Ask the Scholar

Document scope · 1 page
doc
Scholar
Ask about this object, its catalog metadata, its source description, or the page inventory. For page-specific OCR and visual context, open one of the page chats.

Scholar Source Context

Document identity
localId
1534972
label
1975/12/22 - Economic Policy Board
core
doc
dtoType
document
pageCount
1
Source metadata
id
1534972
contentType
document
title
1975/12/22 - Economic Policy Board
collections
James M. Cannon Files (Ford Administration)
James Cannon's Meetings Files
subjects
Business
Taxation
imageCount
1
hasImages
yes
source
import
hasTranscription
no
Source extras
naId
1534972
coverageEndDate
logicalDate
1975-12-31
month
12
year
1975
coverageStartDate
logicalDate
1975-12-01
month
12
year
1975
levelOfDescription
fileUnit
recordType
description
ocrSource
nara-archive
Single page context
seq
1
pageIndex
0
type
document
mediaId
05e33edf63dbb6bd
ocrText
The original documents are located in Box 54, folder "1975/12/22 - Economic Policy Board" of the James M. Cannon Files at the Gerald R. Ford Presidential Library. Copyright Notice The copyright law of the United States (Title 17, United States Code) governs the making of photocopies or other reproductions of copyrighted material. Gerald Ford donated to the United States of America his copyrights in all of his unpublished writings in National Archives collections. Works prepared by U.S. Government employees as part of their official duties are in the public domain. The copyrights to materials written by other individuals or organizations are presumed to remain with them. If you think any of the information displayed in the PDF is subject to a valid copyright claim, please contact the Gerald R. Ford Presidential Library. Digitized from Box 54 of the James M. Cannon Files at the Gerald R. Ford Presidential Library ECONOMIC POLICY BOARD MEETING Monday, December 22, 1975 The Roosevelt Room 8:30 a.m. December 19, 1975 ECONOMIC POLICY BOARD EXECUTIVE COMMITTEE Proposed Agenda Monday, December 22, 1975 1. 1976 Collective Bargaining Negotiations CWPS 2. Status of State of the Union Preparation Seidman Tuesday, December 23, 1975 1. Report of Task Force on Tax Policy and In- Jones ternational Investment Wednesday, December 24, 1975 No Executive Committee Meeting Thursday, December 25, 1975 No Executive Committee Meeting Friday, December 26, 1975 No Executive Committee Meeting Saturday, December 20, 1975 Special Session on Economic Forecast for 1977 Budget Roosevelt Room 9:00 a.m. Principals Only OF THE TREASURY THE DEPARTMENT OF THE TREASURY WASHINGTON, D.C. 20220 1789 ASSISTANT SECRETARY December 19, 1975 MEMORANDUM FOR EXECUTIVE COMMITTEE, ECONOMIC POLICY BOARD FROM: Sidney Lones Subject: Preliminary Report of EPB Task Force on Tax Policy and International Investment on the Allocation of Research and Development Expenses Between United States and Foreign Source Income. Attached is the Task Force report which: 1. Summarizes the background of the problem; 2. Identifies 8 issues to be resolved; 3. Presents arguments associated with the various options under each issue; and 4. Presents revenue estimates and examples of the specific problems. Attachment December 22, 1975 ALLOCATION OF RESEARCH AND DEVELOPMENT EXPENSE BETWEEN UNITED STATES AND FOREIGN SOURCE INCOME The Internal Revenue Code has provided since the 1920's that all current deductions must be allocated to domestic or foreign source income unless they can- not definitely be related thereto, in which case they must be allocated between domestic and foreign source income on a ratable basis. The Treasury has a responsibility to issue regula- tions that clarify and explain existing statutory law. Several years ago, following taxpayer concern that agents lacked guidelines for allocating expenses, the Treasury Department prepared more detailed allocation regulations which were issued in proposed form in April 1973. This Fall, IRS agents were instructed by the Internal Revenue Service not to cite or apply the proposed regulations. The most controversial aspects of these proposed regula- tions concern the allocation of current research and development expenses. The Internal Revenue Code, since 1954, has permitted a United States business to elect either to claim a current business deduction for research and development expenses or to capitalize and amortize such expenses. The current business deduction alternative was legislated in part to provide an incentive for research and development ex- penditures. Patents and other forms of technology developed by a United States company may be used in foreign branch operations (which is relatively uncommon), sold or. licensed to an unrelated third party, or transferred to a foreign affiliate. Under current U.S. tax law, a transfer to an affiliate may take the form of (i) a sale for payment at fair market value, (ii) a royalty license for an arm's length royalty, or (iii) a tax-free transfer to a subsidiary where the property is to be used in manufacturing operations by the subsidiary. - 2 - The proposed Treasury regulations require allocation of current research and development expenses against royal- ties from licensees and dividends from subsidiaries where the research gives rise or "can reasonably be expected" to give rise to foreign source income, for example, divi- dends from subsidiaries that have received tax-free trans- fers of technology in the past. It should be stressed that deductions for research and development are allowed by the United States in all cases. The issue here is the extent to which these de- ductions should be allocated to foreign as opposed to