IN THE HIGH COURT OF DELHI AT NEW DELHI
% Judgment delivered on: 22 .07.2016
+ ITA 381/2013
SUMITOMO CORPORATION INDIA PVT. LTD.
THROUGH MASAHIRO MARUYAMA ..... Appellant
versus
COMMISSIONER OF INCOME TAX ..... Respondent
WITH
+ ITA 738/2015
SUMITOMO CORPORATION INDIA PVT. LTD.
THROUGH MR. MITSUTAKA YASUDA ..... Appellant
versus
COMMISSIONER OF INCOME TAX ..... Respondent
WITH
+ ITA 382/2013
SUMITOMO CORPORATION INDIA PVT. LTD.
THROUGH MASAHIRO MARUYAMA ..... Appellant
versus
COMMISSIONER OF INCOME TAX ..... Respondent
AND
+ ITA 702/2014
SUMITOMO CORPORATION INDIA PVT. LTD.
THROUGH MR. MITSUTAKA YASUDA ..... Appellant
versus
COMMISSIONER OF INCOME TAX ..... Respondent
ITA 381/2013 & Connected Matters Page 1 of 32
Advocates who appeared in these cases:
For the Appellants : Mr. C.S. Aggarwal, Senior Advocate With Mr.
Prakash Kumar and Mr. Himanshu Sinha,
Advocates.
For the Respondents : Mr. Ashok K Manchanda, Senior Standing
counsel.
CORAM:
JUSTICE S.MURALIDHAR
JUSTICE VIBHU BAKHRU
JUDGMENT
VIBHU BAKHRU, J
1. The Assessee has preferred these appeals under Section 260A of the
Income Tax Act, 1961 (hereafter ,,the Act) impugning the orders passed by
the Income Tax Appellate Tribunal (hereafter 'the Tribunal) in respect of
Assessment Years (hereafter ,,AYs) 2007-08, 2008-09, 2009-10 and 2010-
11.
2. The controversy in these appeals relates to the Transfer Pricing
Adjustments directed by the Tribunal in respect of the commission earned
by the Assessee with respect to certain international transactions with its
Associated Enterprises (hereafter 'AEs') which are referred to as "indenting
transactions". The Tribunal has directed that the Arms Length Price
(hereafter ,,ALP) in respect of such transactions be determined on the basis
of the average rate of commission earned by the Assessee in respect of
ITA 381/2013 & Connected Matters Page 2 of 32
transactions with unrelated parties ('Non-AEs'). The Assessee claims that
the said direction is patently erroneous as the indenting transactions with
Non-AEs are not comparable with indenting transactions with its AEs; the
volume of the indenting transactions with Non-AEs is only a small fraction
of such transactions and the concerned products are also different. It is also
the case of the Assessee that the Tribunal has not followed any particular
method in directing the determination of ALP and as such, the same is
wholly arbitrary. The Assessee urges that it is incumbent on the Tribunal to
first determine the most appropriate method for determining the ALP and,
thereafter, compute the ALP in conformity with the discipline of that
method.
3. The controversies involved in these appeals as well as the material
facts are similar. The questions of law framed in these appeals are also
identical and read as under:
"(1) Whether the Income tax Appellate Tribunal was
right in applying and computing arms length price with
associated enterprise on indenting transactions by
applying average rate of commission with non-associated
enterprise in spite of difference in the turnover and the
purported segments and no such correction/computation
on this account was made by the Transfer Pricing
Officer?
ITA 381/2013 & Connected Matters Page 3 of 32
(2) Whether the Income Tax Appellate Tribunal has
disregarded the assessees claim that they had followed
Transactional Net Margin Method? (This question will
include the submission of the appellant that the Transfer
Pricing Officers order does not adopt any specified
method)"
4. For the purposes of addressing the above questions, only the facts
obtaining in ITA No. 381 relevant to AY 2007-08 (ITA No.381/2013) are
referred herein.
5. The Assessee was incorporated in 1997 and has its offices in Delhi,
Mumbai and Chennai. The Assessee is a subsidiary of Sumitomo
Corporation Japan (hereafter the ,,SCJ) which is one of the largest general
trading companies (Sogo Shosha) of Japan. SCJ is the flagship company of
the Sumitomo group which is a large conglomerate of companies. The
Assessee has seven operating divisions and deals in various products.
6. The Assessee filed its return of income for AY 2007-08 on
25.10.2007 declaring a total income of Rs.15,35,40,749/-.
