdictions with Tax Haven Characteristics

Criticisms have been made by a range of commentators that many countries are tax havens or have aspects of tax havens and have been overlooked. These jurisdictions include major countries such as the United States, the UK, the Netherlands, Denmark, Hungary, Iceland, Israel, Portugal, and Canada. Attention has also been directed at a number of states in the United States including Delaware, Nevada, South Dakota and Wyoming²³. Finally, there are a number of smaller countries or areas in countries, such as Campione d'Italia, an Italian town located within Switzerland, that have been characterized as tax havens.

A country not on the list in Table 1, but which is often considered a tax haven, especially for corporations, is the Netherlands, which allows firms to reduce taxes on dividends and capital gains from subsidiaries and has a wide range of treaties that reduce taxes.²⁴ In 2006, for example, Bono and other members of the U2 band moved their music publishing company from Ireland to the Netherlands after Ireland changed its tax treatment of music royalties.²⁵ A 2010 newspaper report explained the role of the Netherlands in facilitating movement to tax havens through provisions such as the various "Dutch sandwiches," which allow money to be funneled out of other countries that would charge withholding taxes to non-European countries, to be passed on in turn to tax havens such as Bermuda and the Cayman Islands.²⁶ Issues have recently been raised in the Netherlands government about its role in tax avoidance.²⁷ The European Commission also began investigating, in June 2014, whether certain arrangements in Ireland, Luxembourg, and the Netherlands constitute prohibited state aid; the inquiry was later expanded to all member states.²⁸ In addition, the European Union has agreed to add an anti-abuse clause to its provision to prevent double taxation within member states, which may have implications for these arrangements in the future.²⁹ Although new laws have been passed, commentators still point to the Netherlands as a major tax haven.³⁰

Some have identified the United States and the United Kingdom (UK) as having tax haven characteristics. Luxembourg Prime Minister Jean-Claude Junker urged other EU member states to challenge the United States for tax havens in Delaware, Nevada, and Wyoming.³¹ One website offering offshore services mentions, in their view, several overlooked tax havens which include the United States, United Kingdom, Denmark, Iceland, Israel, and Portugal's Madeira Island.³² (Others on their list and not listed in Table 1 were Hungary, Brunei, Uruguay, and Labuan [Malaysia]).³³ In the case of the United States the article mentions the lack of reporting requirements and the failure to tax interest and other exempt passive income paid to foreign entities, the limited liability corporation which allows a flexible corporate vehicle not subject to taxation, and the ease of incorporating in certain states (Delaware, Nevada, and Wyoming). Issues have recently been raised in the Netherlands about its role in tax avoidance

Another website includes in its list of tax havens Delaware, Wyoming, and Puerto Rico, along with other jurisdictions not listed in Table 1: the Netherlands, Campione d'Italia, a separate listing for Sark (identified as the only remaining "fiscal paradise"), the UK, and a coming discussion for Canada.³⁴ Sark is an island country associated with Guernsey, part of the Channel Islands, and Campione d'Italia is an Italian town located within Switzerland.

The Economist reported a study by a political scientist experimenting with setting up sham corporations; the author succeeded in incorporating in Wyoming and Nevada, as well as the UK and several other places.³⁵ Michael McIntyre discusses three U.S. practices that aid international evasion: the failure to collect information on tax exempt interest income paid to foreign entities, the system of foreign institutions that act as qualified intermediaries (see discussion below) but do not reveal their clients, and the practices of states such as Delaware and Wyoming that allow people to keep secret their identities as stockholder or depositor.³⁶ Some of these problems have been addressed, in part, by FATCA.

In a meeting in late April 2009, Eduardo Silva, of the Cayman Islands Financial Services Association, claimed that Delaware, Nevada, Wyoming, and the UK were the greatest offenders with respect to, among other issues, tax fraud. He suggested that Nevada and Wyoming were worse than Delaware because they permit companies to have bearer shares, which allows anonymous ownership. A U.S. participant at the conference noted that legislation in the United States, S. 569 (111^th Congress), would require disclosure of beneficial owners in the United States.³⁷

Nicholas Shaxson, in his book Treasure Islands, organizes tax havens into four categories: (1) continental European havens such as Switzerland and Luxembourg; (2) a British zone of influence (which includes the City of London³⁸ as well as countries formally related to the UK, such as Jersey, Guernsey, the Isle of Man, Bermuda, and many of the islands in the West Indies and Caribbean, and those influenced by the UK); (3) a U.S. zone of influence (the United States itself, some of its states, along with the Virgin Islands, Marshall Islands, Liberia, and Panama), and (4) other jurisdictions.³⁹ Anthony van Fossen, in his study of Pacific Island tax havens, indicates that connection with the UK and specifically with the City of London is a contributor to a successful tax haven. While the United States has limited the activities of some islands in its sphere of influence, one of the most important tax havens in this area is the Marshall Islands, which specializes in flags of convenience.⁴⁰

In addition, any country with a low tax rate could be considered as a potential location for shifting income to. In addition to Ireland, three other countries in the OECD not included in Table 1 have tax rates below 20%: Iceland, Poland, and the Slovak Republic.⁴¹ Most of the eastern European countries not included in the OECD have tax rates below 20%.⁴²

The Tax Justice Network probably has the largest list of tax havens, and it includes some specific cities and areas.⁴³ In addition to the countries listed in Table 1, its initial 2005 list included in the Americas and Caribbean, New York and Uruguay; in Africa, Mellila, Sao Tome e Principe, Somalia, and South Africa; in the Middle East and Asia, Dubai, Labuan (Malaysia), Tel Aviv, and Taipei; in Europe, Alderney, Belgium, Campione d'Italia, City of London, Dublin, Ingushetia, Madeira, Sark, Trieste, Turkish Republic of Northern Cyprus, and Frankfurt; and in the Indian and Pacific oceans, the Marianas. Jordan is the only country listed in Table 1 that is not included in the Tax Justice Network's list. Currently, it has a list of corporate tax havens; of the top 20 havens, five are not listed in Table 1 (United Arab Emirates, United Kingdom, Belgium, China, and Hungary). It also has a financial secrecy index: of the top 20, 10 are not in Table 1 (United States, Japan, Netherlands, United Arab Emirates, United Kingdom, Taiwan, Germany, Thailand, Canada, and Qatar).⁴⁴ The United States is often included in lists of financial secrecy jurisdictions that can affect other countries because of state laws that shield owners. The recent leak of the Pandora papers found numerous instances of trusts, with the top five states South Dakota, Florida, Delaware, Texas, and Florida.⁴⁵ Delaware is the state estimated to have the largest number of shell corporations, and Alaska, Wyoming, Nevada, and South Dakota are also noted as states that have laws favoring shell corporations and trusts.⁴⁶

Ronen Palan, Richard Murphy, and Christian Chavagneux report 11 different lists of tax havens. Although the Tax Justice Network is the largest such list, a few countries not on this list appear on others.⁴⁷ Eight countries appeared on all lists: the Bahamas, Bermuda, the Cayman Islands, Guernsey, Jersey, and Malta. Palan, Murphy, and Chavagneux also suggest adding Belgium to the Netherlands and Luxembourg as a location for holding companies in Europe. In addition, they discuss the aspects of rules in the United States and the United Kingdom that might justify identification as a tax haven.

Johannesen and Zucman, in a study focusing on bank secrecy provide a list of tax havens in their online appendix.⁴⁸ Compared to Table 1, they exclude Ireland, Jordan, Lebanon, Maldives, and Tonga, but include Austria, Belgium, Chile, Malaysia, Trinidad and Tobago, and Uruguay.

Corpnet, a research group at the University of Amsterdam, uses a large network data system to identify offshore financial centers (OFCs).⁴⁹ It divides them into two groups: sinks, where excess earnings end, and conduits, where these earnings flow to sinks. Of the top 20 sink OFCs, all but two, Taiwan and Guyana, appear in Table 1. Of the top five conduits, three (Switzerland, Singapore, and Ireland) appear, but two do not (the Netherlands and the UK).