domestic source income for such purposes as calculating the foreign tax credit. The United States credit for foreign income taxes imposed on foreign source income is limited by law to the amount of United States tax imposed on foreign source income. If deductions are not allocated to foreign source income, then those deductions reduce only United States tax on United States source income. Meanwhile, United States tax on the foreign source income could be fully offset by foreign income tax. The long-term effect is that foreign countries may be able to increase their tax revenues at the expense of the United States Treasury. Depending on foreign laws, this principle can of course also apply to technology transfers to the United States. At the present time royalties paid to the United States are approximately ten times those paid from the United States. If deductions are allocated to foreign source income by the United States, but not recognized by the foreign country when it imposes its tax, the immediate result is a higher effective worldwide tax rate, with the taxpayer in the middle. Under the proposed regulations, items of expense which can clearly be determined to relate to either domestic or foreign source income are allocated on that basis. Where - 3 - expenses are not clearly related to either domestic or foreign source income, they are to be allocated between the two sources on some comparable basis. For example, research and development expenses may be allocated between domestic sales income of the parent and dividends it receives from a foreign subsidiary on the basis of the number of units sold by the parent and by the foreign subsidiary or on the basis of gross receipts of the parent and the subsidiary. A few companies, for example, firms in the automobile industry, have historically instituted cost sharing agree- ments with their foreign affiliates and are reimbursed for portions of domestic research and development expense. These industries believe that they have little problem with the proposed regulations. The feasibility of such arrangements may be affected by specific characteristics of a particular industry, such as the comparability of products sold at home and abroad. A number of other com- panies have not had such agreements and have not allocated any research and development expense to foreign source in- come on the theory that it was basically incurred for domestic use. Others have allocated varying amounts with- out uniform guidance. The proposed regulations have the greatest impact on high technology industries with large research budgets, extensive foreign source income, and foreign taxes equal to the foreign tax credit limit. If applied in 1974 and assuming foreign countries do not permit increases in allowable deductions, the proposed regulations could result in the loss of $1.1 - $1.5 billion in foreign tax credits and a corresponding increase in the U.S. tax burden (Appendix 9). This burden would fall on high technology industries. Some firms have suggested that increased allocations and the attendant U.S. tax burden would force them to perform less research and, over time, to move research to their foreign subsidiaries, since that would assure full deductions abroad. It is exceedingly difficult to estimate the number of research jobs that would be shifted abroad. However, in connection with the tax impact and the shift of research to foreign countries, the following points should be noted: - 4 - - The main goal of the proposed allocation regulations is to clarify and explain existing statutory law. The pro- posed regulations do have a revenue raising impact, but this is not their main goal. - During the past two decades, some multinational firms have already devoted larger portions of their research budgets to research in Canada, Western Europe, and Japan. This change has been induced by the availability of skilled foreign scientists, the advantages of proximity to foreign markets, and by local tax and subsidy measures designed to encourage R&D activity. - The interaction of United States withholding taxes on royalties paid abroad and foreign withholding taxes on dividends paid to the United States may in certain cases make shifts of research abroad undesirable. The major issues are set forth below in the order appropriate for resolution. Pro and con arguments, and methods of implementation, are given in the appendices. Implementation of major policy decisions in this area may require new legislation to achieve desired results. - 5 - ISSUES TO BE RESOLVED 1. How should R&D deductions be allocated? OPTIONS: A. R&D expenses should be allocated entirely to U.S. source income (Appendix 1-A). B. Only "clearly related" R&D expenses should be allocated to foreign source income. An example of clearly related R&D expense would be the development of a specific process undertaken at the request of a foreign affiliate. All other R&D expenses should be allocated to United States source income (Appendix 1-B). C. "Clearly related" R&D expenses should be allocated to domestic or foreign source income. All R&D that is not clearly related to either domestic or foreign source income should be allocated between United States income and foreign source income on some reasonable basis (Appendix 1-c). D. The proposed regulations should be withdrawn, with no new regulatory or legislative guid- ance for Internal Revenue Service agents or taxpayers (Appendix 1-D). RECOMMENDATION: 2. If R&D is to be allocated between domestic and foreign source income under Option 1-c, how should this be done? OPTIONS: A. On the basis of gross income from foreign and domestic