This is not the document you are looking for? Use the search form below to find more!

Report home > World & Business

Arm's Length Standard

0.00 (0 votes)
Document Description
Taxpayers won an unequivocal victory in Veritas Software Corp. v. Commissioner, 133 T.C. No. 14 (filed December 10,2009), in which the U.S. Tax Court held that the IRS’s $1.675 billion income adjustment of the cost sharing buy-in paymentreceived by Veritas Software Corp. (the Taxpayer) from its Irish affiliate was arbitrary, capricious, and unreasonable. Thecourt further held that the Taxpayer’s comparable uncontrolled transactions (CUT) method, with adjustments determined bythe court, was the best method to calculate the buy-in payment.
File Details
  • Added: February, 24th 2010
  • Reads: 1806
  • Downloads: 118
  • File size: 135.15kb
  • Pages: 20
  • Tags: arms length standard, tax, capital, canada, india, hongkong
  • content preview
Submitter
  • Username: shinta
  • Name: shinta
  • Documents: 4332
Embed Code:

Add New Comment




Related Documents

India – New Rule Allows Use of ‘Other Method’ for Determining Arm’s Length Principle

by: nair & co., 1 pages

The Indian Tax Administration has introduced a new transfer pricing method which uses the ‘other method’ for determining Arm’s length Principle (ALP) for international transactions.

Popper's Double Standard of Scientificity in Criticizing Marxism

by: terttu, 17 pages

This paper is a consideration of Popper's criticism of Marxism in terms of the criterion of falsifiability. It attempts to show that Popper's criticism, taken together with his alternative proposal ...

China's Transition to a Market Economy: How Far across the River?

by: shinta, 55 pages

China's two-decade reform since 1979 has evolved in two stages with the November 1993 decision marking a turning point. The essence of this decision is to replace the planning system with ...

Practical Problems In Transfer Pricing

by: atsushi, 17 pages

PRACTICAL PROBLEMS IN TRANSFER PRICING By NEERU AHUJA DELOITTE HASKINS & SELLS November’ 2003 Contents Definition of Transfer Pricing Operational ...

Want to Know How to Stop Drinking?

by: Nick, 1 pages

For an individual who desires to find out how to stop consuming, the initial stage is usually the realization that consuming has turn out to be a dilemma that needs to be addressed. As soon as that ...

Standard Deviation Example

by: vistateam123, 4 pages

Standard Deviation is used to measure the deviation of data from its mean. It is equals to square root of variance; standard deviation tells us that how many data items are distributed above and ...

Livind Life As A Godly Single Woman

by: godlydating, 1 pages

Living life as a godly single woman in today’s time can prove to be very challenging, but you can make the decision not to allow it to be challenging. It is no secret that the world’s ...

Living Life As A Godly Single Woman

by: godlydating, 1 pages

Living life as a godly single woman in today's time can prove to be very challenging, but you can make the decision not to allow it to be challenging. It is no secret that the world's moral standard ...

Transfer Pricing Guidelines to Help Companies Avoid Tax Penalties in Italy

by: nair & co., 1 pages

Transfer Pricing provisions generally require international transactions between related parties to be carried out at arm’s length. And as transfer pricing continues to be an area of focus with ...

Singapore Transfer Pricing

by: rikvinsingapore, 11 pages

This is an overview of transfer pricing mechanisms, providing guidelines to follow arm’s length principle and documentation to be maintained for the purpose of audits

Content Preview



Global Transfer Pricing

Arm’s Length Standard

December 2009/January 2010

In this issue:

Tax Court rejects IRS’s cost sharing buy-in valuation analysis ................................................................................................. 1 
GE Capital Canada prevails in guarantee fee case ................................................................................................................. 6 
Hong Kong introduces transfer pricing guidelines in DIPN 46 ............................................................................................... 7 
Taxpayers beware: Year-end competent authority deadlines looming ................................................................................. 11 
India issues rules on new dispute resolution panel .............................................................................................................. 13 
Recent transfer pricing developments in Slovakia ................................................................................................................ 14 
2009 year-end transfer pricing actions: Business not as usual ............................................................................................. 17 



Tax Court rejects IRS’s cost sharing buy-in valuation analysis

Taxpayers won an unequivocal victory in Veritas Software Corp. v. Commissioner, 133 T.C. No. 14 (filed December 10,
2009), in which the U.S. Tax Court held that the IRS’s $1.675 billion income adjustment of the cost sharing buy-in payment
received by Veritas Software Corp. (the Taxpayer) from its Irish affiliate was arbitrary, capricious, and unreasonable. The
court further held that the Taxpayer’s comparable uncontrolled transactions (CUT) method, with adjustments determined by
the court, was the best method to calculate the buy-in payment.

The Tax Court’s opinion, based on the 1996 cost sharing regulations, focused heavily on the IRS’s litigation missteps, the
lack of credibility of the IRS’s experts, and the strength of the Taxpayer’s factual evidence. But the Tax Court also clearly
rejected many of the arguments the IRS has been asserting since the release of the Coordinated Issues Paper on Cost
Sharing Arrangements Buy-in Adjustments on September 27, 2007 (the Coordinated Issues Paper); thus, the case will likely
impact the resolution of billions of dollars of proposed IRS adjustments related to cost sharing buy-in payments.

