CRS Analyzes Likely Results of ETI Repeal Bills

The 108th Congress has considered a set of alternative tax proposals that could affect how U.S. firms compete in the world economy and how they allocate investment between U.S. and foreign locations. A principal impetus for the bills was a dispute between the United States and the European Union (EU) over the U.S. extraterritorial income (ETI) tax benefit for exporting. The EU complained that the provision is an export subsidy and thus violates the World Trade Organization (WTO) agreements; a succession of WTO rulings supported the EU position and the WTO authorized the EU to impose retaliatory tariffs on U.S. goods. All of the major international tax proposals that were introduced in the 108th Congress addressed the dispute by simply repealing the ETI benefit.

The bills differed in how they addressed the economic impact of ETI's repeal. H.R. 1769 and S. 970 (the Crane/Rangel/Hollings proposal) proposed to replace ETI with a tax benefit for domestic production, including both export and other types of income. Chairman Thomas of the House Ways and Means Committee proposed H.R. 2896, containing tax cuts for investment abroad as well as benefits for domestic investment; the bill was approved by the Ways and Means Committee in October, 2003. Also in October the Senate Finance Committee approved S. 1637, containing an alternative mix of tax cuts for domestic and overseas investment. The full Senate approved the measure in May, 2004. In June, Representative Thomas introduced a modified version of his earlier bill as H.R. 4520; the full House approved the bill on June 17. A conference committee version of the legislation was approved by both chambers in later October; the President signed it into law as Public Law 108-357.

A focus of the debate over the proposals was their prospective impact on domestic U.S. employment, driven in part by concern over the impact of ETI's repeal on U.S. jobs. Economic analysis indeed suggests that in the short run, repeal of ETI would result in the loss of a certain amount of jobs in the U.S. export sector, although ETI's role in the economy is probably not large. The alternative investment incentives provided by the bills would also likely result in the shift of labor and other resources to sectors that qualify for the benefits, for example, domestic manufacturing. Economic theory, however, also indicates that in the long run the bills' likely impact on the overall level of domestic employment would be minimal; in the long run the economy tends towards full employment and labor released by one sector will be absorbed by other sectors. In analyzing international taxes, traditional economics instead focuses on how taxes affect the location of investment -- on firms' decision of whether to invest at home or abroad -- and on the efficiency and economic welfare effects that follow. Each bill contains both provisions that favor domestic investment and changes that favor foreign locations; each bill essentially pulls the system in different directions at the same time. The net impact of the respective plans is uncertain, although it is likely that none would change the basic nature of the U.S. system in providing a patchwork of incentives and disincentives towards location of investment and in its mix of efficiency effects. This report will not be updated.

The Proposals

H.R. 1769 and S. 970: The Crane/Rangel/Hollings Proposal

H.R. 2896 and H.R. 4520

S. 1637: The Senate Bill

Other Proposals

Economic Effects

Investment Effects

Efficiency and Welfare Effects

The issue of how U.S. federal taxes apply to the international economy has risen to a prominent place in tax policy debates on numerous occasions since World War II. For example, the 1960s saw Congress approve the tax code's Subpart F provisions, whose purpose was to restrict the ability of multinationals to avoid U.S. taxes by shifting earnings to offshore tax havens. In the 1970s, taxation of income from overseas investment was prominent in congressional tax policy debates, and the Carter Administration proposed elimination of both the Domestic International Sales Corporation (DISC) export tax benefit and the deferral tax benefit for overseas business operations. (The proposals were later withdrawn.) And in 1986, an important feature of the landmark Tax Reform Act was a wide ranging set of reforms in the rules for taxing foreign source income.

But major tax legislation in the international area has, with a few exceptions, not occurred since 1986.¹ Two factors, however, may bring an end to the hiatus. One is the dispute between the United States and the EU over the ETI export benefit that has occurred within the World Trade Organization (WTO). In response to EU complaints, the WTO has ruled that the ETI benefit is an export subsidy and is thus prohibited by the WTO agreements. Under WTO procedures, the United States is required to bring its tax code into WTO compliance or face WTO-sanctioned retaliatory tariffs. Designing an appropriate response to the ETI/WTO dilemma thus poses a time- sensitive policy challenge for Congress. A second factor pressing international taxation into the policy debate is the increased integration of the U.S. economy with that of the world at large. U.S. businesses increasingly are competitors in international as well as domestic markets; capital investment flows increasingly freely across national boundaries. Prominent policymakers in both Congress and the executive branch have suggested that this increased openness of the U.S. economy calls for profound changes in how the United States taxes international transactions. Other observers, however, have suggested that the basic economic principles underlying the tax systems have not changed, and a fundamental reorientation of tax policy is not required.

The most direct economic effect of the taxes that apply to international income is on the flow of investment between the domestic U.S. economy and foreign locations; taxes affect the attractiveness of investing overseas compared to domestic investment and thus can directly affect the extent to which U.S. businesses operate abroad. It is thus not surprising that the policy debate over international taxation has tended to focus on the potential impact of various proposals on the balance between foreign and domestic investment, and on the employment and income effects that might follow. Each of the most prominent legislative proposals in international taxation addresses the ETI controversy in basically the same way: each would phase out the ETI export tax benefit without replacing it with a redesigned export subsidy. But in line with the policy debate, the proposals differ in the tax changes they prescribe to make up for ETI's impact on employment, investment, and the ability of U.S. firms to compete.

To understand the implications of the legislative proposals for the structure of the U.S. international tax system, we look next at the basic components of the U.S. tax system in its international context.