sources (Appendix 2-A). - 6 - B. on the basis of comparable items of income, e.g., at the election of the taxpayer com- parison of foreign and domestic: (1) sales, (2) net income, (3) number of units sold; or (4) actual and constructive royalties (Appendix 2-B). C. R&D expenses should be allocated only to items of income which can reasonably be expected to benefit from R&D expenditures. No allocations would be made to dividend income since all transfers of technology to foreign affiliates would be an arm's length license or sale and would be treated as tax- able events (Appendix 2-c). RECOMMENDATION: 3. While a current business deduction is allowed by the Code for R&D, should the taxpayer have the option of capitalizing the expense and amortizing it over time for purposes of allocating it between U.S. and foreign source income? (Appendix 3). RECOMMENDATION: 4. When should major changes in allocation methods be implemented? OPTIONS: A. The allocation method should apply prospectively (Appendix 4-A). B. The allocation method should apply prospectively after a grace period (Appendix 4-A). - 7 - C. The allocation method should be retroactive (Appendix 4-B). RECOMMENDATION: 5. Should the allocation method apply only in those cases where foreign governments agree to permit deduction of the allocated expenses? (Appendix 5) RECOMMENDATION: 6. Should the same principles be applied to United States branches and subsidiaries of foreign corporations, thus per- mitting United States branches and subsidiaries a deduction for a proportionate share of research expenses incurred by foreign home offices or parent companies? (Appendix 6) RECOMMENDATION: 7. Should any revised allocation regulations be submitted in proposed form for further public comment prior to being issued in final form? (Appendix 7) RECOMMENDATION: 8. Should any change in the allocation method be postponed until more of the economic effects of the change can be studied, including the effects on transfers abroad or possible reduction in R&D? (Appendix 8) . RECOMMENDATION: APPENDIX 1-A OPTION: R&D expenses should be allocated entirely to U.S. source income. PRO: Since some allocation is presently required, this option would provide a clear United States tax incentive for U.S.- - based multinational firms to conduct their R&D activities in the United States. There would be no reduction of the foreign tax credit, even though the research results were used abroad. This option acknowledges that R&D activities principally benefit the country of location, directly and through spillover effects, and it is least likely to encourage a shift of R&D abroad. Since this option requires no allocation it re- duces potential double taxation. CON: This option makes no attempt to allocate R&D expenses in a manner which reasonably reflects the economic benefits conferred through such R&D. Rather, this option enables foreign countries to raise their effective tax rates at the expense of the United States Treasury. That is, the total allocation of R&D expenses against United States income would result in a transfer of tax revenues from the United States to foreign treasuries. All R&D expenses would be re- flected in lower U.S. taxes on U.S. source income. Foreign countries need recognize no expenses associated with the development of know-how. The net result is that foreign taxes could be aggressively raised at the expense of the U.S. Treasury. IMPLEMENTION: This option would require new statutory language. The present statute requires that all current deductions, including R&D, be allocated between domestic and foreign source income. An exception to the statute would be needed for R&D expenses. APPENDIX 1-B OPTION: Only "clearly related" R&D expenses should be allocated to foreign source income. All other R&D expenses would be allocated to U.S. source income. PRO: This option ensures that at least the expense of R&D performed in the United States at the request of a foreign affiliate or unrelated firm would be charged against the income re- ceived by the U.S. company from that firm. This seems reasonable, since otherwise the tax claims of the foreign government could prevent the United States from recouping the lost tax revenue attributable to the R&D deduction. This option presumes that R&D expenses which are not "clearly related" to foreign source income should be allocated to U.S. source income. This treatment recognizes that R&D conducted in the United States principally benefits industry located here. This option would cause less of a "double tax" burden and would be less of an induce- ment for the transfer of R&D activities abroad than the proposed regulations. CON: Only a small portion of total R&D expenses are "clearly related" to foreign source income. thus, the bulk of R&D results would be made available to foreign countries with no foreign tax recognition given to the underlying expense. Foreign countries would be able to raise their effective tax rates at the expense of the United States Treasury. IMPLEMENTATION: This option would require new statutory language. In the Treasury's view, the present statute requires that deductions which are not definitely related to foreign or U.S. source income be ratably allocated between the two. APPENDIX 1-c OPTION: All R&D that cannot be clearly allocated to either domestic or foreign source income should be allocated between United States income and foreign source income on some reasonable basis. PRO: This option provides for a reasonable allocation of R&D costs between domestic and foreign source income. Foreign tax authorities are thereby encouraged to give proper tax