Facts

In November 1999, Veritas-US and Veritas-Ireland entered into the following intercompany agreements: (1) assignment of
certain European sales agreements, (2) Agreement for Sharing Research and Development Costs (the RDA), and a
Technology License Agreement (the TLA). During 1999 and 2000, Veritas-Ireland paid Veritas-US approximately $172
million for the rights transferred pursuant to the TLA. In 2002, the parties agreed to reduce this amount to $118 million.

In March 2006, the IRS issued a notice of deficiency based on a buy-in valuation of $2.5 billion that employed the forgone
profits method, the market capitalization method, and an analysis of the Taxpayer’s arm’s length acquisitions.


Arm’s Length Standard
1 of 20
Copyright ©2009, Deloitte Touche Tohmatsu.
December 2009/January 2010

All rights reserved.

During pretrial proceedings, the IRS abandoned the original analysis and submitted a new analysis that reduced the amount
of the buy-in by $825 million to $1.675 billion. The new analysis valued in the aggregate all the alleged intangibles and
other property transferred to Veritas-Ireland, and was based on a discounted cash flow or “income method” using a
perpetual life. The new analysis was based on Veritas-Ireland’s actual results through 2006 and projections using an
assumed growth rate after that date.1

Tax Court opinion

The Tax Court, based on well settled law, held that the IRS position is presumptively correct unless it is arbitrary, capricious,
and unreasonable.

The IRS original position was arbitrary, capricious and unreasonable – The Tax Court found the IRS original position
failed to meet this standard because the IRS, without any explanation, conceded $825 million of the buy-in amount set
forth in its original analysis and failed to offer even a token defense of its original position.

The Court’s criticisms of the IRS revised position – The IRS’s revised position appears to be consistent with its position as
stated in the Coordinated Issues Paper.2 The Tax Court either rejected most of the IRS’s arguments contained in the
Coordinated Issues Paper or found that the analysis of the IRS expert did not support those positions in the present case.

• Allocation included intangibles created by cost sharing payments – The Tax Court held that Treas. Regs. §
1.482-7(g)(2) only requires taxpayers to make a buy-in payment for preexisting intangibles. No payment is required
for intangible property that is subsequently developed as a result of a cost sharing arrangement. The court
determined and the IRS agreed that the IRS discounted cash flow or income method, which used a perpetual
useful life for preexisting intangibles, was flawed because it took into account items of income other than from
preexisting intangibles. The court held that the IRS valuation was in violation of its own regulations that limited the
buy-in payment to preexisting intangibles.
• The IRS’s “akin” to a sale of the business theory is specious – The Coordinated Issues Paper suggests that the
income method, which is a form of discounted cash flow analysis typically used in business valuations or purchase
price allocations, may be the best method for determining buy-in payments. One of the IRS’s principal reasons for
using the income method is that it values all the transferred rights in the aggregate, which the IRS says may be
more reliable because it takes into account the synergistic value of the transferred intangibles. The IRS valuation
approach is “akin” to a sale. The IRS expert’s report in support of its revised analysis appears to be based on a
similar analysis.

The IRS position was doomed when its own expert stated at trial that he did not have an opinion as to whether his
methodology captured the synergistic value of the transferred intangibles. In disagreeing with the IRS’s theory that valuing
the transferred intangibles and other property in the aggregate was more reliable, the Tax Court concluded the opposite,
because the IRS’s approach valued short-term intangibles as though they had a perpetual life, and took into account
subsequently developed intangibles rather than preexisting intangibles as required by the regulations. As a result, the court
rejected the IRS’s income method and aggregation approach on the basis that it did not produce the most reliable result.

In a footnote, the court also noted that the “akin to a sale” theory may violate Reg. §1.482-1(f)(2)(ii)(A), which provides that
the IRS will evaluate the results of a transaction as actually structured by the taxpayer, unless its structure lacks economic
substance. The court noted that the TLA, which had economic substance, was structured as a license of preexisting
intangibles, not a sale of a business.





1 The IRS position, which used actual data through 2006 rather than the higher original projections, appears to be
inconsistent with their position that a taxpayer cannot apply the commensurate with income standard. The case contains no
discussion of this issue.
2 The Tax Court record in this case is sealed, so that actual analysis prepared by the IRS has not been made public.

Arm’s Length Standard
2 of 20
Copyright ©2009, Deloitte Touche Tohmatsu.
December 2009/January 2010

All rights reserved.

The IRS’s allocation took into account items not transferred or of insignificant value – The IRS claimed that Veritas-US
transferred its customer base, customer lists, distribution channels, and access to its marketing and R&D teams. The court
found that such items were either not transferred or had insignificant value.

• Distribution channels, customer base, and customer lists – The court concluded, based on the evidence
presented by the Taxpayer, that these items were not well developed outside the United States at the time of
transfer and, therefore, such assets were weak and had little value.
• Access R&D and Marketing teams – The IRS’s expert testified that his buy-in analysis did not consider the access
to the R&D or marketing teams. In addition, the IRS expert conceded that if he assumed that the agreement
relating to the sharing of R&D expenses was arm’s length, a fact that the parties stipulated, then access to the R&D
team would have zero value. Based on the IRS expert’s testimony, the court concluded that there was no evidence
that access to the R&D and marketing teams were transferred or had value.