7. The Assessee reported the following international transactions for
FY 2006-07:-
ITA 381/2013 & Connected Matters Page 4 of 32
S. No. Type of International transaction Total value of
transaction (Rs.)
1. Purchase of goods 102,825,122
2. Sale of goods 1,294,774
3. Rendering of support services 304,525,711
4. Interest Earned 722,621
5. Services received 10,335,041
6. Reimbursement of expenses (payment) 628,502
7. Reimbursement of Expenses (receipts) 14,036,868
8. The Assessee claimed that the transactions of purchase and sale of
goods with its AEs were on principal to principal basis. And, the income
from rendering support services was in relation to transactions (referred to
as indenting transactions) where the Assessee only rendered assistance by
following up with the customers and the sale/purchase of goods was done
directly by the AE. The Assessing Officer (hereafter ,,the AO) made a
reference to the Transfer Pricing Officer (hereafter ,,TPO).
The Transfer Pricing (TP) approach of the Assessee
9. In terms of Section 92E of the Act read with Rule 10E of the Income
Tax Rules, 1962 (hereafter 'the Rules'), the Assessee furnished the transfer
ITA 381/2013 & Connected Matters Page 5 of 32
pricing report in respect of its international transactions with AEs. The
Assessee considered Transactional Net Margin Method (hereafter 'TNMM')
as the most appropriate method and selected the ratio of gross profit to
operating costs - Berry Ratio - as the Profit Level Indicator (hereafter 'PLI').
The Assessee computed its gross profit on trading transactions (sales on
principal to principal basis) by reducing the cost of sales from the aggregate
value of sales made to AE as well as Non-AEs. The gross profit on trading
segment so computed was then added to commission earned to compute the
total gross profits. This amount was taken as the numerator and was
divided by operating expenses to compute the Berry Ratio (the PLI selected
by the Assessee) at 1.79%
10. The Assessee claimed that its transaction with the AEs were on arms
length basis and supported this claim by the data relating to a set of 23
comparable companies. The weighted average arithmetic mean (adjusted
using data for financial years 2004-05, 2005-06 and 2006-07) of the PLI of
these comparable companies was computed at 1.18%.
Proceedings before the TPO/AO
11. The TPO noticed that the Assessee's transactions could be classified
into two types - "Indent sales" and "Proper sales". In respect of "Indent
ITA 381/2013 & Connected Matters Page 6 of 32
sales", the Assessee merely indents for the goods which are supplied
directly by the supplier to the purchaser; the Assessee only receives
commission on the value of the invoice or the quantity of goods supplied.
In case of "Proper sales", the Assessee purchases the goods and sells the
same. The purchases made are against confirmed orders and thus, the
transactions of purchase and sale are back to back. The Assessee acquires
the title to goods only for a brief moment; this is described as a "flash title".
Such sale transactions are on a profit margin.
12. The TPO examined the transfer pricing report submitted by the
Assessee and noticed that the PLI used by the Assessee did not take into
account the cost of sales. The TPO held that the use of such ratio (Berry
ratio) as the PLI was not permissible under Rule 10B(1)(e) of the Rules
which contained provisions for computation of ALP by TNMM.
According to the TPO, the TNMM could be applied only by determining
the net profit margin in relation to the costs incurred, sales affected or
assets employed. He reasoned that since the denominator used by the
Assessee for computing the Berry ratio excluded the cost of goods, the PLI
so worked out was not in accordance with Rule10B(1)(e) of the Rules. The
TPO further proceeded to hold that the Assessee had itself indicated that
ITA 381/2013 & Connected Matters Page 7 of 32
functions performed and risks undertaken in respect of the international
transactions of "Proper sales" and "Indent sales" were similar and,
accordingly, held that the indenting transactions with AE ought to be
compared with the trading transactions entered into by the Assessee with
Non-AEs.
13. He held that the commission earned by the Assessee ought to be
expressed as a percentage of FOB price of goods sourced through the
Assessee because the Assessee had played a major role in identifying
suppliers, support in after sales services, business promotion, etc. He
further held that the Assessee had assumed significant risks and the
commission/service income model did not account for the fair
compensation for the value addition made by the Assessee.