recognition to the expenses associated with royalty income. The U.S. taxpayer is not asked to bear the entire bur- den of R&D expenses, when many of the findings are used abroad. CON: This option immediately raises the problem of defining a "reasonable" basis of alloca- tion and therefore may give rise to sub- stantial disputes and litigation over what- ever definition is chosen. Other questions this option raises include: Should a mar- ginal cost or a full cost approach be used?; should R&D be capitalized and amortized for purposes of allocation, or deducted currently?; what about R&D expense that produces no findings?; and so forth. Should the same method of appor- tionment be applied to income from a foreign branch, to royalties from related and unrelated foreign licensees, and to dividends from foreign subsidiaries to whom technology has been or may be transferred? Any allocation formula is inherently arbitrary and may not meet the circumstances of particular firms. Worse, the chosen formula may attribute too much R&D expense to foreign source income, and thereby encourage R&D activities to leave the United States. IMPLEMENTATION: This option can be implemented by the proposed regulations or any modification of the pro- posed regulations which provides for a reason- able allocation of R&D expense between domestic and foreign source income. APPENDIX 1-D OPTION: The proposed regulations should be withdrawn, with no new regulatory or legislative guidance. PRO: The statute has been on the books for many years. During this time, United States cor- porations have worked out their own systems for allocating R&D expense between domestic and foreign income. Any general rule will do violence to the circumstances of particular companies, and would invite extensive litiga- tion. CON: The existing law is arbitrarily applied from company to company, depending on historical accident, and past and present audits. The IRS agents have no clear guidance, and thus allocation methods for a particular firm may change with a change in agents. This is an entirely unsatisfactory arrangement from the standpoint of tax administration. Firms are entitled to greater certainty and uniformity in the tax treatment of a major expense. The Treasury has an obligation to clarify statutory law. In the absence of regulations, Congress may act in a manner unfavorable to corpora- tions with large R&D expense. IMPLEMENTATION: The implementation will require withdrawal of the presently proposed regulations. APPENDIX 2-A OPTION: Allocation should be on the basis of gross income from foreign and domestic sources. PRO: Allocation on the basis of gross income has long been used by taxpayers and recognized by the courts as an acceptable method. It is a simple and relatively easy method to apply. Many firms insist that they will successfully challenge in the courts any allocation method other than on the basis of gross income. The gross income method has the characteristic of assigning the bulk of R&D expense to U.S. source income since gross income from foreign sources is usually "net" types of income (dividends, interest, and royalties), while gross income from domestic sources is usually "gross" types of income (business receipts). Accordingly, this method will have less of an impact on the level and location of R&D activity than the proposed regulations. CON: Allocation on the basis of gross income mixes apples and oranges. Foreign and domestic gross income are entirely different in type: Foreign gross income is essentially net profit (i.e. dividends, interest, royalties), while domestic gross income is essentially gross receipts (sales). Thus allocation on the basis of gross income would assign an in- ordinately large amount of R&D expense to the production of U.S. source income. Moreover, the factual connection between R&D and gross income is weak. The fruits of R&D show up, not necessarily in gross income, but rather in actual and constructive royalties and royalty-type income. IMPLEMENTATION: This option could be implemented by reformulat- ing the proposed regulation to adopt alloca- tion on the basis of gross income. However, the draftsmen of the proposed regulation regard allocation of the basis of gross income as a highly inappropriate method, inconsistent with the intent of the statute. APPENDIX 2-B OPTION: Allocation should be on the basis of comparable items of income. PRO: This option would divide R&D expense between domestic and foreign income on the basis of similar types of income flows. There would be no problem of mixing apples and oranges. The underlying theory is that R&D expenses, a type of overhead cost, would be related to the kinds of income that they can reasonably be expected to produce. Moreover, if the corporation is permitted to elect between alternative methods of allocation, provided only that the alloca- tion formula involves similar types of income at home and abroad, it can select that method which best reflects its own experience. CON: This option does not properly match allocated expenses with related income. Rather, in most cases, R&D expenses will be allocated currently, while the income generated through such ex- penditures will not be realized, if at all, until the future. Moreover, this option exacerbates this mismatch by treating foreign subsidiaries on a consolidated basis for expense allocation purposes, while at the same time treating them on a separate basis for other pur- poses of the tax law. This option does not recognize the value to the United States of locating R&D facilities in this country. The benefits of U.S. -based R&D extend beyond the income