Foreign Parent (FP) and U.S. Subsidiary (USS) enter into a CSA to develop synthetic fertilizers and insecticides. FP and USS share costs on the basis of each controlled participant’s current sales of fertilizers and insecticides. The market shares of the controlled participants have been stable for fertilizers, but FP’s market share for insecticides has been expanding. The controlled participants’ projections of RAB shares are reliable with regard to fertilizers, but not reliable with regard to insecticides; a more reliable projection of RAB shares would take into account the expanding market share for insecticides ...

The facts are the same as in Example 2, except that FS’s market share is rapidly expanding because of the business failure of a competitor in its geographic area. The controlled participants’ RAB shares are not reliably projected by current sales of cleaning products. FS’s benefit projections should take into account its growth in market share ...

U.S. Parent (USP) and Foreign Subsidiary (FS) enter into a CSA to develop new and improved household cleaning products. Both controlled participants have sold household cleaning products for many years and have stable worldwide market shares. The products under development are unlikely to produce unusual profits for either controlled participant. The controlled participants determine RAB shares on the basis of each controlled participant’s current sales of household cleaning products. In this case, the controlled participants’ RAB shares are reliably projected by current sales of cleaning products ...

(i) Foreign Parent (FP) and U.S. Subsidiary (USS) enter into a CSA to develop a new car model. The controlled participants plan to spend four years developing the new model and four years producing and selling the new model. USS and FP project total sales of $4 billion and $2 billion, respectively, over the planned four years of exploitation of the new model. The controlled participants determine RAB shares for each year of 662⁄3% for USS and 331⁄3% for FP, based on projected total sales. (ii) USS typically begins producing and selling new car models a year after FP begins producing and selling new car models. In order to reflect USS’s one-year lag in introducing new car models, a more reliable projection of each participant’s RAB share would be based on a projection of all four years of sales for each participant, discounted to present value ...

The following examples illustrate the principles of this paragraph (e)(2)(iii): ...

The reliability of an estimate of RAB shares also depends upon the reliability of projections used in making the estimate. Projections required for this purpose generally include a determination of the time period between the inception of the research and development activities under the CSA and the receipt of benefits, a projection of the time over which benefits will be received, and a projection of the benefits anticipated for each year in which it is anticipated that the cost shared intangible will generate benefits. A projection of the relevant basis for measuring anticipated benefits may require a projection of the factors that underlie it. For example, a projection of operating profits may require a projection of sales, cost of sales, operating expenses, and other factors that affect operating profits. If it is anticipated that there will be significant variation among controlled participants in the timing of their receipt of benefits, and consequently benefit shares are expected to vary significantly over the years in which benefits will be received, it normally will be necessary to use the present value of the projected benefits to reliably determine RAB shares. See paragraph (g)(2)(v) of this section for best method considerations regarding discount rates used for this purpose. If it is not anticipated that benefit shares will significantly change over time, current annual benefit shares may provide a reliable projection of RAB shares. This circumstance is most likely to occur when the CSA is a long-term arrangement, the arrangement covers a wide variety of intangibles, the composition of the cost shared intangibles is unlikely to change, the cost shared intangibles are unlikely to generate unusual profits, and each controlled participant’s share of the market is stable ...

...

U.S. Parent (USP), Foreign Subsidiary 1 (FS1), and Foreign Subsidiary 2 (FS2) enter into a CSA to develop computer software that each will market and install on customers’ computer systems. The controlled participants measure benefits on the basis of projected sales by USP, FS1, and FS2 of the software in their respective geographic areas. However, FS1 plans not only to sell but also to license the software to unrelated customers, and FS1’s licensing income (which is a percentage of the licensees’ sales) is not counted in the projected benefits. In this case, the basis used for measuring the benefits of each controlled participant is not the most reliable because all of the benefits received by controlled participants are not taken into account. In order to reliably determine RAB shares, FS1’s projected benefits from licensing must be included in the measurement on a basis that is the same as that used to measure its own and the other controlled participants’ projected benefits from sales (for example, all controlled participants might measure their benefits on the basis of operating profit) ...

