Intercompany Transfer Pricing Regulations Under Section 482

DEPARTMENT OF THE TREASURY

Internal Revenue Service

26 CFR Parts 1 and 602

[TD 8552]

RIN 1545-AL80

Intercompany Transfer Pricing Regulations Under Section 482

AGENCY: Internal Revenue Service (IRS), Treasury.

ACTION: Final regulations.

-----------------------------------------------------------------------

SUMMARY: This document contains final regulations relating to

intercompany transfer pricing under section 482 of the Internal Revenue

Code. These regulations reflect the changes made to section 482 by the

Tax Reform Act of 1986, and provide guidance implementing the

amendment.

EFFECTIVE DATES: Effective July 8, 1994, except Secs. 1.482-OT through

1.482-6T are removed effective October 6, 1994.

FOR FURTHER INFORMATION CONTACT: Sim Seo of the Office of Associate

Chief Counsel (International), IRS. (202) 622-3840 (not a toll-free

number).

SUPPLEMENTARY INFORMATION:

Paperwork Reduction Act

The collections of information contained in these final regulations

have been reviewed and approved by the Office of Management and Budget

in accordance with the requirements of the Paperwork Reduction Act (44

U.S.C. 3504(h)) under control number 1545-1364. The estimated average

annual burden per recordkeeper is .8 hour. The estimated average annual

reporting burden per respondent is 1 hour.

Comments concerning the accuracy of this burden estimate, and

suggestions for reducing this burden, should be sent to the Internal

Revenue Service, Attn: IRS Reports Clearance Officer, PC:FP,

Washington, DC 20224, and to the Office of Management and Budget, Attn:

Desk Office for the Department of the Treasury, Office of Information

and Regulatory Affairs, Washington, DC 20503.

Background

Section 482 was amended by the Tax Reform Act of 1986, Pub. L. 99-

514, 100 Stat. 2085, 2561, et. seq. On January 21, 1993, temporary

regulations relating to the evaluation of intercompany transfer pricing

under section 482 were published in the Federal Register (58 FR 5263).

A notice of proposed rulemaking (INTL-401-88) cross-referencing the

temporary regulations was published in the Federal Register for the

same day (58 FR 5310).

Written comments responding to the notice of proposed rulemaking

were received, and a public hearing was held on August 16, 1993. After

consideration of all the comments, the proposed regulations under

section 482 are adopted as revised by this Treasury decision, and the

corresponding temporary regulations are removed.

Explanation of Revisions and Summary of Comments

Introduction

The Tax Reform Act of 1986 (the Act) amended section 482 to require

that consideration for intangible property transferred in a controlled

transaction be commensurate with the income attributable to the

intangible. The legislative history of the Act indicates that the

change was intended to assure that the division of income between

related parties reasonably reflects the economic activities that each

undertakes. See H.R. Rep. 99-281, 99th Cong., 2d Sess. (1986) at II-

637. The legislative history also expresses concern that insufficiently

stringent standards had been used in determining whether an

uncontrolled transaction is sufficiently comparable to a controlled

transaction. In particular, the legislative history observed that

industry norms for transfers of less profitable intangibles frequently

are not realistic comparables for transfers of so-called ``high

profit'' intangibles. See H.R. Rep. 99-426, 99th Cong., 1st Sess.

(1985) at 424. Finally, the Conference Committee report recommended

that the IRS conduct a comprehensive study and consider whether the

regulations under section 482, which had been issued in 1968 (the 1968

regulations) should be ``modified in any respect.''

The White Paper

In response to this directive, the IRS and the Treasury Department

issued a study of intercompany pricing (Notice 88-123, 1988-2 C.B. 458)

on October 18, 1988 (the White Paper). The White Paper proposed two

approaches for implementing the ``commensurate with income'' standard

with respect to transfers of intangible property. The first was based

on either an ``exact comparable'' method or an ``inexact comparable''

method. The second was an income-based approach that also included two

methods: the basic arm's length return method (the BALRM), and the

BALRM with profit split. The BALRM generally assigned an average rate

of return to the assets and functions devoted to the routine activities

associated with the controlled transaction. When high profit

intangibles were involved, any residual profit would be divided on the

basis of the estimated relative values of the intangibles that each

party contributed to the activity.

The 1992 Proposed Regulations

The IRS issued proposed regulations under section 482 on January

30, 1992 (INTL-0372-88; INTL-0401-88, 57 FR 3571) (the 1992

regulations). The most significant change proposed was the introduction

of three new pricing methods for transfers of intangible property: the

matching transaction method (the MTM), the comparable adjustable

transaction method (the CATM) and the comparable profit interval (the

CPI). The CPI also could be used with respect to transfers of tangible

property.

The MTM and the CATM were based on exact and inexact comparable

transactions, respectively. The CPI was an income-based method under

which the operating income resulting from a controlled transaction was

compared with the operating income of comparable uncontrolled

taxpayers. The consideration charged in the controlled transaction

would be considered arm's length if the taxpayer's operating income

fell within a range of results derived from the uncontrolled taxpayers

over a three-year period. Finally, the 1992 regulations required that

the result derived from the CATM be verified by the CPI.

In addition to providing new methods for transfers of intangibles,

the 1992 regulations implemented the ``commensurate with income''

standard by providing that these methods could be applied to adjust the

consideration charged in the year of examination (periodic adjustments)

unless one of three narrow exceptions applied.

Other changes made by the 1992 regulations included modification of

the rules with respect to transfers of tangible property, principally

by providing for the use of CPI with respect to tangible property,

requiring that the results derived from the resale price or cost plus

methods be verified by application of the CPI, introducing new cost

sharing regulations, introducing a limited comparable profit split

method, and relaxing the fixed priority of methods set forth in the

1968 regulations.

Commenters criticized several aspects of the 1992 regulations,

including the inclusion of the CPI in the regulations, the requirement

that the results of most methods be verified by the CPI, the tight

standards of comparability for applying the MTM, the narrow scope of

the exceptions from periodic adjustments, the narrow scope for the

profit split method, and the lack of a safe harbor. Commenters

generally approved of the introduction of a range of acceptable results

and the use of a multi-year average under the CPI, and the relaxation

of the strict priority of methods.

The 1993 Temporary and Proposed Regulations

In response to comments, the IRS issued revised temporary and

proposed regulations on January 21, 1993 (TD 8470; INTL 401-88, 58 FR

5263) (the 1993 regulations). With the exception of the provisions on

cost sharing, the 1993 regulations replaced all the provisions

contained in the 1992 regulations and added some new provisions. In

addition, the 1993 regulations also modified other provisions under the

1968 regulations that had not been affected by the 1992 regulations.

The 1993 regulations also adopted a structure different from that set

forth under the 1968 regulations. Section 1.482-1T sets forth general

rules applicable to all the subsequent provisions of the regulations

under section 482. Most important of these rules is extensive guidance

to be applied in determining whether an uncontrolled transaction is

sufficiently comparable to serve as a basis for application of a

pricing method. Determining comparability under these rules generally

requires consideration of functions, risks, contractual terms, economic

conditions and products. Detailed guidance is provided as to how these

factors are to be assessed. Finally, the relative importance of any of

these factors varies depending upon the method applied.

Section 1.482-1T also provides a ``best method rule'' to be used in

determining which method provides the most accurate measure of an arm's

length result in a given case. The best method rule adopts a flexible

approach under which the determination of which method is most accurate

depends on the facts and circumstances of the case. Factors to take

into account in this determination include the completeness and

accuracy of available data, the degree of comparability between the

controlled and uncontrolled transactions, and the extent of adjustments

required to apply a method. The best method rule also provides that

when two methods provide inconsistent results and the best method rule

does not otherwise indicate which of the two analyses should be

preferred, an additional factor to take into account is whether a third

method provides a result that is consistent with the result of one of

the first two methods.

Section 1.482-1T also sets forth special rules to deal with issues

presented by market penetration strategies, different geographic

markets, ``location savings,'' aggregation of transactions, analysis of

contractual terms, multiple year analyses, collateral adjustments,

coordination with section 936, and consideration of alternatives. Under

the latter rule, the district director is instructed to determine an

arm's length price based upon the structure actually adopted, and not

to restructure the transaction as long as its form was consistent with

its substance. The district director may, however, consider

alternatives reasonably available to the taxpayer in determining

whether a purported comparable transaction actually represents a

reliable indicator of the terms to which the taxpayer would have agreed

under arm's length conditions.

In addition, Sec. 1.482-1T provides that no allocation will be made

if the taxpayer's result falls within a range of arm's length results.

Under this rule, two or more valid applications of any single method

create a range of acceptable results within which the taxpayer's result

are considered to satisfy the arm's length standard. Results falling

outside the range are subject to adjustment to any point within the

range (generally the midpoint).

Section 1.482-1T also provides a safe harbor for small taxpayers.

That provision would permit a small taxpayer (generally defined as a

group having less than $10 million in U.S. or foreign sales) to elect

to determine its U.S. taxable income in accordance with annual

published measures of profitability.

Finally, the 1993 regulations include proposed regulations to be

added to Sec. 1.482-1 dealing with foreign legal restrictions. In

general, this rule provides that a foreign legal restriction preventing

or limiting payment of an arm's length amount will be respected for

purposes of determining an arm's length consideration only if there is

evidence of a comparable uncontrolled transaction in which unrelated

parties agreed to enter a similar transaction subject to the

restriction. In other cases, the foreign legal restriction will be

disregarded in determining an arm's length amount, but the taxpayer is

permitted to elect a deferred method of accounting to defer recognition

of additional income until such time as the restriction is lifted,

subject to the consistent deferral of related expenses.

In addition to revising Sec. 1.482-1, the 1993 regulations

restructured the rules relating to transfers of tangible and intangible

property. The 1993 regulations contain separate subsections for these

rules, which were all contained in Sec. 1.482-2 in the 1968

regulations. Specifically, Sec. 1.482-3T governs transfers of tangible

property, Sec. 1.482-4T governs transfers of intangible property, and

Sec. 1.482-5T sets forth the comparable profit method (the CPM), which,

as under the 1993 regulations, applies to transfers of tangibles or

intangibles. In addition, the 1993 regulations contain a proposed set

of profit split rules (to be added as Sec. 1.482-6).

The section on transfers of tangible property provides five

principal methods: The comparable uncontrolled price (CUP) method, the

resale price method, the cost plus method, the CPM, and (when

authorized by the regulations) profit split. The CUP method was

modified to provide that this method may be applied only if there are

at most minor differences between the controlled and uncontrolled

transactions. The 1968 regulations provided that CUP could be used only

if the transactions were ``so nearly identical'' that any differences

could be reflected by a reasonable number of adjustments. Since the CUP

method is likely to achieve the highest degree of comparability of any

method potentially applicable to a transfer of tangible property, the

1993 regulations state that the CUP method generally provides the most

reliable measure of an arm's length result when it can be applied. The

rules under the resale price and cost plus methods were not

substantially changed from their predecessors in the 1968 regulations.

Section 1.482-3T also provides that when none of the enumerated methods

can be applied, other (unspecified) methods may be used. In order to

employ an unspecified method, taxpayers are required to disclose the

use of such method on the tax return and to prepare contemporaneous

documentation explaining why the method provides the most accurate

measure of an arm's length result. Finally, Sec. 1.482-3T provides

rules coordinating the application of the tangible and intangible rules

in cases involving transfers of so-called imbedded intangibles.

Section 1.482-4T combine the MTM and CATM from the 1992 regulations

into a single method known as the comparable uncontrolled transaction

(CUT) method. Unlike the CATM, the results of the CUT method are not

subject to mandatory check by the CPI. As with the CUP method under

Sec. 1.482-3T, the regulations provide that this method ordinarily

provides the most reliable measure of an arm's length result. The

mandatory CPI check in the 1992 regulations is replaced in the 1993

regulations by requiring that the intangibles transferred in the

controlled and uncontrolled transactions have substantially the same

profit potential. In addition, to apply this method, the property

transferred must be from the same class of intangible property and

relate to the same class of products or services. Further, the

underlying circumstances of the two transactions must be sufficiently

similar that reliable adjustments may be made to account for the effect

of any differences.

Section 1.482-4T also permits the application of unspecified

methods, subject to the same constraints on taxpayer use that are

imposed under Sec. 1.482-3T. The preamble of the proposed regulation

also requested comment on the merits of including a method that would

measure an arm's length result for the transfer of an intangible by

discounting the projected costs and benefits to the licensor or

licensee, employing valid measures of the cost of capital such as those

derived from the capital asset pricing model.

The 1993 regulations broaden the exceptions from periodic

adjustments that were contained in the 1992 regulations by providing

two exceptions. The first applies if the consideration for the

intangible was determined to be arm's length under the CUT Method in

the first year when substantial periodic consideration was paid, the

taxpayer's actual profits from the intangible remained within a band

between 80 and 120 percent of the profits projected at the time of the

controlled transactions, and certain other conditions are met. The

second exception is similar but applies to methods other than the CUT

Method.

The 1993 regulations also provide rules for identifying the owner

of an intangible for purposes of section 482 (the developer-assister

rule). These rules generally track rules provided in prior regulations,

under which the owner normally is considered to be the controlled

taxpayer that bears the greatest share of the risk of developing the

intangible. The party that bears the greatest risk of development

generally is determined by identifying costs of development. Under this

rule the owner for purposes of income allocation under section 482

would not necessarily be the legal owner.

Section 1.482-5T describes the CPM. The CPM may be applied to

transfers of tangible and intangible property. In broad terms the CPM

is similar to the CPI set out in the 1992 regulations. However, it no

longer serves as a mandatory check on the results provided by certain

other methods. In addition to the general comparability factors and

other considerations that must be applied before a method can be

considered to provide a reasonable and reliable benchmark, Sec. 1.482-

5T provides that CPM ordinarily is inappropriate if the tested party

owns ``valuable non-routine intangible'' property. This limitation was

added to the other limitations generally applicable to all methods

because CPM could be expected to understate the income attributable to

such property due to the difficulty in locating uncontrolled taxpayers

that possess comparable intangible property.

The 1993 regulations do not define the term ``valuable non-routine

intangible.'' The preamble to the regulations, however, states that in

general the term would encompass intangible property ``that is central

to the conduct of a business activity and without which the business

activity could not be conducted.'' The preamble to the proposed

regulations requested comment on possible more precise definitions.

The CPM generally is applied to the taxpayer with the simplest and

most easily compared operations (the tested party). In identifying

potential comparables, the regulations provide that the standard of

comparability is not as strict as other methods. Some diversity in

terms of the functions and products is permitted, although the degree

of comparability affects the reliability of the results in relation to

the results of other methods under the best method rule. It also

affects the derivation of the arm's length range under this method. In

addition, the regulations describe a number of adjustments, including

adjustments to achieve accounting consistency, that should be made when

possible to the results of the uncontrolled comparables to enhance

comparability.

Like the CPI under the 1992 regulations, a result will satisfy the

arm's length standard under the CPM if it falls within a range of

results, based on a single profit level indicator derived from

uncontrolled comparables. Profit level indicators include the rate of

return on capital employed (i.e., rate of return on assets) and

financial ratios such as operating profit to gross sales and gross

profit to operating expenses (Berry ratio).

Unlike the other methods in the 1993 regulations, the arm's length

range may be constructed under the CPM in one of two ways. First, if

reliable adjustments for all material differences that would affect

profitability are made, the arm's length range includes all the results

obtained, as under the other methods. In other cases, however, the

range is limited by statistical methods. The range so determined

consists either of the interquartile range or the range constructed

under some other statistically valid method. No additional guidance was

provided as to how the range would be established if the interquartile

range were not used.

