Cite as: 537 U. S. 437 (2003)
Thomas, J., dissenting
Because I believe that § 1.861-8(e)(3) does not apply to a DISC, I need not decide here whether § 1.861-8(e)(3) is consistent with the text of § 861(b) and may be properly applied in other contexts. I am puzzled, however, by the Court's assertion that the Secretary is free to determine that certain expenses "can be properly apportioned on a categorical basis," ante, at 451, and the implication that the Secretary has authority to require "ratable apportionment of expenses that could be, but perhaps in fairness should not be, treated as direct costs." Ibid. By its terms, § 861(b) appears to contemplate two types of expenses: (1) those that can definitely be allocated to some item or class of gross income and (2) those that cannot. 26 U. S. C. § 861(b) (providing for the deduction of "the expenses, losses, and other deductions properly apportioned or allocated thereto and a ratable part of any expenses, losses, or other deductions which cannot definitely be allocated to some item or class of gross income" (emphasis added)). Moreover, on its face, the statute does not appear to permit expenses to be "deemed" related to an item or class of gross income, even though in actual fact they are not so related. Yet, § 1.861-8(e)(3) relies on the notion of "deemed relationships." The regulation states that the methods of allocation and apportionment established there "recognize that research and development is an inherently speculative activity, that findings may contribute unexpected benefits, and that the gross income derived from successful research and development must bear the cost of unsuccessful research and development." 26 CFR § 1.861-8(e)(3)(i)(A) (1979). The regulation then proceeds to require the allocation of R&D expenses based on 2-digit SIC groups. But neither the regulation nor the Court attempt to reconcile the statutory text with the regulation's determination to allocate certain R&D expenses to items or classes of gross income that admittedly did not benefit from that research.
463
Page: Index Previous 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 NextLast modified: October 4, 2007