Foreign Parent (FP) and U.S. Subsidiary (USS) enter into a CSA to develop materials that will be used to train all new entry-level employees. FP and USS determine that the new materials will save approximately ten hours of training time per employee. Because their entry-level employees are paid on differing wage scales, FP and USS decide that they should not measure benefits based on the number of entry-level employees hired by each. Rather, they measure benefits based on compensation paid to the entry-level employees hired by each. In this case, the basis used for measuring RAB shares is the most reliable because there is a direct relationship between compensation paid to new entry-level employees and costs saved by FP and USS from the use of the new training materials ...

The facts are the same as in Example 6, except that FP distributes its fertilizers directly while USS sells to independent distributors. In this case, sales of USS and FP are not the most reliable basis for measuring RAB shares unless adjustments are made to account for the difference in market levels at which the sales occur ...

(i) Foreign Parent (FP) and U.S. Subsidiary (USS) manufacture and sell fertilizers. They enter into a CSA to develop a new pellet form of a common agricultural fertilizer that is currently available only in powder form. Under the CSA, USS obtains the rights to produce and sell the new form of fertilizer for the U.S. market while FP obtains the rights to produce and sell the new form of fertilizer in the rest of the world. The costs of developing the new form of fertilizer are divided on the basis of the anticipated sales of fertilizer in the controlled participants’ respective markets. (ii) If the research and development is successful, the pellet form will deliver the fertilizer more efficiently to crops and less fertilizer will be required to achieve the same effect on crop growth. The pellet form of fertilizer can be expected to sell at a price premium over the powder form of fertilizer based on the savings in the amount of fertilizer that needs to be used. This price premium will be a similar premium per dollar of sales in each territory. If the research and development is successful, the costs of producing pellet fertilizer are expected to be approximately the same as the costs of producing powder fertilizer and the same for both FP and USS. Both FP and USS operate at approximately the same market levels, selling their fertilizers largely to independent distributors. (iii) In this case, the controlled participants’ basis for measuring RAB shares is the most reliable ...

U.S. Parent (USP) and Foreign Subsidiary (FS) enter into a CSA to develop new anesthetic drugs. USP obtains the right to market any resulting drugs in the United States and FS obtains the right to market any resulting drugs in the rest of the world. USP and FS determine RAB shares on the basis of their respective total anticipated operating profit from all drugs under development. USP anticipates that it will receive a much higher profit than FS per unit sold because the price of the drugs is not regulated in the United States, whereas the price of the drugs is regulated in many non-U.S. jurisdictions. In both controlled participants’ territories, the anticipated operating profits are almost entirely attributable to the use of the cost shared intangibles. In this case, the controlled participants’ basis for measuring RAB shares is the most reliable ...

The facts are the same as in Example 3, except that to supply the particular needs of the U.S. market USS manufactures the feedstock with somewhat different properties than FP’s feedstock. This requires USS to employ a somewhat different production process than does FP. Because of this difference, USS would incur significant construction costs in order to adopt any new process that may be developed under the cost sharing agreement. In this case, units produced is not the most reliable basis for measuring RAB shares. In order to reliably determine RAB shares, the Commissioner measures the reasonably anticipated benefits of USS and FP on a direct basis. USS’s reasonably anticipated benefits are its reasonably anticipated total savings in electricity costs, less its reasonably anticipated costs of adopting the new process. FS’s reasonably anticipated benefits are its reasonably anticipated total savings in electricity costs ...

The facts are the same as in Example 2, except that currently USS pays $3 per unit of feedstock produced for electricity while FP pays $6 per unit of feedstock produced. In this case, units produced is not the most reliable basis for measuring RAB shares and dividing the IDCs because the participants do not expect to have a similar decrease in costs per unit of the feedstock produced. The Commissioner determines that the most reliable measure of RAB shares may be based on units of the feedstock produced if FP’s units are weighted relative to USS’s units by a factor of 2. This reflects the fact that FP pays twice as much as USS for electricity and, therefore, FP’s savings of $3 per unit of the feedstock (50% reduction of current charge of $6) would be twice USS’s savings of $1.50 per unit of feedstock (50% reduction of current charge of $3) from any new process eventually developed ...