But there are exceptions to this general structure. First, the United States grants tax credits for foreign taxes paid. While the United States taxes its residents' worldwide income, it concedes that the country of source has the primary right to tax that income and permits its corporate and non-corporate taxpayers to credit foreign income taxes they pay against U.S. taxes they would otherwise owe. In so doing, the United States, in effect, accepts the responsibility for alleviating the double-taxation that would result when the U.S. worldwide tax jurisdiction overlaps the normal practice of host countries in taxing income earned within their borders. Importantly, however, to protect the U.S. tax base, the U.S. foreign tax credit is limited to offsetting U.S. tax on foreign income; foreign taxes cannot be credited against U.S. tax on U.S. income. The tax credit's limitation and associated rules give rise to some of the most complex parts of the tax code, as described further in the sections below on the particular proposals.

Along with the foreign tax credit, another exception to U.S. worldwide taxation is the so-called "deferral" principle. While the United States taxes foreign income earned directly by branches of U.S. corporations -- branches that are not separately incorporated abroad -- the United States does not tax foreign-chartered corporations on their foreign-source income. Thus, if a U.S. firm conducts foreign operations through a subsidiary firm chartered abroad, the foreign income is not subject to U.S. tax until the income is remitted to the U.S. parent as dividends or other income (at which point it enters the U.S. tax jurisdiction as income of a U.S.-resident corporation). U.S. tax on the subsidiary's income is thus tax-deferred as long as the income is reinvested abroad.

Deferral poses a tax incentive for U.S. firms to invest abroad in countries with relatively low tax rates and reduces U.S. tax revenues. Since 1962, however, the tax code's Subpart F provisions have denied deferral's benefit to certain types of income, generally income from passive investment and other income whose source is thought to be easy to manipulate in order to reduce taxes.

Where do exports fit in? As noted at the report's outset, a principal impetus for the proposals is the dispute between the United States and the EU over the ETI benefit. Under the United States' residence-based tax system, U.S. taxes would normally apply to export income in full. If a U.S. corporation were to sell exports directly, U.S. worldwide taxation would ordinarily ensure full U.S. taxation. If a U.S. firm were to sell exports through a related foreign subsidiary outside the U.S. tax jurisdiction, U.S. "transfer pricing" rules (rules governing the allocation of income among related firms) would restrict the extent to which export income could be allocated abroad to a foreign subsidiary outside the U.S. tax jurisdiction. To the extent flexibility in the application of transfer pricing permits the allocation of income to foreign subsidiaries, Subpart F apparently rules out much of the potential for deferral to apply.

Notwithstanding these rules, however, several provisions of the U.S. tax code provide tax benefits for U.S. exports. The ETI exclusion is one of these; its provisions permit U.S. exporters to exclude between 15% and 30% of their export income from tax. As mentioned above, several WTO rulings and actions by the EU have led to the possibility of retaliatory tariffs being applied by the EU to U.S. products.²

Tax legislation can have an impact on the flow of investment to and from the United States by modifying any of the multiple rules that make up this structure. As described below, however, a number of important components of the current legislative proposals seek to boost domestic investment by providing tax benefits that are available only in the United States, and by changing the treatment of domestic investment rather than altering provisions for foreign- source income. It is the relative tax treatment of domestic and overseas investment that matters, not the treatment of foreign investment in isolation. Thus, tax provisions that apply primarily to domestic investment can affect the flow of investment abroad. To illustrate, current law's modified accelerated cost recovery system (MACRS) of depreciation is available only for domestic investment; alteration of MACRS can alter the relative treatment of domestic and foreign investment, and thus affect investment flows. (Indeed, as described below, H.R. 2896 would liberalize MACRS rules.) Also, changes that are applicable to U.S. business generally, such as reductions in the corporate tax rate, can change the relative treatment of domestic and foreign investment because only a fraction of overseas investment is included in the U.S. tax base.

We turn now to the content of the current proposals.

H.R. 1769 and S. 970: The Crane/Rangel/Hollings Proposal

On April 11, 2003, Representatives Crane and Rangel introduced H.R. 1769; on May 1, Senator Hollings introduced an identical bill in the Senate as S. 970. The proposal was intended to resolve the FSC/ETI controversy by gradually repealing ETI while phasing in a new tax benefit restricted to domestic production. The transition period would have been 2003-2009. According to the bill's sponsors, the proposal was designed to be roughly neutral with respect to tax revenue, with the revenue gain from ETI's repeal offsetting the revenue loss from the production benefit.

In general, the fully phased-in domestic production benefit consisted of a deduction from a firm's taxable income that was equal to 10% of the firm's "qualified production activities income" as defined by the bill. (For a firm taxed at the maximum 35% corporate tax rate, this would have had the same effect as a 3.5 percentage point rate reduction.) The bill defined qualified production activities income, in turn, as that portion of the firm's taxable income generated by domestic (and not foreign) production multiplied by what the bill defined as the firm's "domestic foreign fraction." Under the bill, the numerator (top) of the domestic foreign fraction was the value of the firm's domestic production and the denominator was the value of its worldwide production, with "value" being similar to the economic concept of "value added" -- that is, the amount contributed to the price of a good by the cost of inputs at various points in the production process. In effect, then, a firm's deduction was smaller, dropping from 10% towards zero, the more intensive its foreign operations. In the policy debate over the ETI bills, this provision was sometimes referred to as a "haircut," and was contained in several of the Senate proposals, as described below.