14. The TPO rejected the use of Berry Ratio as the PLI for several
reasons. First of all, he held that the Assessee had developed unique
intangibles like supply chain intangibles and human assets intangibles,
which according to him had resulted in huge commercial and strategic
advantage to the AE and had enhanced the profit potential of the AE. The
Berry Ratio could not be used in cases where the Assessee was using
valuable and unique intangibles. Secondly, he held that Berry Ratio was
ITA 381/2013 & Connected Matters Page 8 of 32
very sensitive to cost base and it was difficult to accurately compute the
cost base of comparable companies on the basis of the available data as
different companies had accounted for their costs differently.
15. The TPO finally concluded that the gross margin earned by the
Assessee in its trading segment with Non AEs - computed at 4.45% - ought
to be taken as the rate of commission on the FOB price of the goods
sourced through the Assessee in respect of indenting transactions with the
AEs. He, accordingly, computed the arms length commission income from
indenting transaction with AEs at Rs.85,68,44,783/- (being 4.45% of
Rs.19,25,49,38,946) and accordingly directed enhancement of Assessee's
income by an amount of Rs.55,26,16,748/- after reducing the commission
of Rs.30,42,28,035/- as declared by the Assessee.
16. Pursuant to the order dated 28th October, 2010 passed by the TPO,
the AO issued a draft assessment order on 23 rd December, 2010. The
Assessee filed objections against the draft assessment order before the
Dispute Resolution Panel (hereafter 'DRP') which were not accepted and
the DRP issued its directions on 27th September, 2011. The AO passed the
final assessment order on 25th October, 2011 pursuant to the DRPs
directions.
ITA 381/2013 & Connected Matters Page 9 of 32
Proceedings before the Tribunal
17. Aggrieved by the final assessment order, the Assessee preferred an
appeal before the Tribunal urging several grounds. The Assessee, inter
alia, contended that the functions performed and the risks undertaken in
respect of principal to principal transactions with Non-AEs were not similar
to the indent based transactions with AEs and that the TPO had erred in
proceeding on the basis that the said transactions were comparable. The
Assessee claimed that in respect of indent transactions, the credit risks and
foreign exchange fluctuation risks were negligent and the Assessee's
function was merely to follow up on behalf of the customers and not to deal
with them.
18. The Assessee further claimed that the TPO had "erred in comparing
indent based transactions of AEs with principal based transactions of Non-
AEs and not with indent transaction of Non-AE after allowing appropriate
adjustments" and, therefore, the addition was misconceived, misplaced and
unsustainable.
ITA 381/2013 & Connected Matters Page 10 of 32
19. The Assessee contended that the TPO/DRP had erred in disregarding
the transfer pricing approach adopted by the Assessee for determining the
ALP of its international transactions. It was urged that the Assessee's use of
TNMM with Berry Ratio as the PLI had been discarded without any valid
justification.
20. The Tribunal referred to the tabular statement wherein the TPO had
computed the gross profit margin from trading transactions with AEs at
4.80%; gross profit margin on trading transactions with Non-AEs at 4.45%;
and commission earned at 1.61%. The Tribunal also referred to the order
of the TPO wherein he had referred to the Assessee's letter dated 19 th
October, 2010 in which the Assessee had bifurcated the commission earned
between commission from AEs and Non-AEs; the commission from Non-
AEs was declared as 2.26% on value of goods while the commission from
AE transactions was computed at 1.58%.
21. The Tribunal accepted the Assessee's contention that the nature of
indenting transactions were different from trading transactions. The trading
transactions involved certain risks and finances whereas in respect of
indenting transactions, the Assessee did not incur any financial obligation
or carry any significant risks. The Tribunal found that the indent business
ITA 381/2013 & Connected Matters Page 11 of 32
of the Assessee was nothing but trade facilitation, both in form as well as in
substance. It further noted that there was no material on record to regard the
indent transactions as trading transaction. The Tribunal further proceeded to
note and accept the Assessee's contention that it would be appropriate to
compare commission/service income earned by the Assessee in respect of
transactions with AEs with the similar transactions with Non-AEs.
However, the Tribunal rejected the Assessee's claim for an appropriate
adjustment on account of difference in volumes as well as the associated
risks. The Tribunal held that in the facts and circumstances of the case, no
adjustment as to the extent of volume was necessary as the Assessee had
entered into separate contracts for each transaction and it was not the
Assessee's case that each of such separate transaction with an AE was
greater in volume as compared to a similar transaction in the Non-AE
segment. The Tribunal then proceeded to direct that the commission
computed at the rate of 2.26% (i.e. the rate of commission in respect of
transactions with Non-AEs) be taken as the bench mark for determining
the ALP for commission earned in the AE segment.