Foreign Parent (FP) and U.S. Subsidiary (USS) both produce a feedstock for the manufacture of various high-performance plastic products. Producing the feedstock requires large amounts of electricity, which accounts for a significant portion of its production cost. FP and USS enter into a CSA to develop a new process that will reduce the amount of electricity required to produce a unit of the feedstock. FP and USS currently both incur an electricity cost of $2 per unit of feedstock produced and rates for each are expected to remain similar in the future. The new process, if it is successful, will reduce the amount of electricity required by each company to produce a unit of the feedstock by 50%. Switching to the new process would not require FP or USS to incur significant investment or other costs. Therefore, the cost savings each company is expected to achieve after implementing the new process are $1 per unit of feedstock produced. Under the CSA, FP and USS divide the costs of developing the new process based on the units of the feedstock each is anticipated to produce in the future. In this case, units produced is the most reliable basis for measuring RAB shares and dividing the IDCs because each controlled participant is expected to have a similar $1 (50% of current charge of $2) decrease in costs per unit of the feedstock produced ...

Controlled parties A and B enter into a CSA to develop product and process intangibles for already existing Product P. Without such intangibles, A and B would each reasonably anticipate revenue, in present value terms, of $100M from sales of Product P until it becomes obsolete. With the intangibles, A and B each reasonably anticipate selling the same number of units each year, but reasonably anticipate that the price will be higher. Because the particular product intangible is more highly regarded in A’s market, A reasonably anticipates an increase of $20M in present value revenue from the product intangible, while B reasonably anticipates an increase of only $10M in present value from the product intangible. Further, A and B each reasonably anticipate spending an additional amount equal to $5M in present value in production costs to include the feature embodying the product intangible. Finally, A and B each reasonably anticipate saving an amount equal to $2M in present value in production costs by using the process intangible. A and B reasonably anticipate no other economic effects from exploiting the cost shared intangibles. A’s reasonably anticipated benefits from exploiting the cost shared intangibles equal its reasonably anticipated increase in revenue ($20M) plus its reasonably anticipated cost savings ($2M) less its reasonably anticipated increased costs ($5M), which equals $17M. Similarly, B’s reasonably anticipated benefits from exploiting the cost shared intangibles equal its reasonably anticipated increase in revenue ($10M) plus its reasonably anticipated cost savings ($2M) less its reasonably anticipated increased costs ($5M), which equals $7M. Thus A’s reasonably anticipated benefits are $17M and B’s reasonably anticipated benefits are $7M ...

The following examples illustrates this paragraph (e)(2)(ii): ...

Other bases for measuring anticipated benefits may in some circumstances be appropriate, but only to the extent that there is expected to be a reasonably identifiable relationship between the basis of measurement used and additional revenue generated or net costs saved by the use of cost shared intangibles. For example, a division of costs based on employee compensation would be considered unreliable unless there were a relationship between the amount of compensation and the expected additional revenue generated or net costs saved by the controlled participants from using the cost shared intangibles ...

Operating profit of each controlled participant from the activities in which cost shared intangibles are exploited, as determined before any expense (including amortization) on account of IDCs, may be used as an indirect basis for measuring anticipated benefits. This basis of measurement will more reliably determine RAB shares to the extent that such profit is largely attributable to the use of cost shared intangibles, or if the share of profits attributable to the use of cost shared intangibles is expected to be similar for each controlled participant. This circumstance is most likely to arise when cost shared intangibles are closely associated with the activity that generates the profit and the activity could not be carried on or would generate little profit without use of those intangibles ...