The transition aspects of H.R. 1769/S. 970 consisted of a phaseout of the ETI benefit and phase in of the domestic production benefit, both over the period 2003-2009. The phaseout of ETI depended on the ETI benefit a firm received in a specific "base year," the year 2001, rather than on exports over the phase-out period. The benefit during the transition years was a gradually declining percentage of the 2001 benefit (although the base amount would be inflated in the last few years of the transition period, presumably to reflect economic growth and inflation). More specifically, for 2003 -- 2005 a firm could claim the full amount of its base 2001 benefit; for 2006 and 2007 it could claim 75% of the base benefit, and for 2008 it would receive half the benefit. The benefit would have been completely eliminated for 2009 and beyond.

H.R. 2896 and H.R. 4520

On July 25, 2003, Chairman Thomas of the House Ways and Means Committee introduced H.R. 2896, a bill substantially broader than H.R. 1769/S. 970. The bill was approved by the Ways and Means Committee on October 28. In broad outline the bill proposed to repeal ETI over a three-year transition period and enact in its stead a mix of tax reductions for domestic as well as foreign operations. The bill also contained several revenue-raising items apart from its repeal of the ETI benefit. With a few differences, the international provisions and revenue-raising items of H.R. 2896 were the same as those initially proposed by Representative Thomas in the 107th Congress, as part of H.R. 5095. H.R. 2896, however, differed from H.R. 5095 in the addition of several substantial tax benefits for domestic rather than foreign investment.

According to estimates by the Joint Committee on Taxation, the bill would have reduced U.S. tax revenue by $21.1 billion over five years and by $59.8 billion over 10 years.⁴ The largest portion of the estimated revenue loss was attributable to the bill's proposed incentives for domestic investment; they accounted for $89.5 billion, or two-thirds of the bill's estimated $134.5 billion gross revenue loss over 10 years. The largest of the domestic tax cuts, in turn, was a proposal to cut in the maximum corporate tax rate applicable to domestic production to 32% from current law's 35%. Among its other provisions was a proposal to cut the tax rate applicable to the middle range of corporate income, a two-year extension of an increase in the "expensing" investment benefit for small business, and a relaxation of alternative minimum tax (AMT) restrictions on the deduction of losses ("net operating losses," or losses as defined under the tax code), and more generous rules for depreciation of leasehold improvements.

The proposal's tax cuts for foreign-source income accounted for $41.2 billion, or about one-third of the bill's estimated 10-year gross revenue loss. The bill's modifications generally provided more generous rules relating to the foreign tax credit's limitation and restricted the applicability of Subpart F, thereby expanding the scope of the deferral benefit. For the foreign tax credit, the most important change was an alteration of the rules for allocating interest expense between domestic and foreign sources. The bill implemented a new allocation formula sometimes called "worldwide" allocation that would generally result in a smaller portion of interest expense being allocated to foreign sources. This is an allocation that, in turn, would increase maximum creditable foreign taxes for some firms. In addition, the bill proposed to reduce to two the number of separate foreign tax credit limitations a firm would be required to calculate. (Prior law required up to nine separate limitations.) This change would increase the ability of firms to credit foreign taxes paid with respect to one stream of foreign income against U.S. tax due on a different stream of income. With deferral and Subpart F, the largest proposed change was a relaxation of "foreign base company" sales and service income rules. In general terms, this is income attributable to sales and services income generated by transactions between related corporations that are organized in different countries, and is subject to current taxation rather than under Subpart F. The bill treated the countries of the European Union as a single country for purposes of Subpart F's base company sales and service rules, thus restricting the scope of income classified as sales and service base company income.

Taken alone, the bill's revenue-raising provisions would have increased revenue by an estimated $23.5 billion over five years -- an amount equal to about 51% of the bill's gross revenue-losing items. The principal revenue-raiser was repeal of ETI, which would have increased revenue by an estimated $11.9 billion over five years. The remaining revenue-raising items were a set of provisions designed to restrict the use of tax shelters, proposals designed to restrict tax avoidance through a technique known as "earnings stripping," and provisions designed to reduce the corporate tax savings available from reorganizing to include a foreign parentcorporation ("inverting"). Prominent among the tax shelter proposals were provisions related to penalties and disclosure requirements. The earnings stripping provisions placed restrictions on deductible interest and similar payments between related firms. The inversion provisions stopped short of taxing foreign parent corporations in the same manner as U.S. corporations but applied taxes to certain gains of the inverted corporation and its officers.

The version of H.R. 2896 that was approved by the Ways and Means Committee differed in a number of respects from the bill as it was originally introduced. For example, the Committee bill did not include a proposal that was in the initial version of bill that would provide a temporary 80% tax deduction for foreign-source income firms repatriate to the United States rather than reinvesting abroad. (As noted below, however, a similar provision was contained in the Senate bill.) In addition, the original bill would have substantially liberalized depreciation allowances, provided more generous net operating loss rules, and extended and modified the research and experimentation tax credit. The Committee bill contained none of these provisions, but instead reduced the maximum tax rate for manufacturing income, as described above.

As described below, the Senate approved an ETI bill in May, 2004, before the full House began consideration of the Ways and Means bill. On June 4, Representative Thomas introduced a modified version of the 2003 Ways and Means bill as H.R. 4520. While the new bill had the same general thrust as H.R. 2896 -- repeal of ETI with a mix of domestic and foreign tax cuts -- it contained some changes. In the domestic area, the proposal's central provisions -- the tax rate reductions -- were the same as in H.R. 2896: the bill reduced the top tax rate to 32%. The bill also retained the temporary increase in expensing benefits, relaxation of AMT restrictions on losses, and more generous leasehold depreciation. The new bill added a set of provisions that would extend a list of relatively narrow temporary tax benefits and tax-reducing provisions (often called the "extenders") such as the research and experimentation tax credit and the work opportunities tax credit. The bill also contained a provision that would permit individual taxpayers to deduct state and local sales taxes rather than income taxes.