The 1993 regulations also contain a set of proposed regulations

(Sec. 1.482-6) providing profit split methodologies. Three methods are

described: the residual allocation rule, the capital employed

allocation rule and the comparable profit split. In addition, other

profit splits may be used if they provide an economically valid basis

for the allocation of the combined profit or loss of the relevant

business activity. The basic objective of the profit split methods is

to estimate an arm's length return by comparing the relative economic

contributions that two parties make to the success of an activity (the

relevant business activity), and dividing the returns from the relevant

business activity between them on the basis of the value of such

contributions.

The residual allocation rule is similar to the BALRM with profit

split described in the White Paper. It consists of two basic steps.

First, using other methods such as the CPM, market returns for routine

functions are estimated and allocated to the parties that performed

them. The remaining, residual amount then is allocated between the

parties on the assumption that this residual is attributable to

intangible property contributed to the activity by the controlled

taxpayers. Based on this assumption, the residual is divided based on

the estimate of the relative value of the parties' contributions of

such property. Since fair market value of the intangible property

usually would not be readily ascertainable, the regulations permit use

of other measures of the relative values of intangible property,

including capitalized research and development expenses.

The capital employed allocation rule may be applied only if all the

controlled taxpayers participating in the relevant business activity

assumed an approximately equal level of risk with respect to their

capital employed. Comment was requested on the feasibility of measuring

relative levels of risk. This method divides the combined operating

profit from the relevant business activity by allocating an equal

return to each controlled taxpayer's capital employed. Capital employed

may be measured by either book or fair market value, as long as all

assets are valued on the same basis. Further, if book value is used and

there are intangible assets that have no book value, some other measure

(such as capitalized research and development expense) must be used.

The comparable profit split rule is similar to the profit split

rule set forth in the 1992 regulations. It essentially may be applied

only if it is possible to locate two unrelated parties that are each

comparable to one of the controlled taxpayers and that deal with one

another in a comparable manner. In such a case the combined operating

profit from the relevant business activity is divided among the

controlled taxpayers in the same percentage as it was divided among the

unrelated parties.

There are a number of substantive and procedural restrictions on

the use of the profit split method. These restrictions were imposed

because the profit split method relies either wholly or in part on

internal data rather than data derived from uncontrolled taxpayers, and

it is therefore likely that other methods will provide a more reliable

measure of an arm's length result under the best method rule.

There are three substantive limitations on the use of the profit

split method. The profit split method may be applied only if both

controlled taxpayers own valuable non-routine intangible property, the

intangibles contribute significantly to the combined operating profit

derived from the relevant business activity, and there were significant

transactions between the controlled taxpayers.

The most important administrative requirement is that the taxpayer

must make a binding election to apply the profit split method, which

can be revoked only with the consent of the Commissioner. In addition,

the taxpayer also is required to document the combined profit or loss

attributable to the relevant business activity to the satisfaction of

the district director and explain in such documentation why the profit

split method provides the best method for determining an arm's length

result. Further, prior to electing the profit split method the taxpayer

must execute a pricing agreement setting forth the method chosen, and

the method must be applied consistently from year to year. The district

director also is required to meet all of the substantive, but not the

procedural requirements, to apply the profit split method.

Comments on the 1993 Regulations

The IRS received comments on the 1993 regulations from many

taxpayers and industry and professional groups. In addition, comments

were received from several tax treaty partners, both individually and

through the international forum of the Organization for Economic

Cooperation and Development (OECD).

The commenters generally approved of the introduction of the best

method rule and the flexibility implied by reliance on comparability,

rather than a hierarchy of methods, to identify the method that would

provide the most accurate measure of an arm's length result. Commenters

also generally approved of the extension of the concept of an arm's

length range to all the methods, the proposal of a set of profit split

methods, the inclusion of exceptions from periodic adjustments that

were broader than the exceptions provided under the 1992 regulations,

and the elimination of the requirement that the CPM be a mandatory

check on the results of most other methods.

Commenters also expressed concerns with various aspects of the 1993

regulations. In particular, the most critical comments focused on the

CPM and profit split portions of the regulation. With respect to the

CPM, some commenters expressed a belief that the method was

inconsistent with the arm's length standard and should be eliminated.

Others felt that it could provide useful evidence in certain cases and

therefore should be retained, but the scope for its application should

be circumscribed further than it was in the 1993 regulations. In

addition, some commenters were concerned that examiners might apply the

CPM without regard to the evidence provided by other methods.

Contributing to this concern was the fact that the comparability

standards under other methods were generally tighter than under the

CPM, potentially making the CPM more readily available. Finally,

commenters evidenced some confusion over certain language in

Sec. 1.482-5T; some interpreted the scope language under that section

as indicating that the CPM was preferred to other methods, in

contradiction to the best method rule.

With respect to the profit split method, many commenters urged that

the elective and other procedural requirements for its use by taxpayers

be eliminated. Commenters expressed some ambivalence with respect to

the requirement that both parties to the transaction own valuable non-

routine intangibles in order to employ the profit split method. Some

were concerned that this requirement would make profit split

unavailable in some cases in which it might otherwise provide the most

reliable measure of an arm's length result. Others were uncertain of

its effect given the absence of an explicit definition of the term in

the regulations. Many suggestions were received as to possible

definitions of this term, which was relevant not only under profit

split but also under the CPM.

In addition to the comments received on the CPM and profit split

methods, numerous concerns were expressed with regard to other aspects

of the regulations. The most significant of these included the

following. In connection with the provisions relating to tangible

property, a large number of commenters expressed concerns about the

limited role accorded to evidence provided by ``inexact comparables,''

particularly under the CUP method, arguing that in some instances the

evidence provided by an ``inexact comparable'' would be more reliable

than other available information. Some also perceived the modified

comparability standard under the CUP method as being more restrictive

than the standard under the 1968 regulations, and several comments

objected to the restrictions placed on the use of unspecified methods.

In connection with the provisions relating to intangibles, several

commenters objected to the high comparability standard under the CUT

method, particularly the requirement that profit potential of the

controlled transaction and the uncontrolled transaction be

substantially the same. Further, the continued availability of periodic

adjustments was viewed by some as potentially conflicting with the

arm's length standard. Others criticized the failure of the developer-

assister rule to give sufficient weight to legal ownership in

identifying the owner of an intangible, and objected to an example

illustrating the potential use of alternatives in applying the arm's

length standard.

Commenters also requested further guidance as to the interaction of

the tangible and intangible property rules. Finally, some commenters

requested that the thresholds for application of the safe harbor be

liberalized to make it more widely available, and others requested that

the published measures of profitability allow electing taxpayers to

report less income than they would be expected to report under the

otherwise applicable methods.

Further discussion of the comments received is included in the

following description of the changes reflected in the final

regulations.

The Final Regulations

While the final regulations reflect numerous modifications in

response to the comments received on the 1993 regulations, both the

format and the substance of the final regulations are generally

consistent with the 1993 regulations. The changes adopted are intended

to clarify and refine those provisions of the 1993 regulations that

required improvement, without fundamentally altering the basic policies

reflected in the 1993 regulations.

The most noteworthy feature of the 1993 regulations in comparison

to earlier versions of the regulations under section 482 was the

emphasis on comparability, and the resulting flexibility. This feature

of the 1993 regulations was generally well received by taxpayers and

foreign governments. The final regulations adhere to this emphasis, and

in some cases increase it. For instance, so-called ``inexact''

comparables potentially may be used under all the methods in the final

regulations, while they were generally not taken into account under the

1993 regulations. Further, the elective and other procedural barriers

to the use of profit split and ``other'' (i.e., unspecified) methods

have been removed, and the limitations regarding the presence of

valuable non-routine intangibles under CPM and profit split have been

eliminated. These restrictive rules were contained in the 1993

regulations primarily out of a concern that in their absence taxpayers

or the IRS might employ methods that did not provide the best measure

of an arm's length result.

By removing these restrictions, the final regulations are intended

to maximize the extent to which relevant information may be taken into

account in evaluating taxpayers' results under the arm's length

standard. As a consequence, however, the emphasis on comparability and

the importance of the best method rule are increased; because ex ante

restrictions will no longer prohibit the use of potentially less

reliable information or methodologies, it is critically important that

the best method rule be properly applied to select the most reliable

measure of an arm's length result from the available evidence. Thus,

taxpayers and the IRS will be required to exercise considerable

judgment in applying the arm's length standard. To assist taxpayers and

the IRS in exercising this judgment, the discussion of the factors to

consider in applying the best method rule has been substantially

expanded. Set forth below is a more detailed explanation of the

provisions in the final regulations.

Section 1.482-1

With one exception, the scope and purpose provision of the

regulations (Sec. 1.482-1(a)(1)) is substantially similar to its

counterpart in the 1993 regulations. The final regulations delete the

statement that section 482 places uncontrolled and controlled taxpayers

on a parity by determining the controlled taxpayer's true taxable

income ``in a manner that reasonably reflects the relative economic

activity undertaken by each taxpayer.'' The definition of true taxable

income in Sec. 1.482-1(i)(9) already incorporates the notion that,

under section 482, the controlled taxpayer should earn the amount of

income that would have resulted had it dealt with other controlled

taxpayers at arm's length. Because a transaction at arm's length

naturally would reflect the ``relative economic activity undertaken,''

this definition incorporates that concept, and it is unnecessary to

include the additional language in this provision.

The provision authorizing the IRS to make allocations (Sec. 1.482-

1(a)(2)) is unchanged from the 1993 regulations.

The provision regarding the taxpayer's use of section 482

(Sec. 1.482-1(a)(3)) has been revised to clarify that, although the

taxpayer is generally barred from invoking the provisions of section

482, the taxpayer may report an arm's length result on its original tax

return, even if such result reflects prices that are different from the

prices originally set forth in the taxpayer's books and records. In

response to comments, the requirement in the 1993 regulations that such

differences be eliminated through the use of ``compensating

adjustments'' has been deleted. Section 482 is concerned only with

whether the taxpayer reports its true taxable income, and whether or

not this result is consistent with the taxpayer's books, or is

corrected in the books, is generally irrelevant to this inquiry.

However, the absence of a requirement to eliminate book and tax

differences for section 482 purposes has no effect on the mechanisms

otherwise provided for reporting and reconciling such differences

(e.g., Schedule M-1 of Form 1120). Further, the limited exception

provided by this rule does not permit taxpayers to apply section 482 at

will; thus, for example, a taxpayer may not rely on section 482 to

reduce its taxable income on an amended return.

Section 1.482-1(b) summarizes some key principles that guide

application of section 482. Section 1.482-1(b)(1) states that the

governing principle under section 482 is the arm's length standard.

Under this standard controlled taxpayers are expected to realize from

their controlled transactions the results that would have been realized

if uncontrolled taxpayers had engaged in the same transactions under

the same circumstances. This expression of the arm's length standard

differs from that set forth in the 1993 regulations, which stated that

the arm's length standard was satisfied if the results of a controlled

transaction were consistent with the results of ``comparable

transactions between uncontrolled taxpayers.'' The latter definition

has been replaced because it is inconsistent with the notion underlying

the arm's length standard that controlled and uncontrolled taxpayers

should be placed on the same (rather than a merely similar) footing.

However, this provision recognizes that in most cases identical

transactions between unrelated parties will not be located, and it

therefore will be appropriate to consider uncontrolled transactions

that are comparable rather than identical.

Section 1.482-1(b)(2) provides rules for determining the type of

method that will be applied to evaluate whether controlled transactions

are at arm's length, given that different methods apply to different

types of transactions (e.g., transfers of property or services). In

some cases it may be necessary to apply more than one method to a

single transaction when the transaction is most reliably evaluated

under more than one method. This provision is identical to its

counterpart in the 1993 regulations, except that it, along with a

number of other provisions, has been revised in response to comments to

make clear that it applies to taxpayers as well as the district

director. Thus, such provisions apply with equal force to the district

director and to a taxpayer that seeks to apply the final regulations

for purposes of determining and reporting its true taxable income on

its original return, consistent with Sec. 1.482-1(a)(3).

Section 1.482-1(c) contains the best method rule. Although similar

in purpose and substance to its counterpart in the 1993 regulations,

this provision contains considerably more detail and guidance for

application than its predecessor, and has been given more prominence in

the structure of the regulation. The best method rule guides the

application of all the methods in the final regulations. Whenever the

available data creates the possibility that more than one method could

be applied to a controlled transaction (or that one method could be

applied in more than one way), the best method rule must be applied to

determine which of those methods (or applications) will be selected.

The best method rule provides that an arm's length result must be

determined under the method that, given the facts and circumstances,

provides the ``most reliable measure'' of an arm's length result. This

formulation modifies the 1993 regulation's reference to the ``most

accurate measure,'' to conform to the factors that are taken into

account in applying the best method rule. These considerations are

couched in terms of reliability rather than accuracy.

In deciding which of two or more methods provides the most reliable

measure of an arm's length result, Sec. 1.482-1(c)(2) provides that

there are two primary factors to consider: comparability and the

quality of data and assumptions. In addition, as under the 1993

regulations, in some cases it may be relevant to consider whether the

results of a particular method are consistent with the results under

another method.

Section 1.482-1(c)(2)(i) discusses the role of comparability under

the best method rule. Although the results of identical transactions

between unrelated parties under identical circumstances provide the

most objective basis for determining the true taxable income of a

controlled taxpayer, it rarely is possible to locate uncontrolled

transactions with this degree of similarity to the controlled

transactions. Therefore, the regulations contemplate use of

uncontrolled transactions that are comparable, rather than identical,

to the controlled transaction.

In most cases there will be more than one potential comparable

uncontrolled transaction to which the controlled transaction may be

compared, and it will be possible to apply more than one method. In

such cases it is necessary to evaluate the relative degree of

comparability of each such uncontrolled transaction under the

comparability criteria relevant to the application of the method; the

method that employs the uncontrolled comparables with the highest

degree of comparability to the controlled transaction is more reliable

than methods that employ uncontrolled comparables with a lesser degree

of comparability, assuming data and assumptions of equal quality.

Methods relying on uncontrolled transactions with the highest degree of

comparability are preferred because, as the degree of comparability is

improved, the number of differences that could render the analysis

unreliable is reduced. Adjustments can and should be made for any

differences if the reliability of the analysis is improved by making

the adjustment. Under this approach, the comparable uncontrolled price

and comparable uncontrolled transaction methods ordinarily will provide

the most reliable measure of true taxable income when they can be

applied to closely comparable uncontrolled transactions, because such

analyses can be expected to achieve a higher degree of comparability

than other methods.

In determining which method provides the highest degree of

comparability, it is necessary to consider the comparability factors

discussed in Sec. 1.482-1(d) (Comparability). In addition, it is

necessary to consider further guidance on comparability set out under

each method, as certain differences have a greater effect on

comparability under some methods than under others. Finally, the

reliability of data and assumptions, discussed below, will affect the

ability to determine the degree of comparability between an

uncontrolled transaction and the controlled transaction.

Section 1.482-1(c)(2)(ii) discusses data and assumptions, which

constitute the second factor that must be considered in applying the

best method rule. This provision is divided into three components:

completeness and accuracy of data, reliability of assumptions, and

sensitivity of results to deficiencies in data and assumptions.