Sales by each controlled participant in the business activities in which cost shared intangibles are exploited may be used as an indirect basis for measuring its anticipated benefits. This basis of measurement will more reliably determine RAB shares to the extent that each controlled participant is expected to have a similar increase in net profit or decrease in net loss attributable to cost shared intangibles per dollar of sales. This circumstance is most likely to arise if the costs of exploiting cost shared intangibles are not substantial relative to the revenues generated, or if the principal effect of using cost shared intangibles is to increase the controlled participants’ revenues (for example, through a price premium on the products they sell) without affecting their costs substantially. Sales by each controlled participant are unlikely to provide a reliable basis for measuring RAB shares unless each controlled participant operates at the same market level (for example, manufacturing, distribution, etc.) ...

Units of items used, produced, or sold by each controlled participant in the business activities in which cost shared intangibles are exploited may be used as an indirect basis for measuring its anticipated benefits. This basis of measurement will more reliably determine RAB shares to the extent that each controlled participant is expected to have a similar increase in net profit or decrease in net loss attributable to the cost shared intangibles per unit of the item or items used, produced, or sold. This circumstance is most likely to arise when the cost shared intangibles are exploited by the controlled participants in the use, production, or sale of substantially uniform items under similar economic conditions ...

Indirect bases for measuring anticipated benefits from participation in a CSA include the following: ...

In order to estimate a controlled participant’s RAB share, the amount of each controlled participant’s reasonably anticipated benefits must be measured on a basis that is consistent for all such participants. See paragraph (e)(2)(ii)(E) Example 9 of this section. If a controlled participant transfers a cost shared intangible to another controlled taxpayer, other than by way of a transfer described in paragraph (f) of this section, that controlled participant’s benefits from the transferred intangible must be measured by reference to the transferee’s benefits, disregarding any consideration paid by the transferee to the controlled participant (such as a royalty pursuant to a license agreement). Reasonably anticipated benefits are measured either on a direct basis, by reference to estimated benefits to be generated by the use of cost shared intangibles (generally based on additional revenues plus cost savings less any additional costs incurred), or on an indirect basis, by reference to certain measurements that reasonably can be assumed to relate to benefits to be generated. Such indirect bases of measurement of anticipated benefits are described in paragraph (e)(2)(ii) of this section. A controlled participant’s reasonably anticipated benefits must be measured on the basis, whether direct or indirect, that most reliably determines RAB shares. In determining which of two bases of measurement is most reliable, the factors set forth in § 1.482-1(c)(2)(ii) (Data and assumptions) must be taken into account. It normally will be expected that the basis that provided the most reliable estimate for a particular year will continue to provide the most reliable estimate in subsequent years, absent a material change in the factors that affect the reliability of the estimate. Regardless of whether a direct or indirect basis of measurement is used, adjustments may be required to account for material differences in the activities that controlled participants undertake to exploit their interests in cost shared intangibles. See Examples 4 and 7 of paragraph (e)(2)(ii)(E) of this section ...

...

(i) USP, a US company, wholly owns foreign subsidiary, FS. USP and FS enter into a CSA at the start of Year 1. The CSA’s total IDCs are $100,000 in each year for Years 1 through 4. In Year 1, USP correctly estimates its RAB share as 50%, based on information available at the time, and therefore correctly computes $50,000 as its cost contribution for Year 1. (ii) In Year 4, USP correctly estimates its RAB share to be 70%, based on information available at the time and, therefore, correctly computes $70,000 as its cost contribution for Year 4. (iii) In Year 4, USP also files an amended return for Year 1 in which USP deducts a cost contribution of $70,000, asserting that, for this purpose, it should revise its Year 1 estimated RAB share to 70% based on the information that is now available to it in Year 4. The Commissioner determines that USP is incorrect for two reasons. First, a RAB share determined for a particular purpose (here, to determine USP’s IDC shares and thus USP’s cost contributions in Year 1) should not be revised based on information not available to USP until Year 4. See paragraph (e)(1)(i) of this section. Second, more generally, USP is not permitted to file an amended return for this purpose under § 1.482-1(a)(3). Therefore, for both of these reasons, Commissioner adjusts USP’s amended return for Year 1 by disallowing $20,000 of the $70,000 deduction ...