Garcia-Bernardo and Janský provide a list of the top countries associated with profit shifting: Cayman Islands, Netherlands, China, Hong Kong, Bermuda, British Virgin Islands, Switzerland, Puerto Rico, Ireland, Singapore, and Luxembourg.⁵⁰

Methods of Corporate Tax Avoidance

Prior to the 2018, U.S. multinationals were not taxed on income earned by foreign subsidiaries until it was repatriated to the U.S. parent as dividends, although some passive and related company income that is easily shifted was (and remains) taxed currently under anti-abuse rules referred to as Subpart F. (Foreign affiliates or subsidiaries that are majority U.S. owned are referred to as controlled foreign corporations, or CFCs, and many of these related firms are wholly owned.) Taxes on income that is repatriated (or, less commonly, earned by branches and taxed currently) were allowed a credit for foreign income taxes paid. (A part of a parent company treated as a branch is not a separate entity for tax purposes, and all income is part of the parent's income.)

The Tax Cuts and Jobs Act (TCJA; P.L. 115-97), enacted in 2017, eliminated the tax on dividends of foreign subsidiaries and instead imposed a minimum tax on foreign source income aimed at limiting profit-shifting, the tax on global intangible low-taxed income, or GILTI.⁵¹ This regime allowed a deduction for 10% of tangible assets, aimed at approximating income from tangible investments, and allowed a 50% (37.5% after 2025) deduction of the remainder. TCJA lowered the corporate tax rate from 35% to 21% leading to an effective tax rate on GILTI of 10.5% (13.125% after 2025). TCJA also enacted a deduction for foreign derived intangible income (FDII) of U.S. companies, intended to make firms largely indifferent between holding intangibles used to serve foreign markets in the United States or abroad. Income eligible for the deduction was domestic income reduced by 10% of intangible assets, with a deduction for the remainder of 37.5% (20.875% after 2025), multiplied by the share of sales of goods and services that was exported.

Credits are allowed for foreign taxes, limited to the amount of tax imposed by the United States, so that they, in theory, cannot offset taxes on domestic income. For GILTI only 80% of foreign taxes is allowed as credits. The limit is imposed on an overall basis, allowing excess credits in high-tax countries to offset U.S. tax liability on income earned in low-tax countries, although separate limits apply to passive, active, GILTI, and branch income. Other countries either employ this system of deferral and credit or, in most cases, exempt income earned in foreign jurisdictions. Most countries have some form of anti-abuse rules similar to Subpart F.

If a firm can shift profits to a low-tax jurisdiction from a high-tax one, its taxes will be reduced without affecting other aspects of the company. Tax differences also affect real economic activity, which in turn affects revenues, but it is this artificial shifting of profits that is the focus of this report.⁵²

Because the United States taxes all income earned in its borders as well as imposing a residual tax on income earned abroad by U.S. persons, tax avoidance relates both to U.S. parent companies shifting profits abroad to low-tax jurisdictions and the shifting of profits out of the United States by foreign parents of U.S. subsidiaries. In the case of U.S. multinationals, one study suggested that about half the difference between profitability in low-tax and high-tax countries, which could arise from artificial income shifting, was due to transfers of intellectual property (or intangibles) and most of the rest through the allocation of debt.⁵³ However, a study examining import and export prices suggests a very large effect of transfer pricing in goods (as discussed below).⁵⁴ Some evidence of the importance of intellectual property can also be found from the types of firms that repatriated profits abroad following a temporary tax reduction enacted in 2004; one-third of the repatriations were in the pharmaceutical and medicine industry and almost 20% in the computer and electronic equipment industry.⁵⁵

The TCJA added a provision that was aimed in part at U.S. subsidiaries of foreign parents, in the form of the base erosion and anti-abuse tax (BEAT) applying to large firms that have significant related party payments. BEAT imposed a tax on a base increased by certain payments to foreign related companies, such as royalties and interest. It, however, excluded the cost of goods sold, and also excluded purchases of services if a certain mark-up method is used. The BEAT tax rate is 10% (12.5% after 2025) and is paid if larger than the regular minimum tax. BEAT is temporarily allowed certain domestic tax credits, but not the foreign tax credit, and no credits are allowed after 2025.

Allocation of Debt and Earnings Stripping

One method of shifting profits from a high-tax jurisdiction to a low-tax one is to borrow more in the high-tax jurisdiction and less in the low-tax one. This shifting of debt can be achieved without changing the overall debt exposure of the firm. A more specific practice is referred to as earnings stripping, where either debt is associated with related firms or unrelated debt is not subject to tax by the recipient. As an example of the former earnings stripping method, a foreign parent may lend to its U.S. subsidiary. Alternatively, an unrelated foreign borrower not subject to tax on U.S. interest income might lend to a U.S. firm.

The U.S. tax code currently contains provisions to address interest deductions and earnings stripping. It applies an allocation of the U.S. parent's interest for purposes of the limit on the foreign tax credit. The amount of foreign source income is reduced when part of U.S. interest is allocated and the maximum amount of foreign tax credits taken is limited, a provision that affects firms with excess foreign tax credits.⁵⁶ There is no allocation rule, however, applying directly to GILTI, so that a U.S. parent could operate its subsidiary with all equity finance in a low-tax jurisdiction and take all of the interest on the overall firm's debt as a deduction. A proposal now under consideration in President Biden's budget and in several congressional proposals would introduce such an allocation rule, so that the share of worldwide interest deducted in the United States would be proportionate to the U.S. share of worldwide income.⁵⁷

The United States has thin capitalization rules that apply generally and can limit interest deductions. (Most of the United States' major trading partners have similar rules.) A section of the Internal Revenue Code (163(j)) disallows deductions for net interest exceeding 30% of adjusted taxable income (currently earnings before taxes, interest depreciation, and amortization, or EBITDA, but expanded to earnings before taxes and interest, or EBIT after 2022).

The possibility of earnings stripping received more attention after a number of U.S. firms inverted; that is, arranged to move their parent firm abroad so that U.S. operations became a subsidiary of that parent. The American Jobs Creation Act of 2004 (AJCA; P.L. 108-357) addressed the general problem of inversion by treating firms that subsequently inverted as U.S. firms if the former U.S. shareholders owned at least 80% of the new firm. If U.S. shareholders owned 60% to 80%, a tax would be imposed on the transfer of assets. During consideration of this legislation there were also proposals for broader earnings stripping restrictions as an approach to this problem that would have reduced the excess interest deductions. This general earnings stripping proposal was not adopted. However, the AJCA mandated a Treasury Department study on this and other issues; that study focused on U.S. subsidiaries of foreign parents and was not able to find clear evidence on the magnitude.⁵⁸

Noted in the Treasury's mandated study, there is relatively straightforward evidence that U.S. multinationals allocate more interest to high-tax jurisdictions, but it is more difficult to assess earnings stripping by foreign parents of U.S. subsidiaries, because the entire firm's accounts are not available. The Treasury study focused on this issue and used an approach that had been used in the past of comparing these subsidiaries to U.S. firms. The study was not able to provide conclusive evidence about the shifting of profits out of the United States due to high leverage rates for U.S. subsidiaries of foreign firms but did find evidence of shifting for inverted firms.

A study of profit shifting worldwide estimated that 28% of profit shifting was due to allocation of debt in high-tax countries, with 72% due to transfer pricing.⁵⁹

Inversions have recently become an issue. Although some firms inverted following the 2004 legislation based on an activity exception, that approach was limited by regulation. In 2014, a number of U.S. firms inverted, or considered inversion, by merging with a smaller foreign firm. Regulatory changes largely eliminated inversions, although further provisions to limit them were adopted in the TJCA. There are proposals for further restrictions.⁶⁰

Transfer Pricing

The second major way that firms can shift profits from high-tax to low-tax jurisdictions, and the one that appears most important, is through the pricing of assets, goods, and services sold between affiliates. To properly reflect income, prices of assets, goods, and services sold by related companies should be the same as the prices that would be paid by unrelated parties. By lowering the price of assets, goods, and services sold by parents and affiliates in high-tax jurisdictions and raising the price of purchases, income can be shifted.

An important and growing issue of transfer pricing is with the transfers to rights to intellectual property, or intangibles. If a patent developed in the United States is sold or licensed to an affiliate in a low-tax country income will be shifted if the royalty or other payment is lower than the true value of the license. For many goods there are similar products sold or other methods (such as cost plus a markup) that can be used to determine whether prices are set appropriately. Intangibles, such as new inventions or new drugs, tend not to have comparables, and it is very difficult to know the royalty that would be paid in an arms-length price. Therefore, intangibles represent particular problems for policing transfer pricing.