Completeness and accuracy of data defines the ability to identify

and quantify material differences. When data regarding a controlled or

uncontrolled transaction is relatively incomplete, it is more difficult

to determine if there are material differences between the

transactions. Thus, the fact that no material difference between an

uncontrolled and a controlled transaction has been identified does not

necessarily mean that the two transactions are highly comparable unless

the data on both transactions is sufficiently comprehensive that it is

possible to conclude that it is unlikely that any such differences

exist. In addition, the completeness and accuracy of data will affect

the ability to reliably estimate the effect of a material difference

once such a difference is identified. Thus, merely identifying and

adjusting for the effect of a material difference does not render the

controlled and uncontrolled transactions highly comparable unless the

data on both transactions is sufficiently complete and accurate that

the difference has a definite and reasonably ascertainable effect.

Another factor that affects the reliability of an analysis is the

reliability of the assumptions upon which the analysis is based. All

methods rely on assumptions. For instance, adjustments for differences

in payment terms reflect the assumption that such differences would

have an effect on price at arm's length. While this assumption is

relatively sound, other assumptions may be less reliable. In

particular, assumptions under the profit split method (such as the

assumption that the value of intangible assets is related to the cost

of development) may not always be as reliable as the assumptions under

methods that rely more closely on direct market indicators.

After assessing the completeness and accuracy of data and the

reliability of the assumptions upon which the analysis is based,

Sec. 1.482-1(c)(2)(ii)(C) provides that it is necessary to determine

the effect that any deficiencies in the data and assumptions have on

the reliability of the result. Some deficiencies will be more important

than others. Thus, a difference in risks borne might be expected to

affect all methods to some extent. Further, some deficiencies will have

a more adverse impact on some methods than on other methods, because

different methods rely more heavily on different types of

comparability. Thus, an inability to reliably allocate research and

development expenses would have a serious effect on the reliability of

a residual profit split under Sec. 1.482-6(c)(3), but would have little

effect on an analysis under the CUT method.

Finally, Sec. 1.482-1(c)(2)(iii) provides that in some cases it may

be relevant to consider whether the results obtained under a method are

consistent with the results obtained under another method. This

situation will arise when, after considering the comparability and the

quality of the data and assumptions under two different methods (or

under two different applications of the same method), it is not

possible to determine which of the competing analyses provides a more

reliable measure of an arm's length result. In such a case, it may be

relevant to compare the results with the result obtained under a third

method (or, given two applications of a single method, a third

application of that method).

Section 1.482-1(d) provides general guidance for determining

comparability. General guidance on comparability is provided in this

section of the regulations because the factors described are relevant

under all the methods. This provision is substantially similar to the

guidance provided in Sec. 1.482-1T(c) of the 1993 regulations. It

provides that in determining the degree of comparability between a

controlled and uncontrolled transaction, the functions, contractual

terms, risks, economic conditions, and property or services in the two

transactions must be compared.

Section 1.482-1(d)(2) provides that for two transactions to be

considered comparable, they need not be identical, but must be

sufficiently similar that the uncontrolled transaction provides a

reliable measure of an arm's length result. Further, if there are

material differences between the transactions, adjustments must be made

to account for such differences if the effect of the differences can be

ascertained with sufficient accuracy to improve the reliability of the

results. A ``material difference'' is defined as a difference that

would materially affect price or profit. Thus, adjustments for

differences that would have only a de minimis or minor effect on price

or profit are not required (although such adjustments will tend to

increase the reliability of the result). Further, the extent (i.e., the

number and magnitude) and reliability of any adjustments will affect

the reliability of the result. The number and magnitude of adjustments

affects reliability because as the number or magnitude of adjustments

increases, the potential for error also increases. Although the

standard of comparability under Sec. 1.482-1(d)(2) creates the

possibility that there may be a material difference between the

controlled and uncontrolled transactions for which an adjustment has

not been made, this only would occur if an adjustment was not possible

and no better analysis could be employed.

There are several differences between Sec. 1.482-1(d)(2) and its

counterpart in the 1993 regulations (Sec. 1.482-1T(c)(2)(i)). First, in

response to comments, the definition of comparability in the final

regulations contemplates the use of so-called ``inexact'' comparables

under all methods. The 1993 regulations only contemplated the use of

such analyses under the CPM. Of course, inexact comparables only will

be used if the best method rule indicates that these analyses provide

the most reliable measure of an arm's length result.

A second difference between the final and 1993 regulations is the

standard for making adjustments. The 1993 regulations provided that

adjustments ``may'' be made to account for material differences if such

differences have a ``definite and reasonably ascertainable effect'' on

prices or profits, and that if such differences can be reflected by

such adjustments, the result constitutes an arm's length result for the

controlled transaction. This language therefore seemed to preclude

adjustments when the effect of the difference did not have a definite

and reasonably ascertainable effect. Interpreted literally, this

standard could prevent adjustments that, although not perfectly

precise, nonetheless would improve the reliability of the analysis.

Accordingly, the final regulations provide that adjustments for

material differences should be made to the extent that they improve the

reliability of a result. In some cases it may be possible to make

adjustments that improve reliability even though the difference for

which the adjustment is made does not have a ``definite and reasonably

ascertainable'' effect on price or profit. Such adjustments

nevertheless should be made. The provision adds that the extent and

reliability of the adjustments will affect the reliability of the

result in relation to the reliability of applications of other methods.

A third important difference between the 1993 regulations and the

final regulations is the definition of the minimum level of

comparability. The 1993 regulations provided that an uncontrolled

comparable could be used to determine an arm's length result only if it

provided a ``reasonable and reliable benchmark.'' This phrase has been

replaced by the phrase ``a reliable measure'' of an arm's length

result. To some readers the word ``benchmark'' indicated that

unadjusted industry averages could serve as the basis for adjustments.

This concern was most evident with respect to the comments received

with respect to the CPM.

Section 1.482-1(d)(3) discusses the five factors that affect

comparability: functions, contractual terms, risks, economic conditions

and property or services. Of these factors, the discussions of

functions, economic conditions and property or services are

substantially similar to the discussions of these factors in the 1993

regulations. The discussions of contractual terms and risks differ in

some respects from the discussions of these factors in the 1993

regulations.

Contractual terms were covered in two provisions of the 1993

regulations: Sec. 1.482-1T (c)(3)(iii) and (d)(3)(ii). The final

regulations consolidate the discussion of these provisions into

Sec. 1.482-1(d)(3)(ii) and make some clarifying changes.

The discussion of risk in Sec. 1.482-1(d)(3)(iii), particularly as

it relates to identifying the party that bears a risk, is somewhat

different from its predecessor in the 1993 regulations (Sec. 1.482-

1T(c)(3)(ii)). In general, the determination of which party bears a

risk will be made in accordance with the provisions of Sec. 1.482-

1(d)(3)(ii)(B) (Identifying contractual terms). Thus, to the extent

that taxpayers allocate risks by contract and their conduct is

consistent with such contract, their allocation of risk will be

respected, unless the contract is executed after the impact of the risk

is known or knowable. In cases where the allocation of risk is not

clear from the parties' contractual arrangements, several factors may

be particularly relevant to determine which party bore the risk. These

factors include whether the parties' conduct is consistent over time;

which controlled taxpayer would ultimately bear the consequences of a

risk; and the extent to which each taxpayer controls any activities

that influence the outcome of a particular risk. The rules on

documentation of risks contained in the 1993 regulations has been

revised because some readers thought they implied that the district

director could arbitrarily allocate risks in the absence of express

documentation allocating the risk.

Section 1.482-1(d)(4) sets forth rules for certain special

circumstances affecting comparability. First, Sec. 1.482-1(d)(4)(i)

describes the extent to which market share strategies will be respected

in determining an arm's length result for a controlled transaction. As

under the 1993 regulations, these strategies may be recognized in

certain situations in which a company is attempting to gain entry into

a market or to increase market share. In such circumstances the amount

charged in the controlled transaction, or the expenses borne by a

controlled taxpayer, may for a short time differ from what normally

would be observed at arm's length. The reference in the 1993

regulations to using such strategies to meet competition in an existing

market has not been included in the final regulations because companies

in competitive markets are routinely faced with the problem of meeting

competition, which is reflected in a normal arm's length price.

A market share strategy will be respected only if certain

conditions are satisfied. These conditions are that the costs incurred

to implement the strategy are borne by the controlled taxpayer that

would derive the benefits from engaging in the strategy (and there is a

reasonable likelihood that the strategy will bear fruit), the strategy

is pursued for a reasonable period of time given the industry in

question, and the strategy and related matters are documented before

the strategy was implemented. The taxpayer must provide documentation

establishing that it has satisfied these conditions. These conditions

and the documentation requirement are similar to those imposed under

the 1993 regulations. In addition, a market share strategy ``will be

taken into account only if it can be shown that an uncontrolled

taxpayer engaged in a comparable strategy under comparable

circumstances for a comparable period of time. . . .'' This

requirement ensures that the strategy is consistent with the behavior

of parties operating at arm's length. It does not, however, require

that the taxpayer locate a comparable uncontrolled transaction that

would satisfy the standards of the CUP method in order to take

advantage of this rule. The critical component of this requirement is

that there be evidence that uncontrolled taxpayers engage in similar

behavior under comparable circumstances. A taxpayer could, for example,

satisfy this requirement by providing evidence of an uncontrolled

taxpayer in a different industry engaging in such a strategy, given

evidence that the circumstances otherwise were comparable.

Section 1.482-1(d)(4)(ii) addresses certain issues presented by

differences in geographic markets. This section generally conforms to

the analogous provision in the 1993 regulations by providing that while

it is permissible to derive uncontrolled comparables from geographic

markets that are different from the market in which the controlled

transaction occurred, adjustments must be made to the extent possible

to reflect the effect of any differences between the markets, and the

failure to accurately adjust for such differences will affect the

reliability of the analysis under the best method rule. In addition,

Sec. 1.482-1(d)(4)(ii)(C) provides that ``location savings'' from

operating in a low-cost jurisdiction must be allocated among controlled

taxpayers consistent with the allocation of such savings that would

occur between unrelated parties, taking into account the competitive

conditions in the low-cost market. As a result, some or all of the

location savings might inure to the benefit of the other party to the

controlled transaction.

Finally, Sec. 1.482-1(d)(4)(iii) describes certain transactions

that are not ordinarily accepted as comparables. Transactions not in

the ordinary course of business, and transactions arranged with a

principal purpose of establishing an arm's length result, ordinarily

will not constitute comparable transactions for purposes of section

482. This provision is generally consistent with Sec. 1.482-

1T(c)(4)(iii), except that the reference to ``isolated transactions''

has been deleted, as most transactions involving intangible property

may be viewed as isolated. In addition, the statement that transfers of

tangibles should be significant in number and amount in order to

constitute comparables has also been deleted. Transfers of some

property may be few in volume, but nonetheless be in the ordinary

course of business and provide a useful basis for determining an arm's

length result. Moreover, even large differences in volume are not per

se bars to the use of a potential comparable. Rather, such differences

should be taken into account as comparability factors under Sec. 1.482-

1(d)(3).

Section 1.482-1(e) describes the arm's length range. This provision

corresponds to Sec. 1.482-1T(d)(2)(i) of the 1993 regulations. As under

the 1993 regulations, the arm's length range is derived from two or

more uncontrolled transactions. Under the 1993 regulations the range

included all valid applications of a particular method. A modified rule

has been included in the final regulations to reflect the possible use

of inexact comparables under all of the methods. Given this relaxed

standard of comparability, it would be inappropriate to derive a range

from a mix of exact and inexact comparables, because to do so would

accord the same weight to results with potentially widely varying

degrees of reliability. Therefore, Sec. 1.482-1(e)(2)(i) provides that

the range is derived from two or more uncontrolled transactions ``of

similar comparability and reliability.''

Section 1.482-1(e)(2)(ii) expands upon this rule, providing that

uncontrolled comparables with significantly lower levels of

comparability and reliability than others will be disregarded. Thus, it

is necessary in every case in which more than one uncontrolled

transaction is available to compare the relative levels of

comparability and reliability of the uncontrolled transactions, and to

discard those uncontrolled transactions that do not have approximately

the same level of comparability and reliability as the most comparable

and reliable of the uncontrolled transactions.

Under Sec. 1.482-1(e)(2)(iii), the arm's length range will be

established in one of two ways, depending upon the extent to which

material differences between the uncontrolled comparables and the

controlled transaction can be identified, and the reliability of

adjustments made to account for such differences. First, under

Sec. 1.482-1(e)(2)(iii)(A), the arm's length range will consist of the

results of all the uncontrolled comparables (i.e., all the uncontrolled

comparables of similar comparability and reliability) if certain

requirements are met. These requirements are that every identified

material difference have a definite and reasonably ascertainable effect

on prices or profits; that appropriate adjustments for such differences

be made; and that the data be sufficiently complete that it is likely

that there are no unidentified material differences. Given equal

degrees of high comparability, it is impossible to conclude which of

the uncontrolled comparables provides a more reliable measure of an

arm's length result, and it is inappropriate to draw distinctions

between them by excluding some from the range. Therefore all the

results are included in the arm's length range.

Second, Sec. 1.482-1(e)(2)(iii)(B) provides that if the standards

of Sec. 1.482-1(e)(2)(iii)(A) are not met, then the reliability of the

analysis must be enhanced, if possible, by applying valid statistical

techniques to the uncontrolled comparables that are of similar

comparability and reliability. In this case the range generally

consists of the interquartile range, i.e., the 25th to the 75th

percentile of the results derived from the uncontrolled comparables, or

an equivalent range determined pursuant to other valid statistical

methods corresponding to a level of confidence equal to that provided

by the interquartile range. Finally, this section reemphasizes that all

adjustments for material differences that improve the reliability of

the result must be made to the extent possible. Section 1.482-

1(e)(2)(iii)(C) provides guidance for determining the interquartile

range.

The 1993 regulations contemplated the use of statistical techniques

to derive the arm's length range only under the CPM. The use of these

techniques has been extended to all methods under the final regulations

to reflect the fact that the standards of comparability have been

relaxed to permit the use of inexact comparables.

Unlike the uncontrolled comparables used in establishing the range

under Sec. 1.482-1(e)(2)(iii)(A), the comparables under Sec. 1.482-

1(e)(2)(iii)(B) have, or may have, material differences for which

adjustments have not been made. The justification for including all the

results in the arm's length range under Sec. 1.482-1(e)(2)(iii)(A) is

that all the uncontrolled comparables are of approximately equal

comparability, and it is inappropriate to draw distinctions between

them by excluding some from the range. This is not true, however, of

the comparables used to derive the arm's length range under Sec. 1.482-

1(e)(2)(iii)(B). Although these comparables will appear to be of equal

comparability, the presence of either unidentified material differences

or identified material differences that do not have a definite and

reasonably ascertainable effect, means that they are actually unlikely

to be equally comparable. Therefore, to include all the results in the

arm's length range under Sec. 1.482-1(e)(2)(iii)(B) would mean that

uncontrolled comparables of differing degrees of comparability and

reliability would be included in the range, violating the rule set

forth in Sec. 1.482-1(e)(2)(ii) (Selection of comparables). If results

of varying degrees of comparability and reliability were included in

the range, the analysis would be distorted, because results with

different degrees of comparability and reliability would be accorded

equal weight.