22. The Assessee has impugned the above decision before us in ITA
381/2013.
ITA 381/2013 & Connected Matters Page 12 of 32
Submissions
23. Mr C.S. Aggarwal, learned Senior Advocate, appearing for the
Assessee contended that the fundamental issue in these appeals related to
the determination of the most appropriate method to be adopted for
determining the ALP. He submitted that the Assessee in its TP studies
found TNMM to be the most appropriate method for determination of the
ALP and this method was also accepted in the preceding years. However,
the TPO had rejected the same and made the ALP adjustment without
reference to any particular method. He contended that the Tribunal also fell
in error in making an adjustment without reference to any particular
method.
24. Mr Aggarwal submitted that even if the TPO or the Tribunal found
that Berry ratio was not an acceptable PLI, the said authorities could have
substituted the same with an appropriate PLI but could not have rejected the
TNMM as that was accepted as the most appropriate method in the
preceding years, that is, 2003-04 to 2006-07. He contended that there was
no material difference in the business model of the Assessee and, thus,
there was no reason for the TPO to depart from the method adopted in the
preceding years.
ITA 381/2013 & Connected Matters Page 13 of 32
25. Next, Mr Aggarwal contended that the Tribunal had erred in
accepting the Assessee's submission that the commission from Non-AEs be
compared with the commission from AEs for two reasons. First of all, this
was a submission in the alternative; and secondly, this proposition was
coupled with a claim to make an economic adjustment on account of
volume and the difference in products. He further submitted that the
products in respect of which commission was earned in the Non-AE
segment were different from the special products in the AEs segment. He
handed over a tabular statement in support of his contention. This statement
indicated the products in respect of which indenting transactions were
entered into in the Non-AE and the AE segments. He pointed out that in
respect of products classified under the automotive, chemicals (plastic),
construction, machinery, minerals and energy, power, steel pipes, etc.
divisions, there were no indenting transactions in the Non-AE segment and,
therefore, there would be no comparable standards on the basis of which an
ALP adjustment could be made.
26. Mr Manchanda, learned Senior Standing Counsel countered the
submissions made by Mr Aggarwal. He earnestly contended that the
Tribunal had applied the Comparable Uncontrolled Price (CUP) Method
ITA 381/2013 & Connected Matters Page 14 of 32
and had used an internal comparable transaction, which was from the same
company and the same industry, for determining the ALP. He submitted
that this was the most reliable method for computing the ALP and the
Assessee could not be heard to dispute the same. He earnestly contended
that the above method was proposed by the Assessee with certain economic
adjustments which, the Tribunal found were not justified. He submitted that
the Assessee had canvassed an adjustment on account of volume of
transactions between AE and Non-AEs segment and in support of its claim
had produced the brokerage rates for transactions in the security markets.
The Tribunal had rejected the same as it found that each transaction was a
separate transaction in terms of a separate contract and, therefore, any
discount on volumes was not warranted.
27. With respect to the contention that the products dealt in the AE
segment were different from those in the Non-AEs segment, Mr
Manchanda, contended that there were only two product categories,
namely, telecom and transport where there were no comparables in the
Non-AEs segment. He urged that the commission earned by the Assessee
in respect of these two segments was at the rate in excess of 5% which was
above the average rate of commission of 2.26% in the Non-AE segment. He
ITA 381/2013 & Connected Matters Page 15 of 32
argued that if the said two product categories were excluded, the ALP
adjustment would increase. He further submitted that in order for the
Assessee to make good his claim that a comparison between AE segment
and Non-AEs segment was to be made product wise, the Assessee was
required to produce every contract/agreement and invoice in the two
segments which the Assessee had failed to do.
28. Mr Manchanda further sought to contend that the Assessee had also
earned service fee for services which had been excluded while calculating
the commission earned in the Non-AE segment. He contended that if such
fees was included in the commission earned in the Non-AE segment, the
profit margin calculated in respect of such segment would increase
substantially from 2.26% and would result in a higher ALP adjustment.
