Indian Intercompany Loans Litigation, Journal of International Taxation, Jun 2013
... These rates were based on market rates that prevailed in India on financial ...
Indian Intercompany Loans Litigation, Journal of International Taxation, Jun 2013 Journal of International Taxation (WG&L)
Indian Intercompany Loans Litigation Author: J. HAROLD McClure J. Harold McClure is a Senior Manager with Thomson Reuters ONESOURCE Transfer Pricing. Dr. McClure has written previously for the Journal. [pg. 60] The Indian tax courts have wrestled with one legal issue and two economic issues regarding the extension of intercompany loans by Indian parent companies to their foreign affiliates. The decision by the Mumbai Bench of the Income Tax Appellate Tribunal (ITAT) in Tata Autocomp Systems Limited 1 is the most recent Indian court case involving interest-free loans. The legal question that the taxpayer raised was whether an affiliate can extend interest-free loans to a related party. The Indian courts also addressed the treatment of interest-free loans in four previous cases, discussed below. Tata Autocomp Systems Limited extended advances to its German affiliate, TACO Kunstsofftechnik GmbH, and chose not to charge interest on what it admitted was an intercompany loan. The taxpayer offered several rationales for not charging interest to the German affiliate, including the contention that the affiliate “was suffering such a severe financial crisis that its existence was threatened and it would not have been possible to recover interest.” The Indian tax court in this case, as well as in the others, held that TACO Kunstsofftechnik GmbH had to pay an arm's-length interest rate to Tata Autocomp Systems Limited, with the economic questions relating to what would be an arm's-length rate. Perot Systems TSI v. DCIT involved a similar issue where an Indian entity extended interest-free financing to its affiliates in Bermuda and Hungary. The Indian tax authorities said that the Indian entity should receive interest from these affiliates for the intercompany loans. The tax authorities also argued for a 4% interest rate based on the LIBOR rate plus a 1.64% premium. Similar issues were raised in three other cases:
Tech Mahindra Limited v. DCIT. 2
Siva Industries & Holdings Ltd. v. ACIT. 3
VVF Limited v. Deputy Commissioner of Income Tax. 4
The conflicting arguments between the various taxpayers and the Indian tax authorities present an opportunity to discuss the basic fundamentals involved in evaluating what would represent an arm's-length interest rate. Interest rates can be seen as being determined in a global market place, with differences in market-determined rates existing because loans have different characteristics with respect to the following:
Date of the loan.
Term or maturity of the loan.
Currency of denomination.
Borrower's credit standing.
As noted below, the conflicting arguments between the various taxpayers and the Indian tax authorities indicate [pg. 61] some apparent confusion regarding how to address the last two items on the list. A review of these cases should shed some light on how these economic issues are to be resolved. Attempts to justify a lack of intercompany interest on the ground that the related-party borrower is “suffering such a severe financial crisis that its existence was threatened and it would not have been possible to recover interest” is an invitation for the tax authorities to assert a poor credit standing and hence a high arm's-length rate.
Tata Autocomp—Evaluating the Interest Rate Charged In Tata Autocamp, the date of the deemed loan appears to be August 10, 2006, and the currency was the euro. While the ITAT did not specify the term of the loan, it seems that the ruling implicitly assumed a short-term loan. The Indian tax authorities initially proposed a 10.25% interest rate and later argued for a 12% rate. These rates were based on market rates that prevailed in India on financial instruments that were denominated in rupees. The Indian tax authorities argued that rates that prevailed in the lender's location would constitute the most appropriate application of the comparable uncontrolled price (CUP) method. The taxpayer submitted that the arm's-length rate should be 4.15%, based on a letter from a German bank dated August 10, 2006, suggesting that it would lend to the German affiliate at a rate equal to the Euribor rate (Euro Interbank Offered Rate) plus 0.8%. The ITAT accepted the taxpayer's position and rejected the Indian tax authorities' argument. Its decision is consistent with the proposition that it is the currency of denomination that matters and not the locality of either the related-party lender or related-party borrower. The German affiliate was in a position to undertake third-party loans in euros from any lender globally. German interest rates tended to be lower than Indian interest rates during this period in part because Germany had a much lower inflation rate than India. Thus, an entity that borrowed funds denominated in euros would not pay high interest rates simply because its lending affiliate received high interest rates on rupeedenominated loans. 5 The ITAT did not consider the borrower's credit standing. Its ultimate determination implicitly assumed that the German affiliate had a very good credit rating even though one of the taxpayer's rationales for not paying interest was its contention that the German affiliate was suffering such a severe financial crisis. Credit ratings and credit spreads are addressed below.
