The Organisation of Economic Co-operation and Development (OECD), through its Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations, would have the multinational enterprises (MNEs) and tax administrations select the “most appropriate” transfer pricing method.1 The multinational enterprise or the tax administration, in seeking to apply this most appropriate transfer pricing method, might apply the transactional profit split method,2 and then would seek to apply the “most appropriate” allocation key.3 The reader should be aware of the fact that there may be more than one viable and appropriate allocation key. The OECD does not assert that transfer pricing is an exact science.
The TransferPricingConsortium.com and the Inland Revenue Board of Malaysia,4 through its Tax Academy, joint-ventured a two-day transfer pricing presentation to update tax agents with transfer pricing techniques.5 These sessions took place on September 19 through 20, 2011, in Malaysia. A significant portion of the presentation and a question-and-answer session addressed the development of transfer pricing allocation keys, as reflected within section 2.134 through section 2.149 of the July 2010 OECD guidelines.
As part of this transfer pricing presentation, counselors Robert Feinschreiber and Margaret Kent provided an example of a small condominium residential leasing company in Malaysia and Singapore. This example illustrates how MNEs and tax administrations might apply the transactional profit split method and apply alternative allocation keys. The analysis reflects an elaboration of an example presented in the Malaysia Tax Academy with the Malaysian tax authorities. The example illustrates that the tax administration can expend less professional time and can earn a substantially greater income stream by applying allocation keys to the transactional profit split method.
Feinschreiber and Kent selected the real estate rental leasing example for the Malaysian tax authorities for these reasons:
The OECD guidelines specifically direct us to the transactional profit split method in situations in which a MNE meets one of these five criteria:
A business can apply the transactional profit split method in situations in which the MNE:
Services are “specialized” if the providers of these services require both training and expertise. That is a dictionary definition of the term “specialized services,” but there is an additional specialized services facet when addressing these specialized services for tax purposes. Here the OECD guidelines mean “specialized services” as being in contradiction to the words “ordinary” and “mundane services,” but the guidelines also mean that “specialized services” are the opposite of “subsidiary or incidental” services. Consider the next illustrations.
When a tax administration is examining an international real estate brokerage, the tax administration might view the listing and selling (or leasing) operations of the international real estate brokerage as a performance of a specialized service (i.e., the real estate listing agents and selling agents are the crux of the real estate business). The international real estate brokerage might need various types of service assistance, including, for example, accounting assistance. Such a tax administration might treat accounting services undertaken for the international real estate brokerage as being secondary to the international real estate brokerage activities.
Correspondingly, in a different fact pattern, the tax administration might examine an accounting firm. The tax administration might view the providing of accounting service as the performance of a specialized service in this particular instance. The accounting firm in that scenario might need secondary assistance to expand its accounting offices. Such a tax administration might treat the real estate activities that the accounting firm undertakes—such as negotiating leaseholds for the accounting firm—as being secondary to the accounting agency’s prime function.
The differentiation between specialized services and subsidiary services is an important one, regardless of how the tax administration categorizes specialized services for a particular industry. The tax administration is likely to peg the taxes that the international real estate brokerage might pay based on the performance of the specialized service the real estate brokerage provides (i.e., the listing, leases, or selling the real property) based on the success of the brokers and dealers in the real estate brokerage. In contrast, the tax administration might treat services performed by other than such nonspecialized personnel as being costs or expenses that the international real estate brokerage incurs.
Tax administrations normally treat the intercompany transactions of nonspecialized personnel as being determined as part of the cost that the enterprise actually incurred. The tax administrations normally would not insist on any add-ons (such as overhead or profits) for incurring such non-specialized services costs. In contrast, tax administrations normally would insist on including add-ons for primary, specialized services costs, as being profit centers.
The next step in determining the tax allocation is for the tax administration or the international real estate brokerage to aggregate the entire income of the entire business for all of its relevant transactions.14 The international real estate brokerage or the tax administration might aggregate these income amounts by applying harmonized tax accounting standards (i.e., by using the International Financial Reporting Standards) making this determination across all jurisdictions in which the enterprise performs these specialized services.
