Publications

Until recently, transfer pricing had not been a particularly significant concern for the Hong Kong Inland Revenue Department (IRD). Because of a restricted definition of income and a relatively low corporate tax rate,¹ Hong Kong traditionally has been used in global corporate tax planning as a profit-center jurisdiction rather than a jurisdiction where profits should be avoided. The favorable arbitrage made it such that, while the IRD would from time to time pay lip service to the issue,² in reality, it had little incentive, or need, to develop a transfer pricing framework.

Times have changed. Hong Kong's tax rate is no longer particularly low when compared to the zero rate offered by many offshore centers. In addition, a decision of the highest tribunal in Hong Kong and the rapid expansion of the tax treaty network have stirred the IRD out of its complacency.

 Ngai Lik : the coming of age of transfer pricing in Hong Kong

The Hong Kong Court of Final Appeal (CFA) July 2009 decision in Ngai Lik Electronics Co. Ltd. v. CIR³ effectively marked the coming of age of transfer pricing in Hong Kong, with the highest court in the land ruling that the IRD must follow transfer pricing practices when invoking its power under Section 61A of the Inland Revenue Ordinance (IRO).⁴

In Ngai Lik, the taxpayer's group was involved in manufacturing and distribution of electronic audio products. In the late 1980s, at a time when much of the manufacturing base of Hong Kong moved to the mainland, the principals of the taxpayer established factories in the mainland to provide the taxpayer with processing facilities for the labor-intensive part of its operations. As was typical in those days, the mainland factories provided land and labor while the taxpayer was responsible for everything else, from sourcing raw material to providing machinery, technical know-how and management expertise, from marketing to selling the final products. For tax purposes, the taxpayer offered all of its profits for taxation in Hong Kong, after deducting the processing fee paid to the mainland factories.

In the 1990s, the principals of the taxpayer undertook to restructure their group. First, three British Virgin Islands companies were incorporated to take over control of the mainland factories. Then the taxpayer and the BVI companies entered into a complex series of services and supply agreements that, once in place, had (ostensibly) turned the taxpayer from a manufacturer to merely a trader and service provider.  As planned, even though nothing much had changed functionally in the operations of the business, the result would still be that:

• the BVI companies would be carrying on, on their own, all manufacturing activities at the mainland factories (that is, outside Hong Kong), while
• the taxpayer's involvement would be limited to trading the final products as well as providing certain ancillary manufacturing services to the BVI companies.

For Hong Kong tax purposes, on the basis of the new arrangements, the taxpayer offered substantially less profit for taxation.

As one can imagine, the IRD was not particularly pleased with the new tax position adopted by the taxpayer, particularly when it noted that the taxpayer continued to employ the same personnel, who continued to perform the same tasks (including overseeing the work in the mainland) and sell the same products to the same customers. In due course, the IRD claimed, on the basis of Section 61A IRO, that an arbitrary 50 percent of the overall profits of the business should be assessable to tax in Hong Kong.⁵ The taxpayer duly objected and the matter eventually wound its way to the CFA.

Key findings of CFA

From a transfer pricing perspective, the key findings of the CFA's decision are its comments on how the IRD must exercise its power when reconstructing the tax consequences of a transaction against which Section 61A IRO applies. Under Section 61A(2) IRO, the IRD is authorized to review a transaction as if it or any part thereof had not taken place or in such other manner "as appropriate to counteract the tax benefit which would otherwise be obtained." On how the latter must be exercised, the CFA commented:

The power must therefore be exercised on the basis of a reasonably postulated hypothetical transaction which produces an assessment designed rationally to counteract the tax benefit.  The assessment cannot be raised in some arbitrary amount or arrived at upon some basis that is unreasonable or not rationally related to the tax benefit in question. Such an assessment would not be a proper exercise of the power conferred by section 61A(2).

The court then ruled that the assessments against the taxpayer, by using an arbitrary 50 percent inclusion (as per the manufacturing exemption – see endnote 5), had not been raised on an appropriate basis and were therefore invalid. It commented that

To counteract that tax benefit a reasonable approach which obviously recommends itself would have been to raise an assessment on the profits which would hypothetically have been earned if the taxpayer had purchased the goods at arm's length prices instead of at the prices fixed annually.

.  .  .  .It appears that the Commissioner adopted that 50% figure simply on the basis of the Revenue's practice regarding the apportionment of profits attributable to offshore manufacturing operations in certain cases. But the question whether the Commissioner properly exercised her section 61A(2) powers in raising the additional assessments is a question of law to which the Revenue's practice does not provide an answer. When asked to explain the basis of the figures adopted in those assessments, Mr Ho frankly stated that the figure was arbitrary. As Mr Barlow correctly submitted, the Ordinance does not authorise arbitrary assessments to be made under section 61A(2).