Other Cases and Currency of Denomination In Tech Mahindra, the taxpayer extended credit to its U.S. affiliates. Similar to the taxpayer's reasoning in Tata Autocomp Systems, Tech Mahindra argued that the U.S. affiliate should not pay interest to the Indian parent because the U.S. affiliate had liquidity problems. The Indian tax authority disagreed with this legal contention and argued for a 10% interest rate based on the intercompany rate that Tech Mahindra Limited had extended to its German affiliate. The Indian court rejected this position, as pricing controlled transactions are not seen as reliable evidence for arm's-length pricing. A KPMG newsletter offered one interpretation of what this ruling implied: 6 The Mumbai Tribunal held that the arm's length price in case of interest on extended credit period granted to an Associated Enterprise shall be determined on the basis of USD LIBOR and not on any other currency denominated loan rate....The Tribunal has also made a crucial point that the
arm's length interest rate should be taken from the country of the borrower/debtor, i.e., the rate of interest to be used for benchmarking shall be the rate of interest in respect of the currency in which the underlying transaction has taken place in consideration of economic and commercial factors around the specific currency denominated interest rate. While it is true that the intercompany loan from Tech Mahindra to its German affiliate was denominated in euros, the deemed intercompany loan at issue was a U.S. dollar-denominated loan to a U.S. affiliate. The KPMG newsletter and other discussions of how the Indian courts have treated this issue conflate the borrower's national location with the issue of currency denomination. The actual decision, however, is instructive as it cites the following from Siva Industries & Holdings Ltd. v. ACIT: The assessee has given the loan to the Associated Enterprises in US dollars. The assessee is also receiving interest from the Associated Enterprises in Indian rupees. Once the transaction between the assessee and the Associated Enterprises is in foreign currency and the transaction is an international transaction, then the transaction would have to be looked upon by applying the commercial principles in regard to international transaction. If this is so, then the domestic prime lending rate would have no applicability and the international rate fixed being LIBOR would come into play. In the circumstances, we are of the view that it [is the] LIBOR rate which has to be considered while determining the arm's length interest rate in respect of the transaction between the assessee and the Associated Enterprises. As it is noticed that the average of the LIBOR rate for 1.4.2005 to 31.3.2006 is 4.42% and the assessee has charged interest at 6% which is higher than the LIBOR rate, we are of the view that no addition on this count is liable to be made in the hands of the assessee. In other words, Siva clearly focused on the currency of denomination. That decision also posits that financial transactions occur in a global market place, which implies that the key comparability criteria are limited to the terms of the loan (date, maturity, and currency of [pg. 62] denomination) and the borrower's credit standing. The KPMG newsletter implicitly assumes that the currency of denomination follows the location of the borrower, but consider an example where a German entity extended a euro-denominated loan to its U.S. affiliate on August 10, 2006 (the date relevant for the Tata Autocomp litigation). On that date, the interest rate on ten-year U.S. government bonds was only 2.3%, while the interest rate on ten-year German government bonds was almost 4%. An appropriate analysis would use the German government bond interest rate since the loan was euro denominated even though the borrower was a U.S. affiliate.
Credit Spreads The evaluation of whether an intercompany interest rate is at arm's length can be reduced to whether the credit spread implied by the intercompany interest rate is consistent with the estimated credit rating for the related-party borrower. This point can be illustrated by analyzing market rates on August 10, 2006, which, as noted, was the deemed date for the Tata Autocomp litigation. At the time of the deemed loan, there was only a slight upward tilt to the German term structure. Interest rates on three-month German Treasury bills averaged 3.25% during August 2006, while interest rates on long-term German government bonds averaged 3.88%. 7 Euribor interest rates on August 10, 2006, rose with their term, with the interest rate for three-month Euribor at 3.1% and the interest rate for the 12-month Euribor at 3.6%. The taxpayer's position that 4.15% represented a 0.8% premium over Euribor is consistent with market rates for the sixmonth Euribor on August 10, 2006.