Investment in intangibles is favorably treated in the United States because costs, other than capital equipment and buildings, are expensed for research and development, which is also eligible for a tax credit. In addition, advertising to establish brand names is also deductible. Overall these treatments tend to produce an effective low, zero, or negative tax rate for overall investment in intangibles. Thus, there are significant incentives to make these investments in the United States. On average, the benefit of tax deductions or credits when making the investment tend to offset the future taxes on the return to the investment. However, for those investments that tend to be successful, it is advantageous to shift profits to a low-tax jurisdiction, so that there are tax savings on investment and little or no tax on returns. As a result, these investments can be subject to negative tax rates, or subsidies, which can be significant.

Transfer pricing rules with respect to intellectual property are further complicated because of cost sharing agreements, where different affiliates contribute to the cost.⁶¹ If an intangible is already partially developed by the parent firm, affiliates contribute a buy-in payment for the rights to that asset for given markets. It is very difficult to determine arms-length pricing in these cases where a technology is partially developed and there is risk associated with the expected outcome. Following the buy-in payment, the foreign affiliate can contribute to further research in the United States and obtain the rights to the technology going forward. For a firm that has already developed a successful product, such as a cell phone, it is unlikely they would make such a cost sharing arrangement with an unrelated party. One study found some evidence that firms with cost sharing arrangements were more likely to engage in profit shifting.⁶²

One problem with shifting profits to some tax haven jurisdictions is that, if real activity is necessary to produce the intangible these countries may not have labor and other resources to undertake the activity. However, firms had developed techniques to take advantage of tax laws in other countries to achieve both a productive operation while shifting profits to no-tax jurisdictions. An example is the double Irish, Dutch sandwich method that has been used by some U.S. firms, including, as exposed in news articles, Google.⁶³ In this arrangement, the U.S. firm transfers its intangible asset to an Irish holding company. This company has a subsidiary sales company that sells advertising (the source of Google's revenues) to Europe. However, sandwiched between the Irish holding company and the Irish sales subsidiary is a Dutch subsidiary, which collects royalties from the sales subsidiary and transfers them to the Irish holding company. The Irish holding company claims company management (and tax home) in Bermuda, with a 0% tax rate, for purposes of the corporate income tax. This strategy allows the Irish operation to avoid even the low Irish tax of 12.5% and, by using the Dutch sandwich, to avoid Irish withholding taxes (which are not due on payments to European Union companies). More recently, European countries have complained about companies such as Google, Apple, Amazon, Facebook, and Starbucks using this strategy in some cases.

Ireland eliminated the double Irish arrangements as well as other arrangements that resulted in no foreign tax.⁶⁴ Profits can also be shifted directly to a tax haven, as in the case of Yahoo, where the Dutch intermediary can transfer profits directly to the tax haven (in this case, the Cayman islands) because it does not collect a withholding tax, as would be the case with France or Ireland.⁶⁵

Contract Manufacturing

When a subsidiary is set up in a low-tax country and profit shifting occurs, as in the acquisition of rights to an intangible, a further problem occurs: this low-tax country may not be a desirable place to actually manufacture and sell the product. For example, an Irish subsidiary's market may be in Germany and it would be desirable to manufacture in Germany. But to earn profits in Germany with its higher tax rate does not minimize taxes. Instead the Irish firm may contract with a German firm as a contract manufacturer, who will produce the item for cost plus a fixed markup. Subpart F taxes on a current basis certain profits from sales income, so the arrangement must be structured to qualify as an exception from this rule. There are complex and changing regulations on this issue.⁶⁶

Check-the-Box, Hybrid Entities, and Hybrid Instruments

Another technique for shifting profit to low-tax jurisdictions was greatly expanded with the check-the-box provisions. These provisions were originally intended to simplify questions of whether a firm was a corporation or a partnership. Their application to foreign circumstances through the disregarded entity rules has led to the expansion of hybrid entities, where an entity can be recognized as a corporation by one jurisdiction but not by another. For example, a U.S. parent's subsidiary in a low-tax country can lend to its subsidiary in a high-tax country, with the interest deductible because the high-tax country recognizes the firm as a separate corporation. Normally, interest received by the subsidiary in the low-tax country would be considered passive or tainted income subject to current U.S. tax under Subpart F. However, under check-the-box rules, the high-tax corporation can elect to be disregarded as a separate entity. Thus, from the perspective of the United States, there would be no interest income paid because the two are the same entity. Check-the-box and similar hybrid entity operations also can be used to avoid other types of Subpart F income, for example from contract manufacturing arrangements. According to David R. Sicular, this provision, which began as a regulation, has been, albeit temporarily, codified (called the look-through rules).⁶⁷ The look-through rules expand the scope of check-the-box to more related parties and circumstances. They began as temporary provisions but have been extended numerous times. Currently, they expired at the end of 2025.

Hybrid entities relate to issues other than Subpart F. For example, a reverse hybrid entity formerly could be used to allow U.S. corporations to benefit from the foreign tax credit without having to recognize the underlying income. As an example, a U.S. parent could have set up a holding company in a county that was treated as a disregarded entity, and the holding company could have owned a corporation that was treated as a partnership in another foreign jurisdiction. Under flow-through rules, the holding company was liable for the foreign tax and, because it was not a separate entity, the U.S. parent corporation was therefore liable, but the income could have been retained in the foreign corporation that was viewed as a separate corporate entity from the U.S. point of view. In this case, the entity was structured so that it was a partnership for foreign purposes but a corporation for U.S. purposes.⁶⁸ Provisions in P.L. 111-226 eliminated this practice.

In addition to hybrid entities that achieve tax benefits by being treated differently in the United States and the foreign jurisdiction, there were also hybrid instruments that can avoid taxation by being treated as debt in one jurisdiction and equity in another.⁶⁹ The Tax Cuts and Jobs Act, however, contained provisions limiting hybrid instruments and entities tax benefits by disallowing a deduction by a related party for an interest or royalty payment to a recipient in a foreign country if that payment is not taxed (or is included in income and then deducted) in the foreign country.

Cross Crediting and Sourcing Rules for Foreign Tax Credits

Income from a low-tax country that is received in the United States can escape taxes because of cross crediting: the use of excess foreign taxes paid in one jurisdiction or on one type of income to offset U.S. tax that would be due on other income. In some periods in the past the foreign tax credit limit was proposed on a country-by-country basis, although that rule proved to be difficult to enforce given the potential to use holding companies. Foreign tax credits have subsequently been separated into different baskets to limit cross crediting. Currently, there are four baskets: active, passive, GILTI, and branch.

Because firms could choose when to repatriate income under prior law, they could arrange realizations to maximize the benefits of the overall limit on the foreign tax credit. That is, firms that had income from jurisdictions with taxes in excess of U.S. taxes could also elect to realize income from jurisdictions with low taxes and use the excess credits to offset U.S. tax due on that income. Studies suggest that between cross crediting and deferral, U.S. multinationals typically paid virtually no U.S. tax on foreign source income.⁷⁰ Limited data are available to determine the tax rates in the new GILTI regime adopted by the TCJA. However, two popular tax havens, Bermuda and the Cayman islands, that have no corporate tax showed negligible taxes in 2018, after the tax revisions, as well as 2016, before revisions were being considered. The tax rate on U.S. CFCs of large companies was 1.9% in 2016 and 0.5% in 2018 for Bermuda and 0.4% in the Cayman Islands in both years.⁷¹

The Magnitude of Corporate Profit Shifting

This section examines the evidence on the existence and magnitude of profit shifting and the techniques that are most likely to contribute to it.