Reasoning and Conclusion
29. Before proceeding to address the issues, it would be relevant to refer
to Section 92C of the Act. The relevant extracts of which are set out
below:-
"92C. (1) The arm's length price in relation to an international
transaction or specified domestic transaction shall be
determined by any of the following methods, being the most
ITA 381/2013 & Connected Matters Page 16 of 32
appropriate method, having regard to the nature of transaction
or class of transaction or class of associated persons or
functions performed by such persons or such other relevant
factors as the Board may prescribe, namely :--
(a) comparable uncontrolled price method;
(b) resale price method;
(c) cost plus method;
(d) profit split method;
(e) transactional net margin method;
(f) such other method as may be prescribed by the Board.
(2) The most appropriate method referred to in sub-section (1)
shall be applied, for determination of arm's length price, in the
manner as may be prescribed :
Provided that where more than one price is determined by the
most appropriate method, the arm's length price shall be taken
to be the arithmetical mean of such prices:
Provided further that if the variation between the arm's length
price so determined and price at which the international
transaction or specified domestic transaction has actually been
undertaken does not exceed such percentage not exceeding
three per cent of the latter, as may be notified by the Central
Government in the Official Gazette in this behalf, the price at
which the international transaction or specified domestic
transaction has actually been undertaken shall be deemed to be
the arm's length price:
Provided also that where more than one price is determined by
the most appropriate method, the arm's length price in relation
to an international transaction or specified domestic transaction
undertaken on or after the 1st day of April, 2014, shall be
computed in such manner as may be prescribed and
accordingly the first and second proviso shall not apply.
ITA 381/2013 & Connected Matters Page 17 of 32
Explanation.--For the removal of doubts, it is hereby clarified
that the provisions of the second proviso shall also be
applicable to all assessment or reassessment proceedings
pending before an Assessing Officer as on the 1st day of
October, 2009."
30. It is apparent from the above that ALP has to be computed by the
most appropriate method as is referred to in Section 92C(1). Sub-rule (1)
of Rule 10C of the Rules postulates that the most appropriate method would
be one which is best suited to the facts and circumstances of each particular
international transaction and which provides the most reliable measure of
an ALP in relation to that transaction. It is, thus, necessary that before an
exercise is undertaken for making an ALP adjustment, the Assessee/TPO
must identify the most appropriate method for computation of ALP. Sub-
rule (2) of Section 10C of the Rules that postulates that the following
factors shall be taken into account for selecting the most appropriate
method:-
"(2) In selecting the most appropriate method as specified in
sub-rule (1), the following factors shall be taken into account,
namely:--
(a) the nature and class of the international transaction
[or the specified domestic transaction];
(b) the class or classes of associated enterprises entering
into the transaction and the functions performed by
ITA 381/2013 & Connected Matters Page 18 of 32
them taking into account assets employed or to be
employed and risks assumed by such enterprises;
(c) the availability, coverage and reliability of data
necessary for application of the method;
(d) the degree of comparability existing between the
international transaction [or the specified domestic
transaction] and the uncontrolled transaction and
between the enterprises entering into such
transactions;
(e) the extent to which reliable and accurate adjustments
can be made to account for differences, if any,
between the international transaction [or the specified
domestic transaction] and the comparable
uncontrolled transaction or between the enterprises
entering into such transactions;
(f) the nature, extent and reliability of assumptions
required to be made in application of a method."
31. The Assessee had, for reasons indicated in its transfer pricing report,
adopted TNMM as the most appropriate method with Berry ratio as the
PLI. Although, the TPO found fault in the use of Berry ratio - according to
him, the same was not permissible under Rule 10B(1)(e) of the Rules - he
did not proceed to select the most appropriate method for computation of
ALP. This, in our view, would be essential as the reliability of the
determination of the ALP is in turn dependent on the effectiveness of the
method in relation to the controlled transaction being tested. In the present
ITA 381/2013 & Connected Matters Page 19 of 32
case, the dispute essentially relates to the commission earned by the
Assessee in respect of transaction with its AEs.
32. We are inclined to accept Mr Aggarwal's contention that although the
TPO had discarded the method adopted by the Assessee, it had not
followed any particular method in making the ALP adjustment. It appears
that the TPO has adopted a hybrid method. He imputed the character of
trading transactions to the indenting transactions entered into by the
Assessee with its AEs. Having done so, he compared the profit margin
realized by the AE from such transactions with profit margin realized by
the AE from a comparable uncontrolled transaction. The said approach was
rejected by the Tribunal - and, in our view, rightly so - as it was not
permissible for TPO to re-characterize the tested transaction.