In the foreign area, the new bill's differences from H.R. 2896 were more pronounced. The measure retained H.R. 2896's revised interest expense rules for the foreign tax credit and consolidated the number of separate foreign tax credit limitations. H.R. 4520 did not contain, however, the previous bill's relaxation of Subpart F base company sales rules and it added a temporary 85% tax deduction that applied to dividends repatriated from foreign subsidiaries. The size of the deduction's tax reduction would have been similar to the temporary 5.25% reduced tax rate contained in the Senate-passed bill, as described below.

The bill's revenue-raising items also differed in certain respects from H.R. 2896. The new measure did not contain earnings- stripping provisions, but added new restrictions on the use of leasing transactions to transfer tax benefits, and included several fraud-related provisions in the area of energy taxation.

According to the Joint Committee on Taxation, the bill would have reduce revenue by an estimated $32.0 billion over FY2004-FY2009 and by $35.3 billion over FY2004-FY2014. Its expected revenue loss in the near-term was thus larger than that of H.R. 2896 in the near term, but smaller in the long run.

S. 1637: The Senate Bill

Senators Grassley and Baucus proposed S. 1637 on September 18, 2003; the Senate Finance Committee approved an expanded and modified version of the bill on October 1. The full Senate began debate on the bill in March 2004 and approved the bill on May 11. (On July 15 the Senate approved H.R. 4520, amended to include the contents of S. 1637 rather than the provisions in the House version of the bill.) In general terms, the bill proposed to phase out ETI over a four-year period, and -- like the House-passed bill -- implement a mix of tax benefits both for domestic and overseas investment. Like the House bill, the Senate proposal contained a set of revenue-raisers, although they differed in their particulars. Unlike the House bill, the Senate bill was nearly revenue "neutral," raising slightly more revenue than it would have lost over its first ten years. According to Joint Tax Committee estimates, the bill as passed by the full Senate would have increased tax revenues by $2.8 billion over ten years. (The bill would have reduced revenue by an estimated $14.6 billion over five years.)

The bill's principal tax benefit for domestic investment was a deduction/ratereduction similar in design to that of the Crane/Rangel proposal; the deduction would have been phased in over five years. When fully phased in, it would have consisted of a deduction equal to nine percent of income from property "manufactured, produced, grown, or extracted" in the United States. Until 2013, the percentage was scheduled to be reduced by an amount related to a firm's overseas operations in a manner similar to the "haircut" of the Crane/Rangel proposal. For a firm subject to the maximum 35% corporate tax rate and with no foreign operations, the deduction would have had the same effect as a rate reduction of slightly over three percentage points. The bill also provided a more generous phase-out procedure for the Section 179 "expensing" allowance for equipment investment and an extension of the net operating loss carryback period to three years from current law's two. In addition, the Senate bill provided a temporary (one-year) reduction to 5.25% in the tax rate on dividends repatriated to U.S. parent firms from overseas subsidiaries. As noted above, a similar provision was included in the House-passed version of H.R. 4520.

On the international side, there was substantial overlap between the House and Senate bills (H.R. 4520 and S. 1637), but also some differences. The bills contained similar revisions in the foreign tax credit-related rules for allocating interest expense -- perhaps the most important of the legislation's international provisions. Differences included the Senate bill's carryback and carryforward provisions for the foreign tax credit and the House bill's consolidation of separate foreign tax credit limitations.

Prominent revenue-raising items in the bill -- aside from ETI's repeal -- included provisions aimed at tax shelters, corporate governance provisions, and more stringent rules for both corporate and individual expatriation.

Other Proposals

Aside from the initial Crane-Rangel proposal and the two committee-passed bills, a number of other proposals were been introduced that contained variations on some of the same concepts. Senator Hatch introduced S. 1475, the Promote Growth and Jobs in the USA Act, on July 29, 2003. Like the Ways and Means and Senate bill, S. 1475 proposed to phase out the ETI export benefit over a short transition period. And like the Committee bills, S. 1475 proposed a set of tax cuts that would apply to foreign-source income and a set of benefits that would be restricted to domestic investment. Many, but not all, of the foreign provisions of S. 1475 were substantially the same as the foreign provisions of H.R. 2896; the domestic provisions were similar, but not identical. In contrast to the Ways and Means Committee bill, the Hatch proposal contained no earnings stripping, corporate inversion, or tax shelter provisions. Detailed revenue estimates are not available for the bill, but Senator Hatch indicated that the bill would likely reduce revenue by approximately $200 billion (presumably over 10 years).⁵

The Hatch plan's tax benefits for domestic investment included more generous depreciation and expensing allowances for domestic investment in equipment. The bill proposed a one-year extension of JGTRRA's bonus depreciation, but would have increased the first year allowance to 100% of an asset's cost from 50%. S. 1475 would also have extended JGTRRA's $100,000 expensing allowance for equipment for one year. The bill also would have made the research and experimentation tax credit permanent, and provided an additional alternative method of calculating the credit.

S. 1475's foreign proposals included more generous foreign tax credit-limitation calculations and a scaling back of Subpart F's restrictions of the deferral benefit. Like H.R. 2896, the Hatch plan adopted the "worldwide" method of allocating interest and reduced the number of foreign tax credit baskets to two. With Subpart F, the bill would have removed foreign base company sales and service income, although it included certain restrictions related to transfer pricing. (As noted above, the Committee-approved version of H.R. 2896 would have restricted the scope of the base company provisions rather than repealing them altogether.)

