party or transfer secret processes to that uncontrolled party, directly or indirectly through a member of the controlled group of corporations to which the taxpayer belongs. Past experience with research and experimentation shall be considered in determining reasonable expectations.

(i) Definition of uncontrolled party. For purposes of this paragraph (c)(2) the term uncontrolled party means a party that is not a person with a relationship to the taxpayer specified in section 267(b), or is not a member of a controlled group of corporations to which the taxpayer belongs (within the meaning of section 993(a)(3) or 927(d)(4)).

(ii) Licensed products. In the case of licensed products, if the amount of sales of such products is unknown (for example, where the licensed product is a component of a large machine), a reasonable estimate based on the principles of section 482 should be made.

(iii) Sales of intangible property. In the case of sales of intangible property, regardless of whether the consideration received in exchange for the intangible is a fixed amount or is contingent on the productivity, use, or disposition of the intangible, if the amount of sales of products utilizing the intangible property is unknown, a reasonable estimate of sales shall be made annually. If necessary, appropriate economic analyses shall be used to estimate sales.

(3) Sales of controlled parties. For purposes of the apportionment under paragraph (c)(1) of this section, the sales from the product category (or categories) of the taxpayer shall be taken fully into account and the sales from the product category (or categories) of a corporation controlled by the taxpayer shall be taken into account to the extent provided in this paragraph (c)(3) for determining the taxpayer's apportionment, if such corporation can reasonably be expected to benefit directly or indirectly (through another member of the controlled group of corporations to which the taxpayer belongs) from the taxpayer's research expense connected with the product category (or categories). A corporation controlled by the taxpayer can reasonably be expected to benefit from the taxpayer's research expense if the taxpayer can be expected to license, sell,

or transfer intangible property to that corporation or transfer secret processes to that corporation, either directly or indirectly through a member of the controlled group of corporations to which the taxpayer belongs. Past experience with research and experimentation shall be considered in determining reasonable expectations.

(i) Definition of a corporation controlled by the taxpayer. For purposes of this paragraph (c)(3), the term a corporation controlled by the taxpayer means any corporation that has a relationship to the taxpayer specified in section 267(b) or is a member of a controlled group of corporations to which the taxpayer belongs (within the meaning of section 993(a)(3) or 927(d)(4).

(ii) Sales to be taken into account. The sales from the product category (or categories) of a corporation controlled by the taxpayer taken into account shall be equal to the amount of sales that bear the same proportion to the total sales of the controlled corporation as the total value of all classes of the stock of such corporation owned directly or indirectly by the taxpayer, within the meaning of section 1563, bears to the total value of all classes of stock of such corporation.

(iii) Sales not to be taken into account more than once. Sales from the product category (or categories) between or among such controlled corporations or the taxpayer shall not be taken into account more than once; in such a situation, the amount sold by the selling corporation to the buying corporation shall be subtracted from the sales of the buying corporation.

(iv) Effect of cost-sharing arrangements. If the corporation controlled by the taxpayer has entered into a bona fide cost-sharing arrangement, in accordance with the provisions of §1.4827, with the taxpayer for the purpose of developing intangible property, then that corporation shall not reasonably be expected to benefit from the taxpayer's share of the research expense.

(d) Gross income methods-(1)(i) In general. In lieu of applying the sales method of paragraph (c) of this section, the remaining amount of the deduction for research and experimentation, not apportioned under paragraph (a)(4) or

(b)(1)(ii) of this section, shall be apportioned as prescribed in paragraphs (d)(2) and (3) of this section, between the statutory grouping (or among the statutory groupings) of gross income and the residual grouping of gross in

come.

(ii) Optional methods to be applied to all research and experimental expenditures. These optional methods must be applied to the taxpayer's entire deduction for research and experimental expense remaining after applying the exception in paragraph (a)(4) of this section, and may not be applied on a product category basis. Thus, after the allocation of the taxpayer's entire deduction for research and experimental expense under paragraph (a)(2) of this section (by attribution to SIC code categories), the taxpayer must then apportion as necessary the entire deduction as allocated by separate amounts to various product categories, using only the sales method under paragraph (c) of this section or only the optional gross income methods under this paragraph (d). The taxpayer may not use the sales method for a portion of the deduction and optional gross income methods for the remainder of the deduction separately allocated.

(2) Option one. The taxpayer may apportion its research and experimental expenditures ratably on the basis of gross income between the statutory grouping (or among the statutory groupings) of gross income and the residual grouping of gross income in the same proportions that the amount of gross income in the statutory grouping (or groupings) and the amount of gross income in the residual grouping bear, respectively, to the total amount of gross income, if the conditions described in paragraph (d)(2)(i) and (ii) of this section are both met.

(i) The amount of research and experimental expense ratably apportioned to the statutory grouping (or groupings in the aggregate) is not less than fifty percent of the amount that would have been so apportioned if the taxpayer had used the method described in paragraph (c) of this section; and

(ii) The amount of research and experimental expense ratably apportioned to the residual grouping is not less than fifty percent of the amount

that would have been so apportioned if the taxpayer had used the method described in paragraph (c) of this section.

