China UN Prac-cal Manual on Transfer Pricing for Developing Countries Chapter 10.3 (May, 2013)

Transcription

1 China UN Prac-cal Manual on Transfer Pricing for Developing Countries Chapter 10.3 (May, 2013) Richard T. Ainsworth Director, Graduate Tax Program, BU School of Law October 24, 2014 Room 209

2 What has been Happening UN, Prac%cal Manual on Transfer Pricing for Developing Countries (May, 2013) UN intent is to show developing countries how to perform transfer pricing analysis effec-vely & efficiently budget & revenue issues for developing countries China believes it has some solu-ons (new approaches) Major Chinese TP Law changes (eff ) Prior law - two systems, domes-c business v MNEs Domes-c = 24.53% effec-ve tax rate MNE s = 14.89% effec-ve tax rate 1. Enterprise Income Tax Law [EIT] (March 16, 2007): Chapter 6: Special Tax Adjustments is on transfer pricing 2. Implementa-on Regula-ons for Special Tax Adjustments (SAT Rules) (December 6, 2007) 3. Circular 2 Guo Shui Fa [2009] No. 2 Implementa%on Measures of Special Tax Adjustments

3 China & OECD Prior to 2009 China substan-ally followed OECD OECD methods were follow in prac-ce: interna-onal standards Guo Shui Fa No. 2 (Circular 2) [January 8, 2009] takes a different approach to TP regula-on Instruc-ons to auditors on how to do TP audits (not direc-ons to Taxpayers on how to comply) Differences are now becoming clearer: Secret comparables = OK Guo Shui Han [2005] No. 239 Dependent comparables (no express prohibi-on) No capital intensity adjustments Guo Shui Han [2005] No. 745 Loca-on Specific Advantages UN Chapter Premium from Chinese Market UN Chapter Royal-es for Chinese R&D UN Chapter High- Technology status VAT & CIT UN Chapter Recent differences

4 New TP Documenta-on Requirements Ini-a-ng the nine filing forms in 2009 [EIT & Circular 2] Exempt, from TP study if below RMB20 million in intercompany transac-ons in a year. Middle group subject to TP study documenta-on (but not before a return) Above RMB200 million (related party gross purchase/sales) in a year need a full TP study in advance of filing a return But not if less than RMB 200m Purchase & Sales [and] RMB 40m in other related party transac-ons Excep-ons NOTE: toll manufacturers use customs prices

5 Basic Idea of Transfer Pricing Total Profit = $225 Costs $25 Parent 100% Sub. Sale Price $250 3 rd Party Profit $50 $100 What is this price? $75 $125 Profit $175 $125

6 Only Two Things in Transfer Pricing Related Par-es: The issue is control do you control the pricing? US economic control OECD tended to be more legal control (ownership), but now is closer to economic control China is China moving in another direc-on? (see next) Methods: Comparable Uncontrolled Price (CUP) Resale Price Method (RPM) Transac-onal Methods Cost- Plus Method (C+) Profit Split (P- S) Comparable Profits Method (CPM) Profit Methods Aka: Transac-onal Net Margin Method (TNMM) China accepts these methods (with changes)

7 Related Party (8 tests) EIT, Art (1) Ownership or control 25% equity ownership common stock; 3 rd party owns/controls 25% shares both (2) Senior management appointment Over 50% of senior management [or one execu-ve on board of directors] Two enterprises with more than 50% senior management appointed by same party (3) Debt One party holds debt equal to 50% or more of capital [other than a bank]; 10% of total debt is guaranteed by other party [other than a bank]. (4) All transac-ons between HQ and Chinese subsidiary (5) Licensing of intangibles Normal produc-on & opera-on are dependent on intangibles licensed from other (6) Raw materials/ Spare parts Raw materials/ spare parts are under the control of other party (7) Service provision or receipt determina-on One en-ty s provision or receipts of services is determined by other party (8) Actual control Actual control over another enterprise s produc-on, opera-on, or trading ac-vi-es through rela-onships associated with other interests (including family rela-onships)

8 UN Prac-cal Manual s Three Issues that Impact Methods Lack of appropriate comparables Quan-fica-on & alloca-on of Loca-on Specific Advantages (LSA) Iden-fica-on & valua-on of intangibles

9 Issue [1] Impac-ng Methods Lack of Comparables All OECD methods [except profit- splits] require comparables, however Developing countries have limited data on public companies (e.g., first movers). This results in developed country comparables being used. The theory is (one big market):» Commercial Globaliza-on» Free movement of capital Significant adjustments are needed in China this is common the geographic comparability issue (Japan & Korea) If adjustments cannot be made: (a) Sta-s-cal methods are used in US and (some) OECD (b) Profit Splits are preferred in China (c) See OECD Guidelines (next) 1995 v. 2010