Transfer pricing practitioners will recognize in the decision the basic tenets of their discipline: a requirement for an objective approach in assessing related-party transactions and the use of arm's length as the cornerstone of that approach. The gauntlet had been thrown down: the IRD would have to apply a transfer pricing approach to instances when assessments are issued based on hypothetical circumstances (such as happens when Section 61A IRO is invoked).

The treaty network also plays a key role

In addition to Ngai Lik, the recent development of a tax treaty network is playing a key role in nudging the IRD toward a formal approach to transfer pricing. By now, Hong Kong has 10 tax treaties in force, 11 at various stages of implementation and ongoing discussions about yet more with many other jurisdictions.

So far, all Hong Kong tax treaties closely follow the Organization for Economic Cooperation and Development Model Treaty and, in particular, its Article 9 dealing with associated enterprises. Under Article 9, when, in the context of transactions between connected entities,

conditions are made or imposed between the two enterprises in their commercial or financial relations which differ from those which would be made between independent enterprises, then any profits which would, but for those conditions, have accrued to one of the enterprises, but, by reason of those conditions, have not so accrued, may be included in the profits of that enterprise and taxed accordingly.⁶

In other words, each jurisdiction is committed to treat related entities as if they were independent enterprises and allocate profits accordingly.  Again, transfer pricing practitioners will recognize familiar language in Article 9 and it follows that the more commitment Hong Kong is making to other jurisdictions in the form of tax agreements, the more pressing the need to adopt a formal approach to transfer pricing.

DIPN 46

In December 2009, the IRD issued Departmental Interpretation and Practice Note No. 46, "Transfer Pricing Guidelines—Methodologies and Related Issues," as its first stand-alone public foray to explain how it would approach transfer pricing issues. DIPN 46 is instructive on many fronts.

The authors note the following:

• The IRD does not intend to reinvent the wheel and will closely follow the principles and guidelines developed by the OECD. On that basis, taxpayers in Hong Kong can be (relatively!) confident that, to the extent their transfer pricing policies are developed in accordance with the pronouncements of the OECD, the resulting margins should pass muster for the purposes of the IRD.
• There is no intention (at least for the time being) to introduce specific transfer pricing provisions to the IRO. For now, the IRD is satisfied that it finds sufficient authority to effect transfer pricing adjustments within existing provisions of the IRO.

DIPN 46 refers in particular to Sections 16, 17, 20 and 61A IRO.

Sections 16 and 17: general deduction of expenses

Sections 16 and 17 are relevant to general deduction of expenses in calculating assessable profits. The IRD contends that they provide authority to adjust a deduction if the quantum is found to be unreasonable.

One can wonder if the IRD is right that there is a reasonableness test imbedded within Sections 16 and 17 IRO.  In fact, some comments in the Ngai Lik decision cast doubt on the extent of the authority of the IRD to use those sections to adjust the quantum of an otherwise admissible deduction.⁷

Section 20 and transfer pricing

Section 20 IRO has long been identified as the only transfer pricing inspired provision of the tax legislation. However, it is an awkwardly drafted section that applies when

a non-resident person carries on business with a resident person with whom he is closely connected and the course of such business is so arranged that it produces to the resident person either no profits which arise in or derive from Hong Kong or less than the ordinary profits which might be expected to arise in or derive from Hong Kong, the business done by the non-resident person in pursuance of his connection with the resident person shall be deemed to be carried on in Hong Kong, and such non-resident person shall be assessable and chargeable with tax in respect of his profits from such business in the name of the resident person as if the resident person were his agent, and all the provisions of this Ordinance shall apply accordingly.

Historically, the IRD has rarely invoked Section 20 IRO, at least on a stand-alone basis,⁸ and, indeed, it is barely discussed in DIPN 46, its seminal guidance on transfer pricing. We suspect that the IRD has little faith in the section as a tool to counteract inappropriate transfer pricing practices and, over time, it will surely be overshadowed by the other available options.

Section 61A: the IRD’s best weapon

In the absence of a formal section dealing specifically with transfer pricing, Section 61A IRO is ultimately the best weapon of the IRD.  However, the section is not particularly user-friendly for the IRD because it requires a sufficiently highly placed IRD officer⁹ to confirm that a transaction has been entered into for the sole or dominant purpose of obtaining a tax benefit based on an analysis of no less than seven different factors.¹⁰ Once that hurdle is overcome and a transaction is found to violate Section 61A IRO, the IRD is then provided with significant authority to rewrite the tax consequences of the transaction (that is, as we discussed earlier in our review of the Ngai Lik decision, as if the transaction had not been entered into or as may be appropriate to counteract the tax benefit).