The final step in determining the tax allocation is for each tax administration, or the international real estate brokerage as a group, to allocate the worldwide operations in that jurisdiction.15 The international tax brokerage or the tax administration might apply a formula or develop allocation keys. The transactional profit split method does not entirely eliminate the risk of double taxation because the tax administrations in each location might select differing—and conflicting—allocation keys.16
The MNE or the tax administration’s application of the transactional profit split method presupposes that there are no true comparables.17 The OECD guidelines recognize that the transactional profit split method enables the taxpayer to take into account the associated enterprise’s specific and possibly unique facts and circumstances that are not present in independent enterprises.18 Nevertheless, the OECD views the transactional profit split method as being an arm’s length approach, treating the transactional profit split method as reflecting what independent enterprises would have done if faced with the same circumstances.19
The OECD requires the taxpayer to evaluate each party to the transaction when the taxpayer applies the transactional profit split method rather than looking at each tested party alone. The OECD views this two-sided or multisided valuation in a positive light because it will be less likely that either party to the controlled transaction will be left with an “extreme and improbable result.”20 The OECD guidelines view the dual or multisided analysis as being particularly important for a taxpayer when analyzing the contributions of the associated enterprises as to the intangible property that the associated enterprise employs in the controlled transactions.21
The OECD visualizes the transactional profit split method as operating in this manner:22
The next step in the transfer pricing process is to select potential allocation keys from the four allocation categories: allocation keys based on revenue, allocation keys based on assets or capital, allocation keys based on costs, and other allocation keys. The OECD guidelines implicitly define an allocation key as a basis on which the MNE or the tax administration can allocate operating income. The OECD guidelines provide for four types of allocation keys:23
The OECD contemplates that the MNE or the tax administration will first determine which activities of business are “key functions” in developing the allocation key process. Functions are “key” if these functions are the essential functions of the business and generate value to this business. A functional analysis is crucial in determining which functions of the business are key.
As transfer pricing practitioners, we view “key functions that generate value” as those functions that address the integrated production of highly specialized goods, that create unique intangibles, that provide specialized services, or that participate in the global trading of financial instruments. We view the term “strong correlation” standard within the transactional profit split method as indicating causality (i.e., as effort that the enterprise expends to achieve a specified result). It might be appropriate for the MNE or the tax administration to view this “strong correlation” concept in “but for” terms. Thus, a specific activity provides a key function to the business because, without the presence of this function, the business would not generate its combined operating profits.24
International real estate brokerages are at the cusp of a rapidly growing multinational services industry. At the heart of this services industry, international real estate brokerages are providing “specialized services” to a myriad of clients, performing these specialized services in various locations across the world. By the term “international real estate brokerage,” we mean a real estate enterprise having real estate operations in more than one country, serving as a broker or as an agent for the purpose of buying, leasing, or selling real estate for their clients in more than one country. The term “specialized services” is an important one for tax purposes, as this term determines the applicable tax category —treating this service as a profit center or as a cost.
The selection of the “most appropriate” allocation key should normally depend on the MNE and each tax administration being willing to develop harmonized tax accounting standards.25 We assume in this example that the MNE and each tax administration have access to this harmonized accounting. One of the early steps in the audit process is for a tax administration to review financial data to make sure that the taxpayer harmonized the accounting data or, separately, harmonized the accounting results if the taxpayer fails to harmonize the data.26
The first essential step that a MNE or tax administration must undertake as part of its transfer pricing analysis is to understand the operations of the business under review. The MNE or the tax administration should rely on the functional analysis of the enterprise, together with the assets utilized, risks assumed, contractual terms, economic circumstances, and characteristics of the property being transferred.27
The MNE or the tax administration first ascertain the following facts in our real estate condominium leasing fact pattern:
The real estate enterprise provides the listed data to the Malaysian tax administration regarding the residential leases. The real estate enterprise would provide these data to the Singapore tax administration if Singapore requests this information. (Note: All amounts are in U.S. dollars.)