Evidence on the Scope of Profit Shifting

There is ample, and simple, evidence that profits appear in countries inconsistent with an economic motivation. This section first examines the profit share of income of controlled corporations compared to the share of gross domestic product and how it has changed recently.⁷² The first set of countries, acting as a reference point, includes the remaining G-7 countries that are also among the United States' major trading partners. These countries account for 15% of pretax profits and 31% of rest-of-world gross domestic product. The second group of countries includes larger countries from Table 1 (with gross domestic product [GDP] of at least $15 billion), plus the Netherlands, which is widely considered a tax conduit for U.S. multinationals because of its holding company rules. These countries account for about 27% of earnings and 5% of rest-of-world GDP. The third group of countries includes smaller countries listed in Table 1, with GDP less than $10 billion. These countries account for 17% of earnings and less than one-tenth of 1% of rest-of-world GDP.⁷³

As indicated in Table 2, income-to-GDP ratios in the large G-7 countries in 2010 ranged from 0.2% to 3.3%, the larger amounts reflecting in part the United States' relationships with some of its closest trading partners. Overall, this income as a share of GDP is 0.7%. Outside the UK and Canada, this income as a share of GDP is around 0.3% to 0.6% and does not vary with country size (Japan, for example, has over twice the GDP of Italy). Canada and the UK also have appeared on some tax haven lists, and the larger income shares could partially reflect that fact.⁷⁴ There has been relatively little change in the aggregate between 2004 and 2010 (the latest year IRS data on earnings of multinational firms are available).

Table 3 reports the share for the larger tax havens listed in Table 1 for which data are available, plus the Netherlands. In general, U.S. source profits as a percentage of GDP are considerably larger than those in Table 2. Although the shares fluctuated over time, they are particularly large in Luxembourg, Ireland and Singapore. In most cases, the shares are well in excess of those in Table 2.

Table 4 examines the small tax havens listed in Table 1 for which data are available. In Bermuda, the British Virgin Islands, and the Cayman Islands profits are consistently multiples of total GDP. In other jurisdictions in Table 4, profits are a large share of output. Some of the increase in Jersey may reflect the movement of some operations by a large U.S. company from Ireland.⁷⁵ These numbers clearly indicate that the profits in these countries do not appear to derive from economic motives related to productive inputs or markets but rather reflect income easily transferred to low-tax jurisdictions.

The data do not indicate a change in the location of profits between 2016 before consideration of the TCJA and 2018 after it was enacted.

Evidence of profit shifting has been presented in many other studies. Grubert and Altshuler report that profits of controlled foreign corporations in manufacturing relative to sales in Ireland are three times the group mean.⁷⁶ GAO reported higher shares of pretax profits of U.S. multinationals than of value added, tangible assets, sales, compensation, or employees in low-tax countries such as Bermuda, Ireland, the UK Caribbean, Singapore, and Switzerland.⁷⁷ Costa and Gravelle reported similar results for tax havens using subsequent data.⁷⁸ Martin Sullivan reports the return on assets for 1998 averaged 8.4% for U.S. manufacturing subsidiaries, but with returns of 23.8% in Ireland, 17.9% in Switzerland, and 16.6% in the Cayman Islands.⁷⁹ More recently, he noted that of the 10 countries that accounted for the most foreign multinational profits, the 5 countries with the highest manufacturing returns for 2004 (the Netherlands, Bermuda, Ireland, Switzerland, and China) all had effective tax rates below 12% while the 5 countries with lower returns (Canada, Japan, Mexico, Australia, and the United Kingdom) had effective tax rates in excess of 23%.⁸⁰ A number of econometric studies of this issue have been done.⁸¹ Studies in the next section focusing on the cost of profit shifting also provide evidence.

Estimates of the Cost and Sources of Corporate Tax Avoidance

There are no official estimates of the cost of international corporate tax avoidance, although a number of researchers have made estimates, nor are there official estimates of the cost of individual tax evasion.⁸² In general, the estimates are not reflected in the overall tax gap estimate. The magnitude of corporate tax avoidance has been estimated through a variety of techniques and not all are for total avoidance. Some address only avoidance by U.S. multinationals and not by foreign parents of U.S. subsidiaries. Some focus only on a particular source of avoidance.⁸³

Estimates of the potential revenue cost of income shifting by multinational corporations vary considerably, with some estimates in excess of $100 billion annually. The only study by the IRS in this area is an estimate of the international gross tax gap (not accounting for increased taxes collected on audit) related to transfer pricing based on audits of returns. They estimated a cost of about $3 billion, based on examinations of tax returns for 1996-1998.⁸⁴ This estimate would reflect an estimate not of legal avoidance, but of noncompliance, and for reasons stressed in the study has a number of limitations. One of those is that an audit does not detect all noncompliance, and it would not detect avoidance mechanisms which are, or appear to be, legal.

Some idea of the potential magnitude of the revenue lost from profit shifting by U.S. multinationals might be found in the estimates of the revenue gain from taxing foreign income in full to be $45.4 billion in FY2021.⁸⁵ This number may be low because of the pandemic, as their estimate rises to $62.6 billion in FY2022, and $73.1 in FY2023. If most of the profit in low-tax countries has been shifted there to avoid U.S. tax rates, the projected revenue gain from taxing foreign source income in full would provide an idea of the general magnitude of the revenue cost of profit shifting by U.S. parent firms. The Administration's estimates for raising the tax rate on GILTI to 21%, ending the deduction for tangible assets and imposing a per country foreign tax credit limit, which is tantamount to taxing this income at the current rate, were estimated at $53.4 billion for FY2023.⁸⁶ These estimates could be either an overstatement or an understatement of the cost of tax avoidance. They could be overstated because some of the profits abroad accrue to real investments in countries that have lower tax rates than the United States and thus do not reflect artificial shifting. They could be an understatement because they do not reflect the tax that could be collected by the United States rather than foreign jurisdictions on profits shifted to low-tax countries. For example, Ireland has a tax rate of 12.5% and the United States has a 35% rate, so taxing that income in full (absent behavioral changes) would only collect the excess of the U.S. tax over the Irish tax on shifted revenues, or about two-thirds of lost revenue.

Earlier Academic Studies

Altshuler and Grubert estimated for 2002 that the corporate tax could be cut to 28% if deferral were ended, and based on corporate revenue in that year the gain was about $11 billion.⁸⁷ That year was at a low point because of the recession; if the share had remained the same, the gain would have been around $26 billion for FY2014.⁸⁸ The projection of the effects of deferral in tax expenditures has increased much faster than revenues, however.

Researchers have looked at differences in pretax returns and estimated the revenue gain if returns were equated. This approach should provide some estimates of the magnitude of overall profit shifting for multinationals, whether through transfer pricing, leveraging, or some other technique. Martin Sullivan, using Commerce Department data, estimates that, based on differences in pretax returns, the cost for 2004 was between $10 billion and $20 billion. Sullivan subsequently reports an estimated $17 billion increase in revenue loss from profit shifting between 1999 and 2004, which suggests that earlier number may be too small.⁸⁹ Sullivan suggests that the growth in profit shifting may be due to check-the-box. Sullivan subsequently estimated a $28 billion loss for 2007 which he characterized as conservative.⁹⁰ Charles Christian and Thomas Schultz, using rate of return on assets data from tax returns, estimated $87 billion was shifted in 2001, which, at a 35% tax rate, would imply a revenue loss of about $30 billion.⁹¹ Adjusted proportionally to revenue, that amount would be $70 billion in 2014. As a guide for potential revenue loss from avoidance, these estimates suffer from two limits. The first is the inability to determine how much was shifted out of high-tax foreign jurisdictions rather than the United States, which leads to a range of estimates. At the same time, if capital is mobile, economic theory indicates that the returns should be lower, the lower the tax rate. Thus the results could also understate the overall profit shifting and the revenue loss to the United States.

Simon Pak and John Zdanowicz examined export and import prices, and estimated that lost revenue due to transfer pricing of goods alone was $53 billion in 2001.⁹² This estimate should cover both U.S. multinationals and U.S. subsidiaries of foreign parents, but is limited to one technique. Kimberly Clausing, using regression techniques on cross-country data, which estimated profits reported as a function of tax rates, estimated that revenues of over $60 billion are lost for 2004 by applying a 35% tax rate to an estimated $180 billion in corporate profits shifted out of the United States.⁹³ She estimates that the profit-shifting effects are twice as large as the effects from shifts in actual economic activity. This methodological approach differs from others that involve direct calculations based on returns or prices and is subject to the econometric limitations with cross-country panel regressions. In theory, however, it had an overall of coverage of shifting (that is both outbound by U.S. parents of foreign corporations and inbound by foreign parents of U.S. corporations and covering all techniques).