In a footnote that likely will have implications in other cases, the court went on to state that even if evidence of the value of
the R&D and marketing teams existed, they would not be taken into account in calculating the buy-in because they do not
have “substantial value independent of the services of any individual” and, therefore, are not covered by the definition of
intangible property described in Sec. 936(h)(3)(B) or Regs. §1.482-4(b). The court went on to note without comment that in
December 2008, the IRS issued temporary cost sharing regulations that reference “assembled workforce,” and that the
Obama Administration recently proposed to amend the statute to include “workforce in place” in the section 482 definition
of an intangible.

Employed the wrong useful life, discount rate, and growth rate – With respect to useful life, the court rejected the IRS’s
assumption that the preexisting intangibles had an infinite useful life. The IRS’s expert acknowledged that “if you had 1999
products that you left untouched, that technology would age and eventually become obsolete,” and that the preexisting
product intangibles would “wither on the vine” within two to four years without ongoing R&D. The court also noted that
the Taxpayer provided evidence that the useful life of the preexisting intangibles was four years, and that even with
substantial ongoing R&D the products had finite life cycles.

The court found other errors in the IRS’s expert’s computation of the discount rate, including the determination of the Beta,
the equity risk premium, and the risk-free return.

Finally, the court concluded that the IRS expert applied unrealistic growth rates into perpetuity. The court noted that the
growth rates used by the IRS expert exceeded Veritas-Ireland’s actual growth rate between 2004 and 2006, and that the
expert could not support the growth rates that were used.

The Court accepts the taxpayer’s CUT method with adjustments –

• Valuation of technology – The Coordinated Issues Paper takes the position that uncontrolled licenses of make-
and-sell rights generally are not comparable to the rights transferred in a cost sharing buy-in, because those
transactions usually do not provide the right to further develop the intangibles. The Tax Court clearly rejected that
position. In the view of the court, the focus of the buy-in payment is on preexisting intangibles, not subsequently
developed intangibles.

To determine the appropriate starting royalty rate for the buy-in payment, the Taxpayer used the CUT method. The
Taxpayer’s CUT analysis was based on internal comparable agreements between Veritas and third parties involving
bundled products, in which the Veritas software was incorporated into another company’s product, and
unbundled products in which the software was sold separately.

After reviewing the comparability criteria described in the regulations, the court concluded that the bundled license
agreements were not comparable because Veritas gained credibility and improved brand identity when its products
were sold bundled with another company’s products. The court concluded that the CUTs involving unbundled
products were sufficiently comparable to the controlled transaction, and that the CUT method is the best method
to determine the requisite buy-in payment.

The court calculated a starting royalty rate of 32 percent of sales by calculating the mean royalty rate from 90
licenses involving unbundled products. The court determined that the preexisting intangibles had a useful life of

Arm’s Length Standard
3 of 20
Copyright ©2009, Deloitte Touche Tohmatsu.
December 2009/January 2010

All rights reserved.

four years based on the testimony of Veritas’s and the IRS’s experts. Based on the evidence, the court noted that
unrelated parties that license static technology that is neither subject to updates nor rights to new versions agree
to a ramp down of the royalty over the life of the agreements. Based on the comparable agreements, the court
then reduced the royalty rates starting in year 2 at a rate of 33 percent per year (i.e., 32 percent in year 1; 21
percent in year 2; 14 percent in year 3; and 10 percent in year 4). The court declined to base its ramp down on the
Veritas analysis of changes in the lines of code.

• Valuation of trademark – The Tax Court determined that Veritas’s product and other trademarks and trade names
were well known, respected, and valuable even though they were not extensively registered outside the United
States. The IRS did not submit an expert report valuing the trademarks and trade names. It is unclear from the
opinion exactly how Veritas’s expert estimated the 0.5 percent to 1.0 percent range of royalties and the maximum
life of seven years. Although the court did not find the Taxpayer’s expert convincing with respect to the trademark
valuation, lacking other evidence, the court applied the expert’s upper-end valuation of $9.6 million.

• Valuation of sales agreements – The court noted that it did not have sufficient evidence to determine the value
of the sales agreements transferred to Veritas-Ireland and ordered the parties to address this issue in their
judgment calculations pursuant to the court’s decision. Based on the court’s statement of facts, it is unlikely that
substantial value will be attributed to these agreements.

• Determination of discount rate – Veritas based its discount rate analysis on the traditional capital asset pricing
model (CAPM). Veritas used its own Beta in the computation because the industry Beta was dominated by
Microsoft, “a stronger and more established business than Veritas-US,” and the historic risk premium. Veritas did
not increase the discount rate by industry average tax rate or its specific tax rate. Based on Veritas’s analysis, the
court agreed that 20.47 percent was a reasonable estimate of an appropriate discount rate.