10 OECD, Transfer Pricing Guidelines (1979 nothing); (1995) & (2010) [1.15/1.33] To be comparable means that none of the differences (if any) between the situa-ons being compared could materially affect the condi-on being examined in the methodology (e.g. price or margin), or that reasonably accurate adjustments can be made to eliminate the effect of any such differences. [1.16/1.35] In all cases adjustments must be made [blue is there in 1995, but MODIFIED in 2010 to add where possible ] to account for differences between the controlled and uncontrolled situa-ons that would significantly affect the price charged or return required by independent enterprises. Therefore, in no event can unadjusted industry average returns themselves establish arm's length condi-ons. Next: Issue 2

11 Issue [2] impac-ng Methods Loca-on Specific Advantages [Examples at &.7] Automo&ve Market- for- technology: favorable industrial policy Consumer preference for foreign brands (imports) Inelas-c demand for auto Capacity constraints on domes-c supplies Duty savings High quality/low cost parts 50/50 JV Chinese partner Local distribu-on network In-mate knowledge of mkt. Right market access Foreign partner Control opera-ons Control supply chain of parts

12 Two Examples One example from real fact pawern TNMM with an OM as PLI Second example from China chapter of UN Manual TNMM with full cost mark up (FCMU) essen-ally a C+ with annual aggregate data that s not the best Both demonstrate how TP methods deal with LSAs from a Chinese perspec-ve

13 TNMM Basic Distributor Example from private prac-ce PLI = Opera-ng Margin (OM) Foreign Parent Opera-ng Profit (before Financial expenses) Opera-ng Revenue China [In RMB] Sub. Profit/loss Sales COGS Taxes Gross Profit Selling Exp. Admin. Exp. Asset loss Opera-ng profit OM Formula A B C D=A- B- C E F G H=D- E- F- G H/A Amount 144,496, ,462, ,111 6,737,046 2,703,415 1,772,823 29,864 2,230, % Opera-ng Profit = (Opera-ng revenue) less (Opera-ng costs) [NOTE: not including financial expenses] Opera-ng Revenue = Net Revenue Opera-ng costs = Total Costs [NOTE: not including financial expense] Seven Developed Country comparables Unrelated [In USD thousands] Op Rev. 876, ,213 14,892 1,510, , , ,775 Highest Upper Quar-le Median Lower Quar-le Lowest Op Costs (Total Costs) 854, ,849 13,074 1,444, , , , % 4.47% 2.59% 0.76% 0.59%t PLI OM 2.59% 0.59% 12.21% 3.80% 0.86% 5.14% 0.67% 1.54%

14 How broad is a TNMM (1)? If we remove administra-ve expenses of $500,000 [no change] PLI = Opera-ng Margin (OM) Foreign Parent Opera-ng Profit (before Financial expenses) Opera-ng Revenue China [In RMB] Sub. Profit/loss Sales COGS Taxes Gross Profit Selling Exp. Admin. Exp. Asset loss Opera-ng profit OM Formula A B C D=A- B- C E F G H=D- E- F- G H/A Amount 144,496, ,462, ,111 6,737,046 2,703,415 1,272,823 29,864 2,730, % Opera-ng Profit = (Opera-ng revenue) less (Opera-ng costs) [NOTE: not including financial expenses] Opera-ng Revenue = Net Revenue Opera-ng costs = Total Costs [NOTE: not including financial expense] Seven Developed Country comparables [In USD thousands] Op Rev. 876, ,213 14,892 1,510, , , ,775 Highest Upper Quar-le Median Lower Quar-le Lowest Op Costs (Total Costs) 854, ,849 13,074 1,444, , , , % 4.47% 2.59% 0.76% 0.59%t PLI OM 2.59% 0.59% 12.21% 3.80% 0.86% 5.14% 0.67% 1.89%

15 How broad is a TNMM (2)? if we remove selling $2,000,000 & administra-ve expenses $1,000,000 [no change] PLI = Opera-ng Margin (OM) Foreign Parent Opera-ng Profit (before Financial expenses) Opera-ng Revenue China [In RMB] Sub. Profit/loss Sales COGS Taxes Gross Profit Selling Exp. Admin. Exp. Asset loss Opera-ng profit OM Formula A B C D=A- B- C E F G H=D- E- F- G H/A Amount 144,496, ,462, ,111 6,737, , ,823 29,864 5,730, % Opera-ng Profit = (Opera-ng revenue) less (Opera-ng costs) [NOTE: not including financial expenses] Opera-ng Revenue = Net Revenue Opera-ng costs = Total Costs [NOTE: not including financial expense] Seven Developed Country comparables [In USD thousands] Op Rev. 876, ,213 14,892 1,510, , , ,775 Highest Upper Quar-le Median Lower Quar-le Lowest Op Costs (Total Costs) 854, ,849 13,074 1,444, , , , % 4.47% 2.59% 0.76% 0.59%t PLI OM 2.59% 0.59% 12.21% 3.80% 0.86% 5.14% 0.67% 3.97%