(i) USP and FS plan to conduct research to develop Product Lines A and B. USP and FS reasonably anticipate respective benefits from Product Line A of 100X and 200X and respective benefits from Product Line B, respectively, of 300X and 400X. USP and FS thus reasonably anticipate combined benefits from Product Lines A and B of 400X and 600X, respectively. (ii) USP and FS could enter into a separate CSA to develop Product Line A with respective RAB shares of 331⁄3 percent and 662⁄3 percent (reflecting a ratio of 100X to 200X), and into a separate CSA to develop Product Line B with respective RAB shares of 426⁄7 percent and 571⁄7 percent (reflecting a ratio of 300X to 400X). Alternatively, USP and FS could enter into a single CSA to develop both Product Lines A and B with respective RAB shares of 40 percent and 60 percent (in the ratio of 400X to 600X). If the separate CSAs are chosen, then any costs for activities that contribute to developing both Product Line A and Product Line B will constitute IDCs of the respective CSAs as required by paragraphs (d)(1) and (2) of this section ...

The following examples illustrate the principles of this paragraph (e)(1): ...

A controlled participant’s RAB share must be determined by using the most reliable estimate. In determining which of two or more available estimates is most reliable, the quality of the data and assumptions used in the analysis must be taken into account, consistent with § 1.482-1(c)(2)(ii) (Data and assumptions). Thus, the reliability of an estimate will depend largely on the completeness and accuracy of the data, the soundness of the assumptions, and the relative effects of particular deficiencies in data or assumptions on different estimates. If two estimates are equally reliable, no adjustment should be made based on differences between the estimates. The following factors will be particularly relevant in determining the reliability of an estimate of RAB shares: (A) The basis used for measuring benefits, as described in paragraph (e)(2)(ii) of this section. (B) The projections used to estimate benefits, as described in paragraph (e)(2)(iii) of this section ...

A controlled participant’s share of reasonably anticipated benefits is equal to its reasonably anticipated benefits divided by the sum of the reasonably anticipated benefits, as defined in paragraph (j)(1)(i) of this section, of all the controlled participants. RAB shares must be updated to account for changes in economic conditions, the business operations and practices of the participants, and the ongoing development of intangibles under the CSA. For purposes of determining RAB shares at any given time, reasonably anticipated benefits must be estimated over the entire period, past and future, of exploitation of the cost shared intangibles, and must reflect appropriate updates to take into account the most reliable data regarding past and projected future results available at such time. RAB shares determined for a particular purpose shall not be further updated for that purpose based on information not available at the time that determination needed to be made. For example, RAB shares determined in order to determine IDC shares for a particular taxable year (as set forth in paragraphs (b)(1)(i) and (d)(4) of this section) shall not be recomputed based on information not available at that time. Similarly, RAB shares determined for the purpose of using a particular method such as the acquisition price method (as set forth in paragraph (g)(5)(ii) of this section) to evaluate the arm’s length amount charged in a PCT shall not be recomputed based on information not available at the date of that PCT. However, nothing in this paragraph (e)(1)(i) shall limit the Commissioner’s use of subsequently available information for purposes of its allocation determinations in accordance with the provisions of paragraph (i) (Allocations by the Commissioner in connection with a CSA) of this section ...

...

...

A CSA produces results that are consistent with an arm’s length result within the meaning of § 1.482-1(b)(1) if, and only if, each controlled participant’s IDC share (as determined under paragraph (d)(4) of this section) equals its RAB share, each controlled participant compensates its RAB share of the value of all platform contributions by other controlled participants, and all other requirements of this section are satisfied ...

See paragraph (b)(1)(i) of this section regarding the requirement that controlled participants, as defined in section (j)(1)(i) of this section, share intangible development costs (IDCs) in proportion to their shares of reasonably anticipated benefits (RAB shares) by entering into cost sharing transactions (CSTs) ...