(3) Option two. If, when the amount of research and experimental expense is apportioned ratably on the basis of gross income, either of the conditions described in paragraph (d)(2)(i) or (ii) of this section is not met, the taxpayer may either

(i) Where the condition of paragraph (d)(2)(i) of this section is not met, apportion fifty percent of the amount of research and experimental expense that would have been apportioned to the statutory grouping (or groupings in the aggregate) under paragraph (c) of this section to such statutory grouping (or to such statutory groupings in the aggregate and then among such groupings on the basis of gross income within each grouping), and apportion the balance of the amount of research and experimental expenses to the residual grouping; or

(ii) Where the condition of paragraph (d)(2)(ii) of this section is not met, apportion fifty percent of the amount of research and experimental expense that would have been apportioned to the residual grouping under paragraph (c) of this section to such residual grouping, and apportion the balance of the amount of research and experimental expenses to the statutory grouping (or to the statutory groupings in the aggregate and then among such groupings ratably on the basis of gross income within each grouping).

(e) Binding election-(1) In general. A taxpayer may choose to use either the sales method under paragraph (c) of this section or the optional gross income methods under paragraph (d) of this section for its original return for its first taxable year to which this section applies. The taxpayer's use of either the sales method or the optional gross income methods for its return filed for its first taxable year to which this section applies shall constitute a binding election to use the method chosen for that year and for four taxable years thereafter.

(2) Change of method. The taxpayer's election of a method may not be revoked during the period referred to in paragraph (e)(1) of this section without the prior consent of the Commissioner.

After the expiration of that period, the taxpayer may change methods without the prior consent of the Commissioner. However, the taxpayer's use of the new method shall constitute a binding election to use the new method for its return filed for the first year for which the taxpayer uses the new method and for four taxable years thereafter. The taxpayer's election of the new method may not be revoked during that period without the prior consent of the Commissioner.

(i) Short taxable years. For purposes of this paragraph (e), the term taxable year includes a taxable year of less than twelve months.

(ii) Affiliated groups. In the case of an affiliated group, the period referred to in paragraph (e)(1) of this section shall commence as of the latest taxable year in which any member of the group has changed methods.

(f) Special rules for partnerships—(1) Research and experimental expenditures. For purposes of applying this section, if research and experimental expenditures are incurred by a partnership in which the taxpayer is a partner, the taxpayer's research and experimental expenditures shall include the taxpayer's distributive share of the partnership's research and experimental expenditures.

(2) Purpose and location of expenditures. In applying the exception for expenditures undertaken to meet legal requirements under paragraph (a)(4) of this section and the exclusive apportionment for the sales method and the optional gross income methods under paragraph (b) of this section, a partner's distributive share of research and experimental expenditures incurred by a partnership shall be treated as incurred by the partner for the same purpose and in the same location as incurred by the partnership.

(3) Apportionment under the sales method. In applying the remaining apportionment for the sales method under paragraph (c) of this section, a taxpayer's sales from a product category shall include the taxpayer's share of any sales from the product category of any partnership in which the taxpayer is a partner. For purposes of the preceding sentence, a taxpayer's share of sales shall be proportionate to the tax

payer's distributive share of the partnership's gross income in the product category.

(g) Effective date. This section applies to taxable years beginning after December 31, 1995. However, a taxpayer may at his or her option, apply this section in its entirety to all taxable years beginning after August 1, 1994.

(h) Examples. The following examples illustrate the application of this section:

Example 1 (i) Facts. X, a domestic corporation, is a manufacturer and distributor of small gasoline engines for lawn mowers. Gasoline engines are a product within the category, Engines and Turbines (SIC Industry Group 351). Y, a wholly owned foreign subsidiary of X. also manufactures and sells these engines abroad. During 1996, X incurred expenditures of $60,000 on research and experimentation, which it deducts as a current expense, to invent and patent a new and improved gasoline engine. All of the research and experimentation was performed in the United States. In 1996, the domestic sales by X of the new engine total $500,000 and foreign sales by Y total $300,000. X provides technology for the manufacture of engines to Y via a license that requires the payment of an arm's length royalty. In 1996, X's gross income is $160,000, of which $140,000 is U.S. source income from domestic sales of gasoline engines and $10,000 is foreign source royalties from Y, and $10,000 is U.S. source interest income.

(ii) Allocation. The research and experimental expenditures were incurred in connection with small gasoline engines and they are definitely related to the items of gross income to which the research gives rise, namely gross income from the sale of small gasoline engines in the United States and royalties received from subsidiary Y, a foreign manufacturer of gasoline engines. Accordingly, the expenses are allocable to this class of gross income. The U.S. source interest income is not within this class of gross income and, therefore, is not taken into account.