33. We find no infirmity with the Tribunal's finding that indenting
transactions reported by the Assessee were plainly in the nature of
facilitating trade where the Assessee was required to do nothing more than
to follow up the customers for facilitation of the transaction. The Assessee
was not required to raise any invoice for sale and purchase and its financial
commitment and risk were inconsiderable.
ITA 381/2013 & Connected Matters Page 20 of 32
34. However, we find that the Tribunal erred in proceeding to determine
the ALP on the basis of the rate of commission reported by the Assessee in
respect of indenting transactions with Non-AEs, without further
examination as to the similarity between the two transactions. The Tribunal
effectively used the CUP Method for imputing the ALP of Assessee's
indenting transaction with AEs. This may well be the most appropriate
method to be used for determining the ALP. However, if the Tribunal
thought that this was the case, it was necessary for the Tribunal to conduct
a further in-depth inquiry as to the relevant uncontrolled transactions. It is
well settled that in applying the CUP Method, a very high degree of
similarity between the controlled and uncontrolled transactions is required.
It is the Assessee's case that volume of such transactions in the Non-AEs
segment was insignificant as compared to the transactions in the AE
segment against such transactions were only in a few product categories. If
the average rate of commission on such transactions was to be applied to
the FOB value of the goods involved in the indenting transactions with
AEs, the Tribunal would have to satisfy itself that there is no significant
variation in the rate of commission between different products. This would
confirm that the dissimilarity between the product categories did not have a
ITA 381/2013 & Connected Matters Page 21 of 32
vital bearing on the rate of commission. The Tribunal did not conduct any
such enquiry and it is material to note that the TPO also did not conduct
any such exercise. In our view, this methodology was used by the Tribunal
at a stage at which - given the extent of the examination required - it may
not be feasible.
35. One of the principal issues before the Tribunal concerned the
applicability of TNMM with Berry ratio as the PLI, as the most appropriate
method. Mr Aggarwal had sought to contend before us that the TPO had
rejected the PLI of Berry ratio but had not rejected the TNMM as the most
appropriate method and, therefore, it was incumbent upon him to replace
the PLI with whichever ratio he considered appropriate as had been done in
the preceding years. He contended that on principles of consistency, he was
required to follow the TNMM method. There is much merit in the
contention that a method once considered appropriate should be
consistently applied unless for good reasons, the TPO decides otherwise.
However, this is a salutary guiding principle and would not fetter the TPO
from independently examining the transfer pricing approach reported by the
Assessee. The purpose of imputing ALP to international transactions is to
ensure that the real income of the Assessee in respect of international
ITA 381/2013 & Connected Matters Page 22 of 32
transactions (and with effect from 1st April, 2013 certain domestic
transactions) are charged to tax under the Act. It is thus, implicit that the
exercise to determine such income be undertaken for each assessment year.
36. The special provisions for assessing income from international
transactions having regard to ALP is of a recent vintage and was introduced
by the Finance Act, 2001. The provisions under Chapter X of the Act have
undergone significant changes over a period of time. The principles for
computation of ALP are also evolving and as such, we are not persuaded to
accept that the TPO was required to simply follow the transfer pricing
methodology adopted in the preceding years. It is also well settled that
principles of res judicata do not apply in assessment proceedings as
assessment for each year is a separate proceeding and inquiry into the ALP
in respect of international transactions under Section 92 of the Act is in aid
of assessing the income chargeable to tax for the year under consideration.
37. We may now also consider Mr Aggarwal's contention that Berry
ratio had been accepted as the appropriate PLI in respect of Sogo Shosha
establishments and, therefore, the same should also be accepted in the case
of the Assessee. The term ,,Sogo Shosha' is used in respect of large general
trading companies that include within their fold a large network of
ITA 381/2013 & Connected Matters Page 23 of 32
subsidiary and affiliated companies, thus, enabling the said companies to
leverage their network for their business. It is reported that these companies
account for a substantial portion of the Japan's overall trade across the
world. However, it is not necessary that the trading arrangement between
Sogo Shosha enterprises and their affiliates/ subsidiaries in India be
identical or similar. It is also not possible to assume - without it being
established as a fact - that all international transactions entered into by
Indian enterprises with their related Sogo Shosha enterprises would be on
identical footing. Thus, it is not apposite to determine the ALP without
examining the nature of international transaction in each case.