Implications of Tax Court’s decision

IRS appeal is likely – The Tax Court’s decision rejected the major IRS positions contained in the Coordinated Issues Paper
on Cost Sharing Arrangements Buy-in Adjustments. The IRS has proposed billions of dollars in adjustments based on those
theories. Therefore, the IRS is very likely to appeal the Tax Court decision to the Ninth Circuit Court of Appeals. Given the
fact that additional procedures are required in the Tax Court before the decision becomes final, however, and the time
required to generally file and hear an appeal in the Ninth Circuit, it is unlikely that a final decision on this case will be
rendered in less than three years.3

Pending adjustments under 1996 regulations – The IRS National Office of Appeals was expected to issue the long-
awaited and much delayed appeals settlement guidelines with respect to cost sharing buy-in payments. The guidelines were
supposed to provide guidance to appeals officers regarding the settlement of cost sharing buy-in payments. As a result of
the Tax Court’s decision, we would expect that those guidelines will be further delayed, perhaps indefinitely.

There is only one other case pending in the Tax Court in which the taxpayer is challenging the IRS position on buy-in
payments, Medtronic, Inc. That case has not been scheduled for trial. In light of the Tax Court’s decision in Veritas, it is
unclear whether the case will move forward, await the Ninth Circuit’s decision in Veritas, or be settled. Assuming
Medtronic, Inc. does not go to trial, the chance of another case being decided in the next few years is remote, given the
time it takes to file a case and reach a court decision.4

Since it will likely be several years before the Ninth Circuit issues its opinion or another court issues an opinion on cost
sharing buy-in payments, the IRS must address the Tax Court’s decision in Veritas when examining cost sharing buy-in
payments computed under the 1996 regulations. Therefore, as a practical matter, in many cases the large adjustments
made by IRS field economists based on the income method using an infinite life are not likely to be sustained in full.





3 The time between the Tax Court decision in Xilinx and the Ninth Circuit’s decision in that case was almost four years.
4 The time between filing the petition and the Tax Court’s decision in Veritas was 3.5 years.

Arm’s Length Standard
4 of 20
Copyright ©2009, Deloitte Touche Tohmatsu.
December 2009/January 2010

All rights reserved.

There are many factual aspects of Veritas that taxpayers seeking the best settlement position may need to prove. For
example, in Veritas, the taxpayer presented evidence on both useful life and the royalty ramp down, crucial factors in
determining the value of the technology intangibles. In addition, the court reviewed each of the intangibles transferred to
Veritas-Ireland and the value assigned to each of those intangibles. In some cases, taxpayers may have to examine whether
their buy-in approach took those intangibles into account and whether they have adequate support for the valuation of
those intangibles. The court also examined the underlying factual support for the Taxpayer’s CUT method in Veritas, and
made certain adjustments to the Taxpayer’s CUTs. Taxpayers should be prepared to factually support the assumptions they
make in their valuation analyses, whether the method used is the CUT, the residual profits method, or any other method.
Although the court did not accept the IRS’s income method, the Taxpayer in Veritas was still required to support the
appropriateness of its method. In Appeals, taxpayers can expect any settlement to be very fact-specific, and the stronger
their factual support, the more likely a satisfactory resolution will be reached.

Impact of Tax Court’s decision on temporary regulations – The Treasury Department issued temporary cost sharing
regulations on January 5, 2009. The Tax Court’s decision has no direct bearing on those regulations. However, some of the
court’s determinations could have an impact on the interpretation of those regulations. For example, the court’s
determination that a valuation using the income method based on a perpetual life took into account items of income other
than preexisting intangibles may be equally applicable in determining the life of transferred Prior and Contemporaneous
Transactions (PCTs) under the temporary regulations. In addition, the court’s determination that valuing the intangibles in
the aggregate was not reliable may be equally applicable to the aggregation rules in the temporary regulations.5 Finally, the
court’s conclusion that access to the U.S. based R&D and Marketing teams may not have any value independent of the
services of any individual may limit the value of those items under the temporary regulations.

In the Ninth Circuit’s recent decision in Xilinx, Inc. v. Commissioner, 567 F.3d 482 (9th Cir. 2009) the court permitted the
IRS and Treasury to promulgate regulations defining the arm’s length standard differently from its plain meaning. Xilinx has
requested the Ninth Circuit to rehear the case because the court’s decision appears to be inconsistent with the testimony in
that case that arm’s length parties do not share stock option expense. If the Ninth Circuit were to require that the
regulations be consistent with the arm’s length standard, the Tax Court’s decision could have added implications under the
temporary regulations.

Proposed legislation – The President’s Fiscal Year 2010 Budget Proposals contain a proposal that would “clarify” that (i)
work force in place, goodwill, and going concern value are intangibles for purposes of sections 482 and 367(d), and (ii) that
when multiple intangible properties are transferred, the commissioner may value them on an aggregate basis when doing
so achieves a more reliable result.

The Joint Committee on Taxation, in its commentary on the Administration’s transfer pricing proposal, described other
proposals Congress could consider in the transfer pricing area that would go beyond the President’s proposals.6 It is possible
that the Tax Court’s decision could be seen by some in Congress and/or the Treasury Department as evidence that the
current transfer pricing rules are in need of revision.