produced within the corporate family. Accordingly, it is inappro- priate to allocate R&D expense strictly on the basis of corporate income flows. Further, this option entails considerable accounting com- plexity, as suggested by the examples in the proposed regulations. IMPLEMENTATION: This option is similar to the proposed regulation, except that it provides for the elective use of actual and constructive royalties as a basis for allocation. APPENDIX 2-C OPTION: R&D expenses should be allocated only to items of income which can reasonably be expected to benefit from R&D expenditures. No allocations would be made to dividend income. All transfers would be taxable events. PRO: This approach eliminates the difficult con- ceptual problem of determining in which cases, and to what extent, R&D should be allocated to dividends from foreign sub- sidiaries. Since the foreign affiliate will have purchased or obtained a license of the property at a fair market value, the earnings produced by the foreign sub- sidiary are generated out of its own capital and assets for which it has paid value. Thus, there is no conceptual need for re- quiring further allocations of R&D against dividends subsequently paid by the subsidiary. This option is consistent with the United States tax principles which provide for tax deferral and for the separate identity of subsidiaries. It is also consistent with the broad scheme of transfers abroad of in- come-producing property. Unlike the gross- to-gross method, it more finely tunes the allocation of R&D expenses to the appropriate income without sacrificing that method's administrative simplicity. At the same time it avoids the complexity and unrealistic formulas engendered by some of the other approaches. The double tax impact of the pro- posed regulations would be substantially mitigated. CON: There can be problems of valuation of a patent, or know-how for recognition of gain on transfer. This is a concept not now in use, some aspects of which would require additional analysis be- fore it could be made operative. IMPLEMENTION: It is possible that all technology transfers could be made taxable under the present authority of the Commissioner. However, since there would be some question as to such authority, and since taxpayers may object that this has not been existing practice, it may be necessary to obtain - 2 - a statutory amendment. Other aspects of this approach may be accomplished under existing law. APPENDIX 3 OPTION: While a current business expense deduction is allowed by the Code for R & D, should the taxpayer have the option of capitalizing the expense and amortizing it over time for purposes of allocating it between U.S. and foreign source income? PRO: This option is addressed to one of the major objections to the proposed regulations - that R&D activities do not generate foreign source income until after the findings have been ex- ploited domestically and thus it is unfair to require an immediate allocation of the full R&D expenditure to foreign source income. By permitting R&D expenses to be capitalized and then amortized over time, the allocation to foreign source income can be made to correspond more closely to the actual generation of for- eign source income by the R&D expenditure. Foreign tax authorities may thus be encouraged to give proper recognition to the allocation. This makes sound economic sense, although the accounting profession has recently held that R&D expenditures should be expensed rather than capitalized for purposes of determining net income. This opinion, however, does not extend to the allocation issue. Moreover, the impact of allocation could be phased in slowly over time, since the amortized amount of 1977 R&D expense, for example, would be a small portion of total 1977 R&D outlays. CON: This option merely postpones the full effect of R&D allocation without resolving the under- lying questions of the proper method of allo- cation. It provides taxpayers with the best of both worlds: they could claim a current deduction for R&D expense and yet capitalize R&D expense for purposes of allocation. More- over, the option would create administrative complexity for both taxpayers and the Treasury. - 2 - For example, what time period would be allowed for amortization? How should the amortized expenditures be apportioned among the time periods (straight line, industry experience, or some other basis) ? IMPLEMENTATION: The option, if made elective by the taxpayer, might be implemented by appropriate modifi- cation of the proposed regulations. However, the Internal Revenue Service believes that statutory language might be required to per- mit the use of a capitalization and amortiza- tion approach to allocation. APPENDIX 4-A OPTION: The allocation method should apply prospectively or should apply prospectively after a grace period. PRO: Since the final allocation regulations could represent a significant departure from present practices, they should apply prospectively in order to minimize any undue hardships on tax- payers and permit them time to accomodate their activities and recordkeeping to the new require- ments. Whether or not a grace period should also be provided depends on the choice of method and the degree by which such method differs from existing practices. CON: The final regulations are merely an amplifica- tion of the statute and previously enunciated policy. They reflect the allocation of expenses which taxpayers should have been making in the past. Hence, making their application prospec- tive absolves those taxpayers who did not com- ply in the past and unfairly prejudices those taxpayers who