38. Insofar as the use of Berry ratio as a PLI is concerned, the TPO had
rejected the same for three reasons. First of all, he held that the same is not
permissible under Rule 10B(1)(e) of the Rules; secondly, he held that the
Assessee had acquired substantial intangibles in the form of supply chain
intangibles and human resources intangibles and Berry ratio was not an
apposite PLI in cases where an Assessee used substantial intangibles for its
business. Thirdly, the TPO held that the rate of commission on indenting
transaction was determined in reference to the value of goods and not on
the basis of any cost incurred by the Assessee.
ITA 381/2013 & Connected Matters Page 24 of 32
39. In our view, the decision of the TPO that Rule 10B(1)(e) of the Rules
does not permit use of Berry ratio is not sustainable. The TPO had reasoned
that under Rule 10B(1)(e) of the Rules, only the total costs incurred, sales
effected or assets employed could be used as denominator of the ratio
chosen as the PLI. This is plainly erroneous. Rule 10B(1)(e)(i) of the Rules
reads as under:-
"10(e) transactional net margin method, by which,--
(i) the net profit margin realised by the enterprise from an
international transaction or a specified domestic
transaction entered into with an associated enterprise is
computed in relation to costs incurred or sales effected or
assets employed or to be employed by the enterprise or having
regard to any other relevant base;"
40. It is clear from the plain language of the above quoted clauses that
the net profit margin realised could be computed having regard to "any
other relevant base". Berry ratio is a ratio of operating profits to operating
expenses. In cases where operating expenses is considered as a relevant
base, there would be no difficulty in using Berry ratio as the PLI in terms of
Rule 10B(1)(e)(i) of the Rules.
ITA 381/2013 & Connected Matters Page 25 of 32
41. Insofar as the other two reasons are concerned, it is necessary to
understand the substratal rationale of using Berry ratio as the PLI. The said
ratio was used by the Internal Revenue Service (IRS) in the USA in the
case of E.I. Du Pont DE Nemours & Co. v. United States: 608 F.2d 445
(1979) to sustain their stand that substantial part of the profits of a
subsidiary in Switzerland were rightly allocated to Du Pont. Charles H.
Berry, an economist (since deceased) provided necessary evidence in
support of IRS's stand by using the ratio of Operating Profits to Selling,
General and Administration Expenses to show that more than fair share of
profits had been transferred to Du Pont's Swiss subsidiary. This ratio came
to be known by the name of the economist who had used it in the
aforementioned case. The aforesaid case concerned the allocation of profits
between Du Pont De Nemours, an American Company engaged in
manufacture of chemicals, and its wholly owned subsidiary established in
Switzerland (Du Pont International S.A. referred to as 'DISA'). The said
subsidiary was established in 1959. At the material time, Section 482 of the
Internal Revenue Code empowered the Secretary of the Treasury (or his
delegate) to "distribute, apportion, or allocate gross income, deductions,
credits, or allowances between or among such organizations, trades, or
ITA 381/2013 & Connected Matters Page 26 of 32
businesses, if he determines that such distribution, apportionment, or
allocation is necessary in order to prevent evasion of taxes or clearly to
reflect the income of any of such organizations, trades, or businesses."
42. The Treasury Regulations as in force during the relevant period
contemplated determination of ALP for sale by one controlled entity to
another by four methods (in order of preference); comparable uncontrolled
price method; the resale price method; cost plus method; and any other
appropriate method. Du Pont sold its chemical products to DISA and also
arranged for resale of such chemicals to the legitimate consumers. The said
transactions resulted in DISA reporting a profit margin of 35%. Du Pont
submitted the set of 21 comparable entities which it claimed performed
functions similar to DISA. IRS, on the other hand, introduced evidence to
show that six of the companies identified by Du Pont as similar to DISA
had reported average selling cost which were much higher than DISA. The
Court agreed with the evidence produced by IRS's expert that what a
business spends to provide services would be a reasonable indication of the
magnitude of those services.
ITA 381/2013 & Connected Matters Page 27 of 32
43. Berry ratio was used in the above context to show that the average
ratio in the case of 21 distributors was 129.3% and in the case of DISA, it
was 281.5% for the year 1959 and 397.1% in the year 1960. It is relevant
to note that in that case also the Commissioner of Revenue did not
reallocate the profits between DISA and Du Pont on the basis of Berry ratio
but on the basis of applying the resale price method. Berry ratio was only
introduced as an expert evidence by IRS to defend the challenge to such
reallocation. In other words, the said ratio was only used to show that DISA
had made extraordinarily high profits and IRS had rightly exercised its
jurisdiction to reallocate such profits of DISA to Du Pont. The Du Pont's
challenge to the said ratio as being inappropriate measure was rejected by
the Court in the following words:-
"Whatever the general limits of any particular gauge of
industry profitability, plaintiff cannot escape the basic thrust
of defendant's proof. Defendant has shown that DISA made
extraordinarily high profits which the Commissioner
reallocated to an economically reasonable level."