The bottom line, then, is that, in spite of the potential shortcomings of the IRO to address transfer pricing issues, the IRD will, for now, eschew the introduction of a formal regime.  Time will tell whether one will eventually be needed.

The IRD does not propose (for the time being) to impose mandatory transfer pricing documentary and reporting obligations, despite the fact that such obligations now are standard in several jurisdictions. Instead, DIPN 46 speaks of "expectations" of the IRD that, in the context of a transfer pricing audit, as part of the general obligation to maintain appropriate books and records,¹¹ a taxpayer would have on hand all manners of information on its business and transfer pricing policies. DIPN 46 mentions:

(a) information about associated enterprises involved in the controlled transactions, transactions at issue, functions performed, information derived from independent enterprises engaged in similar transactions

(b) information about the controlled transactions – for example, nature and terms, economic conditions, property involvement, product and service flows, changes in trading conditions, and renegotiations of existing arrangements

(c) information relating to comparable companies having transactions similar to the controlled transactions

(d) information on associated enterprises, such as an outline of the business, structure of the organization, ownership links within the multinational group, amount of sales and operating results from the last few years preceding the transaction, and the level of the taxpayer's transactions with foreign associated enterprises

(e) information on pricing, business strategies, and special circumstances – for example, factors that influence the setting of prices or the establishment of any pricing policies for the taxpayer and the whole multinational enterprise group like those of markup on cost, deducting related costs from sales prices to end users in the market where the foreign related parties are conducting a wholesale business, or employing an integrated pricing or cost contribution policy on a group basis

(f) information on the factors that lead to the development of such pricing policies, and, where applicable, consistency with transactional conditions in the open market

(g) explanation of the selection, application, and consistency with the arm's-length principle of the transfer pricing method used

(h) special circumstances concerning any setoff transactions that have an effect on determining the arm's-length price, and details of any particular management strategies or circumstances particular to the type of business that may temporarily alter pricing structures – for example, market entry, market share, new product or defensive strategies

(i) general commercial and industry conditions

(j) information about functions performed, taking into account assets used and risks assumed, and

(k) documents showing the process of negotiations for determining or revising prices in controlled transactions

It is not a coincidence that the information "expectations" of the IRD are similar to the formal documentation requirements found in other jurisdictions.¹²

It also should also be noted that, to date, the IRD has not developed a separate unit focusing exclusively on transfer pricing, nor has it specifically hired economists or other specialists in this area.  Rather, the IRD officer in charge of a particular audit is expected to deal with any transfer pricing issues he or she may encounter in the course of the audit.

IRD audit practices

How an audit may begin

Since there is no specific contemporaneous documentation filing requirement in Hong Kong related to transfer pricing, the filing or failure to file such documentation would not trigger a transfer pricing audit.¹³ Rather, transfer pricing audits in Hong Kong usually start innocently, when the IRD queries particular aspect of the tax filing of the taxpayer. 

We often have seen a simple letter asking questions on a particular item of deduction become the lynchpin for a series of inquiry letters which, in no time, turns a simple inquiry into a full-blown audit covering many years of assessment¹⁴ and delving into minute details of the business of both the taxpayer in Hong Kong and its related entities around the world.

Information requests and replies

The IRD can be extremely punctilious in requesting information. Unfortunately, there is little taxpayers can do but comply with its requests since the IRD essentially controls the process and is backed up by all the coercive measures of the IRO to gather information in the context of an audit.¹⁵ Furthermore, we have found that the IRD will not be amenable to discuss a settlement unless it is satisfied that it understands all the nooks and crannies of a particular transaction or business. It follows that withholding information from the IRD rarely serves any useful purpose and often creates negative inferences that may induce the IRD to harden its position.

In the case of transfer pricing inquiries, it is clear that the best way to address the IRD's expectations will be to prepare formal transfer pricing studies. Indeed, a transfer pricing study can be structured to include all, or at least most, of the documentary expectations of the IRD. Perhaps more importantly, the transfer pricing study likely will become the battleground on which future disputes with the IRD will be debated.

By controlling the preparation of the transfer pricing study, the taxpayer has a head start in the analysis of the information and will be better able to address the eventual queries of the IRD.

Transfer pricing methods

Many multinational enterprises employ international tax structures where affiliates in Hong Kong perform specific routine functions, without employing valuable unique assets and without incurring any significant risks. For example, stripped-risk buy-sell distributors, contract sales and marketing service providers, and contract manufacturing service companies all are common in Hong Kong. Absent some specific transactions with third parties from which to develop data under one of the traditional methods, such taxpayers often rely on the transactional net margin method (TNMM) and various profit ratios developed from studies of comparable companies to benchmark the appropriate arm's-length profit levels for the Hong Kong affiliate.