Clausing and Reuven Avi-Yonah estimate the revenue gain from moving to a formula apportionment based on sales that is on the order of $50 billion per year because the fraction of worldwide income in the United States is smaller than the fraction of worldwide sales.⁹⁴ While this estimate is not an estimate of the loss from profit shifting (since sales and income could differ for other reasons), it is suggestive of the magnitude of total effects from profit shifting. A similar result was found by another study that applied formula apportionment based on an equal weight of assets, payroll, and sales.⁹⁵

A later study by Clausing indicated that the revenue loss from profit shifting may have been as high as $90 billion in 2008, although an alternative data set indicates profit shifting of $57 billion.⁹⁶ For the last five years, the first method yielded losses ranging from 20% to 30% of profits. Using the second method, the range was 13% to 20%. If rising proportional to revenue, the 2014 level would be $66 billion to $104 billion.

More Recent Studies

Alex Cobham and Petr Janský estimated a loss of $50 to $80 billion for 2012.⁹⁷ Their study also estimated worldwide profit shifting and their method was to examine differentials in profitability compared to other measures of real economic activity. For the same year, Maria Alvarez-Martinex estimated a loss for the United States of 36 billion euros, which would be $47 billion at the exchange rate at that time.⁹⁸ Alvarez-Martinex used a general equilibrium model based on estimated behavioral responses from other literature to tax rate differentials. Gabriel Zucman, using two different methodologies, found the revenue cost for profit shifting could range from $55 billion to $133 billion for 2013.⁹⁹ The lower number is from estimating the share of profits booked in tax havens and not repatriated, which would be assumed to be taxed fully if made subject to U.S. taxes. The second examines the decline in effective tax rate over time, and the residual, after accounting for other factors (about two-thirds), is attributed to profit shifting.

Thomas R. Tørsløv, Ludvig S. Wier, and Gabriel Zucman estimated that losses were equal to 14% of revenue collected in 2015, indicating a loss of $47 billion given revenue collections of $329 billion in that year.¹⁰⁰ This study used data on the profitability of foreign affiliates and compared that profitability with local firms to estimate profit shifting.

Kimberly Clausing estimated that the revenue loss from profit shifting was over $100 billion in 2017, using new country-by-country data to estimate the responsiveness of profits to tax rates.¹⁰¹

The only estimate found of the revenue loss from profit shifting after the TCJA was enacted, by the Tax Justice Network, indicates a loss of $77 billion.¹⁰² This study is a worldwide study based on the misalignment of profits and economic activity.

It is very difficult to develop a separate estimate for U.S. subsidiaries of foreign multinational companies because there is no way to observe the parent firm and its other subsidiaries. An exception is for studies that are multi-country. Several studies have documented that these firms have lower taxable income and that some have higher debt to asset ratios than domestic firms. There are many other potential explanations these differing characteristics, however, and domestic firms that are used as comparisons also have incentives to shift profits when they have foreign operations. No quantitative estimate has been made.¹⁰³ However, some evidence of earnings stripping for inverted firms was found.¹⁰⁴

Importance of Different Profit Shifting Techniques

Some studies have attempted to identify the importance of techniques used for profit shifting. Grubert has estimated that about half of income shifting was due to transfer pricing of intangibles and most of the remainder to shifting of debt.¹⁰⁵ In a subsequent study, Altshuler and Grubert find that multinationals saved $7 billion more between 1997 and 2002 due to check-the-box rules.¹⁰⁶ Some of this gain may have been at the cost of high-tax host countries rather than the United States, however.

Some of the estimates discussed here conflict with respect to the source of profit shifting. The Pak and Zdanowich estimates suggest that transfer pricing of goods is an important mechanism of tax avoidance, whereas Grubert suggests that the main methods of profit shifting are due to leverage and intangibles. The estimates for pricing of goods may, however, reflect errors, or money laundering motives rather than tax motives. Much of the shifting was associated with trade with high-tax countries; for example, Japan, Canada, and Germany accounted for 18% of the total.¹⁰⁷ At the same time, about 14% of the estimate reflected transactions with countries that appear on tax haven lists: the Netherlands, Taiwan, Singapore, Hong Kong, and Ireland.

A study by Jost Hekemeyer and Michael Overesch based on an analysis of 25 empirical studies found that transfer pricing was considerably more important than debt, accounting for an estimated 72% of the total, although their review covered studies on non-U.S. multinationals.¹⁰⁸ The growing importance of firms holding substantial intangible assets may point to a growing important of transfer pricing of intangibles. Hekemeyer and Overesch also found that reported profits on average decrease by 0.8% with a one percentage point change in the tax differential between two locations.

Some evidence that points to the importance of intangibles and the associated profits in tax haven countries can be developed by examining the sources of dividends repatriated during the "repatriation holiday" enacted in 2004.¹⁰⁹ Under the tax regime prior to the TCJA, retaining profits abroad was the method used to avoid U.S. taxes on profits earned in low-tax countries. This provision allowed, for a temporary period, dividends to be repatriated with an 85% deduction, leading to a tax rate of 5.25%. The pharmaceutical and medicine industry accounted for $99 billion in repatriations or 32% of the total. The computer and electronic equipment industry accounted for $58 billion or 18% of the total. Thus these two industries, which are high tech firms, accounted for half of the repatriations. The benefits were also highly concentrated in a few firms. According to a recent study, five firms (Pfizer, Merck, Hewlett-Packard, Johnson & Johnson, and IBM) are responsible for $88 billion, over a quarter (28%) of total repatriations.¹¹⁰ The top 10 firms (adding Schering-Plough, Du Pont, Bristol-Myers Squibb, Eli Lilly, and PepsiCo) accounted for 42%. The top 15 (adding Procter and Gamble, Intel, Coca-Cola, Altria, and Motorola) accounted for over half (52%). These are firms that tend to, in most cases, have intangibles either in technology or brand names.

Finally, as shown in Table 5, which lists all countries accounting for at least 1% of the total of eligible dividends (and accounting for 87% of the total), most of the dividends were repatriated from countries that appear on tax haven lists.

Methods of Avoidance and Evasion by Individuals

Individual evasion of taxes may take different forms, and they are all facilitated by the growing international financial globalization and ease of making transactions on the Internet. Individuals can purchase foreign investments directly (outside the United States), such as stocks and bonds, or put money in foreign bank accounts and simply not report the income (although it is subject to tax under U.S. tax law). There has been little or no withholding information on individual taxpayers for this type of action. They could also use structures such as trusts or shell corporations to evade tax on investments, including investments made in the United States, which may take advantage of U.S. tax laws that exempt interest income and capital gains of nonresidents from U.S. tax. Rather than using withholding or information collection, the United States has largely relied in the past on the Qualified Intermediary (QI) program where beneficial owners are not revealed. To the extent any information gathering from other countries is done it is through bilateral information exchanges rather than multilateral information sharing. The European Union had developed a multilateral agreement but the United States does not participate.

New developments in information exchange may affect individual tax evasion both in the United States and abroad. In 2010, Congress enacted the Foreign Account Tax Compliance Act (FATCA) as part of the Hiring Incentives to Restore Employment Act (HIRE; P.L. 111-147).¹¹¹ FATCA recently become effective and requires foreign financial institutions to report information on asset holders or be subject to a 30% withholding rate. Its effectiveness is yet to be determined, although revenue projections when enacted did not predict a significant effect.

One hundred twelve countries (but not the United States) signed a multilateral information exchange agreement that set reporting standards, which should eventually lead to fuller exchange of tax information by most countries.¹¹² The OECD also developed a common reporting standard (CRS) for the exchange of information, with 91 countries participating.¹¹³ The United States does not participate and does not provide reciprocal information under FATCA, which might require congressional action. The United States, due to laws enacted by some states, is viewed by some as one of the most important financial secrecy jurisdictions, although not necessarily as a tax haven (although states accommodating secrecy may cause revenue losses to other states by avoiding state income taxes).