After the MNE and/or the tax administration ascertain the applicable database, they next should analyze the database to ascertain potential allocation keys. The MNE or the tax administration, in our residential leasing transfer pricing allocation example, might feasibly consider four types of allocation keys based on the real estate agency’s available data:
The OECD guidelines apply the “strong correlation” standard to the enterprise’s creation of combined profits. This strong correlation standard applies to revenue-based allocation keys, to asset-based allocation keys,33 to cost-based allocation keys,34 and to additional allocation keys.35 We apply two steps to implement this procedure:
The MNE or the tax administration then analyzes 11 potential allocation keys in four categories:
This commission-generated allocation key assumes that real estate agents expend effort to achieve their commissions. Here the real estate enterprise generated commissions of $420,000 in Malaysia and $300,000 in Singapore, $720,000 in total. We determine that the commission-generated allocation key meets the strong correlation standard based on the agents’ effort expended and the creation of combined profits. Based on these facts, Hasil could feasibly assert that, in applying the transactional profit split method to the commissions generated as an allocation key, 58.3% of the income ($420,000/($420,000 + $300,000)) was attributable to Malaysian activities.
This leased property allocation key assumes that real estate agents expend proportionately more effort for more valuable properties than they spend for lower-value properties. Here the real estate enterprise leased residential real estate property of $140,000,000 in Malaysia and $90,000,000 in Singapore, $230,000,000 in total. We determine that the leased property allocation key meets the strong correlation standard based on the agents’ effort expended and the creation of combined profits. Based on these facts, Hasil could feasibly assert that, in applying the transactional profit split method to the properties held for lease as an allocation key, 60.9% of the income ($140,000,000/($140,000,000 + $90,000,000)) was attributable to Malaysian activities.
This commission-leased allocation key assumes that real estate agents expend effort as to the property being leased to achieve their commissions. Here the real estate enterprise leased 700 condominium units in Malaysia and leased 300 condominium units in Singapore, 1,000 condominium units in total. We determine that the commission-leased, commission-generated allocation key meets the strong correlation standard based on the agents’ effort expended and the creation of combined profits. Based on these facts, Hasil could feasibly assert that, in applying the transactional profit split to the number of condominium units being leased as an allocation key, 70% of the income (700/(700 + 300)) was attributable to Malaysian activities.
This number of rooms allocation key assumes that the real estate agents will devote proportionately more effort to leasing condominium structures that have more rooms than condominium structures with fewer rooms. Here the real estate enterprise determines that the number of rooms rented in the condominiums were 3,500 rooms in Malaysia and 1,200 rooms in Singapore, 4,700 rooms in total. We determine that the number of rooms allocation key does not meet the strong correlation standard based on the agents’ effort expended and the creation of combined profit. We reject the number of rooms allocation key because the number of rooms does not correlate to the effort that real estate agents undertake or the creation of combined profit. But for those facts, Hasil could have asserted that, in applying the transactional profit split to the number of rooms as an allocation key, 74.5% of the income (3,500 rooms/(3,500 rooms + 1,200 rooms)) would have been attributable to Malaysian activities.
This value of property under lease allocation key postulates that the real estate agents will devote proportionately more time to leasing more valuable housing property than they would to leasing less valuable housing property. Here the real estate enterprise determines that the annual value of property being leased was $7,000,000 in Malaysia and $6,000,000 in Singapore, $13,000,000 in total. We determine that the value of property under lease allocation key meets the strong correlation standard based on the agents’ effort expended and creation of combined profit. Based on these facts, Hasil could feasibly assert that, in applying the transactional profit split to the value of condominium units being leased as an allocation key, 53.8% the income ($7,000,000/($7,000,000 + $6,000,000)) was attributable to Malaysian activities.