— Kerwin Chung (Washington, DC)
Cindy Hustad (San Francisco)
Director
Director
Deloitte Tax LLP
Deloitte Tax LLP
+1 (202) 879 3108
+1 (415) 783 5567
kechung@deloitte.com
chustad@deloitte.com







5 This conclusion may be based on the testimony before the court in this case, rather than enunciation of a broader
principle.
6 Joint Committee on Taxation, Description of Revenue Provisions Contained in the President’s Fiscal Year 2010 Budget
Proposal; Part Three: Provisions Related to the Taxation of Cross-Border Income and Investment (JCS-4-09), September 2009
pg 26-55.

Arm’s Length Standard
5 of 20
Copyright ©2009, Deloitte Touche Tohmatsu.
December 2009/January 2010

All rights reserved.


Alan Shapiro (Chicago)
Partner
Deloitte Tax LLP
+1 (312) 486 9112
ashapiro@deloitte.com


GE Capital Canada prevails in guarantee fee case

The Tax Court of Canada on December 4 issued a long-awaited decision in the case General Electric Capital Canada Inc. vs.
The Queen. The court held in favor of GE Capital Canada Inc., allowing it to maintain as a deductible expense 100 percent
of the guarantee fee paid to its U.S. parent, General Electric Capital Corporation (GECUS) during the 1996-2000 period. In
his decision, Justice Robert Hogan ruled that the 100 basis points (1 percent) guarantee fee was an arm’s length price.

The case stems from a guarantee fee arrangement between a Canadian subsidiary and its U.S. parent. GECUS
unconditionally guaranteed all payments due under all debt securities issued after 1988 by its wholly owned subsidiary,
General Electric Capital Canada Inc. In 1995, GECUS began to charge a 1 percent per annum fee for those financial
guarantees. The CRA issued reassessments totaling $136 million for taxation years ended 1996 to 2000, denying the
deductions for guarantee fees claimed by GE Capital Canada in those years. In April 2006, GE Capital Canada filed a notice
of appeal regarding the CRA’s denial of the deductions.

The decision has far-reaching implications for the pricing of guarantee fees in Canada. The method the court used to arrive
at its conclusion provides a framework for pricing guarantee fees that is a complex, multistep approach requiring many
inputs. These are summarized below.

• Determine the recipient’s stand-alone creditworthiness – The court’s approach starts by ascertaining the
creditworthiness of the guarantee recipient. Distinctions were made between a “status quo” rating, taking into
account the benefits of the common name, the parent’s management team and existing business arrangements,
such as intercompany loans, and a stand-alone rating, which abstracts from such benefits. The judge favorably
cited the status quo rating of BB-/B+ developed by an expert testifying on behalf of GE Capital Canada.
• Adjust to the final rating incorporating implicit support – The court then considered any enhancements to the
initial rating from implicit support. The judge noted that GE and GECUS had a widely advertised AAA credit rating
that they would seek to protect, and, as such, would be economically motivated to provide support to GE Capital
Canada even in the absence of a formal guarantee. In assigning a three-notch uplift to GE Capital Canada’s initial
rating, the judge rejected the notion that the implicit support would result in an equalization of the rating of GE
Capital Canada to AAA. Justice Hogan similarly rejected an approach that would start with the parent’s credit
rating and then notch down.
• Price the upper bound using the yield approach – The judge favorably cited testimony from Jeffrey Werner,
retired Treasurer for GECUS and GE Capital Canada during the years of the disputed guarantee fee, that the
guarantee fee had been priced based on the benefit to the guarantee recipient, and that GE Capital Canada would
not pay more for the guarantee fee than the benefit of the fee. Judge Hogan goes on to equate the benefit with
the interest cost savings, although noting based on expert testimony that additional cost savings were likely to
have been realized, such as on placement fees and a backup credit facility. The judge did not comment on
testimony that the benefit included substantially greater borrowing capacity for GE Capital Canada. Accepting
calculations by an expert for GE Capital Canada, the judge determined an upper bound of 183 basis points for the
guarantee fee.
• Finalize the arm’s length range of guarantee fee – The judge, noting that GE Capital Canada cannot be
expected to pay 100 percent of its interest cost savings, ruled that 100bps “is equal to or below an arm’s length
price. “Judge Hogan would reach his decision on the 100 bps fee without suggesting how to fix the arm’s length
range of guarantee fees.

For the Canadian market, certainly, the decision confirms that guarantee transactions have a more-than-nominal value.
Rejecting various arguments presented by the Crown, the judge affirmed that such transactions need to be priced in
accordance with the arm’s length principle. However, the decision also introduces uncertainty with regard to the
application of the arm’s length principle because of complications attendant to factoring implicit support into the analysis.

Arm’s Length Standard
6 of 20
Copyright ©2009, Deloitte Touche Tohmatsu.
December 2009/January 2010

All rights reserved.

Put in the context of the OECD Guidelines, the decision increases uncertainty with regard to the interpretation of Paragraph
7.13, which deals with issues surrounding passive association.