The premium over Euribor or any LIBOR-based interest rate may understate the credit spread if these rates exhibit what is often referred to as the TED 8 spread, which is usually defined as the difference between the three-month LIBOR rate and the three-month Treasury bill interest rate. A comparison of the average August 2006 three-month Euribor rate and the three-month German Treasury bill rate for the same month to determine if an analysis needs to consider a TED spread shows that during this period the two rates were virtually the same. Thus, it is reasonable to assume that the taxpayer's position was consistent with the German affiliate having a 0.8% credit spread. This credit spread would be consistent with the arm's-length standard only if the credit rating for the borrower was consistent with a high credit rating. As of August 10, 2006, the credit spread implicit for BBB corporate borrowers in the United States was 1.5%. 9 The taxpayer contended, however, that the German affiliate was suffering such a severe financial crisis that its existence was threatened and it would not have been possible to recover interest. The Indian tax authorities could have used this to argue for a significantly higher credit spread had the evidence suggested that the German affiliate had a poor credit rating, but they did not do so. The court upheld a 2% interest rate in Tech Mahindra, asserting that U.S. LIBOR plus 0.8% was reasonable. The court stated that some measure of U.S. LIBOR was 1.2% during 2004-2005, but was not specific as to the date or the term. The interest rate on three-month LIBOR was approximately 1.2% on April 2004, and the interest rate on three-month Treasury bills averaged 0.94% at that time. It is, therefore, reasonable to assert that this 2% deemed intercompany interest rate is consistent with a credit spread near 1%. As in the discussion above, because the taxpayer in Tech Mahindra asserted that its U.S. affiliate had liquidity problems, the Indian tax authorities could have argued for a significantly higher credit spread but they did not do so. In VVF Limited, the taxpayer had extended interest-free financing to its affiliates in Canada and Dubai. Similar to the previous cases, the Indian tax authorities successfully asserted that this intercompany financing represented loans that required an arm's-length interest payment. While the tax authorities argued for a 14% interest rate based on rupee-denominated loan rates, the Indian court ruled that the interest rate should be based on LIBOR plus a reasonable credit spread because the intercompany loans were denominated in dollars, not rupees. The taxpayer presented two potential means for determining the appropriate interest rate:
A Bank of India letter stating that it would be willing to extend loans to these affiliates at rates between LIBOR plus 1.5% and LIBOR plus 3%.
An actual loan from ICICI Bank to the taxpayer at a rate equal to LIBOR plus 3%.
The Indian court accepted the latter evidence as an actual third-party transaction that established the credit spread appropriate for the intercompany loans. A 3% credit spread is often observed when the borrower has a BB credit rating. When third-party loans comparable to the intercompany loan are not readily available, another way to determine the appropriate credit spread is to estimate the credit rating of the related-party [pg. 63] borrower, then observe the credit spreads for third-party borrowers with the same credit rating. Dhaivat Anjaria and Suchint Majmudar discuss this approach and present an illustration of its application. 10 Their example assumes a dollar-denominated loan issued on March 1, 2007, with a maturity just under four years. While the authors claim that their approach justifies a 5% interest rate, the evidence of market rates on that date suggests that the arm's-length rate should be significantly higher.
Their first step is to derive the credit spread that the intercompany interest rate implies by observing the market rate on U.S. government bonds of similar maturity on March 1, 2007. The authors claim that this rate was 4%, which would suggest a 1% implied credit spread. However, market rates for U.S. government bonds with maturities between three and five years were 4.5%, which suggests an implied credit spread equal to only 0.5%. A 0.5% credit spread would be consistent with an arm's-length price if the related-party borrower had a credit rating of AAA. The authors assume, however, that the credit rating for the relatedparty borrower is BB. They also argue that borrowers with credit spreads of BB would face a 1% credit spread, but this assumption is at odds with market evidence on credit spreads. As noted above, borrowers with a BB credit rating would tend to face credit spreads equal to at least 3%. Thus, the arm's-length interest rate for the hypothetical loan that they present would be at least 7.5%.
Conclusion The Indian courts have wrestled with both legal and economic issues regarding intercompany loans. On the question of whether the arm's-length interest rate should be based on rates denominated in the currency of the lender or the borrower, it is clear that the determination should be based on the currency of denomination of the intercompany loan. The Indian courts have sensibly addressed this issue, but have shown a mixed record on credit spreads. Section 10.3 of the United Nation's Practical Manual on Transfer Pricing for Developing Countries is entitled “Emerging Transfer Pricing Challenges in India.” 11 Section 10.3.10 on financial transactions claims that the Indian tax authorities follow "a quite sophisticated methodology" for pricing intercompany loans that revolves around:
Comparison of terms and conditions of the loan agreement.
Determination of the credit rating of lender and borrower.
Identification of comparable third-party loan agreement.
Suitable adjustments to enhance comparability.