have complied. Moreover, prospec- tive application of the regulation leaves un- resolved questions over the existing rule. IMPLEMENTATION: Prospective effect to the allocation regulations may be provided by regulation. APPENDIX 4-B OPTION: The allocation method should apply retro- actively. PRO: The final regulations are merely an amplifica- tion of the statute and previously enunciated policy. They reflect the allocation of expenses which taxpayers should have been making in the past. Hence, making their application retro- active will treat all taxpayers equally. Moreover, retroactive application of the regulation will resolve questions over the existing rule. CON: Since the final allocation regulations could represent a significant departure from present practices, they should not apply retroactively in order to minimize any undue hardships on taxpayers and permit them time to accommodate their activities and recordkeeping to the new requirements. IMPLEMENTATION. Retroactive effect to the allocation regulations may be provided by regulation. APPENDIX 5 OPTION: Should the allocation method apply only in those cases where foreign governments agree to permit deduction of the allocated expenses? PRO: Foreign taxing jurisdictions will generally not permit deduction for additional R&D expenses incurred in the U.S. and allocated to foreign source income; nor will they permit the U.S. parent company to charge a greater royalty to the foreign income-pro- ducing entity. Accordingly, any increase in the R&D expense allocated to foreign source income will reduce the amount of foreign tax that is creditable, and thus may well generate excess and unutilized foreign tax credits resulting in a form of double taxation. In order to mitigate these effects, the allocation of R&D expenses should be limited to cases where deduction is per- mitted by the foreign taxing jurisdiction. CON: This option leaves the determination of United States expense allocations to foreign taxing jurisdictions, and since most foreign govern- ments will not permit additional deductions, the allocation rule would have little effect. Moreover, no effective means is provided by which to encourage foreign governments to permit deduction for the allocated expenses. For example, there would be no incentive for taxpayers to pressure foreign governments to change their rules. IMPLEMENTATION: This option will require new statutory language since the present statute requires allocation in all cases. APPENDIX 6 OPTION: The same allocation principles should be applied to U.S. branches and subsidiaries of foreign corporations. PRO: Permitting U.S. branches and subsidiaries to reimburse R&D expenses incurred abroad by foreign parent companies may encourage other countries to be more willing to allow deductions for R&D expenses incurred in the U.S. or may be used as a bargaining chip to negotiate recipro- cal treatment. Moreover, such a rule would represent a consistent application of the United States position on the allocation of expenses to foreign income. CON: This option raises serious questions of tax policy. For example, how strong is the United States policy which does not permit deduction of expenses unless they benefit the taxpayer and are made on an arm's length basis? The answer will depend upon the final allo- cation rule that is adopted. Under usual United States concepts a branch might be entitled to a pro: rata allocation of expenses, while a subsidiary could not deduct expenses incurred by the parent unless it were entitled to the benefits of the research under an arms length arrangement. Such a rule would also give foreign companies a competitive advantage in those cases where they can deduct the allocated expense for United States tax pur- poses, are not required to allocate by the foreign country, and hence can also deduct the full expense for foreign tax purposes. IMPLEMENTATION: This option may require a statutory change or may be provided for by treaty. APPENDIX 7 OPTION: Any revised allocation regulations should be submitted in proposed form for further public comment. PRO: The magnitude and wide-ranging scope of the expense allocation regulations may have a serious economic impact on United States tax- payers. Thus, to the extent any revised regu- lations are issued which differ materially from those published in the past, taxpayers should be provided an opportunity to comment and to present their problems in order to assure these regu- lations do not inadvertently create irreparable and unwarranted economic harm. CON: Proposed allocation regulations have been circulating for many years, and the basic issues raised by such allocation have been long known. Accordingly, taxpayers have had ample opportunity to make their comments and problems known. Fail- ure to publish these regulations in final form merely delays resolution of the basic problem -- the lack of clear guidance as to the appropriate method of allocating expenses. IMPLEMENTATION: No statutory change is required. This is merely an administrative determination. APPENDIX 8 OPTION: Postpone any changes until more of the economic impact of the change can be studied. PRO: Since we do not presently know the full economic impact of either the present allocation method, the method in the proposed regulations, or the method in any of the alternatives thereto; and since that impact could be substantial, we should not make any decisions until the economic impact can be studied. CON: In the first instance, the Treasury Department has an obligation to issue regulations as guidance for taxpayers in applying the statute, whether or not the