The decision directs taxpayers to the stand-alone rating of the recipient as the starting point. The conversion of the stand-
alone rating to the status quo rating is likely to be problematic for practitioners, given the lack of guidance in the decision.
As to the next step in the analysis, the establishment of the final rating, the judge emphasized that implicit support is strictly
a matter of facts and circumstances. The judge cited public rating agency requirements for more concrete signs of implicit
support to bridge a ratings gap between the recipient and its parent; that is, simply assuming parental support absent those
concrete signs flies in the face of the legal separation of corporations. “The extent of a shareholder’s exposure through the
corporation is limited to the amount of capital the shareholder chooses to invest. In the absence of a guarantee from the
shareholder, creditors can expect nothing more nor less.”

Judge Hogan cited expert testimony that the upgrade for implicit support is not greater than two or three notches. Thus,
the value of implicit support can be perceived as relatively modest (given the 12-13 notch ratings gap between the initial
rating of GE Capital Canada and GECUS).

The case will have repercussions in the Australian market as well, as the Australian Taxation Office (ATO) has been targeting
“excessive” guarantee fees in its audit program. In 2008, the ATO issued a discussion paper that stated that the stand-alone
credit rating is the appropriate starting point; however, in its view, lenders may notch upwards from 1 to 9 notches having
regard to implicit credit support provided by the parent. Although Judge Hogan allowed for upward notching, the rationale
for limitation in this case to just two or three notches will likely be considered closely by both the ATO and multinationals
with guarantee arrangements in Australia. The ATO is expected to publish further rulings and papers on debt financing
issues over the coming months.

The Crown has 30 days from the date of release of the opinion to appeal.

Please contact a Deloitte transfer pricing expert to discuss this decision and its application to your circumstances.

— Muris Dujsic (Toronto)
Mark Fidelman (New York)
Partner
Director
Deloitte Canada
Deloitte Tax LLP
+1 (416) 601 6006
+1 (212) 436 6708
mdujsic@deloitte.ca
mfidelman@deloitte.com


Richard Garland (Toronto)
Kevin Gale (Toronto)
Partner
Senior Manager
Deloitte Canada
Deloitte Canada
+1 (416) 601 6026
+1 (905) 315 6761
rigarland@deloitte.ca
kgale@deloitte.ca


Geoff Gill (Sydney)
Partner
Deloitte Australia
+61 (2) 9322 5358
gegill@deloitte.com.au


Hong Kong introduces transfer pricing guidelines in DIPN 46

The Hong Kong Inland Revenue Department (IRD) on December 4 issued Departmental Interpretation and Practice Notes
No. 46 (DIPN 46), the widely anticipated guidance on transfer pricing. While DIPN 45, issued in May 2009, deals with

Arm’s Length Standard
7 of 20
Copyright ©2009, Deloitte Touche Tohmatsu.
December 2009/January 2010

All rights reserved.

double taxation relief under DTAs7 when a transfer pricing adjustment has been made by another contracting state, DIPN
46 is the first practice note to provide the IRD’s views on transfer pricing matters. This article summarizes the salient points
in DIPN 46 and provides insights on how it reflects the ongoing stance of the IRD in transfer pricing enforcement.

When will IRD seek to impose a transfer pricing adjustment?

The arm’s length principle refers to the allocation of profits and expenses relating to transactions between associated
enterprises, having regard to how independent enterprises would deal with each other under the same circumstances. In
general, when transfer pricing does not follow the arm’s length principle so that the profits or tax liabilities of associated
enterprises are distorted, the IRD will seek to impose transfer pricing adjustments to reallocate profits or adjust deductions
by substituting an arm’s length consideration. While an adjustment may be initiated by the IRD, as authorized under the
Hong Kong Inland Revenue Ordinance (IRO), an adjustment may also be made pursuant to one of Hong Kong’s five DTAs
with Belgium, Luxembourg, mainland China, Thailand, and Vietnam.

Invoking transfer pricing adjustments under DTAs – The “business profits” provision in Article 7 of Hong Kong’s DTAs
provides for the attribution of profits to a permanent establishment (PE) as if the PE were a separate enterprise operating at
arm’s length. Under this approach, a Hong Kong PE would be treated as a separate entity, and its related-party transactions
would be analyzed using transfer pricing principles. It should also be noted that the Hong Kong DTAs with Belgium,
mainland China, and Vietnam would disregard the payment or receipt of commissions, royalties, or interest between a head
office and its branch in determining the profit of a PE. However, it is not clear from DIPN 46 whether the separate entity
approach will prevail over Inland Revenue Rule 5 when dealing with the PE of an enterprise of a non-DTA country.8

The “associated enterprises” provision in Article 9 of Hong Kong’s DTAs provides the IRD with the authority to impose
transfer pricing adjustments if a Hong Kong enterprise’s transactions with an associated enterprise in a DTA state are not
consistent with the arm’s length principle. Under such circumstances, the IRD has the right to adjust upwards the profits of
the Hong Kong enterprise to restore an arm’s length position. It should be noted that the term “associated enterprises” is
widely defined in the DTAs, and no threshold is prescribed.