Tax Provisions Affecting the Treatment of Income by Individuals

The ability of U.S. persons (whether firms or individuals) to avoid tax on U.S. source income that they would normally be subject to arises from U.S. rules that do not impose withholding taxes on many sources of income paid abroad. In general, interest and capital gains are not subject to withholding. Dividends, non-portfolio interest (such as interest payments by a U.S. subsidiary to its parent), capital gains connected with a trade or business, and certain rents are subject to tax, although treaty arrangements widely reduce or eliminate the tax on dividends. In addition, even when dividends are potentially subject to a withholding tax, new techniques have developed to transform, through derivatives, those assets into exempt interest.¹¹⁴

The elimination of tax on interest income was unilaterally initiated by the United States in 1984, and other countries began to follow suit.¹¹⁵ Currently, fears of capital flight are likely to keep countries from changing this treatment. However, it has been accompanied with a lack of information reporting and lack of information sharing that allows U.S. citizens, who are liable for these taxes, to avoid them whether on income invested abroad or income invested in the United States channeled through shell corporations and trusts. Citizens of foreign countries can also evade the tax, and the U.S. practice of not collecting information contributes to the problem.

Based on actual tax cases, Guttenberg and Avi-Yonah describe a typical way that U.S. individuals can easily evade tax on domestic income through a Cayman Islands operation with little expense using current technology. The individual, using the Internet, can open a bank account in the name of a Cayman corporation that can be set up for a minimal fee. Money can be electronically transferred without any reporting to tax authorities, and investments can be made in the United States or abroad. Investments by nonresidents in interest bearing assets and most capital gains are not subject to a withholding tax in the United States.¹¹⁶

In addition to corporations, foreign trusts can be used to accomplish the same approach. Trusts may involve a trust protector who is an intermediary between the grantor and the trustees, but whose purpose may actually be to carry out the desires of the grantor. Some taxpayers argue that these trusts are legal but in either case they can be used to protect income from taxes, including those invested in the United States, from tax, while retaining control over and use of the funds.

Limited Information Reporting Between Jurisdictions

In the past, the international taxation of passive portfolio income by individuals has been easily subject to evasion because there was no multilateral reporting of interest income. Even in those cases in which bilateral information sharing treaties, referred to as Tax Information Exchange Agreements (TIEAs) were in place, they had limits. As pointed out by Avi-Yonah, most of these agreements were restricted to criminal matters, which are a minor part of the revenues involved and pose difficult issues of evidence. Also, these agreements sometimes required that the activities related to the information being sought constitute crimes in both countries, which can be a substantial hurdle in cases of tax evasion. The OECD has adopted a model agreement with the dual criminality requirements.¹¹⁷ TIEAs usually allowed for information only upon request, requiring the United States and other countries to identify the potential tax evaders in advance and they do not override bank secrecy laws.

In some cases the countries themselves have little or no information of value. One article, for example, discussing the possibility of an information exchange agreement with the British Virgin Islands, a country with more than 400,000 registered corporations, where laws require no identification of shareholders or directors, and require no financial records, noted: "Even if the BVI signs an information exchange agreement, it is not clear what information could be exchanged."¹¹⁸

U.S. Collection of Information on U.S. Income and Qualified Intermediaries

Under the QI program, the United States did not require U.S. financial institutions to identify the true beneficiaries of interest and exempt dividends. The IRS set up a QI program in 2001, under which foreign banks that received payments certify the nationality of their depositors and reveal the identity of any U.S. citizens.¹¹⁹ However, although QIs are supposed to certify nationality,¹²⁰ apparently some relied on self-certification.¹²¹ QIs are also subject to audit. However, UBS, the Swiss bank involved in a tax abuse scandal that helped clients set up offshore plans, was a QI, and that event raised some questions about the QI program.

A nonqualified intermediary must disclose the identity of its customers to obtain the exemption for passive income such as interest and or the reduced rates arising from tax treaties, but there are also questions about the accuracy of disclosures.

The FATCA provisions in P.L. 111-147 strengthened the rules affecting qualified intermediaries' identification of asset holders, with backup withholding provisions. The projected revenue gain was quite small (less than $1 billion per year) relative to projected costs (discussed below).

European Union Savings Directive

The European Union, in its savings directive, has developed among its members an option of either information reporting or a withholding tax. The reporting or withholding option covers the member countries as well as some other countries. Three states, Austria, Belgium, and Luxembourg, have elected the withholding tax. While this multilateral agreement aids these countries' tax administration, the United States is not a participant.

FATCA and the Common Reporting Standard

Recently, steps have been taken to provide for the automatic sharing of information. In the Hiring Incentives to Restore Employment (HIRE) Act of 2010, P.L. 111-147, the United States enacted the Foreign Account Tax Compliance Act (FATCA), which required foreign financial institutions to report beneficial owners of accounts or face withholding taxes. The implementation of FATCA took some time as it involved bilateral agreements, but most countries are now participating.¹²² FATCA applies only to institutions that receive payments from the United States. The OECD has adopted the automatic exchange of information (AEOI) using the common reporting standard (CRS) which, as noted above, has 112 signatories. The United States is a notable nonparticipant in CRS, relying on FATCA to provide information about its own citizens but not providing for reciprocity (which would probably require action by Congress). As a result, the United States, and particularly the laws of some of its states (such as Delaware, Nevada, South Dakota, and Wyoming) have caused it to be viewed by some as one of the major tax havens for individuals in other countries. Because of the delay in implementation of both FATCA and CRS, it has been difficult to measure the effectiveness. One study found that these provisions had reduced evasion by 67%.¹²³ A study of FATCA found that it reduced investment into the United States from tax havens during 2012-2015 by 21%.¹²⁴

Estimates of the Revenue Cost of Individual Tax Evasion

A number of different approaches have been used to estimate corporate tax avoidance, however, all of these approaches rely on data reported on assets and income. For individual evasion, estimates are much more difficult because the initial basis of the estimate is the amount of assets held abroad whose income is not reported to the tax authorities. In addition to this estimate, the expected rate of return and tax rate are needed to estimate the revenue cost.

Prior Estimates

Joseph Guttentag and Avi-Yonah estimate a value of $50 billion in individual tax evasion, based on an estimate of holdings by high net worth individuals invested outside the United States at $1.5 trillion.¹²⁵ Using a rate of return of 10% and a tax rate of approximately one-third, they obtain an estimate of $50 billion. They also summarize two other estimates in 2002 of $40 billion for the international tax gap by the IRS and $70 billion by an IRS consultant.

To the extent that the earnings are interest, the 10% rate of return may be too high, while if it is dividends and capital gains, the tax rate is too high. Using a tax rate of 15% (currently applicable to capital gains and dividends) would lead to about $23 billion. In the case of equity investments, if a third of the return is in dividends and half of capital gains is never realized, the tax rate would be 10% or about $15 billion assuming the 10% return. During 2002 and beginning in 2011, however, the tax rate on capital gains and dividends is 20%, indicating a loss of $20 billion rather than $15 billion. For interest, since investors can earn tax free returns in the neighborhood of 4% to 5% on domestic state and local bonds, to yield a 5% after-tax return at a 35% tax rate would require a pretax yield of about 7.7%. The estimate would then be $40 billion.

The Tax Justice Network has estimated a worldwide revenue loss for all countries of $255 billion from individual tax evasion, basically using a 7.5% return and a 30% tax rate.¹²⁶ These assumptions would be consistent with a $33 billion loss for the United States using the $1.5 trillion figure. Their worldwide numbers are consistent with $11 trillion in offshore wealth. Their more recent estimates place wealth at $21 trillion to $32 trillion, which would double or triple these estimates.¹²⁷ Thus the cost for the United States could be much larger approaching $100 billion.

Zucman estimates $1.2 trillion in U.S. financial wealth abroad based on anomalies in investment data, with an estimated tax loss of $36 billion in 2013.¹²⁸ There is no way to know whether the high-profile cases of prosecuting individuals, tax amnesty, or the imminent arrival of FATCA might have reduced these amounts of wealth.

Post FATCA/CRS Estimate

The Tax Justice Network has estimated a loss of $37 billion for the United States and $182 billion for the world.¹²⁹

Alternative Policy Options to Address Corporate Profit Shifting

Because much of the corporate tax revenue loss arises from activities that either are legal or appear to be so, it is difficult to address these issues other than with changes in the tax law. Outcomes would likely be better if there is international cooperation. Currently, the possibilities for international cooperation appear to play a bigger role in options for dealing with individual evasion than with corporate avoidance.