Since it is impossible to directly identify and quantify these

material differences, the regulations require that they be taken into

account indirectly through the use of a statistical range. Results that

differ widely from one another have significant, unaccounted for,

differences. Therefore when it is highly probable that the uncontrolled

comparables are not of roughly equal comparability (it being impossible

to identify all material differences), it is reasonable to assume that

the results diverging significantly from the norm are not comparable to

the controlled taxpayer.

If results fall outside the arm's length range, Sec. 1.482-1(e)(3)

provides that the district director may make adjustments so that the

taxpayer's result falls at any point within the range. When the

interquartile range is used, this point generally will be the median of

the results. In other cases, this point will normally be the mean of

the results.

Section 1.482-1(e)(4) states that the district director may

properly propose an adjustment based on a single comparable

uncontrolled price. However, if the taxpayer subsequently demonstrates

that its results are within a range established by additional equally

comparable transactions, no allocation will be made.

Section 1.482-1(f) sets forth several rules relating to the scope

of review under section 482 that correspond closely to rules provided

under Sec. 1.482-1T(d) of the 1993 regulations. This section provides

that a section 482 allocation may be made whenever the taxable income

of a controlled taxpayer differs from an arm's length amount. Further,

Sec. 1.482-1(f)(1)(i) provides that the discretion of the district

director to utilize section 482 is not limited to cases in which the

taxpayer intentionally distorted its taxable income. Section 1.482-

1(f)(1)(ii) provides that the district director may make an allocation

even if the ultimate income anticipated from a series of transactions

is not realized. Sections 1.482-1(f)(1) (iii) and (iv) relate to the

interaction between section 482 and the nonrecognition and consolidated

return provisions, respectively. They are consistent with guidance set

forth in the 1968 and 1993 regulations.

Section 1.482-1(f)(2) contains rules relating to the determination

of true taxable income. These rules are substantially similar to rules

set forth in Sec. 1.482-1T(d)(3) of the 1993 regulations. Section

1.482-1(f)(2)(i) provides that multiple transactions (generally within

the same product grouping) may be aggregated when they are so

interrelated that it is necessary to view them as a whole. Section

1.482-1(f)(2)(ii) provides that the district director ordinarily will

evaluate controlled transactions based on the structure of the actual

transaction, and will not treat the transaction as if it had been

structured differently. The district director may, however, consider

the alternatives that were available to the taxpayer in determining

whether the terms of the controlled transactions would be acceptable to

an uncontrolled taxpayer faced with similar alternatives. Adjustments

should be made in such cases to reflect material differences between

the alternative and the controlled transactions. As under the 1993

regulations, this authority to examine alternatives is limited to the

determination of an arm's length price, and does not permit the

district director to treat a controlled transaction as if it actually

had been structured differently.

Section 1.482-1(f)(2)(iii) provides that the results of controlled

transactions ordinarily will be compared with the results of

uncontrolled taxpayers derived from the same period as the controlled

transactions. Data from different years may be used, however, under

certain conditions. If data from other years is employed, such data

should be compared to the controlled taxpayer's results from the same

years. Data from multiple years also may be relevant for purposes of

certain enumerated provisions, including analysis of risk, market share

strategy, periodic adjustments, and the CPM. Section 1.482-

1(f)(2)(iii)(C) provides that results from other years may be examined

to determine if the same economic conditions that caused the taxpayer's

result also caused the uncontrolled taxpayer's result. For example, in

determining whether a loss from a controlled transaction is within an

arm's length range based upon losses realized by uncontrolled

taxpayers, it may be relevant to consider data from other taxable years

to determine whether the same conditions that caused the controlled

taxpayer's loss had a similar effect on the controlled taxpayer.

Section 1.482-1(f)(2)(iii)(D) provides that if the application of a

method is based on a multiple year analysis, the district director may

make an adjustment if the taxpayer's average result for the period is

outside the range of average results derived from uncontrolled

comparables for the same period. Comparison of multiple year averages

may provide a more accurate reflection of a taxpayer's transfer pricing

practices over a period than an analysis based on a single year and

reduces the effect of short-term variations that may be unrelated to

transfer pricing. The determination of whether the taxpayer is within

the arm's length range is based on a comparison of the taxpayer's

average result for the relevant years to the results of the

uncontrolled comparables over the same period, which indicates whether

the taxpayer had similar results over a similar period. An adjustment

ordinarily will be equal to the difference, if any, between the

taxpayer's result for the taxable year and the mid-point (generally the

median) of the uncontrolled comparables' results for the taxable year.

However, an adjustment will be made only to the extent that it would

move the controlled taxpayer's multiple year average closer to the

arm's length range for the multiple year period or to any point within

such range. Thus, for example, if a method is applied to a U.S.

taxpayer, and the taxpayer's average result for the multiple year

period is below the arm's length range of average results derived from

uncontrolled comparables for the same period, an adjustment may be made

that is equal to the difference between the controlled taxpayer's

result for the taxable year and the mid-point of the results of the

uncontrolled comparables for that year. However, the adjustment would

not be made to the extent that the adjustment would cause the taxpayer

to have an average result for the multiple year period that exceeds any

point within the arm's length range derived from the average results of

the uncontrolled comparables.

Section 1.482-1(f)(2)(iv) permits evaluation of product lines and

statistical groupings in a manner analogous to that permitted under the

1968 and 1993 regulations. Finally, Sec. 1.482-1(f)(2)(v) follows the

1993 regulations in providing that it is not necessary for the district

director to determine whether the method that a taxpayer employs to

determine the amounts charged in its controlled transactions correspond

to the method that the taxpayer might have used in uncontrolled

transactions. In other words, the focus of this evaluation is the

result achieved rather than the method employed in reaching that

result.

Section 1.482-1(g) provides procedural rules relating to collateral

adjustments. Such adjustments include correlative allocations,

conforming adjustments, and set-offs. Section 1.482-1(g)(2) provides

rules regarding correlative allocations that are generally similar to

the rules provided under the 1968 and 1993 regulations.

Section 1.482-1(g)(3) provides that appropriate adjustments must be

made to conform a taxpayer's accounts to reflect allocations under

section 482. Such adjustments may include the treatment of an allocated

amount as a dividend or a capital contribution. In other cases,

pursuant to applicable revenue procedures (see, e.g., Rev. Proc. 65-

17), amounts may be repaid without further income tax consequences.

Section 1.482-1(g)(4) provides rules relating to setoffs that are

similar to the rules provided under the 1993 regulations and the 1968

regulations. Several requirements are imposed on taxpayers that claim

setoffs. First, as in the 1993 regulations, the taxpayer must establish

that the transaction that is the basis of the set-off was not at arm's

length. Second, the taxpayer must document all adjustments resulting

from the proposed set-off. Finally, the taxpayer must notify the

district director of any claimed set-off within 30 days after the

earlier of the date of a letter by which the district director

transmits an examination report notifying the taxpayer of proposed

adjustments or the date of the issuance of the notice of deficiency.

This requirement corresponds to Sec. 1.482-1A(d)(3) of the 1968

regulations.

In addition to set-offs arising from transactions between the

controlled taxpayers that were the parties to the transaction giving

rise to the original allocation, the temporary regulations permitted

set-offs arising from transactions between one of these controlled

taxpayers and a third controlled taxpayer. In discussions with treaty

partners it proved impossible to reach an acceptable consensus on this

issue, and no such rule was included in the final regulations.

Therefore, as under the 1968 regulations and in accordance with the

practice of most treaty partners, set-offs are permitted only for

transactions between the same two controlled taxpayers that were

parties to the transactions giving rise to the original allocation.

The 1993 regulations contained a provision on compensating

adjustments that has not been included in the final regulations. The

provision in the 1993 regulations (Sec. 1.482-1T(e)(2)) imposed several

restrictions on the ability of taxpayers to adjust reported results to

reflect an arm's length result. This requirement was deleted because it

was not appropriate to impose such restrictions given the penalties

that could be imposed under section 6662(e) on taxpayers that fail to

make such adjustments in certain circumstances. The final regulations

therefore impose no restrictions on taxpayers' ability to report a

result on their original tax return that differs from the result

reflected in the taxpayer's books and records. However, as provided in

Sec. 1.482-1(a)(3), taxpayers may not use section 482 to decrease their

taxable income on an amended return.

Section 1.482-1(h) provides special rules relating to a small

taxpayer safe harbor, foreign legal restrictions and coordination with

section 936. Under Sec. 1.482-1(h)(1), the small taxpayer safe harbor

is reserved. The 1993 regulations contained a small taxpayer safe

harbor that has not been included in the final regulations. This

provision was never implemented because the IRS did not issue the

profit level indicators required to apply the provisions of the safe

harbor. There are three reasons why this provision was not included.

First, treaty partners had expressed concern that the safe harbor might

cause taxpayers to overreport their U.S. taxable income and underreport

their foreign taxable income. They requested that the safe harbor

provide that electing taxpayers be required to report an amount of

profit in the United States that was less than that expected under a

strict application of the arm's length standard. Such an approach was

not acceptable. Second, it would have been necessary to add a number of

anti-abuse provisions in order to eliminate the possibility of

inappropriate use of the provision by large taxpayers. Commenters had

already expressed concern that the existing restrictions were

excessively complex and burdensome given the level of sophistication of

its intended beneficiaries. The final concern was that both taxpayers

and the IRS might give undue weight to the published measures of

profitability in cases not governed by the safe harbor. It was not

possible to address these problems consistently with the overall

objective of alleviating the compliance burden for small taxpayers.

Moreover, the concern regarding the compliance burden on small

taxpayers has been addressed to some extent by the regulations under

section 6662(e), which provide that one of the factors to be taken into

account in determining whether a taxpayer reasonably applied a method

to determine its transfer prices is the taxpayer's experience and

knowledge. Comment is requested on alternative approaches to the small

taxpayer safe harbor that would not suffer from the deficiencies noted

above.

The rules on foreign legal restrictions were originally issued in

proposed form in the 1993 regulations. Section 1.482-1(h)(2) modifies

and finalizes that provision. It provides that a foreign legal

restriction will be taken into account to the extent that such

restriction affects the results of transactions at arm's length. If

there is no evidence that the restriction affected uncontrolled

taxpayers the restriction will be disregarded in determining an arm's

length result, and it will be taken into account only to the extent

provided in Secs. 1.482-1(h)(2) (iii) and (iv), relating to the

deferred income method of accounting. A foreign legal restriction is

generally defined under Sec. 1.482-1(h)(2)(ii) as a restriction that is

publicly promulgated and generally applicable, not imposed as part of a

commercial transaction between the taxpayer and the foreign government,

with respect to which the taxpayer has exhausted all practicable legal

remedies afforded under foreign law, expressly prevents the payment, in

any form, of an arm's length amount within the meaning of section 482,

and was not otherwise circumvented by the controlled taxpayers.

Section 1.482-1(h)(2)(iii) provides that if a provision meets the

definition of a foreign legal restriction and the taxpayer has elected

the deferred income method of accounting, any section 482 allocation

connected with the transaction will be deferrable until the restriction

is removed.

Section 1.482-1(h)(2)(iv) provides that if the requirements of

Sec. 1.482-1(h)(2)(iii) are satisfied, the amount subject to the

restriction will be treated as deferrable until payment or receipt of

the relevant item ceases to be prevented by the foreign legal

restriction. Deductions and credits incurred in open years and that are

chargeable against a deferred amount are subject to deferral under

Sec. 1.461-1(a)(4).

Section 1.482-1(h)(3) provides a coordination rule for section 936

that is identical to Sec. 1.482-1T(f)(3) of the 1993 regulations.

Section 1.482-1(i) defines ten terms that are employed in the

regulations. These are generally identical to their definitions under

the 1993 regulations, with the following exceptions. The definition of

trade or business under Sec. 1.482-1(i)(2) clarifies that employment

for compensation will constitute a separate trade or business from the

employing trade or business. The definition of ``controlled'' in

Sec. 1.482-1(i)(4) has been modified. The definition in the 1993

regulations was misinterpreted to provide that a presumption of control

arises only if income or deductions have been arbitrarily shifted ``as

a result of the actions of two or more taxpayers acting in concert or

with a common goal or purpose.'' This phrase was added to the 1993

regulations as an example of the type of control that could be

considered control for purposes of section 482. The definition has been

amended to make clear that this addition is only an example. The

definition of the term ``controlled taxpayer'' has been clarified to

include the taxpayer that owns or controls other taxpayers.

Section 1.482-1(j)(1) provides that these regulations generally are

effective for taxable years beginning 90 days after publication in the

Federal Register. Section 1.482-1(j)(2) provides that taxpayers may

elect to apply these regulations retroactively to all open years (in

which case they also must be applied to all subsequent years). Section

1.482-1(j)(3) provides that the last sentence of section 482 is

generally effective for taxable years beginning after December 31,

1986, and this sentence, prior to the effective date of the final

regulations, must be applied using any reasonable method not

inconsistent with the statute (including these regulations). Finally,

Sec. 1.482-1(j)(4) provides that the final regulations will not apply

to transfers made or licenses granted prior to November 17, 1985 (in

the case of a foreign transferee) or August 17, 1986 (in the case of

other transferees), unless the property was not in existence on the

relevant date.

Section 1.482-2

The regulations under section 1.482-2 have not been changed.

Section 1.482-2(d), providing that guidance with respect to transfers

of property is set forth in Secs. 1.482-3 through 1.482-6, is now part

of the final regulations.

Section 1.482-3

Section 1.482-3 provides rules for transfers of tangible property.

Six methods are provided: the CUP method; the resale price method; the

cost plus method; the CPM; the profit split method; and unspecified

methods. The method that will be applied in a particular case will be

selected in accordance with Sec. 1.482-1(c) (Best method rule).

Section 1.482-3(b) describes the CUP method. Consistent with the

best method rule, Sec. 1.482-3(b)(2)(ii) provides that the CUP method

generally provides the most direct and reliable measure of an arm's

length result if an uncontrolled transaction either has no differences

from the controlled transaction or there are only minor differences

that have a definite and reasonably ascertainable effect on price, and

appropriate adjustments are made for such differences. Further, unlike

the 1993 regulations, the CUP method potentially may be used when there

are more than minor differences between the controlled and uncontrolled

transactions, or when adjustments for minor differences cannot be made.

In such cases, the method may be employed, but its reliability for

purposes of the best method rule will be reduced.

In determining comparability under this method, product similarity

is the most important factor to consider. Indeed, Sec. 1.482-

3(b)(2)(ii) provides that if there are material product differences for

which reliable adjustments cannot be made, this method ordinarily will

not provide a reliable basis for determining an arm's length result.

Comparability also will be reduced if either the uncontrolled taxpayer

or the controlled taxpayer owns a trademark that is exploited in

connection with the sale of the product. Minor differences in

contractual terms and economic conditions also can have a material

effect on price, so comparability under this method also depends on

close similarity with respect to these factors.

Although all the comparability factors described in Sec. 1.482-1(d)

must be considered, Sec. 1.482-3(b)(2)(ii)(B) provides examples of

several factors that may be particularly relevant to the application of

the CUP method.

Section 1.482-3(b)(2)(iii) provides that the data and assumptions

used to apply the CUP method also will affect the reliability of the

result. This method assumes that a very similar transaction between

uncontrolled taxpayers is a reliable basis for determining the price

that would have been agreed between the controlled taxpayers had they

been dealing at arm's length. Given reasonably complete and accurate

data, this assumption is usually very reliable.