This agent’s commission cost-based allocation key postulates that the real estate agents will devote proportionately more effort to leasing where their aggregate commissions are higher than when the commissions are less. Here the real estate enterprise determines the aggregate commission is $210,000 in Malaysia and $150,000 in Singapore, $360,000 in total. We determine that the agent’s commission allocation key meets the strong correlation standard based on the agents’ effort expended and the creation of combined profit. Based on these facts, Hasil could feasibly assert that, in applying the transactional profit split to the aggregate agents’ commission as an allocation key, 58.3% of the income ($210,000/($210,000 + $150,000)) was attributable to Malaysian activities.
This agent’s commission cost-based allocation key postulates that real estate agents earn commissions based on their activities, and the real estate agents earning higher commissions are more valuable to the real estate enterprise than those agents earning lower commissions. Here the real estate enterprise determines the agent’s commission per agent is $21,000 in Malaysia and $30,000 in Singapore. We determine that the per-agent commission allocation key does not meet the strong correlation standard based on the agents’ effort expended and the creation of combined profit. We reached this decision because of the lack of causality between the agent’s commission structure and the revenue generated.
This agent’s cost allocation key as to real estate gross revenues after commissions postulates that the real estate brokerage earns income based on these commission activities. Here the real estate enterprise determines its income after commissions as being $210,000 in Malaysia and $150,000 in Singapore, $360,000 in total.
We determine that the per-agent commission allocation key meets the strong correlation standard based on the agents’ effort expended and the creation of combined profit. Based on these facts, Hasil could feasibly assert that, in applying the transactional profit split to the aggregate agents’ commission as an allocation key, 58.3% of the income ($210,000/($210,000 + $150,000)) was attributable to Malaysian activities.
This number of real estate agents as an allocation key might correlate to the real estate brokerage’s leasing income. The real estate enterprise has 10 real estate agents in Malaysia, 5 agents in Singapore, 15 agents in total. The number of agents the real estate enterprise adds does not indicate the agents’ effort expended. Further, the number of real estate agents as an allocation key is suspect because the real estate brokerage can hire additional agents who expend little effort and fail to earn commissions. The number of real estate agents does not correlate location of real estate agent with operating commission income in the creation of combined income. We determine that the per-agent commission allocation key does not meet the strong correlation standard based on the number of real estate agents to combined profit. If the real estate enterprise were to use number of real estate agents as an allocation key, Hasil could feasibly assert that, in applying the transactional profit split based on the number of real estate agents as an allocation key, 66.7% of the income (10 agents/(10 agents + 5 agents)) would have been attributable to Malaysian activities.
This number of condominium units per property that customers leased might be an allocation key that conceivably might correlate to the real estate brokerage’s leasing income. Nevertheless, the facts indicate that there is no causality or other connection between the number of condominium units per property the residential customers lease. We determine that the number of condominium units per property being leased as an allocation key does not meet the strong correlation standard based on the agents’ effort expended and the creation of combined profit.
The floor level or flat or apartment that the customers of the real estate agency lease might conceivably correlate to the real estate brokerage’s leasing income. Nevertheless, the facts indicate that there is no causality or other connection to the number of condominium units per property the customers of the real estate agency lease. We determine that the number of condominium units per property being leased as an allocation key or the apartment being leased does not meet the strong correlation standard based on the agents’ effort expended and the creation of combined profit.
Let us combine the results we have previously ascertained:
Rejected Potential Allocation Keys
Remaining Allocation Keys | Malaysia Taxable Percentage | Singapore Taxable Percentage |
Commission generated | 58.3 | 41.7 |
Properties held | 60.9 | 39.1 |
Units leased | 70.0 | 30.0 |
Value of property | 53.8 | 46.2 |
Commission | 58.3 | 41.7 |
Note that, in this example, as in most fact patterns, the business activities present more than one allocation key. Each of the selected allocation keys meets the strong correlation standard, and each allocation key can meet the creation of operating income criterion. The strong correlation test is a bright-line test. The determination of whether a particular allocation key meets the strong correlation test is often a matter of professional opinion. Note that professional opinions might differ.