(iii) Apportionment. (A) For purposes of applying the foreign tax credit limitation, the statutory grouping is general limitation gross income from sources without the United States and the residual grouping is gross income from sources within the United States. Since the related class of gross income derived from the use of engine technology consists of both gross income from sources without the United States (royalties from Y) and gross income from sources within the United States (gross income from engine sales). X's deduction of $60.000 for its research and experimental expenditure must

be apportioned between the statutory and residual grouping before the foreign tax credit limitation may be determined. Because more than 50 percent of X's research and experimental activity was performed in the United States, 50 percent of that deduction can be apportioned exclusively to the residual grouping of gross income, gross income from sources within the United States. The remaining 50 percent of the deduction can then be apportioned between the residual and statutory groupings on the basis of sales of small gasoline engines by X and Y. Alternatively, X's deduction for research and experimentation can be apportioned under the optional gross income method. The apportionment for 1996 is as follows:

(1) Tentative Apportionment on the Basis of Sales

(i) Research and experimental expense to be apportioned between residual and statutory groupings of gross in

come:

(ii) Less: Exclusive apportionment of research and experimental expense to the residual grouping of gross income ($60,000×50 percent): ... (iii) Research and experi

mental expense to be apportioned between residual and statutory groupings of gross income on the basis of sales:

(iv) Apportionment of research and experimental expense to the residual grouping of gross income ($30,000-$500,000/

($500,000+ $300,000)): .......

(v) Apportionment of research and experimental ex

statutory

grouping of gross income

$60,000

$30,000

$30,000

$18,750

pense to the

(B) The total research and experimental expense apportioned to the statutory grouping ($3,000) under the gross income method is approximately 26 percent of the amount apportioned to the statutory grouping under the sales method. Thus, X may use option two of the gross income method (paragraph (d)(3) of this section) and apportion to the statutory grouping fifty percent (50%) of the $11,250 apportioned to that grouping under the sales method. Thus, X apportions $5,625 of research and experimental expense to the statutory grouping. X's use of the optional gross income methods will constitute a binding election to use the optional gross income methods for 1996 and four taxable years thereafter.

Example 2 (i) Facts. Assume the same facts as in Example 1 except that X also spends $30,000 in 1996 for research on steam turbines, all of which is performed in the United States, and X has steam turbine sales in the United States of $400,000. X's foreign subsidiary Y neither manufactures nor sells steam turbines. The steam turbine research is in addition to the $60,000 in research which X does on gasoline engines for lawnmowers. X thus has a deduction of $90,000 for its research activity. X's gross income is $200,000, of which $140,000 is U.S. source income from domestic sales of gasoline engines, $50,000 is U.S. source income from domestic sales of steam turbines, and $10,000 is foreign source royalties from Y.

(ii) Allocation. X's research expenses generate income from sales of small gasoline engines and steam turbines. Both of these products are in the same three digit SIC code category, Engines and Turbines (SIC Industry

194-088 D-01--9

Group 351). Therefore, the deduction is definitely related to this product category and allocable to all items of income attributable to it. These items of X's income are gross income from the sale of small gasoline engines and steam turbines in the United States and royalties from foreign subsidiary Y, a foreign manufacturer and seller of small gasoline engines.

(iii) Apportionment. (A) For purposes of applying the foreign tax credit limitation, the statutory grouping is general limitation gross income from sources outside the United States and the residual grouping is gross income from sources within the United States. X's deduction of $90,000 must be apportioned between the statutory and residual groupings. Because more than 50 percent of X's research and experimental activity was performed in the United States, 50 percent of that deduction can be apportioned exclusively to the residual grouping, gross income from sources within the United States. The remaining 50 percent of the deduction can then be apportioned between the residual and statutory groupings on the basis of total sales of small gasoline engines and steam turbines by X and Y. Alternatively, X's deduction for research and experimentation can be apportioned under the optional gross income methods. The apportionment for 1996 is as follows:

(1) Tentative Apportionment on the Basis of Sales

(i) Research and experimental expense to be apportioned between residual and statutory groupings of gross in

come:

(ii) Less: Exclusive apportionment of the research and experimental expense to

the residual grouping of gross income ($90,000×50 percent):

(iii) Research and experimental expense to be apportioned between the residual and statutory groupings of gross income on the basis of sales: (iv) Apportionment of research and experimental expense to the residual grouping of gross income ($45,000×($500,000+$400,000)/ ($500,000+$400,000+ $300,000)):

$90,000

$45,000

$45,000

$33,750

(B) The total research and experimental expense apportioned to the statutory grouping ($3,375) under the gross income method is 30 percent of the amount apportioned to the statutory grouping under the sales method. Thus, X may use option two of the gross income method (paragraph (d)(3) of this section) and apportion to the statutory grouping fifty percent (50%) of the $11,250 apportioned to that grouping under the sales method. Thus, X apportions $5,625 of research and experimental expense to the statutory grouping. X's use of the optional gross income methods will constitute a binding election to use the optional gross income methods for 1996 and four taxable years thereafter.

Example 3 (i) Facts. Assume the same facts as in Example 1 except that in 1997 X continues its sales of the new engines, with sales of $600,000 in the United States and