economic consequences of those regulations are known beforehand. More- over, in this case delay will not increase our knowledge. It will not be possible to estimate the effects of any expense allocation regulations before they are implemented because taxpayers do not keep their accounts in a manner which permits a determination of the amounts of R&D expense which would be allocated under varying allocation methods. Even if such a determination could be made, no estimate of the degree to which shift of R&D will occur is possible because such shifts are determined by a variety of unquantifiable and unpredictable factors such as the action of foreign governments or the substitutability of research personnel. IMPLEMENTATION: This option may be accomplished administratively. APPENDIX 9 Revenue Estimate Apportionment of Research and Development Expenditures Between United States and Foreign Source Income It is estimated that the apportionment of research and development (R&D) expenses for the year 1974 on the basis of the proposed regulations would have reduced the allowable foreign tax credit and therefore would have increased U.S. Treasury tax revenues by between $1.1 and $1.5 billion. The estimate was derived as follows. Research and Development Expenditures In 1974, U.S. industry spent about $13.9 billion of private funds for research and development. 1/ According to the National Science Foundation, large companies, i.e., those with 10,000 or more employees, account for about 83 percent of the R&D expenditures. 2/ The large companies dominate U.S. investment abroad. Therefore, 83 percent of the $13.9 billion, or about $11.5 billion in 1974 R&D expenditures, is assumed to be affected by the proposed allocation to foreign source income. Sales: Worldwide and Domestic According to Fortune, consolidated sales of large corporations, i.e., those with 10,000 or more employees, totalled about $800 billion in 1974. 3/ This figure needs to be apportioned between domestic and foreign sales. 1/ U.S. Department of Commerce, Bureau of the Census, Statistical Abstract, 1975, p. 548. 2/ U.S. National Science Foundation, Research and Development in Industry 1970, p. 11. 3/ "Fortune Directory of the 500 Largest Industrial Corporations," Fortune, May 1975, pp. 208-235. This is the sales figure for the top 400 corporations since they are the ones with 10,000 or more employees. - 2 - The 1966 and 1970 special surveys of 298 large U.S. multinational companies (MNCs) by the Bureau of Economic Analysis provide data for such an apportionment. The 298 MNCs in the survey consist of 298 U.S. reporters (the U.S. parents of the MNCs) and their 5,237 majority-owned foreign affiliates. Using these data, a recent article estimated MNC consolidated worldwide sales, defined as: (1) sales by the U.S. reporter to unaffiliated U.S. and foreign residents; plus (2) sales by its majority-owned foreign affiliates to unaffiliated U.S. residents and to unaffiliated foreign residents other than sales to minority-owned foreign affiliates of the MNC. 4/ The 1966 and 1970 estimates are: All Industries 1966 1970 Worldwide Consolidated Sales 100.0% 100.0% Sales to U.S. residents as percent of total 78.5 74.7 Sales to foreigners as percent of total 21.5 25.3 Since sales to foreigners grew faster than sales to U.S. residents, the 1974 percentages were estimated using simple extrapolation as: All Industries 1974 Worldwide Consolidated Sales 100.0 Sales to U.S. residents as percent of total 70.2 Sales to foreigners as percent of total 29.8 Thus, it is estimated that the $800 billion in 1974 sales of large corporations was comprised of $562 billion ($800 X 702) in domestic sales and $238 billion ($800 X .298) 4/ Leonard A. Lupo, "Worldwide Sales by U.S. Multinational Companies, " Survey of Current Business, January 1973, pp. 33-39. - 3 - in foreign sales. Allocable R&D Expenditures Assuming the apportionment of R&D expenditures on a sales basis was chosen on the basis of the proposed regu- lations, the apportionment of the $11.5 billion would be as follows: R&D X Foreign Sales = R&D Allocation to Foreign Worldwide Sales Source Income (billions) $11. 5 X $238 = $3.4 billion $800 On this basis about $3.4 billion in R&D expenditures would be apportioned to foreign source income. As extreme assumptions, suppose that: (a) presently no R&D expense is apportioned to foreign source income; (b) all affected companies pay foreign taxes in an amount equal to the U.S. foreign tax credit limit. Then the change in the U.S. foreign tax credit limit represents the additional tax liability to the U.S. Treasury. The change in the foreign tax credit limit is given by: U.S. tax Foreign source income - Apportioned R&D expense Worldwide income X before credits - Foreign source income U.S. tax Change in X = Worldwide income before credits foreign tax credit limit Tax Credit Reduction This apportionment would reduce the limit on the foreign tax credit, and thereby provide the U.S. Treasury with a revenue gain. The foreign tax credit is limited to U.S. tax liability on worldwide income times a fraction, the numerator of which is taxable income from sources outside the U.S. and the denominator of which is total worldwide income. - 4 - This may be rewritten as: U.S. tax before credits - X Apportioned R&D expense Worldwide income = Change in foreign tax credit limit Or: 5/ - .44 X $3.4 billion = - $1.5 billion. The figure of $1.5 billion for 1974 represents an upper estimate of the loss in foreign tax credits and the gain in Treasury revenues. Alternative Method A somewhat lower estimate of $1.1 billion can be derived from the results of a