The MNE or the tax administration might consider employing cost accountants and economists in implementing this transfer pricing analysis. In our example, consider the difference in results if the MNE or the tax administration accepted the number of agents allocation key, resulting in an allocation of 66.7% to Malaysia, or accepted advertising as being an allocation key, resulting in an allocation of 33.3% to Malaysia.
Like other facets of transfer pricing, the determination of an allocation key under the transactional profit split method provides a MNE or the tax administration with approximate transfer pricing results—a range of 53.8% to 70.0% for Malaysia and a range of 30.0% to 46.2% in Singapore. This range can cause double taxation or can cause a tax gap:
The MNE or the tax administration should consider making these six inquiries:
The MNE corporate strategy appears to take the initiative in the first instance in setting pricing. The corporation could develop a transfer pricing strategy that meets the allocation key criteria in both Malaysia and Singapore. The taxpayer in this instance might be able to create a tax gap of 14.2% in this example by taking the first-mover advantage and by taking inconsistent positions when dealing with both Malaysia and Singapore. This tax gap should be transitory—Hasil and the Inland Revenue of Singapore may later establish discovery regimes that indicate the taxpayer’s conflicting transfer pricing regimes.
There might be situations in which the taxpayer, the real estate enterprise, takes the first-mover advantage and establishes an appropriate transfer pricing allocation. If the taxpayer does select such a transfer pricing result that is within the range of the allocation keys, it might be an advantage for Hasil to accept the taxpayer’s transfer pricing results rather than to develop a transfer pricing result that would be in conflict. If the taxpayer fails to select such a transfer pricing result that is within the result of the allocation keys, it then would be prudent for Hasil to challenge the allocation key rather to accept this audit challenge.
Based on the next results, Hasil will find that the transactional profit split method is substantially easier to apply than is the TNMM, and will increase the yield per professional hours spent. Quite simply, it is substantially easier for Hasil to examine a company’s internal records than it is to develop external data comparables. Further, it appears to be advantageous for Hasil to develop its transfer pricing specialization program, focusing on oil and gas activities, on real estate activities, and on agricultural activities and the like. We suggest that Hasil, to further implement the transactional profit split method, should consider hiring international accountants very familiar with the harmonized tax accounting process, cost accountants, and economists, together with transfer pricing experts.
NOTES
1. Promulgated July 2010; Chapter II, Part IA.
2. OECD, Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (Paris: Author, July 2010), 2.108 et seq.
3. Guideline 2.134 et seq.
4. Lemaga Hasil Delam Negeri Malaysia. (Hasil) in Behasa Malay.
5. The Inland Revenue of Malaysia has more than 30 transfer pricing examiners of various levels of experience, 30 of whom attended the joint venture conference.
6. OECD, Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations, 2.141, 2.142.
7. Guideline 1.9.
8. Guideline 1.9.
9. Guideline 2.109.
10. Guideline 1.9; Guideline 2.109.
11. Guideline 1.9.
12. Guideline 2.109.
13. Guideline 8.19.
14. Guideline 2.124.
15. Guideline 2.131.
16. Guideline 2.133.
17. Guideline 1.9, 2.109; R. Feinschreiber and M. Kent, “What You Need to Know about the OECD’s Transactional Profit Split Method,” Corporate Business Taxation Monthly (February 2011): 29.
18. Guideline 2.112.
19. Guideline 3.39.
20. Guideline 2.113.
21. Guideline 2.109.
22. Guideline 2.108.
23. Guideline 2.135.
24. Guideline 2.131.
25. Guideline 2.136.
26. Guideline 2.126.
27. Guideline 1.36.
28. Guideline 2.132.
29. Guideline 2.135.
30. Guideline 2.135.
31. Guideline 2.138.
32. Guideline 2.136.
33. Guideline 2.136.
34. Guideline 2.138.
35. Guideline 2.135.
3.142.197.198