INTM467030 - Establishing the arm's length price: gathering your own evidence

Using the correct method

You should ensure that an appropriate OECD methodology underpins your conclusions on the arm’s length price.

As can be seen from the chapter on the OECD Transfer Pricing Guidelines, there are a number of different methods that the OECD recommends for establishing an arm’s length price - see INTM463000 on these methodologies. A group may not necessarily use an OECD method, but any exercise you undertake to put forward an alternative transfer price should be based on the methods approved by OECD. The OECD methods shadow but do not mirror how prices are set between independents. A group is entitled to set its prices in any way it prefers but the results must be arm’s length.

The OECD Transfer Pricing Guidelines recognise that different situations will call for different methods. It is entirely possible to use more than one method to price a transaction; an MNE is entitled to use a transactional net margin method ('TNMM') rather than say a resale minus method, if both methods produce the arm’s length result. If however two different methods produce two different results, traditional transactional methods (comparable uncontrolled price, cost plus or resale minus) should be used in preference to profit split or TNMM.

There may be instances where the methodology adopted will lead inexorably to non-arm’s length pricing. For example:

Example 1

The Pukka Investments Group manages funds for wealthy clients worldwide. The parent company is incorporated in Luxembourg, but is resident in Bermuda. The group’s two major subsidiaries are based in London and New York. The staff in these two offices look for new business, manage existing clients and actively manage the funds held on behalf of clients. The UK subsidiary makes its CTSA return for the year ended 31 December 1999, and on enquiry the group explain that the UK is remunerated on a cost-plus basis.

The UK subsidiary says it is unable to supply any information about activities in the US or Bermuda. The inspector discovers that one of the UK directors is also a director and shareholder of the parent company. Following the issue of a notice under TMA70/S20(3) the director supplies accounts for Bermuda for the year ended 31 December 1999. The US accounts for the same year are obtained After an internet search, the accounts, including comparative figures for 1998, show the following (summarised) results:

Pukka Investments

Bermuda

Pukka Investments

UK

Pukka Investments

US

1999

1998

1999

1998

1999

1998

£’000

£’000

£’000

£’000

£’000

£’000

Fees from managed funds

36,750

34,300

Other fees

8,600

7,900

Profits on sale of investments

7,800

5,600

Net interest

2,150

1,900

Management fees

20,900

18,150

17,100

15,290

Total income

55,300

49,700

20,900

18,150

17,100

15,290

Management fees

(38,000)

(33,440)

Salaries

(150)

(150)

(12,500)

(10,900)

(11,200)

(10,100)

Establishment

(2,600)

(2,400)

(3,400)

(3,200)

(2,500)

(2,100)

Promotions

(1,100)

(900)

(750)

(600)

Legal

(50)

(50)

(1,300)

(850)

(600)

(700)

General expenses

(100)

(80)

(700)

(650)

(500)

(400)

Total expenses

(40,900)

(36,120)

(16,500)

(15,550)

(13,900)

Profit

14,400

13,580

1,900

1,650

1,550

1,390

Further enquiries establish that the group does not actually own or rent office space in Bermuda. A local bank provides facilities for meetings and administration services, including some asset administration. Although board meetings of the parent company are held in Bermuda, it is established that all the group’s activities are carried out in either London or New York. Although risk is purported to lie with Bermuda, it is established that both UK and New York carry out asset management and investment advisory and management services. All the decisions regarding risk are taken in either the UK or USA. The UK was successfully sued in 2000 for mis-selling some investments.

The cost plus method of rewarding the UK company will not produce the arm’s length reward. A profit-split method should be considered instead.

Example 2

Contrast this with the case of Bodgit & Scarper (Cool Fork Hats) Ltd who have been selling Bodgit & Scarper’s own exclusive range of hats since 1991. The hats are bought from a group company based in Mexico and sold via a UK-wide chain of retail shops. Bodgit & Scarper (Cool Fork Hats) Ltd’s accounts show the following results:


1997
1998
1999
2000
2001

£’000
£’000
£’000
£’000
£’000






Sales
90,000
110,000
130,000
110,000
125,000
Cost of sales
(55,800)
(66,000)
(79,300)
(68,200)
(76,250)
Gross profit
34,200
44,000
50,700
41,800
48,750






GPR
38%
40%
39%
38%
39%






Distribution
(4,500)
(5,500)
(6,500)
(5,500)
(6,250)
Sales & marketing
(22,670)
(29,910)
(33,970)
(26,820)
(32,500)
Administration
(5,500)
(6,500)
(7,500)
(7,500)
(8,000)
Operating profit
1,530
2,090
2,730
1,980
2,000






OPR
1.7%
1.9%
2.1%
1.8%
1.6%

Upon enquiry, Bodgit & Scarper (Cool Fork Hats) Ltd produce a transfer pricing report which shows the results of comparable distributors. The results, for sixteen independent companies, shown a range of net operating margins over the period 1994 to 1998 from 1% to 3%. The prices charged by the Mexican affiliate are adjusted at quarterly intervals to try and ensure that Bodgit & Scarper (Cool Fork Hats) Ltd achieves an operating margin as near to 2% as possible.

The comparables are looked at in some detail. Three more comparable companies are found, and enough information is available to establish the gross margins made by seven of the comparable companies over the period 1994 to 1998. The results show a range of gross margins from 37.5% to 41%.

While the resale minus method is potentially a more accurate way of calculating the arm’s length price, in this case, both the resale minus and TNMM comparable results show a range into which Bodgit & Scarper (Cool Fork Hats) Ltd’s results fall. On the evidence so far, there is nothing to suggest that transfer prices are not arm’s length.