survey of 75 corporations having foreign operations who are included in the Fortune listing of the top 150 U.S. industrial corporations. The survey, conducted by five of the major accounting firms, obtained adequate information from 41 of the 75 corporations. It was estimated that these corporations spent $2.88 billion on research and development and that the proposed regulation would reduce their allowable foreign tax credit by $283 million. If the 400 large corporations which spent $11.5 billion in private funds on R&D in 1974 experienced a similar reduction in their allowable foreign tax credits, the total reduction would be: - $11.5 2.88 X $283 = -$1.1 billion This figure is lower than the estimate based on aggregate data because that estimate made no allowance for the present apportionment of R&D expense to foreign source income, nor did it reflect the fact that the foreign 5/ The factor of 44 is based on data contained in Statistical Abstract, 1975, P. 499. - 5 - taxes paid by some companies are less than the U.S. foreign tax credit limit. Even the alternative estimate of $1.1 billion may be. exaggerated because some firms may now classify doubtful items in the R&D expenditure account in order to produce a large number for public relations purposes. However, if firms are required to allocate R&D expenses to foreign source income, some of the doubtful items presently classified as R&D may be placed elsewhere in the business accounts. More- over, firms may be able to establish that much R&D is "clearly related" to the U.S. market; for example, testing to obtain U.S. approval of a new drug. This characterization would reduce the allocation of R&D expense to foreign source income and there- fore reduce the gain in U.S. Treasury revenues. APPENDIX 10 December 12, 1975 EXAMPLE The following example is illustrative of the problem. A United States company X, manufactures and sells toasters in the United States, and two wholly owned foreign subsidiaries of X, A and B, manufacture and sell toasters abroad. All toaster research and development is carried on by X in the United States. This research produces results which are commercially applicable throughout the world. X transfers patents developed through its R & D to A as a tax-free con- tribution of capital and licenses specific patents and know-how on successful research to B for an annual royalty of five percent of B's gross income. Except for the royalty charges there is no reimburse- ment for the research undertaken in the United States. For 1975, the following additional facts apply: X A B Gross income from $1850 $1000 $1000 manufacturing Royalty income from B 50 -- -- Dividend from A 100 -- - TOTAL GROSS INCOME 2000 1000 1000 R & D expenses (200) -- --- Other expenses (800) (500) (500) Taxable income 1000 500 500 Tax at 50% (500) (250) (250) NET INCOME $ 500 $ 250 $ 250 NUMBER OF TOASTERS PRODUCED 100 50 50 - 2 - Based on these facts, and applying three alternative methods of allocating X's research and development expenses to foreign source income, X's foreign tax credit would be as follows: Allocation Allocation of R & D on based on the the basis of ratio of toasters the ratio of produced abroad foreign source to total world- No alloca- gross income wide toaster tion of to total gross production R & D income Creditable foreign taxes 1/ $ 50 $ 50 $ 50 R & D expenses allocated to foreign source income None 15 2/ 100³/ U.S. foreign tax credit limitation 4/ 75 67.50 25 Excess foreign tax credits None None 25 As illustrated through this example, both under the no allocation and gross income allocation approaches, virtually all of the research and development expense is deducted against United States source income and U.S. taxes on that income are correspondingly reduced. Moreover, dividends paid to the parent company incur no additional U.S. tax because of the foreign tax credit. For footnotes see page 4, - 3 - However, if significant allocations of research and expense are made, the dividend income from the foreign subsidiaries would be substantially reduced and excess foreign tax credits would be generated. Indeed, a full allocation of research costs on the basis of worldwide sales would mean that the sub- sidiaries are not earning the profits claimed by them and foreign taxes would be reduced or even eliminated. Foreign governments would thus resist claims to reimburse the parent. - 4 - Footnotes: 1/ The only creditable foreign taxes available to X are those deemed paid by X with respect to the dividend from A. The formula is: dividend X taxes paid A 'saccu- by A mulated profits for the year Thus, 100 X 250 = $50. 500 2/ The formula is foreign source gross income X R & D total gross income expense Thus, 150 X 200 = $15. 2000 3/ The formula is foreign toaster production X R & D worldwide toaster produc- expense tion Thus, 100 200 X 200 = $100. 4/ Assuming that X elects the overall credit limitation, the formula is foreign source taxable income X U. S. tax total taxable income liability Thus, with no allocation this is: 150 X 500 = $75. 1000 with gross to gross alloca- tion: (150-15) X 500 = $ 67.50 1000 and with units of production allocation: (150-100) x500 = $ 25. 1000 APPENDIX 11 Historical Note on Allocation/Deduction Regulations The existing allocation regulations under section 861 were proposed in 1956 and adopted in 1957. They give minimal guidance to taxpayers and to revenue agents as to the handling of various types of expenses. Somewhat more detailed regulations were proposed on August 2, 1966. These proposed regulations were withdrawn with the issuance of new proposed regulations in April, 1973.