Taxpayers might have an initial concern that the IRD will challenge the use of TNMM because DIPN 46 expresses a preference for traditional transfer pricing methods and effectively states that TNMM is a method of last resort. However, a revision of the OECD transfer pricing guidelines Chapters I-III, published in July 2010 (subsequent to the issuance of DIPN 46), removed most, if not all, of the previous reservations related to TNMM.¹⁶ Moreover, an ever-increasing body of knowledge demonstrates that TNMM is now accepted in many jurisdictions,¹⁷ and in our experience, the IRD now also accepts TNMM. It follows that DIPN 46 is not the final word on the topic of transfer pricing methods. Rather, to the extent that the IRD has committed to follow the guidance of the OECD, taxpayers should keep updated of any OECD development and be prepared to demonstrate those developments to the IRD.

Obtaining agreement with the IRD that TNMM is the best method has certain efficiencies for both the taxpayer and the IRD. Since TNMM focuses on the operating profits of the tested party, detailed questions on all sorts of typical group-related expenses (such as product development and general expenditure, royalty payments, cost of sales, intercompany charges and staff benefits or stock options) are not as relevant as the ultimate profits of the Hong Kong affiliate. In many cases, this will reduce the inconvenience and length of the audit by avoiding questions requiring complex searches of past expense records.

Comparable data and adjustments

When the taxpayer has prepared a transfer pricing study, it would be unlikely, at least currently, that the IRD would develop its own list of comparables from scratch. However, taxpayers nonetheless can anticipate that the IRD may disagree with some of the comparables and insist on their removal for the purposes of calculating an appropriate interquartile range to assess the taxpayer's results.¹⁸ Taxpayers can anticipate that, as in other jurisdictions, when the profit margin of the taxpayer is anywhere within the interquartile range (or above), no adjustment would be required, whereas when the profit margin of the taxpayer is below the lower boundary of the interquartile range, the profits will be adjusted to the median.

It appears that the IRD officers involved in transfer pricing disputes with taxpayers have authority to resolve cases with taxpayers, with appropriate approvals from within their own organization. However, if a resolution with the IRD officer is not forthcoming, then the taxpayer will have to resort to the appeal process, starting with the Board of Review. The BOR is an independent statutory body established under the IRO to determine tax appeals, and its members consist of persons of legal training and experience appointed by the Chief Executive of Hong Kong. At any time prior to the hearing of the appeal, the IRD and the taxpayer still may reach a settlement, which settlement must be endorsed by the BOR.

Decisions of the BOR are in principle final and binding except on questions of law, which may be further appealed within the Hong Kong judicial system. There are then three layers of courts: the Court of First Instance, the Court of Appeal and the Court of Final Appeal, Hong Kong's highest judicial body. Unfortunately for taxpayers, the process of appeal can be quite expensive and time-consuming.

Interest and penalties

Unlike many jurisdictions, the Hong Kong IRO does not provide for interest to be paid on back taxes unless and until an assessment has been issued.  It follows that, when the IRD issues an assessment for years past, only the tax payable can be assessed, regardless of how far back the year of assessment may be. The IRD often tries to use penalty provisions as a means of obtaining commercial restitution equivalent to interest but, in a typical case of assessments arising from adjustments to comparables, it is unlikely that penalties can be sustained. Indeed, the honest dispute over whether some companies are truly comparable involves no misconduct on the part of the taxpayer.

The IRD is growing more sophisticated – prepare to defend your pricing policies

Without a doubt, transfer pricing has landed in Hong Kong. It is still early days and the IRD is not yet completely up to speed with the concepts and organization, but there is no doubt that it will quickly become more experienced and sophisticated as time goes by.  Indeed, the need to respond to aggressive tax planning and the international commitments it is making through the development of its tax treaty network makes this a priority.

Taxpayers now must be prepared to defend their pricing policies by preparing properly constructed transfer pricing studies. These studies are important in that they place the battle on a turf controlled by the taxpayer rather than the IRD.

It remains to be seen whether, in time, the IRD will put together its own transfer pricing studies to counterbalance those of the taxpayers.

For more information about transfer pricing issues in Hong Kong, please contact:

In Hong Kong:

Patrice Marceau

Elaine Lu

In the US:

Eric Ryan 

An earlier version of this article appeared in BNA’s Tax Management Transfer Pricing Report, June 2, 2011, published by The Bureau of National Affairs.