S. 1475 would have provided a tax cut for earnings repatriated from foreign subsidiary corporations. The tax cut would have the effect of reducing the tax rate applicable to repatriations to 15% of the currently applicable rate -- thus reducing the tax rate to 5.25% for a firm normally subject to the top corporate rate of 35%. In contrast to the Senate bill's provision the Hatch proposal's tax cut would have been permanent.

Senator Rockefeller introduced S. 1688 on September 30, 2003. Like the Senate plan, the Rockefeller bill phased out the ETI benefit over 2003-2006, with the transition amount equal to a declining percentage of a firm's 2002 ETI benefit. And like the Committee plan, the bill provided a 9% deduction for domestic production that would have been phased in over 2003-2008. In contrast to the Committee bill, there was no "haircut" reduction for foreign production. The bill did however, propose a tax credit for employers who pay health insurance expenses of retired employees.

Senators Smith and Breaux introduced S. 1922 on November 22, 2003. The bill would have phased out ETI and phase in a 9% domestic production deduction in the same manner and over the same period as the Senate and Rockefeller bills. Like the Rockefeller bill, the Smith-Breaux proposal contained no haircut reduction for foreign production. It also added a link with domestic employment: the maximum production deduction would be limited to 50% of a firm's wages reported on W-2 forms. S. 1964 was introduced by Senators Stabenow and Graham on November 25, 2003. It phased out ETI in the same manner as S. 1637, S. 1688, and S. 1922 and also provided a 9% domestic production deduction. Unlike the other bills, however, the deduction would have been effective beginning the first year after enactment -- that is, without a phase-in period. It also contained no reduction for foreign production and limited the deduction to 50% of wages.

The Conference Agreement on H.R. 4520

On October 6, 2004, House and Senate conferees approved an agreement on H.R. 4520. The full House approved the agreement on October 7, and the Senate on October 11. The President signed the bill on October 22; it became P.L. 108-357. A summary of the provisions contained in the conference agreement is provided by CRS Report RL32652, The 2004 Corporate Tax and FSC/ETI Bill: The American Jobs Creation Act of 2004.

Like the House and Senate bills, the conference agreement was quite broad in scope and focused on business taxed; it contained a broad range of both tax cuts and tax increases in areas of the tax code other than the ETI provisions. However, the centerpiece of the agreement was a repeal of the ETI benefit on the one hand, and provision of a tax benefit for domestic production on the other, along with a number of tax cuts for firms with overseas production. The agreement followed the Senate's version of the domestic production benefit, providing a deduction rather than a taxrate cut for domestic production, although it omitted the "hair cut" provision described above. After a phase-in, the deduction is scheduled to be 9% of taxable income. As with both the House and Senate bills, the largest tax reduction for multinational firms is an alteration of the rules for allocating interest expense in connection with the foreign tax credit limitation.

Other prominent differences between the House and Senate bills included a number of items included in the House bill, but not the Senate legislation. These included a number of tax cuts, including a rate reduction for lower levels of corporate income, an extension of the "expensing" benefit for equipment investment, liberalized depreciation for leaseholds, and an option for individual taxpayers to deduct state and local sales taxes rather than income taxes. Of these, the October conference agreement did not provide the rate- reduction, but included the other three of these items. Likewise, the Senate bill contained a number of investment tax cuts not in the House bill, including tax incentives related to energy and an extension of the allowable carryback period for tax losses (net operating losses, or NOLs). These two items were not in the conference bill.

Notwithstanding this debate, traditional economic theory evaluates international taxes in somewhat different terms. First, while not denying that ETI's repeal could lead to short-term unemployment and transitional costs, theory predicts the economy will adjust and the lost jobs would in the long run be restored in non-export sectors of the economy. Further, this long-run adjustment would occur without provision of alternative investment incentives.⁶ Second, economic theory is skeptical of the usefulness of the concept of competitiveness in setting broad tax policy and focuses instead on how various tax policies affect economic welfare at home and abroad.

Rather than employment or competitiveness, economic theory begins its analysis of international tax policy by looking at the impact of taxes on the location of investment: how do various international tax policies affect firms' decisions on where to employ their investment resources and establish operations? How do those decisions affect capital flows to and from the United States and foreign locations? Business income taxes are, after all, taxes on the return to capital investment, and it is logical that the most fundamental and immediate impact of those taxes will be on how and where capital is used.

Various other effects follow from the impact of taxes on investment: shifts in investment may affect sectoral employment in the short run, although, as noted above, aggregate employment is not altered in the long run. Changes in investment can also alter how income is distributed among economic actors and groups. For example, other factors being equal, the higher the capital/labor ratio in an economy, the larger is the share of income that accrues to labor. This is because the more capital labor has to work with, the more productive labor is, and the higher real wages are. On the other hand, capital income is lower, the higher is the capital/labor ratio. These effects may help explain the traditional split between business and labor interests over the appropriate level of taxes on foreign vis a vis domestic investment.

Although recognizing the existence of these effects, the ultimate focus of economic analysis is on how taxes affect economic efficiency through the allocation of investment, and on how taxes thereby affect economic welfare: the more efficient the economy, the greater is economic welfare. We return to these concepts at the close of this section, but first begin by assessing the plans' most immediate impact: how the proposals may affect the allocation of investment between foreign and domestic locations. Due to the complexity and variety of the proposals in the bills we do not attempt to discern the overall, net impact of each plan, and instead limit our analysis to the likely impact of their most important components. Further, we confine our analysis of specific proposals to the first four ETI bills introduced in the current Congress: H.R. 1769/S. 970 (Crane-Rangel proposal); H.R. 2896 (the Ways and Means bill); S. 1475 (the Hatch bill); and S. 1637 (the Senate plan).