16 How broad is a TNMM (3)? If we add premium sales $10,000,000 (a 7% price increase) PLI = Opera-ng Margin (OM) Foreign Parent Opera-ng Profit (before Financial expenses) Opera-ng Revenue China [In RMB] Sub. Profit/loss Sales COGS Taxes Gross Profit Selling Exp. Admin. Exp. Asset loss Opera-ng profit OM Formula A B C D=A- B- C E F G H=D- E- F- G H/A Amount 154,496, ,462, ,111 16,737,046 2,703,415 1,772,823 29,864 12,230, % Opera-ng Profit = (Opera-ng revenue) less (Opera-ng costs) [NOTE: not including financial expenses] Opera-ng Revenue = Net Revenue Opera-ng costs = Total Costs [NOTE: not including financial expense] Seven Developed Country comparables [In USD thousands] Op Rev. 876, ,213 14,892 1,510, , , ,775 Highest Upper Quar-le Median Lower Quar-le Lowest Op Costs (Total Costs) 854, ,849 13,074 1,444, , , , % 4.47% 2.59% 0.76% 0.59%t PLI OM 2.59% 0.59% 12.21% 3.80% 0.86% 5.14% 0.67% 7.92%

17 In this case A check of the Annual Report for this company expressly states that the product distributed in China shows: (a) Rela-vely constant global sales in units, but (b) A significant profit increase due to price increases in China only (sugges-ve of a market premium) Faced with this evidence a foreign (US) auditor would probably believe that the transfer price for the product sold to the Chinese distributor was below arm s length by RMB 8,229,491 b/c need to have OM = 2.59%, i.e., 154,496,279/4,001,453 Chinese auditor would want to determine the OM as 1.54% as in the original example, with RMB 10,000,000 removed from Sales leaving the original RMB 144,496,279 as OM denominator. See next slide on how the TNMM tells the company to protect the movement of these profits off shore.

18 US SOLUTION Increase COGS! [That is the transfer price to the parent] PLI = Opera-ng Margin (OM) Foreign Parent Opera-ng Profit (before Financial expenses) Opera-ng Revenue China [In RMB] Sub. Profit/loss Sales COGS Taxes Gross Profit Selling Exp. Admin. Exp. Asset loss Opera-ng profit OM Formula A B C D=A- B- C E F G H=D- E- F- G H/A Amount 154,496, ,462, ,111 6,737,046 2,703,415 1,772,823 29,864 2,230, % Opera-ng Profit = (Opera-ng revenue) less (Opera-ng costs) [NOTE: not including financial expenses] Opera-ng Revenue = Net Revenue Opera-ng costs = Total Costs [NOTE: not including financial expense] Seven Developed Country comparables [In USD thousands] Op Rev. 876, ,213 14,892 1,510, , , ,775 Highest Upper Quar-le Median Lower Quar-le Lowest Op Costs (Total Costs) 854, ,849 13,074 1,444, , , , % 4.47% 2.59% 0.76% 0.59%t PLI OM 2.59% 0.59% 12.21% 3.80% 0.86% 5.14% 0.67% 1.44%

19 SECOND EXAMPLE from UN Full Cost Mark Up (FCMU) Chinese cost = $100; Others = $150 change the MU not the base (FC) Foreign China PLI = FCMU Difference is the loca-on savings (difference) x (FCMU) (50) x (8%) = 4 Parent 12 not 8 research R&D Sub. Seven Developed Country comparables Cost base = 100 China wants a mark up on the costs that were saved, because these were real costs paid by Chinese workers. These costs should be no different than tradi-onal costs. So, compensa-on for R&D services must be = 12 Cost base = 150; FCMU median = 8% Next: Issue 3

20 Issue [3] Impac-ng Methods Intangibles LSAs are ozen related to marke-ng intangibles Technical intangibles decline in value over -me, but marke-ng intangibles increase in value. Treas. Reg (g)(6)(vi)(Ex. 2)(dual intangibles) Treas. Reg (g)(2)(Ex. 1) Treas. Reg (c)(3)(iii)(Ex.) Royal-es for intangibles should be adjusted over -me (CWI).

21 Prac-cal Issues & Solu-ons Aggrega-on & Separa-on &.2 China will do both depending on facts Contract R&D: China prefers P- S over C+ believing intangibles reside in R&D subsidiary par-cularly if they have high and new technology status [HNTS] Contract v. Toll Mfg. 10, lack of comparables for toll mfg., thus use contract mfg. (gross undervalua-on). Customs values accepted Convert toll to contract mfg. and then eliminate excess factors Sales, marke-ng & distribu-on mismatch of full func-ons performed (developed v. developing) market differences - China is fast growing & sales are more easily made.

22 Alterna-ve Methods (to TNMM) Example of Electronic Mfg. Services (EMS) En-re business is contract/toll mfg in China HQ outside of China (low tax) Sales of finished product outside of China Risk- based pricing does not work Resis-ng the separa-on of risk from assets or ac-vi-es same development in Indian TP policy Contribu-on analysis formulary with people & assets Aggregate all the related par-es and allocate Residual P- S with the residual going to China

23 China s Solu-ons Summary Loca-on Specific Advantages Market Premiums Alterna-ve Methodologies