Section 1.482-3(b)(5) describes the use of indirect evidence

derived from public exchanges or quotation media to establish a

comparable uncontrolled price under this method. This provision had

been added in response to comments that in some industries in which

commodities are traded in large quantities, it is common for unrelated

parties to set prices based upon publicly available prices or

quotations. Since these quoted prices are not themselves transactional

prices, but may be averages based upon actual transactions, they do not

qualify as applications of the CUP method as that method is otherwise

described in the regulations. This section has been added to permit use

of such indicators in appropriate circumstances.

Such indicators may be employed only if the data is widely and

routinely used in the ordinary course of business in the industry to

establish prices in uncontrolled transactions, the data is used in the

same way by uncontrolled and controlled taxpayers, and adjustments are

made for differences that affect price. Further, such indicators may

not be employed under extraordinary market conditions.

Section 1.482-3(c) describes the resale price method. It is

generally similar to the resale price method provisions of the 1968 and

1993 regulations, although greater guidance has been provided on

comparability factors to consider in applying the method, and the

standards of comparability expressly permit use of ``inexact

comparables'' under this method.

The discussion of comparability considerations emphasizes that,

although all the factors described in Sec. 1.482-1(d)(3) must be

considered, this method is particularly dependent on similarity of

functions performed, risks borne, and contractual terms. Further,

although close product similarity will tend to improve the reliability

of the result, reliable application of the resale price method is less

dependent on product similarity than the CUP method. The reliability of

the analysis also would be reduced if the uncontrolled taxpayer sells

goods that are significantly more (or less) valuable than the goods in

the controlled transaction, or if either the uncontrolled taxpayer or

the controlled taxpayer owns a trademark that is exploited in

connection with the resale of the product. In addition, it may be

necessary to consider certain factors, such as management efficiency

and differences in business experience, that would normally have little

effect on comparability under the CUP method.

The final regulations do not include the statement that in the

absence of comparable transactions, prevailing gross profit margins in

the general industry may be appropriate. Although this statement was

included in the 1968 and 1993 regulations, such a measure of an arm's

length result would be contrary to the rule under Sec. 1.482-1(d)(2)

that unadjusted industry average returns cannot independently establish

an arm's length result.

Section 1.482-3(d) describes the cost plus method. It is generally

similar to the cost plus method provisions of the 1968 and 1993

regulations, although additional guidance has been provided on

comparability factors to consider in applying the method, the rules for

computing the arm's length price and appropriate gross profit have been

clarified, and the standards of comparability expressly permit use of

``inexact comparables'' under this method.

As under the resale price method, the discussion of comparability

considerations emphasizes that, although all the factors described in

Sec. 1.482-1(d)(3) must be considered, this method is particularly

dependent on similarity of functions performed, risks borne, and

contractual terms. Further, although close product similarity will tend

to improve the reliability of the result, reliable application of the

cost plus method is less dependent on product similarity than the CUP

method. As under the resale price method, it may be necessary to

consider certain factors, such as management efficiency and differences

in business experience, that would normally have little effect on

comparability under the CUP method.

Section 1.482-3(d)(3)(iii)(B) emphasizes that in computing the

gross profit markup, items such as inventory and cost allocation must

be accounted for consistently.

Section 1.482-3(e) provides that in addition to the methods

specifically enumerated in Sec. 1.482-3, unspecified methods may be

employed. This provision differs from its counterpart in the 1993

regulations in two significant respects. First, in response to

comments, the procedural requirements that the 1993 regulations imposed

in connection with the use of an unspecified method by the taxpayer

have been deleted.

Second, guidance has been provided on considerations that should be

taken into account in applying an unspecified method. Such methods

should reflect the principle underlying the arm's length standard that

uncontrolled taxpayers compare the terms of a transaction to their

realistic alternatives to the transaction. Therefore, an unspecified

method should provide information on the prices or profits that the

controlled taxpayer could have realized by choosing a realistic

alternative to the controlled transaction. This guidance has been

included because it is a principle that is consistent with all methods

that apply the arm's length standard. For example, the CUP method

identifies an alternative price at which the controlled taxpayer could

have completed the controlled transaction. An example of the

application of this principle is provided in which a bona fide offer is

used to establish an arm's length price. Unspecified methods are not,

however, limited to examination of potential transactions that did not

occur. They should, in general, be based on actual transactions and

other indicia derived from actual or potential market transactions.

Section 1.482-3(f) provides rules coordinating the application of

the tangible property rules with the rules governing transfers of

intangible property. This provision provides more guidance than its

predecessor in the 1993 regulations. It provides that in most cases the

transfer of tangible property with a so-called ``embedded intangible''

will not be considered a transfer of the intangible if the purchaser

does not acquire the right to exploit the intangible other than in

connection with the resale of the tangible property. This provision

responds to commenters who expressed concern that sales of branded

products to controlled taxpayers for resale would routinely have to be

evaluated under the provisions of both Secs. 1.482-3 and 1.482-4. While

in such a case the transaction may be evaluated under Sec. 1.482-3, the

presence of the intangible will affect the analysis of comparability,

because the value of the product may be increased by the presence of an

embedded intangible.

Finally, when a purchaser of a tangible product acquires the right

to commercially exploit an embedded intangible, it may be necessary to

apply Sec. 1.482-3 to determine the arm's length consideration for the

tangible property transferred and Sec. 1.482-4 to determine the arm's

length consideration for the embedded intangible. An example of this

type of transaction could include the transfer of a machine

incorporating a valuable manufacturing process that the purchaser will

exploit in connection with the operation of the machine.

Section 1.482-4

Section 1.482-4 provides rules with respect to the transfer of

intangible property. Four methods are provided: the comparable

uncontrolled transaction (CUT) method, the CPM, the profit split

method, and unspecified methods. The method that will be applied in a

particular case will be selected in accordance with Sec. 1.482-1(c)

(Best method rule).

Section 1.482-4(b) provides a definition of intangible property

that is similar to that provided in the 1968 regulations. It differs

from the 1993 regulations in that the requirement that the property be

``commercially transferrable'' has been deleted. This language was not

included in the definition because it was superfluous: if the property

was not commercially transferrable, then it could not have been

transferred in a controlled transaction. In addition, the reference to

``other similar items'' under Sec. 1.482-4(b)(6) has been clarified to

refer to items that derive their value from intellectual content or

other intangible properties rather than physical attributes.

Section 1.482-4(c) describes the CUT method. This method is similar

but not identical to the CUT method under the 1993 regulations. The CUT

method determines an arm's length royalty for an intangible by

reference to uncontrolled transfers of comparable intangible property

under comparable circumstances. An important comparability factor under

this method is the profit potential of the intangibles in the

controlled and uncontrolled transactions. In response to comments, the

requirement that the profit potential of the intangibles transferred in

the controlled and uncontrolled transactions be ``substantially the

same'' has been relaxed to permit more frequent use of this method, as

long as it provides the most reliable measure of an arm's length result

under the best method rule. In addition, as under all methods, the

discussion of the comparability factors under this method has been

expanded.

As under the 1993 regulations, Sec. 1.482-4(c)(2)(ii) provides,

consistent with the best method rule, that the CUT method generally

provides the most direct and reliable measure of an arm's length result

if the same intangible is transferred in the controlled and

uncontrolled transactions, and there are, at most, only minor

differences between the uncontrolled and the controlled transaction,

these differences have a definite and reasonably ascertainable effect

on price, and appropriate adjustments are made for such differences.

The CUT method also may provide the most reliable measure of an arm's

length result in other cases, as determined under the best method rule

in Sec. 1.482-1(c).

Section 1.482-4(c)(2)(ii) emphasizes that, although all the factors

described in Sec. 1.482-1(d)(3) must be considered, this method is

particularly dependent on similarity in terms of contractual

arrangements and economic conditions. Further, this method cannot be

applied unless the intangible property involved in the controlled and

uncontrolled transactions is comparable within the meaning of

Sec. 1.482-4(c)(2)(iii)(B)(1).

Section 1.482-4(c)(2)(iii)(B)(1) provides that two requirements

must be satisfied in order for the intangible property involved in two

transactions to be comparable. First, as under the 1993 regulations,

the intangibles must be used in connection with similar products or

processes within the same general industry or market. Second, the

intangibles must have similar profit potential. As indicated, this

requirement is not as strict as the profit potential requirement under

the 1993 regulations. Additional guidance is provided concerning the

analysis applied in determining whether the profit potential of two

intangibles is similar within the meaning of this provision.

In order to conclude that the profit potential of two intangibles

is similar, it is necessary to have an acceptably reliable measure of

the profit potential of the two intangibles. Profit potential is most

reliably measured by direct calculations, based on reliable

projections, of the net present value of the benefits to be realized

through use of the intangible. While this information frequently will

be available with respect to the controlled transaction, it normally

will not be available with respect to an uncontrolled transaction

unless one of the controlled taxpayers was a party to it. In

recognition of this difficulty, the regulations provide that in certain

cases it may be acceptable to refer to evidence other than projections

to compare profit potential. Such indirect comparisons of profit

potential will be most useful in cases where it is not possible to

directly calculate the profit potential of the intangibles in either

the controlled or uncontrolled transaction. An example of such a case

could include a transfer of an intangible that relates to a component

of an asset consisting of many components (such as an airplane or

automobile). In such a case it would be difficult to reliably calculate

the net present value of the profit attributable to the intangible that

was transferred in the controlled transaction, because the profit

attributable to the intangible will be difficult to isolate from the

overall profit attributable to the final asset.

As the profit potential increases, the importance of reliably

measuring the profit potential also increases, because the effects of

errors may increase as the overall profitability of the intangible

increases. The reliability of indirect comparisons of profit potential

therefore decreases as the profit potential in the controlled

transaction increases. Consequently, given an indirect measure of

profit potential, it might be concluded that the profit potential of

the intangible property involved in the uncontrolled transaction might

be similar to that of the controlled transaction within the meaning of

this provision when the overall profitability of the intangibles is

relatively small, but the profit potential might not be similar under

the same circumstances if the overall profitability was much greater.

The ultimate determination will depend on the facts and circumstances

of each case.

Finally, Sec. 1.482-4(c)(2)(iii)(B)(2) provides that to apply the

CUT method the circumstances involved in the controlled and

uncontrolled transactions must be similar. This provision is

substantially the same as its counterpart in the 1993 regulations.

Section 1.482-4(d) provides a rule for the use of unspecified

methods that is analogous to the rule on unspecified methods provided

for tangible property under Sec. 1.482-3(e). To the extent that a

method relies on internal data rather than uncontrolled comparables,

its reliability will be reduced. Reliability also will be affected by

the reliability of the data and assumptions used to apply the method,

including any projections.

Section 1.482-4(e) provides a rule for coordination with the

tangible property rules that is analogous to the rule provided under

Sec. 1.482-3(f).

Section 1.482-4(f) provides special rules for transfers of

intangible property. Section 1.482-4(f)(1) provides that when a

controlled taxpayer pays nominal or no consideration for the right to

exploit an intangible, and the transferor retains a substantial

interest in the intangible, the arm's length consideration shall be in

the form of a royalty, unless a different form is more appropriate.

Section 1.482-4(f)(2) provides that if an intangible is transferred

for a period in excess of one year, the consideration charged is

generally subject to an annual adjustment to ensure that it is

commensurate with the income attributable to the intangible. This

provision is required by the 1986 amendment to section 482.

The 1993 regulations contained two exceptions to this rule. The

final regulations retain these exceptions and add three additional

exceptions in response to comments. First, Sec. 1.482-4(f)(2)(ii)(A)

provides that no periodic adjustments will be made if the consideration

for the transfer of an intangible is determined to be an arm's length

amount under the CUT method, and if the uncontrolled transaction that

serves as the basis for the application of the CUT method involved the

transfer of the same intangible under substantially the same

circumstances as those of the controlled transaction. Thus, for

example, the consideration for the transfer of an intangible to a

controlled taxpayer in one country could be determined to be arm's

length based on the transfer of the same intangible to an uncontrolled

taxpayer in another country in which the relevant economic conditions

were substantially similar to those in the first country. In such case

no periodic adjustment would be made if the two transactions occurred

under substantially similar circumstances.

Section 1.482-4(f)(2)(ii)(B) provides an exception, based on the

CUT method, that is similar to the exception provided in Sec. 1.482-

4T(e)(2)(ii)(A) of the 1993 regulations. Section 1.482-4(f)(2)(ii)(C)

provides an exception based on other methods that is similar to the

exception provided in Sec. 1.482-4T(e)(2)(ii)(B) of the 1993

regulations. Both of these exceptions are substantially identical to

their counterparts in the 1993 regulations with one modification. The

rule requiring that the taxpayer's actual profits attributable to the

intangible must be no less than 80 percent and no greater than 120

percent of the projected profits has been liberalized. In the 1993

regulations, the profits subject to this comparison were only the

projected and actual profits for all open years. Under the final

regulations, this comparison applies to all past years, which in many

cases will be a longer period. By enlarging the pool of data that is

taken into account for purposes of this comparison, it generally will

be less likely that the taxpayer's actual profits will fall outside the

band of projected profits based solely on timing variances, and

therefore fewer periodic adjustments will be permitted under these

provisions.

Section 1.482-4(f)(2)(ii)(D) provides an additional exception from

periodic adjustments for extraordinary events. It provides that no

periodic adjustments will be made if the aggregate actual profits fall

outside the permissible band of projected profits, but this variation

from the projected results was due to extraordinary events that could

not reasonably have been anticipated (such as natural or man-made

disaster but does not include more routine events such as the failure

of a market to develop as anticipated), and all the other requirements

of either Sec. 1.482-4(f)(2)(ii) (B) or (C) are satisfied.

Finally, Sec. 1.482-4(f)(2)(ii)(E) provides that if the

requirements of either Sec. 1.482-4(f)(2)(ii) (B) or (C) are satisfied

for the five-year period beginning with the year in which substantial

periodic consideration is first paid, no periodic adjustments will be

made.

Section 1.482-4(f)(3) provides rules regarding the ownership of

intangible property. These rules are required to identify the

controlled taxpayer that should recognize the income attributable to

intangible property. The 1993 regulations provided that, for purposes

of section 482, intangible property generally would be treated as owned

by the controlled taxpayer that bore the greatest share of the costs of

development. This rule was criticized by many commenters, principally

because it disregarded legal ownership. The commenters asserted that

disregarding legal ownership could be inconsistent with the arm's

length standard. For instance, a controlled taxpayer that was treated

as the owner of an intangible for section 482 purposes might not be the

legal owner. At arm's length, the legal owner could transfer the rights

to the intangible to another person irrespective of the developer's

contribution to the development of the intangible. On the other hand,

it would be unlikely that at arm's length an unrelated party would

incur substantial costs adding value to an intangible that was owned by

an unrelated party, unless there was some assurance that the party that

incurred the expenses would receive the opportunity to reap the benefit

attributable to the expenses.

The final regulations recognize these criticisms and adopt a

modified approach to the identification of the owner of an intangible

that is more consistent with legal ownership. Under Sec. 1.482-

4(f)(3)(ii)(A), the legal owner of the right to exploit an intangible

will be considered the owner for purposes of section 482. Legal

ownership does not refer solely to the registered holder of an

intangible: ownership rights may be transferred by explicit or implicit

agreement, and more than one party may be considered a legal owner of

rights in the same intangible. For example, a license agreement would

grant the licensee a set of rights in the intangible for the duration

of the agreement, while the licensor would retain the residual rights

to the intangible after the expiration of the agreement.