Investment Effects

According to economic theory, taxes affect international business investment by altering the relative attractiveness of domestic versus foreign investment. Other factors being equal, if taxes on income from a foreign investment are lower than taxes on an identical domestic project, a firm will have an incentive to undertake the foreign investment; if taxes on domestic investment are low compared to identical foreign investment, firms will have an incentive to undertake the domestic investment. And if taxes are the same in either location, they have no influence -- are "neutral" towards -- the location of investment. Alteration of the relative tax burden on domestic and foreign investment can therefore alter the share of the economy's investment capital that is employed, respectively, at home and abroad.

The ETI benefit does not explicitly apply to foreign or domestic investment but its repeal, the starting point for each of the bills, would nonetheless change this calculus. An export, by definition, is the sale of a good produced in the exporter's home country. Thus, current law's ETI benefit poses an incentive to employ capital in the United States rather than abroad and, in isolation, the provision's repeal would encourage the shift of a certain amount of investment out of the United States to foreign locations as well as to alternative non-export uses within the United States.

But each of the bills also proposed tax benefits targeted directly at domestic investment.⁷ These measures would work in the opposite direction from ETI's repeal, and -- in isolation -- would encourage the shift of investment from foreign locations to the United States. In this category are the deductions for income from domestic production in the Crane/Rangel/Hollings proposal and the Finance Committee bill and the reduced maximum tax rate in the Ways and Means bill for domestic production. Under H.R. 1769, for example, depending on the extent of a firm's foreign operations, a firm could deduct from taxable income up to 10% of qualified income, a provision that would be similar to a statutory tax rate reduction of up to 3.5 percentage points. Further, the deduction would be larger, the less intensive a firm's overseas operations.

The depreciation and research and development proposals of the Hatch bill and the expensing provision of both S. 1475 and H.R. 2896 would similarly favor domestic over foreign investment, since each of these provisions would apply only to domestic assets. For example, foreign research outlays do not qualify for the R&E credit; augmentation of the credit would thus, in isolation, result in the shift of a certain amount of investment from foreign locations to domestic research and development. Similarly, the "bonus" depreciation provided by S. 1475 would not apply to foreign assets nor would the extension of expensing proposed in both S. 1475 and H.R. 2896. The general effect of these provisions, then, would be to cut taxes on a range of domestic investments compared to identical foreign investments. In isolation, these provisions would therefore likely result in an increased share of investment occurring in the United States rather than abroad.

Two additional provisions that would favor domestic over foreign investment are not explicitly targeted at domestic investment: the interest allocation rules contained H.R. 2896, S. 1475, and S. 1636; and H.R. 2896's rate cut for intermediate levels of corporate income. H.R. 2896's reduced maximum tax rate would explicitly apply only to domestic investment, but its rate cut for intermediate levels of income -- a cut not restricted to domestic production -- would also favor domestic investment but in a more indirect way. Because of deferral and the foreign tax credit, a larger share of U.S. investment than foreign investment is subject to U.S. statutory tax rates and a rate cut thus disproportionately benefits domestic investment. The impact of the interest allocation rules is perhaps harder to see: the provision would reduce taxes, after all, only for firms that have foreign operations. However, under both the current and proposed interest allocation rules, the more foreign assets a firm has, the more interest is allocated abroad, reducing foreign tax credits. And the thrust of the proposed rules change is to increase the weight given to foreign assets in the allocation calculation. Thus, the proposal will increase the tax disincentive to invest abroad. Overall, however, a multinational's taxes would nonetheless be reduced, implying that domestic assets would be the investments benefitting from the reduction.

Other foreign tax credit provisions of H.R. 2896 and S. 1475, however, would unambiguously reduce the relative tax burden of foreign compared to domestic investment. The bills' proposed consolidation of foreign tax credit baskets would permit firms to achieve more cross-crediting than under current law; there would accordingly be more situations where high foreign taxes on one stream of income could offset U.S. tax due on more lightly taxed income. As a result, there would be more cases where lightly taxed foreign investments would be shielded from any additional U.S. tax; there would be more situations where firms would have a tax incentive to increase their overseas investment. In isolation, the reduction of baskets would likely increase the level of foreign investment from what would otherwise occur.

Most of the bills' Subpart F provisions would likely also reduce the tax burden on foreign compared to domestic investment. For example, the proposal to expand the Subpart F's look-through rules would remove a range of foreign investments from Subpart F's coverage, thus increasing the scope of foreign investment for which there is a tax incentive. Similar results would likely flow from other proposals to rescind Subpart F coverage, for example, shipping income and gain from the sale of partnership interests.

But what of the largest Subpart F proposals -- the proposals to repeal (S. 1475) or relax (H.R. 2896) Subpart F's foreign base company sales and service rules? At first glance, these provisions seems likely to encourage foreign over domestic investment: they would permit U.S. tax on income allocated to foreign sales subsidiaries to be deferred, thus cutting taxes on investment in foreign sales activity. But any impact in increasing foreign investment would likely be small: if it is assumed that transfer pricing rules result in the accurate allocation of income between domestic parents and foreign subsidiaries, the provisions would likely have only a small effect. The value added by sales activity alone is a small portion of the total value of a product; only a small amount of income would therefore likely be allocated to sales subsidiaries. Further, in addition to the limited impact on foreign investment, some of the repeal's benefit might accrue to domestic investment. If there were to be difficulty in accurately allocating income, firms might be able to shift what is actually U.S. income or currently taxed foreign income to sales subsidiaries. For example, absent workable transfer pricing rules, export income that is actually earned by a U.S. parent corporation could be shifted to a foreign sales subsidiary and benefit from the deferral benefit.⁸ Accordingly, depending on whether the actual source of the shifted income is domestic or foreign, the base company sales and service proposals could increase either foreign or domestic investment; the outcome is thus not clear.