This discussion also suggests that if the lending affiliate is an Indian entity, the currency of denomination should be the Indian rupee. As noted above, the currency of denomination does not necessarily follow the jurisdiction of either the lender or the borrower. While this discussion is correct to note that credit ratings are an important part of the analysis, the relevant issue is the credit rating of the borrower. Finally, the discussion argues for the use of the Indian Prime Lending Rate as an appropriate “external CUP.” Given that this interest rate is currently in excess of 14%, this argument does not seem credible if the related-party borrower has a good credit rating and the currency of denomination is not the Indian rupee. Taxpayers at times try to assert that no interest needs to be paid on an intercompany loan when the related-party borrow is financially distressed. This argument is dangerous not only because it is not persuasive, but also because it suggests that the related-party borrower should be deemed a poor credit risk. Third-party lenders in these situations would extend a loan to such a borrower only if the interest rate were sufficiently high to afford the lender a credit spread commensurate with the poor credit standing. While the general discussion for estimating credit spreads and arm's-length interest rates presented by Anjaria and Majmudar is reasonable, the actual application in their example is questionable. The Indian tax authorities have been invited by the decision in Aithent Technologies Pvt Ltd. v. ITO 12 to apply the approach that this article has outlined. In this litigation, the taxpayer tried to justify the lack of intercompany interest payments based on the claims that no external CUP
could be found to properly evaluate the intercompany interest rate, and that an analysis of the related-party borrower's profits based on the transactional net margin method demonstrated that the borrower did not have excessive profits in spite of its failure to pay interest on the loan. The Tribunal, however, disagreed: The assessee was required to comply with the transfer pricing provisions of [India Income Tax Act 1961] s. 92 to 92F with respect to the transaction of interest-free loan to its subsidiary. The CUP method is the most appropriate method in order to ascertain the ALP [arm's-length price] of such international transaction by taking into account prices at which similar transactions with other unrelated parties have been entered into. For that purpose, an assessment of the credit quality of the borrower and estimation of a credit rating, evaluation of the terms of the loan, e.g., period of loan, amount, currency, interest rate basis, and additional inputs such as convertibility and finally estimation of arm's length terms for the loan based upon the key comparability factors and internal and/or external comparable transactions are relevant. None of these inputs have anything to do with the costs; they only refer to prevailing prices in similar unrelated transactions instead of adopting the prices at which the transactions have been actually entered in such cases, the hypothetical arms length prices, at which these associated enterprises, but for their relationship, would have entered into the same transaction, are taken into account. Whether the funds are advanced out of interest bearing funds or interest free advances or are commercially expedient for the assessee or not, is wholly irrelevant in this context. As the transaction is of lending money, in foreign currency, to its foreign subsidiary, the comparable transaction should also be of foreign currency lending by unrelated parties. In other words, the Tribunal agreed that the evaluation should be based on what government bonds denominated in the foreign currency would be paying in terms of interest rates since the intercompany loan was denominated in the foreign currency. The Tribunal also noted that the evaluation must estimate the credit rating of the related-party borrower to evaluate the appropriate credit spread.
1
ITA No. 7354/MUM/11, April 30, 2012; http://indiankanoon.org/doc/133135055.
2
ITA No. 1176/Mum/2010, June 30, 2011; taxguru.in/income-tax-case-laws/arms-length-pricecase-interest-extended-credit-period-granted-enterprise-determined-basis-usd-libor-currencydenominated-loan-rate.html.
3
ITA No. 2148/Mds/2010, May 20, 2011.
4
ITA No. 673/Mum/06, January 8, 2010.
5
Uncovered interest rate parity—also known as the international Fisher effect—asserts that the
difference in interest between two nations is equal to the expected change in exchange rates between the nation's currencies. Relative purchasing power parity is an economic proposition that the expected change in exchange rates reflects differences in nations' expected inflation rates. These two economic propositions should be seen as first approximations that hold over the long run even though short-run deviations may occur. When a nation such as India experiences persistently higher inflation than nations such as Germany and the United States, its nominal interest rates will tend to be higher, reflecting the expectation that its currency will devalue.
6
KPMG Flash News, KPMG in India, July 6, 2011.
7
International Financial Statistics.
8
TED stands for T-bill and ED, the ticker symbol for Eurodollar futures contract.
9
www.federalreserve.gov reports that long-term corporate borrowers with BBB credit were paying interest at 6.66% while the interest rate on 20-year federal bonds was 5.15%.
10
“India: Intercompany Financial Transactions,” Tax Planning International Asia Pacific Focus, February 28, 2011.
11
The U.N. Committee of Experts on International Cooperation in Tax Matters, U.N. Practical Transfer Pricing Manual for Developing Countries, ch. 10—India; www.un.org/esa/ffd/tax/documents/bgrd_tp.htm. See Spencer, “BRICS, BEPS, and the U.N. Transfer Pricing Manual,” Part 1, 24 JOIT 28 (May 2013); Part 2, 24 JOIT 42 (June 2013). Part 3 of Spencer's article, in a forthcoming issue of the Journal, will cover the India section of the Manual.
12
2010-TII-134-ITAT-DEL-TP, January 12, 2011.