Relief for double taxation under DTAs – Under Article 9 of Hong Kong’s DTAs, if an associated enterprise in a DTA state
has been subject to a transfer pricing adjustment, a request can be made to the IRD to make “appropriate adjustments” to
a Hong Kong enterprise’s profits to eliminate double taxation. As stated in DIPN 45, the IRD holds the view that an
appropriate adjustment will be made only if it agrees with the adjustment in principle and with the amount. Hence, in cases
involving requests for relief under Article 9(2) of a DTA, the IRD will consider making an appropriate adjustment to the
extent the primary adjustment was made in the other DTA state to correct non-arm’s-length transactions, and the amount
of adjustment satisfies the arm’s length principle.

Counteracting tax benefits under IRO Section 61A – According to the recent Court of Final Appeal decision in Ngai Lik,
the IRD is empowered under Section 61A, the anti-avoidance provision, to impose transfer pricing adjustments to
counteract the tax benefits obtained by a taxpayer through non-arm’s-length transactions, when tax avoidance is the “sole
or dominant purpose.”

DIPN 46 includes extensive discussion of tax schemes aiming at tax evasion through related-party transactions without any
commercial reason. The IRD makes it clear that if transfer pricing is used to siphon profits to offshore companies, it will
invoke Section 61A to combat the tax benefit obtained. Moreover, if dishonesty or willful intent is involved in such tax
schemes, the IRD will impose penalties under the IRO of up to 300 percent of any tax underpaid.





7 Double Tax Agreements or Double Tax Arrangements. For a discussion on DIPN 45, please see our Transfer Pricing Alert
titled “Hong Kong Issues Guidance on Relief From Double Taxation Due to Transfer Pricing Adjustments,” issued on 13 May
2009.
8 Inland Revenue Rule 5 deals with the profit tax assessment of a Hong Kong PE (including the Hong Kong branch of a non-
resident enterprise) and sets out the manner in which the taxable profit is ascertained for Hong Kong tax purposes.

Arm’s Length Standard
8 of 20
Copyright ©2009, Deloitte Touche Tohmatsu.
December 2009/January 2010

All rights reserved.

While Section 61A is potentially applicable to any intercompany transaction involving a Hong Kong enterprise, whether
cross-border or domestic, the IRD would still need to show that tax avoidance is the “sole or dominant” purpose of the
transaction. This requirement may limit the potential applicability of Section 61A in practice, as one cannot simply presume
that cases involving non-arm’s-length transaction are for the sole or dominant purpose of tax avoidance.

Authority under other IRO sections – The IRD considers that it also has the authority to enforce the arm’s length principle
under Sections 16(1) and 17(1)(b) of the IRO. Generally, Section 16(1) permits the deduction of outgoings and expenses “to
the extent” they are incurred in the production of chargeable profits; Section 17(1)(b) disallows expenses not for the
purpose of producing chargeable profits. The IRD’s position in DIPN 46 is that it has the authority under these two
provisions to disallow non-arm’s-length payments to an associated enterprise on the grounds that such payments are not
made for purposes of the taxpayer’s trade, but rather for the recipient’s trade.

This position may be controversial. In the Ngai Lik decision, Justice Ribeiro stated that Sections 16(1) and 17(1)(b) do not
require the commissioner to compare amounts deducted against market prices and to disallow deductions considered
excessive. However, this statement was made as dicta, and stopped short of an authoritative determination, as the IRD had
proceeded with the case solely on the basis of section 61A.

IRO Section 20(2) – DIPN 46 indicates that profits transferred to a “closely connected” nonresident enterprise in a transfer
pricing scheme can be assessed under Section 20(2) of the IRO. Section 20(2) is the existing provision in the IRO that
specifically addresses transfer pricing. Under this provision, if a Hong Kong enterprise and a related nonresident enterprise
enter into non-arm’s-length transactions that result in less than “ordinary profits” for the Hong Kong enterprise, the IRD
may treat the nonresident enterprise as carrying on a trade or business in Hong Kong and assess tax with respect to the
related-party transactions in the name of the Hong Kong enterprise as an agent of the nonresident.

While Section 20(2) may be potentially powerful in combating non-arm’s length-transactions, the IRD has seldom invoked
this provision in practice because of its unusual remedy of imposing tax on the nonresident enterprise. DIPN 46 does not
discuss in what way and to what extent the IRD will invoke this provision; thus, it remains to be seen whether the IRD will
seek to combat transfer pricing issues using Section 20(2) going forward.

Arm’s length principle in the OECD Guidelines

DIPN 46 states that the IRD will seek to apply the principles in the OECD Transfer Pricing Guidelines for Multinational
Enterprise and Tax Administrations (the OECD Guidelines), except when they are incompatible with the provisions of the
IRO. However, no situations in which the IRO may conflict with the OECD Guidelines are identified or discussed.