44. Subsequently, in 1990, Berry ratio was included as an acceptable PLI
in certain circumstances under the Treasury Regulations in USA. OECD
Guidelines issued in July 2010 also accepted that Berry ratio to be apposite
in certain circumstances. More recently, Japan has also accepted use of
ITA 381/2013 & Connected Matters Page 28 of 32
Berry ratio for purposes of transfer pricing in certain circumstances in its
tax legislation reforms introduced in March 2013.
45. Traditionally, the denominator of the ratio only comprised of selling,
general and administration expenses. However, the Treasury Legislation of
USA also included depreciation as a part of the Operating Expenses used as
a denominator in the berry ratio. As is apparent, Berry ratio has limited
applicability; it can be used effectively only in cases where the value of
goods have no role to play in the profits earned by an Assessee and the
profits earned are directly linked with the operating expenditure incurred by
the Assessee. In other words, the operating expenditure incurred by the
Assessee effectively captures all functions performed and risks undertaken
by the Assessee. Thus, in cases where an Assessee uses intangibles as a
part of its business, Berry ratio would not be an apposite PLI as the value of
such tangibles would not be captured in the operating cost and, therefore, it
would not be appropriate to compute the ALP based on net profit margin
having regard to the operating cost as a relevant base. Similarly, Berry
ratio would not be an appropriate PLI for determining ALP in cases of
Assessees who have substantial fixed assets since the value added by such
assets would not be captured in Berry ratio.
ITA 381/2013 & Connected Matters Page 29 of 32
46. It can be seen from the above that the Berry ratio can be used only in
very limited circumstances and the limitations that we have listed above are
by no means exhaustive. There is also a view expressed that use of Berry
ratio as a PLI results in indicating less than fair ALPs in tax jurisdiction
where the Assessees have a lower bargaining power. In the aforesaid
context, in our view, the TPO had correctly reasoned that Berry ratio could
not be used as a PLI in cases of Assessees which were using intangibles.
However, we find that there was no cogent material for the TPO to hold
that the Assessee had developed supply chain and human resources
intangibles. In any event, there was no material to conclude that costs of
such intangibles were not captured in the operating expenses.
47. In our prima facie view, the third reason stated by the TPO, that is,
the rate of commission paid to the Assessee is based on the value of the
goods, would be a valid reason to reject the use of Berry ratio because
Berry ratio can only be applied where the value of the goods are not
directly linked to the quantum of profits and the profits are mainly
dependent on expenses incurred. The fundamental premise being that the
operating expenses adequately represent all functions performed and risks
undertaken. For this reason Berry ratio is effectively applied only in cases
ITA 381/2013 & Connected Matters Page 30 of 32
of stripped down distributors; that is, distributors that have no financial
exposure and risk in respect of the goods distributed by them.
48. In the present case, the Assessee asserts that its business comprises
of two segments, trading segment and indenting segment and the functional
risk and the reward in the two segments are different. In the trading
segment, the Assessee earns a higher profit margins (calculated on the
value of the goods traded) while in the indenting segment its profit margins
are lower. Plainly, the use Berry ratio would give unreliable results if the
product mix of the comparables is different from the product mix of the
Assessee. This would make the task of finding a set of comparables fairly
difficult.
49. In view of the above, the first question of law is answered in the
negative and the second question of law is answered in the affirmative; that
is, the questions are answered in favour of the Assessee and against the
Revenue. The impugned orders are accordingly set aside and the matters
are remanded back to the Tribunal to decide it afresh. It will be open for the
Tribunal to further remand the matter to the TPO/AO for a fresh
ITA 381/2013 & Connected Matters Page 31 of 32
examination of the issues relating to Transfer Pricing in accordance with
law. The parties are left to bear their own costs.
VIBHU BAKHRU, J
S.MURALIDHAR, J
JULY 22, 2016
MK/RK
ITA 381/2013 & Connected Matters Page 32 of 32
|