The final provision we assess -- the temporary reduced tax rate for repatriated dividends -- is also ambiguous. The proposal is contained in the Senate bill; while it was included in H.R. 2896 as first introduced, it was not contained in the version of the bill approved by the Ways and Means Committee. Two alternative economic theories of dividend behavior can be applied to the analysis. First, economists analyzing the impact of taxes on repatriated dividends have drawn an analogy between the repatriation decisions of foreign subsidiary corporations and the decisions a domestic corporation makes in deciding whether to pay dividends to its stockholders. Under this so-called "new view" or "trapped equity" theory of dividends, a firm whose foreign operations are mature undertakes new equityfinanced investment by retaining earnings rather than sending new equity capital from the United States abroad. In deciding whether to retain and reinvest its foreign earnings or to repatriate them, such a firm does not factor any taxes that apply to the act of repatriation into the calculation. The reason is this: since the earnings that fund the investment are already abroad, repatriation taxes must inevitably be paid, whether the repatriation occurs currently or at some point in the future. This theory therefore predicts that a permanent reduction or elimination of taxes on repatriated dividends will have no impact on the level of investment abroad or at home; repatriations will not increase. The chief impact will be a windfall increase in the value of the subsidiary's stock in the hands of its parent, and, in turn, an increase in value of parent's stock in the hands of its domestic stockholders.

The reduction in repatriation taxes in the proposals at hand, however, is temporary; the analysis therefore differs. Here, while it is still true that repatriation taxes must be paid whether the repatriations occur sooner or later, the tax will be reduced if they occur within the time frames specified by the proposals. Accordingly, a firm may advance or accelerate repatriations to take advantage of the temporarily reduced rates. This effect, however, is likely to be transitory and may even reverse itself when the reduced rate expires. In the long-run, the aftertax attractiveness of foreign versus domestic investment would not be changed by the proposal, and firms might be expected to temporarily reduce repatriations after the tax cut expires so as to restore its desired long-run level of foreign- employed capital.

Under the "new view," the impact of eliminating the repatriation tax is different for young, growing foreign operations than for mature subsidiaries. Young foreign subsidiaries fund part of their foreign investment by new contributions of capital from their U.S. parent rather than by retaining earnings. In considering the stock of capital it desires to employ abroad, a parent considers that repatriation taxes will ultimately have to be paid on the earnings of that capital; capital sent abroad, in other words, will be "trapped." Accordingly, a permanent reduction in repatriation taxes will increase the desired stock of capital abroad and increase U.S. investment abroad. A temporary reduction would likely have little effect.

As it applies to domestic corporations and their individual stockholders, the new view of dividends has been criticized on a number of grounds. For example, it assumes firms have no method of distributing earnings other than paying dividends, which is counter-factual. Corporations, for example, can and do repurchase their own shares. It is beyond the scope of this analysis to evaluate the applicability of these criticisms to foreign subsidiaries. Under the more traditional view of dividends, however, there is no distinction between young and mature firms; a foreign subsidiary simultaneously receives new capital from its parent and repatriates dividends, a means, perhaps, of signaling its profitability. Under this analysis, a permanent reduction of dividend taxes might cause earnings to be repatriated, but would also result in a new increase in U.S. firms' overseas investment. A temporary tax cut might likewise temporarily increase repatriations, but as under the new view, the increase would likely be temporary and would likely not shift the long-run stock of investment from foreign locations to the United States.⁹

Efficiency and Welfare Effects

Among the variety of economic effects that follow from the impact of taxes on investment, economic theory emphasizes international taxes' impact on economic efficiency and economic welfare. In general, theory holds that aggregate world economic welfare is maximized when the economies' scarce capital and other resources are deployed where they are most productive -- that is, where they earn the highest pre-tax return possible. In general (with a few exceptions) this occurs, again according to theory, when taxes do not interfere with firms' decision of where to employ investment. Thus, taxes maximize economic welfare when they apply equally to identical investments, regardless of their location. With respect to international taxes, traditional economic analysis characterizes a tax system that taxes foreign and domestic investment the same and that is therefore neutral towards investment location as possessing "capital export neutrality" (CXN)

Economic analysis also distinguishes aggregate world economic welfare from the economic welfare that can accrue to a capital exporting country and recognizes that a policy that is neutral towards foreign and domestic investment may maximize world welfare but may not be optimal from the point of view of the capital exporting country (in this case, the United States). Rather, theory suggests that a tax policy that to a degree discourages overseas investment by taxing it more heavily than domestic investment maximizes the economic welfare of the capital exporting country. According to theory, this is the case because at least part of the total, before-tax return to investment accrues to foreign factors of production and foreign governments when investment is undertaken abroad, but the entire return accrues to the domestic economy when investment occurs within the United States.