Under Sec. 1.482-4(f)(2)(ii)(B), ownership of intangible property

that is not legally protected will be determined in a manner similar to

that under the 1993 regulations, i.e., the owner generally will be the

person that bore the greatest share of the costs of development.

Section 1.482-4(f)(2)(iii) provides that allocations may be made

with respect to assistance provided to the owner by other parties that

assisted in the development of the intangible. Assistance does not

include expenditures of a routine nature that an unrelated party

dealing at arm's length would be expected to incur under similar

circumstances. For instance, even in the absence of a license agreement

transferring the right to exploit a trademark to an unrelated

distributor, a distributor may be expected to incur a certain amount of

advertising and other marketing expenses that could increase the value

of the trademark. If an uncontrolled taxpayer would incur such expenses

without express or implicit reimbursement by the owner of the

intangible, then no allocation with respect to similar levels of

expenses would be made under section 482 in the case of a distributor

that is a member of the controlled group to which the legal owner of

the trademark belongs. On the other hand, an allocation could be made

if the expenses were greater than those that an unrelated party would

have incurred without some form of compensation.

Section 1.482-4(f)(4) provides that the arm's length consideration

for the transfer of an intangible is not limited either by prevailing

industry average royalty rates or the consideration paid in

uncontrolled transactions that are not comparable to the controlled

transaction.

Section 1.482-4(f)(5) addresses lump sum payments. This issue was

reserved in the 1993 regulations. The final regulations provide that

lump sum payments are potentially subject to periodic adjustments to

the same extent as license agreements providing for periodic royalty

payments. For purposes of determining if the lump sum payment satisfies

the arm's length standard and if periodic adjustments may be made, the

lump sum must be treated as an advance payment of a stream of royalties

over the life of the agreement. This ``equivalent royalty amount''

serves as the basis for determining if the consideration is arm's

length. In addition, if a periodic adjustment is made pursuant to the

provisions of Sec. 1.482- 4(f)(2), the royalty that was deemed to have

been prepaid for the taxable year in question will be set off against

the arm's length royalty determined for such year, and the difference

will be treated as an additional payment in the year of the allocation

that is of the same character as the initial lump sum payment.

Section 1.482-5

Section 1.482-5 describes the comparable profits method. It is

similar to Sec. 1.482-5T of the 1993 regulations. The CPM may be used

to determine the arm's length consideration for tangible and intangible

property. The CPM relies on the general principle that similarly

situated taxpayers will tend to earn similar returns over a reasonable

period of time. The CPM determines the arm's length consideration for a

controlled transaction by referring to objective measures of operating

profit (profit level indicators) derived from uncontrolled taxpayers

that engage in similar activities with other uncontrolled taxpayers

under similar circumstances.

Many commenters were concerned that the CPM could be applied in a

manner that would be inconsistent with the arm's length standard. This

concern was attributable to the fact that operating profit is normally

affected by more factors than gross profit or price, which are the

measures employed under the other methods provided in Secs. 1.482-3 and

1.482-4. The commenters were concerned that the CPM would be applied

without taking these additional differences into account, and as a

result the comparability achieved under the CPM would be weaker than

the comparability required under other methods. The commenters believed

that in such cases an analysis under the CPM would be suspect. Further,

some commenters concluded that it was permissible under the 1993

regulations to apply the CPM without making adjustments for observed

material differences, which led them to believe that the IRS would

routinely apply the CPM in a manner that did not provide a reasonably

reliable measure of an arm's length result. This concern was heightened

by language in Sec. 1.482-5T(a) of the 1993 regulations stating that

the CPM would ordinarily provide an accurate measure of an arm's length

result unless the tested party owned certain types of intangible

property. Some commenters misinterpreted this language as indicating

that the CPM was preferred to the results obtained under other methods,

irrespective of the relative levels of comparability obtained under the

potentially applicable methods and without regard to the operation of

the best method rule.

The final regulations make it clear that the CPM is subject to the

same considerations as any other method. First, the language providing

that the CPM ``ordinarily will provide an accurate measure of an arm's

length result'' has been deleted.

Second, and more importantly, the final regulations contain a much

more extensive discussion of comparability considerations under the CPM

than did the 1993 regulations. While it may be permissible to apply the

CPM when there is (or may be) a material difference but it is not

possible to make a reliable adjustment for such difference, application

of the CPM in such a case only would be permissible if the other

methods were less reliable than the CPM under the facts and

circumstances.

Given adequate data, methods that determine an arm's length price

(e.g., the CUP method) or gross margin (e.g., the resale price method)

generally achieve a higher degree of comparability than the CPM.

Because the degree of comparability, including the extent and

reliability of adjustments, determines the relative reliability of the

result under the best method rule, the results of these methods will be

selected unless the data necessary to apply them is relatively

incomplete or unreliable. In this regard the CPM generally would be

considered a method of last resort.

This greater emphasis on comparability under the CPM is also

reflected in the treatment of ``valuable non-routine intangibles.'' The

1993 regulations indicated that the CPM ordinarily would not provide an

accurate measure of an arm's length result if the tested party owned

certain highly valuable intangibles because it was unlikely that it

would be possible to locate uncontrolled taxpayers that possessed

similar intangibles in connection with activities similar to those

performed by the controlled taxpayer. In such a case the CPM could

understate the income attributable to the assets of the controlled

taxpayer. As a result of this concern, the 1993 regulations generally

provided that the CPM should not be applied if the tested party owned

``valuable non-routine intangibles'' that it developed or that it

acquired from third parties.

The final regulations have taken a different approach to the

problem of valuable non-routine intangibles. Due to the difficulty in

adequately defining the term and the fact that it would be

inappropriate to deny use of the CPM if a comparable uncontrolled

taxpayer could be located that owned intangibles similar to those owned

by the controlled taxpayer, the restriction contained in the 1993

regulations has been eliminated. In its place intangible property is

expressly mentioned as a factor to consider in determining if an

uncontrolled taxpayer is comparable to the controlled taxpayer. In most

cases in which the controlled taxpayer owns intangible property of the

type described in the 1993 regulations it will not be possible to

locate an uncontrolled comparable that owns similarly valuable

intangible property. The final regulations recognize, however, the

possibility that a comparable uncontrolled taxpayer with such

intangibles might be found; therefore they do not rule out the

possibility of the CPM being applied under such facts.

Section 1.482-5(b)(4) describes the profit level indicators that

are used to evaluate operating profit. They include two types of

measures: the rate of return on capital employed and financial ratios.

In addition, Sec. 1.482-5(b)(4)(iii) provides that other profit level

indicators not specifically described in the regulations may be

employed, provided that they provide the most reliable measure of an

arm's length result within the meaning of the best method rule. Under

this provision, any measure of profit based on objective measures of

profitability derived from uncontrolled taxpayers that engage in

similar business activities under similar circumstances could be

employed. Accordingly, profit level indicators based solely on a

controlled taxpayer's internal data would not be included under this

provision, as they are not objective measures of profitability derived

from transactions between uncontrolled taxpayers.

In determining an arm's length result under the CPM, the taxpayer's

average reported operating profit for the year under review and the

preceding two taxable years ordinarily will be compared to the average

result of the uncontrolled comparables for the same period. Comparison

of multiple year averages may provide a more accurate reflection of a

taxpayer's transfer pricing practices over a period than an analysis

based on a single year and reduces the effect of short-term variations

in operating profit that may be unrelated to transfer pricing. If the

taxpayer's average reported operating profit for this period falls

outside the range of arm's length results determined pursuant to

Sec. 1.482-1(e)(2) (Determination of arm's length range), the district

director may make an adjustment. In most cases, the adjustment will be

made to the median of the uncontrolled comparables results for the

taxable year. Adjustments under section 482 for prior years are taken

into account in determining the tested party's average reported

operating profit for a particular year if a final determination has

been made with respect to such adjustments.

In line with the greater emphasis on comparability under the final

regulations, Sec. 1.482-5(c) contains a discussion of comparability

factors that is substantially more comprehensive than that contained in

the 1993 regulations. Section 1.482-5(c)(2) describes a number of

comparability factors that must be taken into account under the CPM. In

particular, Sec. 1.482-5(c)(2)(ii) provides that, while all the

comparability factors described in Sec. 1.482-1(d)(3) must be

considered, comparability under the CPM is particularly dependent on

resources employed and risks assumed. Further, since resources and

risks are directly related to functions, functional comparability,

while somewhat less important than under the resale price or cost plus

methods, also is an important consideration under the CPM.

Section 1.482-5(c)(2)(iii) provides that product comparability is

not as important a consideration under the CPM as it is under the cost

plus or resale price methods. Conversely, comparability under the CPM

may be adversely affected by other factors that have little effect on

comparability under the CUP method, such as management efficiency.

Determining whether such differences exist may be difficult in some

cases. As with any potential difference, the regulations provide that

objective evidence, such as long-term sales or executive compensation

trends, is required in order to ascertain whether such differences

exist.

Section 1.482-5(c)(2)(iv) provides that adjustments must be made

for all material differences to the extent that such adjustments

improve the reliability of the analysis. In a departure from the 1993

regulations, the final regulations provide that differences in non-

interest bearing liabilities (such as accounts payable) that would

materially affect operating profit generally should be reflected by

adjustments to operating profit to reflect an imputed interest charge

on each party's liability. Such differences are more appropriately

reflected by adjustments to operating profits than by adjustments to

operating assets.

Section 1.482-5(c)(3)(ii) requires that all items that have a

material affect on the profit level indicators be accounted for

consistently, and if necessary to obtain this consistency, adjustments

must be made. An additional factor that may affect reliability under

the CPM is the ability to allocate costs and other items between the

relevant business activity and the other activities of the controlled

taxpayer or the uncontrolled taxpayers.

The definitions of several items under Sec. 1.482-5(d) are

substantially similar to the definitions provided under the 1993

regulations.

Section 1.482-6

Section 1.482-6 describes profit split methods. This provision,

which was in proposed form under the 1993 regulations, has been

finalized. Like the CPM, profit split methods may be applied to

controlled transactions involving tangible or intangible property. The

basic approach of a profit split method is to estimate an arm's length

return by comparing the relative economic contributions that the

parties make to the success of a venture, and dividing the returns from

that venture between them on the basis of the relative value of such

contributions. Two profit split methods are provided: the comparable

profit split and the residual profit split. In addition to these two

methods, the 1993 regulations proposed two other profit split methods:

the capital employed allocation rule and other profit splits. These

methods have not been included in the final regulations for the reasons

set forth below.

Under the 1993 regulations taxpayers were required to satisfy a

number of requirements in order to employ a profit split method. Since

a profit split method either wholly or in part relies on internal data

rather than data derived from uncontrolled taxpayers, other methods

ordinarily will provide more reliable measures of an arm's length

result, and these restrictions were imposed in order to ensure that the

profit split method was applied only in cases where it was likely to be

the most reliable measure of an arm's length result. As noted

previously, these restrictions were both procedural and substantive.

Many commenters objected to these restrictions. They observed that,

contrary to the best method rule, these restrictions could prevent

taxpayers from employing a profit split method in cases where the

method would be likely to provide the most accurate measure of an arm's

length result. In response to these comments, and in accordance with

the greater flexibility associated with the increased reliance on the

best method rule, the final regulations have removed these

restrictions. In particular, for reasons similar to those discussed

above under the CPM, the requirements that a profit split method may be

applied only if both controlled taxpayers own valuable non-routine

intangible property and that the intangibles contribute significantly

to the combined operating profit derived from the relevant business

activity, have been deleted. Further, the requirement that the taxpayer

must make a binding election to apply a profit split method also has

been deleted. The concerns that these and other restrictions were

intended to address (the possibility that a profit split method would

be used when it did not provide the most reliable measure of an arm's

length result) are addressed by the more prominent role played by the

best method rule in selecting a method.

Section 1.482-6(c)(2) describes the comparable profit split method.

It is substantially similar to the comparable profit split rule set

forth under the 1993 regulations. As with other methods described in

the final regulations, the discussion of comparability factors under

this method is more comprehensive than under the 1993 regulations. In

addition, Sec. 1.482-6(c)(2)(ii)(C) identifies several reliability

considerations that are particularly pronounced under this method.

These are reliability of cost and income allocation to the relevant

business activity and accounting consistency. Further, Sec. 1.482-

6(c)(2)(ii)(D) provides that if the data and assumptions with respect

to one of the controlled taxpayers are significantly more reliable than

for the other, a method relying solely on an analysis of the first

controlled taxpayer may be more reliable than the profit split method.

Section 1.482-6(c)(3) describes the residual profit split. Like the

version of this method described in the 1993 regulations, the residual

profit split determines an arm's length consideration in a two-step

process. First, using other methods such as the CPM, market returns for

routine functions are estimated and allocated to the parties that

performed them. The remaining, residual amount then is allocated

between the parties on the assumption that this residual is

attributable to intangible property contributed to the activity by the

controlled taxpayers. Based on this assumption, the residual is divided

based on the estimate of the relative value of the parties'

contributions of such property. Since fair market value of the

intangible property usually will not be readily ascertainable, the

regulations permit use of other measures of the relative values of

intangible property, including capitalized intangible development

expenses.

The final regulations contain an extensive discussion of

comparability and reliability considerations under this method. The

comparability considerations relevant to the first step of the

allocation are analogous to the considerations relevant under the

method employed to determine this portion of the allocation (such as

the CPM). Since the second step ordinarily will not be based on a

market benchmark, the reliability of this method will tend to be

reduced for purposes of the best method rule as the amount of the

residual profit allocated pursuant to the second step increases.

Under the best method rule, methods that determine an arm's length

result based on the results of transactions between uncontrolled

taxpayers are generally considered to be more reliable than methods

(such as the residual profit split) that only rely on such transactions

in part. Therefore, the results of the methods based solely on results

of transactions between uncontrolled taxpayers will be selected under

the best method rule unless the data necessary to apply them is

relatively incomplete or unreliable. In this regard the residual profit

split generally would be considered a method of last resort.

Further, Sec. 1.482-6(c)(3)(ii)(C) identifies several other factors

that may reduce the reliability of this method. In addition to

allocation of costs, income and assets, and accounting consistency,

another factor to take into account under the residual profit split is

the reliability of the estimate of the value of intangible property.

The reliability of this method could be particularly adversely affected

if capitalized costs of development are used to estimate the value of

intangible property because such costs may bear no relation to market

value, calculation of such costs may require allocation of indirect

expenses between the relevant business activity and the controlled

taxpayer's other lines of business, and capitalizing costs requires

assumptions regarding the useful life of intangible property.

Finally, Sec. 1.482-6(c)(3)(ii)(D) provides that the analysis of

both parties to the controlled transaction under this method may, to

some extent, mitigate the reliability concerns attributable to the use

of internal data in allocating the residual profit. However, as under

the comparable profit split, other methods that analyze only one of the

parties to the controlled transaction may be more reliable if the data

and assumptions regarding one of the parties is more reliable than the

data and assumptions regarding the other party.

As indicated previously, the final regulations do not provide for

the use of the capital employed allocation rule or other profit splits.

The capital employed allocation rule generally could be applied only

when the controlled taxpayers were subject to approximately equal

levels of risk with respect to the relevant business activity. This

requirement was imposed because the method allocated the same rate of

return to all the assets employed in the relevant business activity.