Some of the transfer pricing principles discussed in DIPN 46 include:

• OECD transfer pricing methods – DIPN 46 adopts the traditional transaction-based transfer pricing methods,
(comparable uncontrolled price (CUP) method, resale price method (RPM), cost plus method) as well as the profit-
based methods (transactional net margin method (TNMM), and profit split method) provided in the OECD
Guidelines. The appendices to DIPN 46 provide a short summary of each method with illustrative examples.
• Arm’s length ranges – DIPN 46 provides that the use of ranges, such as an interquartile range, would be accepted
in the determination of an arm’s length price. Interestingly, Mainland China does not fully recognize arm’s length
ranges, and transfer pricing adjustments will usually be imposed in a transfer pricing audit to bring the profits of a
Chinese enterprise to the median point, even if its profits are within the arm’s length range but below the median
point. It is not clear how this difference in the application of the arm’s length principle between Hong Kong and
China would be resolved under Article 9(2) or the mutual agreement procedure of the Hong Kong-China DTA.
• Multiple-year analysis under the TNMM – When applying the TNMM, multiple-year data should be used for both
the tested party and the comparables.
• No strict priority of methods – In contrast to the current version of the OECD Guidelines, the TNMM and the
profit split method are not described as methods of last resort. Rather, DIPN 46 provides that the selection of a
transfer pricing method should aim to find the most appropriate method for a particular case, taking into account
the comparability analysis and the availability of information. However it does mention that when both a
transaction-based method and a profit-based method can be applied in an equally reliable manner, the
transaction-based method is preferred.

Arm’s Length Standard
9 of 20
Copyright ©2009, Deloitte Touche Tohmatsu.
December 2009/January 2010

All rights reserved.

• Use of unspecified methods – Enterprises may apply methods not described in the OECD Guidelines (“other
methods”) to establish arm’s length transfer prices, if the traditional transfer pricing methods are not appropriate
to the facts and circumstances of a taxpayer. However, if other methods are used, their selection should be
supported by documentation that explains why the traditional methods were regarded as less appropriate than the
other methods.
• True-up adjustments – In describing the implementation of the arm’s length principle, DIPN 46 states that
enterprises should have “a review process to ensure adjustment for material changes.” This would appear to allow
the practice of current year true-up/down adjustments to reach target levels of profitability established in
benchmarking studies. If this is the case, it would help to clarify the IRD’s restriction on retroactive transfer pricing
adjustments relating to a closed year, as stipulated in DIPN 45.

Treatment of losses

Following the arm’s length principle, DIPN 46 does not specifically restrict the reporting of losses from related-party
transactions. However, the IRD indicates that the arm’s length principle would not be satisfied if an enterprise assumes risks
or functions without an adequate reward, or incurs costs or losses that benefit another group member. Based on DIPN 46,
continued losses from related-party transactions may not be acceptable unless there is contemporaneous documentation
explaining the market penetration strategy, and appropriate future rewards will be allocated to an enterprise incurring the
temporary losses.

Transfer pricing versus source of profits

While DIPN 46 endorses the notion that significant functions and key entrepreneurial risk-taking functions are the key
factors in determining the arm’s length return to an enterprise, it also emphasizes that the sourcing principle will be used
primarily to determine whether or not profit is taxable in Hong Kong. Once it is concluded that the profits of an enterprise
are sourced in Hong Kong, the IRD will not accept further apportionment of profits using transfer pricing principles, despite
the fact that the enterprise may have some significant functions located outside Hong Kong.

The only circumstance in which the IRD is prepared to make a transfer pricing adjustment to reduce the profits of a Hong
Kong enterprise with Hong Kong-source profits is under Article 9(2) of a DTA,, when a primary adjustment has been made
to an associated enterprise by a contracting state.

Intragroup services

DIPN 46 generally follows the OECD Guidelines in determining whether intragroup services have been rendered and
whether the charges are at arm’s length. Service fees may be determined by reference to direct charges that are based on a
specific service, as well as indirect charges that are based on allocations.

In determining whether a profit element is to be included in a service fee, DIPN 46 actually goes beyond the OECD
Guidelines to provide that a profit element should be included if the service provider is in the principal business of rendering
services, or is providing similar services to related and unrelated parties. On the other hand, no markup should be required
when an enterprise renders services merely as part of its general management activity, or when a PE purchases goods for
the enterprise not in the normal course of business activity.

Although the IRD has accepted markups of 5 percent/10 percent on the provision of routine back-office services to group
members in some advance ruling cases, DIPN 46 does not provide “safe harbors” that would assist in determining generally
acceptable markup fees for routine service transactions. Hence, under DIPN 46, a taxpayer cannot be certain that a 5
percent or 10 percent markup would be regarded as appropriate in the absence of a benchmarking study.

Transfer pricing documentation

DIPN 46 provides that transfer pricing documentation is not mandatory under Section 51C, but indicates that the IRD may
call upon enterprises to justify their transfer pricing in an enquiry, audit, or investigation. The extent of information and
documentation required by the IRD would depend on the size and complexity of the business or transaction in question. For
instance, while a brief functional analysis may be appropriate in cases in which transactions involve quoted markets for

Arm’s Length Standard
10 of 20
Copyright ©2009, Deloitte Touche Tohmatsu.
December 2009/January 2010

All rights reserved.

Download
Arm's Length Standard

 

 

Your download will begin in a moment.
If it doesn't, click here to try again.

Share Arm's Length Standard to:

Insert your wordpress URL:

example:

http://myblog.wordpress.com/
or
http://myblog.com/

Share Arm's Length Standard as:

From:

To:

Share Arm's Length Standard.

Enter two words as shown below. If you cannot read the words, click the refresh icon.

loading

Share Arm's Length Standard as:

Copy html code above and paste to your web page.

loading