Several of the issues addressed below, such as hybrid entities and instruments, transfer pricing for intangibles, and debt also have been considered in the OECD action plan on base erosion and profit shifting.¹³⁰

Broad Changes to International Tax Rules

The first set of provisions would introduce broad changes in international tax rules.

Strengthen GILTI and Rules Preventing Corporate Inversions

One approach to mitigate the rewards of profit shifting is to strengthen GILTI, by decreasing deductions and imposing a per country limit on the foreign tax credit. The Biden Administration proposed to increase the GILTI tax rate to 21%, eliminating the deduction for tangible deductions, and imposing a per country foreign tax credit limit. In a separate provision, the corporate tax rate would be increased to 28%, so that GILTI would still not be taxed at full rates.¹³¹ This measure was estimated to raise $553 billion over ten years (or about $55 billion a year). The increased tax rates on GILTI would also allow changes in FDII, which was designed to help encourage holding intangible assets in the United States. The Administration proposal would have eliminated FDII with a revenue gain of $124 billion over 10 years.

Several congressional proposals would also increase the effectiveness of GILTI. S. 20 (Klobuchar), S. 714 (Whitehouse), H.R. 1785 (Doggett), and S. 991 (Sanders) would increase GILTI by taxing income at ordinary rates, eliminating the deduction for tangible assets, and providing for a per-country limit on the corporate tax. Except for S. 991, which also returns the corporate rate to 35%, these proposals for GILTI are the same as the Administration's proposal. The last three bills also repeal the deduction for FDII.

Following reconciliation, the House Build Back Better Act (BBBA; H.R. 5376 ) includes a number of corporate tax revisions, many of them similar to the Biden Administration's proposals and the various congressional proposals discussed above. There were two versions of the BBBA, one with the Ways and Means Committee legislative recommendations and one passed by the House.

The first version increases the corporate tax rate to 26.5% and makes a series of changes affecting GILTI and taxation of foreign source income in general. It accelerates the deduction for GILTI to the 37.5% now scheduled for after 2025, making the tax rate 16.5625%. It also accelerates the lower deduction for FDII, leading to a tax rate of 20.7% and allows a carryforward of unused GILTI and FDII deductions. It reduces the deduction for tangible assets from 10% to 5% and increases the share of foreign taxes credited from 80% to 95%. Foreign oil and gas extraction income would be included in GILTI. Under BBBA, GILTI income and loss applies on a per country basis (so that losses in one country could not offset gains in another country). It allows losses a one-year carryforward. The BBBA proposal applies the foreign tax credit limit on a per-country basis for all baskets: GILTI, passive, and active (it eliminates the branch basket). It no longer allocates interest and head office costs to foreign source income (increasing the limit). Unused foreign tax credits could be carried forward for 5 years rather than 10 years and the one-year carryback would be eliminated. These changes together raise revenues by $221 billion over 10 years.

A different version of the BBBA passed the House, which did not increase the corporate tax rate but instead imposed a minimum tax of 15% on financial income. Most of the international revisions are unchanged although adjustments are made in the GILTI and FDII deductions to directly increase tax rates (because the corporate rate does not increase). These changes increased the rate on GILTI to 15.015% and the rate on FDII to 15.792%. The Senate Finance Committee draft of the proposal contains similar provisions.

Senator Wyden, chairman of the Senate Finance Committee, along with Senators Brown and Warner, had previously proposed draft legislation that would eliminate the deemed deduction for tangible investment from GILTI. It would exempt income in countries with tax rates higher than the U.S. rate and impose a per country limit on foreign tax credits for the remaining countries as well as a per country limit on losses. The amount of any deduction, either GILTI or FDII, is yet to be determined, as is the share of foreign tax credits allowed (80% or more). The proposal would apply the same exclusion for countries with high tax rates and the same limit on the foreign tax credit to Subpart F income, and apply the high tax exclusion to branch income. (Currently, Subpart F income is excluded if taxes are 90% or more of the U.S. rate, with a similar rule applied through regulation to GILTI, but not to branch income. Both are eligible for credits for 100% of foreign taxes paid.) The income eligible for the deduction for FDII would be revised from a provision based on an estimate of intangible income to a percentage (not specified in the proposal) of research costs and certain worker training costs conducted in the United States. The deduction percentages for GILTI and FDII would be equated. Eligible training costs would be defined as apprenticeship and training programs that lead to a postsecondary credential and are provided to non-highly compensated employees.

Some of the issues surrounding strengthening GILTI have focused on the real effects of repeal on the allocation of capital. Traditionally, economic analysis has suggested that eliminating deferral would increase economic efficiency, although recently some have argued that this gain would be offset by the loss of production of some efficient firms from high-tax countries. The elimination of the deduction for tangible assets is focused on this issue.

Taxing GILTI at full rates would largely eliminate the value of the planning techniques discussed in this report. There are concerns, however, that firms could avoid the effects of full taxation by having their parent incorporate in other countries without taxes on foreign subsidiaries. The most direct and beneficial to reducing firms' tax liabilities of these planning approaches, inversion, has been addressed by legislation in 2004.¹³² Other legislative and regulatory changes have taken place as well so that little activity now takes place.¹³³ It would be possible to further limit inversions. The Administration's proposals treat as a U.S. firm any firm where former U.S. shareholders own 50% of the firm and some of the introduced bills contain this provision as well as treating a firm managed and controlled in the United States as a U.S. firm. These changes are also included in S. 1501 and H.R. 2976. The Senate Finance Committee draft adds a provision, not included in the House-passed bill, to tighten the rules for inversions, by treating inverted firms as domestic firms if the ownership is 65% (rather than 80% under current law). It also treats inverted firms as subject to taxes on gains for assets transferred if ownership is 50% (rather than 60% under current law). Mergers with minority ownership would be another method to avoid GILTI, although mergers involve real changes in organization that would not likely be undertaken to gain a small tax benefit. Another possibility is that more direct portfolio investment (i.e., buying shares of stock by individual investors) in foreign corporations will occur. There has been a significant growth in this direct investment, although the evidence suggests this investment has been due to portfolio diversification and not tax avoidance.¹³⁴

The increased tax rates in GILTI would also allow changes in FDII, which was designed to help encourage holding intangible assets in the United States. The Administration proposal would have eliminated FDII with a revenue gain of $124 billion over 10 years.

The OECD/G20 proposal for addressing worldwide profit shifting includes a provision to impose a worldwide minimum tax of 15%, the global base erosion, or GLoBE, tax.¹³⁵ Worldwide adoption of a minimum tax would also reduce profit shifting out of the United States by foreign multinationals. The United States, along with 137 other countries, including the G20, have agreed to this proposal.¹³⁶

Worldwide Allocation of Interest

Most of the major proposals discussed in the previous section also contain a provision for allocating interest deductions in the United States limited to the share of worldwide income. Specifically, the Administration proposal and the BBBA, as well as some other bills would limit the share of interest deducted to 110% of the share of worldwide earnings before interest, taxes, depreciation, and amortization (EBITDA). This provision would directly address profit shifting through borrowing and deducting interest in the United States. Stricter anti-inversion rules would also limit the ability to use debt to shift profits.

Altering or Strengthening BEAT

The Administration proposal would replace the current base erosion and anti-abuse tax (BEAT) with the stopping harmful inversions and ending low-tax developments (SHIELD), which disallows deductions for payments to related firms in tax havens. This proposal was estimated to raise $390.5 billion over 10 years. The BBBA (both the House-passed version and the Senate Finance Committee draft) would alter BEAT. The earlier version increases the BEAT tax rate from 10% (12.5% after 2025) to 12.5% in 2024 and 2025, and 15% after 2025 and allows tax credits. It adds to the base payments to foreign related parties for inventory that is required to be capitalized (such as inventory to produce tangible property) and payments for inventory in excess of cost. Because of the tax credits, the proposal raises only $26.7 billion over ten years. The House-passed version makes the same general changes but increases the BEAT rate to 12.5% in 2023, 15% in 2024, and 18% in 2025 and after. It is estimated to raise $24.9 billion over 10 years.