This result generally would be encountered under arm's length

conditions only if the parties shared all profits in proportion to

their risks. With one exception it has not been possible to describe a

case in which it would be possible to conclude with certainty that two

or more controlled taxpayers face equal levels of risk. The one case

where it undoubtedly is true that the parties face equal levels of risk

is where the parties agree ex ante to share costs and benefits

proportionately. Since, absent a joint venture, such a scenario is

encountered rarely, if ever, under arm's length conditions, and it

describes a situation contemplated under the cost sharing regulations,

the capital employed allocation rule has been deleted from the profit

split provisions of the final regulations. Its possible role as a

variant of the cost sharing provisions will be examined in connection

with the revision of those provisions. Comments are requested as to its

potential use as a variant of the cost sharing method.

Finally, the use of other profit split methods is not discussed

under Sec. 1.482-6 because it was unnecessary to provide an express

rule for use of other profit splits given the liberalized rules for use

of unspecified methods under Secs. 1.482-3 and 1.482-4. Accordingly, a

profit split method other than the comparable profit split method and

the residual profit split will be considered to be an unspecified

method.

Section 1.482-7, relating to cost sharing, is not being finalized

with these regulations. The temporary regulations, which incorporate

the text of the 1968 regulations, continue to apply. However, final

regulations based on the 1992 proposed regulations are anticipated in

the near future.

Finally, Sec. 1.482-8 provides a number of examples illustrating

the application of the best method rule under specific fact patterns.

Like all the examples in these regulations, the examples under

Sec. 1.482-8 are provided solely for purposes of illustrating the

principles contained in the text of the regulations, and are not

themselves statements of principles not contained in the text. The

conclusions reached in these examples are based on the assumed

simplified facts of the examples, and should not be read as general

conclusions.

Special Analyses

It has been determined that this Treasury decision is not a

significant regulatory action as defined in EO 12866. Therefore, a

regulatory assessment is not required. It also has been determined that

section 553(b) of the Administrative Procedure Act (5 U.S.C. chapter 5)

and the Regulatory Flexibility Act (5 U.S.C. chapter 6) do not apply to

these regulations, and, therefore, a Regulatory Flexibility Analysis is

not required. Pursuant to section 7805(f) of the Internal Revenue Code,

the notice of proposed rulemaking preceding these regulations was

submitted to the Small Business Administration for comment on its

impact on small business.

Drafting Information

The principal author of these regulations is Sim Seo, Office of

Associate Chief Counsel (International). However, other personnel from

the IRS and Treasury Department participated in their development.

List of Subjects

26 CFR Part 1

Income taxes, Reporting and recordkeeping requirements.

26 CFR Part 602

Reporting and recordkeeping requirements.

Adoption of Amendments to the Regulations

Accordingly, 26 CFR parts 1 and 602 are amended as follows:

PART 1--INCOME TAXES

Paragraph 1. The authority citation for part 1 is amended by

removing the entries for ``Sec. 1.482-1T'', ``Sec. 1.482-2T'',

``Sec. 1.482-3T'', ``Sec. 1.482-4T'', ``Sec. 1.482-5T'' and adding

entries in numerical order to read as follows:

Authority: 26 U.S.C. 7805 * * * Section 1.482-1 also issued

under 26 U.S.C. 482 and 936. Section 1.482-2 also issued under 26

U.S.C. 482. Section 1.482-3 also issued under 26 U.S.C. 482. Section

1.482-4 also issued under 26 U.S.C. 482. Section 1.482-5 also issued

under 26 U.S.C. 482. * * *

Secs. 1.482-0T through 1.482-6T [Removed].

Par. 2. Sections 1.482-0T through 1.482-6T are removed.

Par. 3. Sections 1.482-0 through 1.482-6 and 1.482-8 are added to

read as follows:

Sec. 1.482-0 Outline of regulations under 482.

This section contains major captions for Secs. 1.482-1 through

1.482-8.

Section 1.482-1 Allocation of income and deductions among

taxpayers.

(a) In general.

(1) Purpose and scope.

(2) Authority to make allocations.

(3) Taxpayer's use of section 482.

(b) Arm's length standard.

(1) In general.

(2) Arm's length methods.

(i) Methods.

(ii) Selection of category of method applicable to transaction.

(c) Best method rule.

(1) In general.

(2) Determining the best method.

(i) Comparability.

(ii) Data and assumptions.

(A) Completeness and accuracy of data.

(B) Reliability of assumptions.

(C) Sensitivity of results to deficiencies in data and

assumptions.

(iii) Confirmation of results by another method.

(d) Comparability.

(1) In general.

(2) Standard of comparability.

(3) Factors for determining comparability.

(i) Functional analysis.

(ii) Contractual terms.

(A) In general.

(B) Identifying contractual terms.

(1) Written agreement.

(2) No written agreement.

(C) Examples.

(iii) Risk.

(A) In general.

(B) Identification of party that bears risk.

(C) Examples.

(iv) Economic conditions.

(v) Property or services.

(4) Special circumstances.

(i) Market share strategy.

(ii) Different geographic markets.

(A) In general.

(B) Example.

(C) Location savings.

(D) Example.

(iii) Transactions ordinarily not accepted as comparables.

(A) In general.

(B) Examples.

(e) Arm's length range.

(1) In general.

(2) Determination of arm's length range.

(i) Single method.

(ii) Selection of comparables.

(iii) Comparables included in arm's length range.

(A) In general.

(B) Adjustment of range to increase reliability.

(C) Interquartile range.

(3) Adjustment if taxpayer's results are outside arm's length

range.

(4) Arm's length range not prerequisite to allocation.

(5) Examples.

(f) Scope of review.

(1) In general.

(i) Intent to evade or avoid tax not a prerequisite.

(ii) Realization of income not a prerequisite.

(A) In general.

(B) Example.

(iii) Nonrecognition provisions may not bar allocation.

(A) In general.

(B) Example.

(iv) Consolidated returns.

(2) Rules relating to determination of true taxable income.

(i) Aggregation of transactions.

(A) In general.

(B) Examples.

(ii) Allocation based on taxpayer's actual transactions.

(A) In general.

(B) Example.

(iii) Multiple year data.

(A) In general.

(B) Circumstances warranting consideration of multiple year

data.

(C) Comparable effect over comparable period.

(D) Applications of methods using multiple year averages.

(E) Examples.

(iv) Product lines and statistical techniques.

(v) Allocations apply to results, not methods.

(A) In general.

(B) Example.

(g) Collateral adjustments with respect to allocations under section

482.

(1) In general.

(2) Correlative allocations.

(i) In general.

(ii) Manner of carrying out correlative allocation.

(iii) Events triggering correlative allocation.

(iv) Examples.

(3) Adjustments to conform accounts to reflect section 482

allocations.

(i) In general.

(ii) Example.

(4) Setoffs.

(i) In general.

(ii) Requirements.

(iii) Examples.

(h) Special rules.

(1) Small taxpayer safe harbor [Reserved].

(2) Effect of foreign legal restrictions.

(i) In general.

(ii) Applicable legal restrictions.

(iii) Requirement for electing the deferred income method of

accounting.

(iv) Deferred income method of accounting.

(v) Examples.

(3) Coordination with section 936.

(i) Cost sharing under section 936.

(ii) Use of terms.

(i) Definitions.

(j) Effective dates.

Section 1.482-2 Determination of taxable income in specific

situations.

(a) Loans or advances.

(1) Interest on bona fide indebtedness.

(i) In general.

(ii) Application of paragraph (a) of this section.

(A) Interest on bona fide indebtedness.

(B) Alleged indebtedness.

(iii) Period for which interest shall be charged.

(A) General rule.

(B) Exception for certain intercompany transactions in the

ordinary course of business.

(C) Exception for trade or business of debtor member located

outside the United States.

(D) Exception for regular trade practice of creditor member or

others in creditor's industry.

(E) Exception for property purchased for resale in a foreign

country.

(1) General rule.

(2) Interest-free period.

(3) Average collection period.

(4) Illustration.

(iv) Payment; book entries.

(2) Arm's length interest rate.

(i) In general.

(ii) Funds obtained at situs of borrower.

(iii) Safe haven interest rates for certain loans and advances

made after May 8, 1986.

(A) Applicability.

(1) General rule.

(2) Grandfather rule for existing loans.

(B) Safe haven interest rate based on applicable Federal rate.

(C) Applicable Federal rate.

(D) Lender in business of making loans.

(E) Foreign currency loans.

(3) Coordination with interest adjustments required under

certain other Internal Revenue Code sections.

(4) Examples.

(b) Performance of services for another.

(1) General rule.

(2) Benefit test.

(3) Arm's length charge.

(4) Costs or deductions to be taken into account.

(5) Costs and deductions not to be taken into account.

(6) Methods.

(7) Certain services.

(8) Services rendered in connection with the transfer of

property.

(c) Use of tangible property.

(1) General rule.

(2) Arm's length charge.

(i) In general.

(ii) Safe haven rental charge.

(iii) Subleases.

(d) Transfer of property.

Section 1.482-3 Methods to determine taxable income in connection

with a transfer of tangible property.

(a) In general.

(b) Comparable uncontrolled price method.

(1) In general.

(2) Comparability and reliability considerations.

(i) In general.

(ii) Comparability.

(A) In general.

(B) Adjustments for differences between controlled and

uncontrolled transactions.

(iii) Data and assumptions.

(3) Arm's length range.

(4) Examples.

(5) Indirect evidence of comparable uncontrolled transactions.

(i) In general.

(ii) Limitations.

(iii) Examples.

(c) Resale price method.

(1) In general.

(2) Determination of arm's length price.

(i) In general.

(ii) Applicable resale price.

(iii) Appropriate gross profit.

(iv) Arm's length range.

(3) Comparability and reliability considerations.

(i) In general.

(ii) Comparability.

(A) Functional comparability.

(B) Other comparability factors.

(C) Adjustments for differences between controlled and

uncontrolled transactions.

(D) Sales agent.

(iii) Data and assumptions.

(A) In general.

(B) Consistency in accounting.

(4) Examples.

(d) Cost plus method.

(1) In general.

(2) Determination of arm's length price.

(i) In general.

(ii) Appropriate gross profit.

(iii) Arm's length range.

(3) Comparability and reliability considerations.

(i) In general.

(ii) Comparability.

(A) Functional comparability.

(B) Other comparability factors.

(C) Adjustments for differences between controlled and

uncontrolled transactions.

(D) Purchasing agent.

(iii) Data and assumptions.

(A) In general.

(B) Consistency in accounting.

(4) Examples.

(e) Unspecified methods.

(1) In general.

(2) Example.

(f) Coordination with intangible property rules.

Section 1.482-4 Methods to determine taxable income in connection

with a transfer of intangible property.

(a) In general.

(b) Definition of intangible.

(c) Comparable uncontrolled transaction method.

(1) In general.

(2) Comparability and reliability considerations.

(i) In general.

(ii) Reliability.

(iii) Comparability.

(A) In general.

(B) Factors to be considered in determining comparability.

(1) Comparable intangible property.

(2) Comparable circumstances.

(iv) Data and assumptions.

(3) Arm's length range.

(4) Examples.

(d) Unspecified methods.

(1) In general.

(2) Example.

(e) Coordination with tangible property rules.

(f) Special rules for transfers of intangible property.

(1) Form of consideration.

(2) Periodic adjustments.

(i) General rule.

(ii) Exceptions.

(A) Transactions involving the same intangible.

(B) Transactions involving comparable intangible.

(C) Methods other than comparable uncontrolled transaction.

(D) Extraordinary events.

(E) Five-year period.

(iii) Examples.

(3) Ownership of intangible property.

(i) In general.

(ii) Identification of the owner.

(A) Legally protected intangible property.

(B) Intangible property that is not legally protected.

(iii) Allocations with respect to assistance provided to the

owner.

(iv) Examples.

(4) Consideration not artificially limited.

(5) Lump sum payments.

(i) In general.

(ii) Exceptions.

(iii) Example.

Section 1.482-5 Comparable profits method.

(a) In general.

(b) Determination of arm's length result.

(1) In general.

(2) Tested party.

(i) In general.

(ii) Adjustments for tested party.

(3) Arm's length range.

(4) Profit level indicators.

(i) Rate of return on capital employed.

(ii) Financial ratios.

(iii) Other profit level indicators.

(c) Comparability and reliability considerations.

(1) In general.

(2) Comparability.

(i) In general.

(ii) Functional, risk and resource comparability.

(iii) Other comparability factors.

(iv) Adjustments for differences between tested party and the

uncontrolled taxpayers.

(3) Data and assumptions.

(i) In general.

(ii) Consistency in accounting.

(iii) Allocations between the relevant business activity and

other activities.

(d) Definitions.

(e) Examples.

Section 1.482-6 Profit split method.

(a) In general.

(b) Appropriate share of profits and losses.

(c) Application.

(1) In general.

(2) Comparable profit split.

(i) In general.

(ii) Comparability and reliability considerations.

(A) In general.

(B) Comparability.

(1) In general.

(2) Adjustments for differences between the controlled and

uncontrolled taxpayers.

(C) Data and assumptions.

(D) Other factors affecting reliability.

(3) Residual profit split.

(i) In general.

(A) Allocate income to routine contributions.

(B) Allocate residual profit.

(ii) Comparability and reliability considerations.

(A) In general.

(B) Comparability.

(C) Data and assumptions.

(D) Other factors affecting reliability.

(iii) Example.

Section 1.482-7T Sharing of costs and risks.

Section 1.482-8 Examples of the best method rule.

(a) In general.

(b) Examples.

Sec. 1.482-1 Allocation of income and deductions among taxpayers.

(a) In general--(1) Purpose and scope. The purpose of section 482

is to ensure that taxpayers clearly reflect income attributable to

controlled transactions, and to prevent the avoidance of taxes with

respect to such transactions. Section 482 places a controlled taxpayer

on a tax parity with an uncontrolled taxpayer by determining the true

taxable income of the controlled taxpayer. This Sec. 1.482-1 sets forth

general principles and guidelines to be followed under section 482.

Section 1.482-2 provides rules for the determination of the true

taxable income of controlled taxpayers in specific situations,

including controlled transactions involving loans or advances,

services, and property. Sections 1.482-3 through 1.482-6 elaborate on

the rules that apply to controlled transactions involving property.

Section 1.482-7T sets forth the cost sharing provisions. Finally,

Sec. 1.482-8 provides examples illustrating the application of the best

method rule.

(2) Authority to make allocations. The district director may make

allocations between or among the members of a controlled group if a

controlled taxpayer has not reported its true taxable income. In such

case, the district director may allocate income, deductions, credits,

allowances, basis, or any other item or element affecting taxable

income (referred to as allocations). The appropriate allocation may

take the form of an increase or decrease in any relevant amount.

(3) Taxpayer's use of section 482. If necessary to reflect an arm's

length result, a controlled taxpayer may report on a timely filed U.S.

income tax return (including extensions) the results of its controlled

transactions based upon prices different from those actually charged.

Except as provided in this paragraph, section 482 grants no other right

to a controlled taxpayer to apply the provisions of section 482 at will

or to compel the district director to apply such provisions. Therefore,

no untimely or amended returns will be permitted to decrease taxable

income based on allocations or other adjustments with respect to

controlled transactions. See Sec. 1.6662-6T(a)(2) or successor

regulations.