A tax policy that restricts overseas investment is called "national neutrality" (NN) in the terminology, because it is "neutral" towards the national economic welfare of the capital exporting country. Finally, businesses frequently emphasize the importance of their ability to compete effectively with foreign firms and recommend a third standard for tax policy, sometimes called "competitive neutrality" or "capital import neutrality" (CMN). Under such a policy, home country taxes (i.e., U.S. taxes) would not apply to foreign source income. Economic theory suggests that such a policy distorts the geographic allocation of capital and maximizes the welfare of neither the world nor the United States. Thus, even though it establishes equal tax burdens when certain comparisons are made (i.e., U.S. firms compared to foreign firms), CMN is not a "neutral" policy in the same sense as CXN or NN.

The current United States method of taxing foreign source income conforms to no single one of the three policies. Instead, it poses a patchwork of incentives, disincentives, and neutrality towards foreign investment, depending on factors such as a taxpayer's overall foreign investment situation, the level of foreign taxes faced by prospective foreign investment (and thus the particular country where investment might occur), and the legal form the investment will take. Further, different facets of the U.S. system influence the system in different directions. For example, application of worldwide taxation to U.S. corporations combined with application of the foreign tax credit are aspects of the U.S. system that are consistent with capital export neutrality. In contrast, the imposition of a limitation on the foreign tax credit is more consistent with national neutrality, although situations where cross-crediting can occur within the limitation can produce either neutrality or pose an incentive to invest abroad. The deferral principle generally nudges the system in the direction of capital import neutrality, while Subpart F restrains this effect.

How, then, would the various proposals affect the system's impact on economic welfare? As with investment, the overall impact of each bill is not clear; each bill contains provisions that would, in isolation, pull the system in different directions. The bills' repeal of ETI would, taken alone, move the system away from NN in the direction of CXN by eliminating a provision the favors domestic over foreign investment. Each of the bills' various domestic incentives, however, are, taken alone, consistent with NN since they favor domestic investment over investment abroad. Thus, for example, H.R. 1769/S. 970's coupling of ETI repeal with a domestic production deduction might, on balance, either nudge the system towards NN or towards CXN, depending on which provision is the most powerful.

The Ways and Means Committee and Hatch proposals have provisions consistent with each of the three policy standards. As with the Crane/Rangel/ Hollings plan, they couple provisions consistent with CXN, repeal of ETI, with a set of domestic incentives consistent with NN, for example, depreciation, expensing, and research credit provisions that are restricted to domestic investment. Also, in a result that is perhaps counter to intuition, the changes in the interest allocation rules are also consistent with NN. Both the Ways and Means Committee and Hatch bills, however, would also scale back Subpart F, a change generally consistent with CMN. The two bills' consolidation of foreign tax credit limitations would nudge the system away from NN and in the direction of either CXN or CMN, depending on the particular investment.

In short, the current U.S. tax system is a hybrid of CXN, NN, and CMN, containing important features consistent with each of these standards. This would likely not change under any of the bills. Like the current system, none of the bills is consistent with any one standard. Thus, their impact on economic efficiency and economic welfare is not certain.

² A second export tax benefit is the so-called "inventory source" or "export source" rule, under which firms with a surfeit of foreign tax credits can use those credits to shield export income from U.S. tax. While the benefit a firm can obtain from the export source rule is potentially larger than that from ETI or FSC before it, because its mechanics depend on foreign tax credits, it can only be used by firms with overseas operations that are subject to foreign tax.

For information on the ETI controversy, see CRS Report RL31660, A History of the Extraterritorial Income (ETI) and Foreign Sales Corporation (FSC) Export Tax-Benefit Controversy, by David Brumbaugh.

³ H.R. 2896 is quite similar to legislation introduced by Representative Thomas in the 107th Congress, H.R. 5095, although with some additions and modifications. For a detailed description of H.R. 5095, see CRS Report RL31574, International Tax Provisions of the American Competitiveness and Corporate Accountability Act (H.R. 5095), by David L. Brumbaugh. For a detailed description of H.R. 2896, see U.S. Congress, Joint Committee on Taxation, Technical Explanation of H.R. 2896, the "American Jobs Creation Act of 2003," JCX-72-03, Aug. 13, 2003, 146 pp. Available on the Committee's website, at [http://www.house.gov/jct/x-72- 03.pdf], visited Sept. 4, 2003.

⁴ U.S. Congress, Joint Committee on Taxation, Estimated Revenue Effects of the Chairman's Amendment in the Nature of a Substitute to H.R. 2896, The "American Jobs Creation Act of 2003," Scheduled for Markup by the Committee on Ways and Means on October 28, 2003 JCX-95-03, Oct. 24, 2003.

⁵ See Senator Hatch's press release that accompanied his introduction of the bill, reprinted in BNA Daily Tax Report TaxCore, July 28, 2003.

⁶ The provision of new investment incentives have at least the potential of increasing rather than reducing transitional unemployment by causing new shifts in where capital is employed.

⁷ For an analysis of how the investment incentives would affect effective tax rates on investment, see CRS Report RL32099, Capital Income Tax Revisions and Effective Tax Rates, by Jane Gravelle.

⁸ Although some observers have expressed concern over such developments, Treasury officials have recently stated that advances in income allocation and transfer pricing rules have made such a scenario "less of a concern." See Samuel C. Thompson, Jr. "A Critical Perspective on the Thomas Bill," Tax Notes, July 22, 2002, pp. 581-584; and Alison Bennett and Katherine M. Stimmel, "Extraterritorial Income" Administration Stresses International Relief in Effort to Replace U.S. Export Tax Regime," BNA Daily Tax Report, July 16, 2003, p. G-9.

⁹ For a more detailed analysis of proposals to reduce repatriation taxes, see CRS Report RL32125, Tax Exemption for Repatriated Foreign Earnings: Proposals and Analysis, by David L. Brumbaugh.