Formula Apportionment and the OECD Pillar One Proposal

Another approach to addressing income shifting is through formula apportionment, which would be a major change in the international tax system. With formula apportionment, income would be allocated to different jurisdictions based on their shares of some combination of sales, assets, and employment. This approach is used by many states in the United States and by the Canadian provinces to allocate income. (In the past, a three factor apportionment was used, but some states have moved to a sales based system.) Studies have estimated a significant increase in taxes from adopting formula apportionment. Slemrod and Shackleford estimate a 38% revenue increase from an equally weighted three-factor system.¹³⁷ A sales-based formula has been proposed by Avi-Yonah and Clausing that they estimate would raise about 35% of additional corporate revenue, or $50 billion annually over the 2001-2004 period.¹³⁸

The ability of a formula apportionment system to address some of the problems of shifting income becomes problematic with intangible assets.¹³⁹ If all capital were tangible capital, such as buildings and equipment, a formula apportionment system based on capital would at least lead to the same rate of return for tax purposes across high-tax and low-tax jurisdictions. Real distortions in the allocation of capital would remain, since capital would still flow to low-tax jurisdictions, but paper profits could not be shifted. An allocation system based on assets becomes more difficult when intangible assets are involved. It is probably as difficult to estimate the stock of intangible investment (given lack of information on the future pattern of profitability) as it is to allocate it under arms-length pricing. In the case of an allocation based on sales, profits that might appropriately be associated with domestic income as they arise from domestic investment in R&D would be allocated abroad. Moreover, new avenues of tax planning, such as selling to an intermediary in a low-tax country for resale, would complicate the administration of such a plan. Whether the benefits are greater than the costs is in some dispute.¹⁴⁰

One problem is that if the United States adopted the system, there could be double taxation of some income and no taxation of other income unless there were a multinational plan. The European Union has been considering a formula apportionment applied to its member states, based on property, gross receipts, number of employees, and cost of employment.¹⁴¹ This proposal and the consequences for different countries are discussed by Devereux and Loretz.¹⁴² If the European Union adopted such a plan it would be easier for the United States to adopt a similar apportionment formula without as much risk of double or no taxation with respect to its major trading partners.

The OECD/G20 proposal (Pillar 1) would apply a limited form of formula apportionment, by allocating a share of the residual profits of large digital companies to the market countries (where digital services are used, or where products are bought and sold in on-line market places).¹⁴³

Eliminate Check-the-Box, Hybrid Entities, and Hybrid Instruments

A number of proposals have been made to eliminate check-the-box and the look-through rules (S. 725, S. 991, and H.R. 1786). A more general change would require legal entities to be characterized in a consistent manner by the United States and the country in which an entity is established. This proposal has been made by McIntyre.¹⁴⁴ Rules requiring that legal entities be characterized in a consistent manner by the United States and by the country in which they are established and that tax benefits arising from inconsistent treatment of instruments be denied would address this particular class of provisions that undermine Subpart F and the matching of credits and deductions with income. President Obama's first budget proposal included a provision that disallows a subsidiary to treat a subsidiary chartered in another country as a disregarded entity.

Foreign Tax Credits: Source Royalties as Domestic Income for Purposes of the Foreign Tax Credit Limit or Create Separate Basket; Restrict Credits for Taxes Producing an Economic Benefit

As noted above, one of the issues surrounding the cross-crediting of the foreign tax credit is the use of excess credits to shield royalties from U.S. tax on income that could be considered U.S. source income. Two options might be considered to address that issue: sourcing these royalties as domestic income for purposes of the credit or putting them into a separate foreign tax credit basket.¹⁴⁵ President Biden's proposal and the BBBA include a provision to restrict the crediting of taxes that are in exchange for an economic benefit (such as payments that are the equivalent of royalties).

Options to Address Individual Evasion

Most of the options for addressing individual evasion involve more information reporting and additional enforcement. There are options that would involve fundamental changes in the law, such as shifting from a residence to a source basis for passive income. That is, the United States would tax this passive income earned in its borders, just as is the case for corporate and other active income. This change involves, however, many other economic and efficiency effects that are probably not desirable. The remainder of the proposals discussed here do not involve any fundamental changes in the tax itself, but rather focus on administration and enforcement.

Strengthening FATCA

FATCA applies only to foreign financial institutions that hold U.S. accounts, and does not apply to other financial institutions that may impede U.S. tax enforcement. The Stop Tax Haven Abuse Act (S. 725 and H.R. 1786) would provide sanctions modeled on anti-money-laundering provisions to encourage cooperation of other financial institutions. These would include prohibiting U.S. banks from dealing with offending foreign banks and ensuring that credit and debit cards issued by the foreign banks do not work in the United States. The bill would also create an evidentiary presumption that individuals who create or finance offshore entities control them. It also would create a presumption that money transferred offshore has not been taxed. The burden would be on the taxpayer to disprove these presumptions. The bill would strengthen FATCA disclosure requirements to ensure that checking accounts and derivatives are disclosed. It would legally require what is already in Treasury guidance that requires banks to comply with FATCA if they discover through money laundering due diligence that a foreign entity is controlled by a U.S. taxpayer. It would also allow the IRS to share taxpayer information with other regulators and law enforcement agencies and require foreign holding companies (passive foreign investment companies) to file tax returns. It would require banks and brokers that discover through money laundering due diligence that the beneficial owner of a foreign account is a U.S. taxpayer to disclose that information to the IRS. Other parts of the bills would extend the scope of money laundering due diligence to investment advisors to hedge funds and private equity funds.

The legislation also would increase the ability to use John Doe summons (where the identity of the taxpayer is not known) by presuming that payments to non-FATCA compliant banks involve tax compliance issues and to make it easier to issue multiple summonses.

Another possible adjustment to FATCA is to lower the minimum amount (currently $50,000) on which accounts must be reported. The common reporting standard used by other countries does not contain these minimum requirements.

Finally, an increase in IRS resources, which might require additional funding, could be needed to take full advantage of the data received by IRS under FATCA. A report by the Government Accountability Office (GAO) made recommendations to IRS for improved use of data. It also made legislative recommendations that overlapping filing requirements and limits between tax compliance and financial crimes be coordinated and that agencies have shared access to data.¹⁴⁶

Using Information from FBAR and Individual Income Tax Reporting

Individuals are required to file a report to the Treasury Department on foreign accounts that exceed $10,000, in a Report of Foreign Bank and Financial Accounts (FBAR) under Financial Crimes Enforcement Network (FinCEN). Individuals are also required to report foreign bank account information on their individual tax return for accounts of $50,000 or more (Form 8938). The GAO report recommended legislation to allow these reports to be coordinated or information shared.¹⁴⁷ The Stop Tax Haven Abuse Act provided that FBAR reports could be used in tax administration, and also changed the amount requiring an FBAR to the highest amount during the reporting period. It also clarified that suspicious activity reports could be used for tax administration.

FATCA and the Common Reporting Standard

Although the common reporting standard (CRS) was modeled on FATCA, the United States does not participate and does not provide reciprocal information to other countries. Although providing this information does not aid the IRS in collecting revenue, it leads to the United States (because of various state laws that do not disclose beneficial owners) to be considered as a major tax haven by other countries. Legislative changes would probably be required to provide full reciprocity and it would increase burdens on U.S. financial institutions. At the same time, if another major bloc of countries (such as the European Union) were to impose withholding taxes on payments to U.S. banks, such reciprocity would be needed to avoid withholding.

Another alternative is to replace FATCA with participation in CRS.¹⁴⁸ This shift would have the advantage of imposing only one type of reporting standard and thus would be simpler for foreign financial institutions. As with reciprocity, it would probably require legislation.

Incentives/Sanctions for Tax Havens

Avi-Yonah and Guttenberg suggest a carrot and stick approach to tax havens.¹⁴⁹ They argue that little of the benefit of tax havens flows to their sometimes needy residents, but rather to the professionals providing banking and legal services, who often live elsewhere. They suggest transitional aid to move away from these offshore activities. For non-cooperating tax havens, they suggest the Treasury use its existing authority to deny benefits of the interest exemption. They suggest that tax havens cannot continue to exist unless the wealthy countries permit it, because funds are not productive in tax havens.