(b) Arm's length standard--(1) In general. In determining the true

taxable income of a controlled taxpayer, the standard to be applied in

every case is that of a taxpayer dealing at arm's length with an

uncontrolled taxpayer. A controlled transaction meets the arm's length

standard if the results of the transaction are consistent with the

results that would have been realized if uncontrolled taxpayers had

engaged in the same transaction under the same circumstances (arm's

length result). However, because identical transactions can rarely be

located, whether a transaction produces an arm's length result

generally will be determined by reference to the results of comparable

transactions under comparable circumstances. See Sec. 1.482-1(d)(2)

(Standard of comparability). Evaluation of whether a controlled

transaction produces an arm's length result is made pursuant to a

method selected under the best method rule described in Sec. 1.482-

1(c).

(2) Arm's length methods--(i) Methods. Sections 1.482-2 through

1.482-6 provide specific methods to be used to evaluate whether

transactions between or among members of the controlled group satisfy

the arm's length standard, and if they do not, to determine the arm's

length result.

(ii) Selection of category of method applicable to transaction. The

methods listed in Sec. 1.482-2 apply to different types of

transactions, such as transfers of property, services, loans or

advances, and rentals. Accordingly, the method or methods most

appropriate to the calculation of arm's length results for controlled

transactions must be selected, and different methods may be applied to

interrelated transactions if such transactions are most reliably

evaluated on a separate basis. For example, if services are provided in

connection with the transfer of property, it may be appropriate to

separately apply the methods applicable to services and property in

order to determine an arm's length result. But see Sec. 1.482-

1(f)(2)(i) (Aggregation of transactions). In addition, other applicable

provisions of the Code may affect the characterization of a

transaction, and therefore affect the methods applicable under section

482. See for example section 467.

(c) Best method rule--(1) In general. The arm's length result of a

controlled transaction must be determined under the method that, under

the facts and circumstances, provides the most reliable measure of an

arm's length result. Thus, there is no strict priority of methods, and

no method will invariably be considered to be more reliable than

others. An arm's length result may be determined under any method

without establishing the inapplicability of another method, but if

another method subsequently is shown to produce a more reliable measure

of an arm's length result, such other method must be used. Similarly,

if two or more applications of a single method provide inconsistent

results, the arm's length result must be determined under the

application that, under the facts and circumstances, provides the most

reliable measure of an arm's length result. See Sec. 1.482-8 for

examples of the application of the best method rule.

(2) Determining the best method. Data based on the results of

transactions between unrelated parties provides the most objective

basis for determining whether the results of a controlled transaction

are arm's length. Thus, in determining which of two or more available

methods (or applications of a single method) provides the most reliable

measure of an arm's length result, the two primary factors to take into

account are the degree of comparability between the controlled

transaction (or taxpayer) and any uncontrolled comparables, and the

quality of the data and assumptions used in the analysis. In addition,

in certain circumstances, it also may be relevant to consider whether

the results of an analysis are consistent with the results of an

analysis under another method. These factors are explained in

paragraphs (c)(2)(i), (ii), and (iii) of this section.

(i) Comparability. The relative reliability of a method based on

the results of transactions between unrelated parties depends on the

degree of comparability between the controlled transaction or taxpayers

and the uncontrolled comparables, taking into account the factors

described in Sec. 1.482-1(d)(3) (Factors for determining

comparability), and after making adjustments for differences, as

described in Sec. 1.482-1(d)(2) (Standard of comparability). As the

degree of comparability increases, the number and extent of potential

differences that could render the analysis inaccurate is reduced. In

addition, if adjustments are made to increase the degree of

comparability, the number, magnitude, and reliability of those

adjustments will affect the reliability of the results of the analysis.

Thus, an analysis under the comparable uncontrolled price method will

generally be more reliable than analyses obtained under other methods

if the analysis is based on closely comparable uncontrolled

transactions, because such an analysis can be expected to achieve a

higher degree of comparability and be susceptible to fewer differences

than analyses under other methods. See Sec. 1.482-3(b)(2)(ii)(A). An

analysis will be relatively less reliable, however, as the uncontrolled

transactions become less comparable to the controlled transaction.

(ii) Data and assumptions. Whether a method provides the most

reliable measure of an arm's length result also depends upon the

completeness and accuracy of the underlying data, the reliability of

the assumptions, and the sensitivity of the results to possible

deficiencies in the data and assumptions. Such factors are particularly

relevant in evaluating the degree of comparability between the

controlled and uncontrolled transactions. These factors are discussed

in paragraphs (c)(2)(ii) (A), (B), and (C) of this section.

(A) Completeness and accuracy of data. The completeness and

accuracy of the data affects the ability to identify and quantify those

factors that would affect the result under any particular method. For

example, the completeness and accuracy of data will determine the

extent to which it is possible to identify differences between the

controlled and uncontrolled transactions, and the reliability of

adjustments that are made to account for such differences. An analysis

will be relatively more reliable as the completeness and accuracy of

the data increases.

(B) Reliability of assumptions. All methods rely on certain

assumptions. The reliability of the results derived from a method

depends on the soundness of such assumptions. Some assumptions are

relatively reliable. For example, adjustments for differences in

payment terms between controlled and uncontrolled transactions may be

based on the assumption that at arm's length such differences would

lead to price differences that reflect the time value of money.

Although selection of the appropriate interest rate to use in making

such adjustments involves some judgement, the economic analysis on

which the assumption is based is relatively sound. Other assumptions

may be less reliable. For example, the residual profit split method may

be based on the assumption that capitalized intangible development

expenses reflect the relative value of the intangible property

contributed by each party. Because the costs of developing an

intangible may not be related to its market value, the soundness of

this assumption will affect the reliability of the results derived from

this method.

(C) Sensitivity of results to deficiencies in data and assumptions.

Deficiencies in the data used or assumptions made may have a greater

effect on some methods than others. In particular, the reliability of

some methods is heavily dependent on the similarity of property or

services involved in the controlled and uncontrolled transaction. For

certain other methods, such as the resale price method, the analysis of

the extent to which controlled and uncontrolled taxpayers undertake the

same or similar functions, employ similar resources, and bear similar

risks is particularly important. Finally, under other methods, such as

the profit split method, defining the relevant business activity and

appropriate allocation of costs, income, and assets may be of

particular importance. Therefore, a difference between the controlled

and uncontrolled transactions for which an accurate adjustment cannot

be made may have a greater effect on the reliability of the results

derived under one method than the results derived under another method.

For example, differences in management efficiency may have a greater

effect on a comparable profits method analysis than on a comparable

uncontrolled price method analysis, while differences in product

characteristics will ordinarily have a greater effect on a comparable

uncontrolled price method analysis than on a comparable profits method

analysis.

(iii) Confirmation of results by another method. If two or more

methods produce inconsistent results, the best method rule will be

applied to select the method that provides the most reliable measure of

an arm's length result. If the best method rule does not clearly

indicate which method should be selected, an additional factor that may

be taken into account in selecting a method is whether any of the

competing methods produce results that are consistent with the results

obtained from the appropriate application of another method. Further,

in evaluating different applications of the same method, the fact that

a second method (or another application of the first method) produces

results that are consistent with one of the competing applications may

be taken into account.

(d) Comparability--(1) In general. Whether a controlled transaction

produces an arm's length result is generally evaluated by comparing the

results of that transaction to results realized by uncontrolled

taxpayers engaged in comparable transactions under comparable

circumstances. For this purpose, the comparability of transactions and

circumstances must be evaluated considering all factors that could

affect prices or profits in arm's length dealings (comparability

factors). While a specific comparability factor may be of particular

importance in applying a method, each method requires analysis of all

of the factors that affect comparability under that method. Such

factors include the following--

(i) Functions;

(ii) Contractual terms;

(iii) Risks;

(iv) Economic conditions; and

(v) Property or services.

(2) Standard of comparability. In order to be considered comparable

to a controlled transaction, an uncontrolled transaction need not be

identical to the controlled transaction, but must be sufficiently

similar that it provides a reliable measure of an arm's length result.

If there are material differences between the controlled and

uncontrolled transactions, adjustments must be made if the effect of

such differences on prices or profits can be ascertained with

sufficient accuracy to improve the reliability of the results. For

purposes of this section, a material difference is one that would

materially affect the measure of an arm's length result under the

method being applied. If adjustments for material differences cannot be

made, the uncontrolled transaction may be used as a measure of an arm's

length result, but the reliability of the analysis will be reduced.

Generally, such adjustments must be made to the results of the

uncontrolled comparable and must be based on commercial practices,

economic principles, or statistical analyses. The extent and

reliability of any adjustments will affect the relative reliability of

the analysis. See Sec. 1.482-1(c)(1) (Best method rule). In any event,

unadjusted industry average returns themselves cannot establish arm's

length results.

(3) Factors for determining comparability. The comparability

factors listed in Sec. 1.482-1(d)(1) are discussed in this section.

Each of these factors must be considered in determining the degree of

comparability between transactions or taxpayers and the extent to which

comparability adjustments may be necessary. In addition, in certain

cases involving special circumstances, the rules under paragraph (d)(4)

of this section must be considered.

(i) Functional analysis. Determining the degree of comparability

between controlled and uncontrolled transactions requires a comparison

of the functions performed, and associated resources employed, by the

taxpayers in each transaction. This comparison is based on a functional

analysis that identifies and compares the economically significant

activities undertaken, or to be undertaken, by the taxpayers in both

controlled and uncontrolled transactions. A functional analysis should

also include consideration of the resources that are employed, or to be

employed, in conjunction with the activities undertaken, including

consideration of the type of assets used, such as plant and equipment,

or the use of valuable intangibles. A functional analysis is not a

pricing method and does not itself determine the arm's length result

for the controlled transaction under review. Functions that may need to

be accounted for in determining the comparability of two transactions

include--

(A) Research and development;

(B) Product design and engineering;

(C) Manufacturing, production and process engineering;

(D) Product fabrication, extraction, and assembly;

(E) Purchasing and materials management;

(F) Marketing and distribution functions, including inventory

management, warranty administration, and advertising activities;

(G) Transportation and warehousing; and

(H) Managerial, legal, accounting and finance, credit and

collection, training, and personnel management services.

(ii) Contractual terms--(A) In general. Determining the degree of

comparability between the controlled and uncontrolled transactions

requires a comparison of the significant contractual terms that could

affect the results of the two transactions. These terms include--

(1) The form of consideration charged or paid;

(2) Sales or purchase volume;

(3) The scope and terms of warranties provided;

(4) Rights to updates, revisions or modifications;

(5) The duration of relevant license, contract or other agreements,

and termination or renegotiation rights;

(6) Collateral transactions or ongoing business relationships

between the buyer and the seller, including arrangements for the

provision of ancillary or subsidiary services; and

(7) Extension of credit and payment terms. Thus, for example, if

the time for payment of the amount charged in a controlled transaction

differs from the time for payment of the amount charged in an

uncontrolled transaction, an adjustment to reflect the difference in

payment terms should be made if such difference would have a material

effect on price. Such comparability adjustment is required even if no

interest would be allocated or imputed under Sec. 1.482-2(a) or other

applicable provisions of the Internal Revenue Code or regulations.

(B) Identifying contractual terms--(1) Written agreement. The

contractual terms, including the consequent allocation of risks, that

are agreed to in writing before the transactions are entered into will

be respected if such terms are consistent with the economic substance

of the underlying transactions. In evaluating economic substance,

greatest weight will be given to the actual conduct of the parties, and

the respective legal rights of the parties (see, for example,

Sec. 1.482-4(f)(3) (Ownership of intangible property)). If the

contractual terms are inconsistent with the economic substance of the

underlying transaction, the district director may disregard such terms

and impute terms that are consistent with the economic substance of the

transaction.

(2) No written agreement. In the absence of a written agreement,

the district director may impute a contractual agreement between the

controlled taxpayers consistent with the economic substance of the

transaction. In determining the economic substance of the transaction,

greatest weight will be given to the actual conduct of the parties and

their respective legal rights (see, for example, Sec. 1.482-4(f)(3)

(Ownership of intangible property)). For example, if, without a written

agreement, a controlled taxpayer operates at full capacity and

regularly sells all of its output to another member of its controlled

group, the district director may impute a purchasing contract from the

course of conduct of the controlled taxpayers, and determine that the

producer bears little risk that the buyer will fail to purchase its

full output. Further, if an established industry convention or usage of

trade assigns a risk or resolves an issue, that convention or usage

will be followed if the conduct of the taxpayers is consistent with it.

See UCC 1-205. For example, unless otherwise agreed, payment generally

is due at the time and place at which the buyer is to receive goods.

See UCC 2-310.

(C) Examples. The following examples illustrate this paragraph

(d)(3)(ii).

Example 1--Differences in volume. USP, a United States

agricultural exporter, regularly buys transportation services from

FSub, its foreign subsidiary, to ship its products from the United

States to overseas markets. Although FSub occasionally provides

transportation services to URA, an unrelated domestic corporation,

URA accounts for only 10% of the gross revenues of FSub, and the

remaining 90% of FSub's gross revenues are attributable to FSub's

transactions with USP. In determining the degree of comparability

between FSub's uncontrolled transaction with URA and its controlled

transaction with USP, the difference in volumes involved in the two

transactions and the regularity with which these services are

provided must be taken into account if such difference would have a

material effect on the price charged. Inability to make reliable

adjustments for these differences would affect the reliability of

the results derived from the uncontrolled transaction as a measure

of the arm's length result.

Example 2--Reliability of adjustment for differences in volume.

(i) FS manufactures product XX and sells that product to its parent

corporation, P. FS also sells product XX to uncontrolled taxpayers

at a price of $100 per unit. Except for the volume of each

transaction, the sales to P and to uncontrolled taxpayers take place

under substantially the same economic conditions and contractual

terms. In uncontrolled transactions, FS offers a 2% discount for

quantities of 20 per order, and a 5% discount for quantities of 100

per order. If P purchases product XX in quantities of 60 per order,

in the absence of other reliable information, it may reasonably be

concluded that the arm's length price to P would be $100, less a

discount of 3.5%.

(ii) If P purchases product XX in quantities of 1,000 per order,

a reliable estimate of the appropriate volume discount must be based

on proper economic or statistical analysis, not necessarily a linear

extrapolation from the 2% and 5% catalog discounts applicable to

sales of 20 and 100 units, respectively.

Example 3--Contractual term imputed from economic substance. (i)

USD, a United States corporation, is the exclusive distributor of

products manufactured by FP, its foreign parent. The FP products are

sold under a tradename that is not known in the United States. USD

does not have an agreement with FP for the use of FP's tradename.

For Years 1 through 6, USD bears marketing expenses promoting FP's

tradename in the United States that are substantially above the

level of such expenses incurred by comparable distributors in

uncontrolled transactions. FP does not directly or indirectly

reimburse USD for its marketing expenses. By Year 7, the FP

tradename has become very well known in the market and commands a

price premium. At this time, USD becomes a commission agent for FP.

(ii) In determining USD's arm's length result for Year 7, the

district director considers the economic substance of the

arrangements between USD and FP throughout the course of their

relationship. It is unlikely that at arm's length, USD would incur

these above-normal expenses without some assurance it could derive a

benefit from these expenses. In this case, these expenditures

indicate a course of conduct that is consistent with an agreement

under which USD received a long-term right to use the FP tradename

in the United States. Such conduct is inconsistent with the

contractual arrangements between FP and USD under which USD was

merely a distributor, and later a commission agent, for FP.

Therefore, the district director may impute an agreement between USD

and FP under which USD will retain an appropriate portion of the

price premium attributable to the FP tradename.

(iii) Risk--(A) Comparability. Determining the degree of

comparability between controlled and uncontrol

This text is long and has been trimmed here. Open the source document for the complete record.