The OECD report on Base Erosion and Profit Shifting (BEPS) concerning the perceived abuse of commissionaire structures Jos Peters is the Senior Tax Partner at Merlyn International Tax Solutions Group Introduction On July 19, 2013, the OECD released a comprehensive report, detailing a two year program for its member states (and other countries if they want to), to combat aggressive tax planning structures, revise some transfer pricing rules and introduce other ways to combat the erosion of these countries’ tax bases and profit shifting. The report is already widely known as the BEPS report. In the coming issues of World Commerce Review I should like to discuss some of the plans which the OECD has unfolded in its BEPS report, to see how effective the new approach might be. It is to be kept in mind in this regard that the OECD is not in a position to actually create new tax laws or change tax treaties. All it can do, and has done over the years in this respect, is to report findings of situations where multinational enterprises (MNE’s) do not seem to report their ‘fair share’ of taxable income in the countries of their operations and to suggest ways on how this situation can be mended. The report has a very short timeframe: most measures should have been adopted before the end of 2014. One of the action points that has been announced, is to bring so-called commissionaire structures under the working of the permanent establishment article of tax treaties. This will be my focus for this WCR issue. An explanation of the commissionaire concept Large MNE’s obviously derive most of their revenue from large markets. These markets are almost always in high tax countries. But revenue streams are not the predominant criterion for the payment of corporate income tax: this tax is payable on the taxable profits of the subsidiaries of these MNE’s in their countries of operation. So there has, logically, always been a demand for tax planning whereby high turnover in a given jurisdiction, even if this would give rise to high profits, can still be construed in such a manner that taxable profit remains low. In the mid-nineties of the previous century, just around the time that transfer pricing rules were introduced to avoid that MNE’s would, by using artificial prices for intra group transfers of goods and services, be able to realize their profits in countries with a relatively low tax burden, the commissionaire concept was discovered by the major firms occupying themselves with international tax planning (including the big four accounting firms which all have a significant international tax division). As is often the case with tax planning, legal instruments that have local significance only, appear to have the ability for much wider use and can be seen as a ‘tax discovery’. One example that I will not further elaborate on here, are the Dutch Cooperative Associations. Even ten years ago these legal entities were hardly used in the Netherlands outside economic sectors like fishing and farming, but nobody can stop other businesses to use a similar set-up if this would be beneficial. So today, the Netherlands knows many hundreds of ‘cooperatives’, because these entities were, at least till early 2013, not subject to dividend withholding tax. So by interposing a ‘coop’ between a foreign entity and a Dutch intermediate holding company or operating company, one could avoid the then 25% (now 15%) Dutch dividend w/h tax in cases where the treaty rate was higher than 0% or in case there was no tax treaty with the Netherlands at all. Commissionaires were also hardly used by MNE’s till someone, 20 years ago, spotted the fact that the newly emerging transfer pricing rules would allow MNE’s, if they would set up their sales subsidiaries in the countries of their major operations as such, to reduce taxable income in these countries despite high turnover and high profitability of the products or services sold. A commissionaire, in the countries where this notion exists as part of a country’s legal system, is an entity that sells goods and/or services in its own name but for the risk and account of an undisclosed principal. Because it operates it its own name, it is not, legally, an agent of its principal: in agency contracts the agent sells goods and/or services for a disclosed principal. The customer, when dealing with an agent of, say, ABC Inc in the USA, knows he is buying from ABC Inc. But if ABC Inc sets up a subsidiary in, say, France, with the name ABC SarL, the customer is not informed of the fact that ABC SarL might operate for the risk and account of ABC Inc. However, from an international tax viewpoint, an agent and a commissionaire are treated quite differently, at least in situations where the agent is ‘dependent’ on his principal. A dependent agent creates a so-called permanent establishment for the principal in his country of operations, whilst a commissionaire (arguably) does not. A commissionaire is perhaps economically an agent of his principal but legally he is not. The difference between both structures is that a dependent agent (an agent that only has one principal) leads to a tax liability for ABC Inc in the country of the agent and ABC Inc will have to allocate part of its sales profits to such agent, which will become taxable in the agency country (confirmed by the tax treaty, if any exists, between the USA and the country of the agent). A commissionaire will only have to report a type of service income: it’s activity is to renders services to its principal in a rather riskless manner (the risks are for the principal) which under prevailing transfer pricing principles leads to a much lower taxable income than an agency structure whereby part of the sales profits of ABC Inc become taxable in the country of the dependent agent as well. So commissionaire structures looked like a promising concept and if they would be accepted by the tax courts, the MNE’s would have achieved the first step of their long list of goals: profit shifting from high tax jurisdictions to a low tax jurisdiction. Ireland, in many cases, was the place of choice, firstly because of its low tax rate (12.5% compared to an average 30+% in the larger European jurisdictions and regions) and secondly for language reasons, I assume. After all, the first commissionaire structures were set up by US based MNE’s. As we all know, the profits did not stay in Ireland too long: a further tax reduction was achieved by substantial intercompany charges from within the MNE to the Irish entity for the technology embedded in the products and/or services, sold by the Irish entity through its commissionaires. Everyone, by now, will have heard about the ‘Double Irish/Bermuda’ structures many US based MNE’s are using to reduce taxation on their foreign profits to very low percentages indeed. One of the BEPS action points is to revise the ‘dependent agent’ meaning in tax treaties, by adding commissionaires to the definition of a dependent agent, because the commissionaire, if he is economically dependent on his principal, represents his principal in an economical sense. This magazine is not the forum to discuss whether such an extension of the definition of a dependent agent can be introduced for existing treaties or should be reserved for new tax treaties only, but everybody will understand that this will become a big issue. Tax treaties are revised once in 25 years on average so the OECD will have to find a solution here if the BEPS program should be operating by the end of 2014... The question for the MNE’s that operate commissionaire structures is, if they should anticipate the OECD action plan concerning commissionaire structures and if so, find out if there is a workable alternative. I believe there is. Limited risk distributors The decision by the large multinational tax consultancy firms to give off a ‘should’ or even stronger opinion on commissionaire structures in the mid-nineties of the last century, was of course not taken lightly and the result of a long lasting debate amongst the member firms in the various countries. Some advisers clearly believed that a commissionaire would not work in their country if it had only one principal. This would make the relationship ‘dependent’ and in a ‘substance over form’ approach, the economical dependency might well override the legal independency resulting from the fact that the customers would not be aware of the fact that the products or services they purchased were purchased for the risk and account of another legal person than the one they did business with. But there were more hurdles: many countries did not acknowledge the concept of a commissionaire in their legal system and relying on a foreign law concept is dangerous: tax authorities and tax judges do not normally want to engage in interpretation of foreign law and will look at the arrangements in their own laws that most closely resemble the foreign contract or structure. The outcome of an analysis of commissionaire structure in countries unfamiliar with the commissionaire concept would also in these cases put additional risk on the possibility that a tax judge would let the economical aspects (‘dependent agent’) prevail over the legal aspects (acting in one’s own name). The essence of a commissionaire structure is profit shifting: by reducing the various business risks of a sales entity in a high tax jurisdiction, the taxable profits of the distributor should go down under the basic transfer pricing principles that each entity in a MNE group should be properly rewarded for: a) Its functions; b) Its risks; c) Its assets in use (including intangible assets) Clearly, if distributor X in country X operates as a commissionaire for principal Y in country Y, this leads to risks shifting. Under proper structuring of the commissionaire agreement, the following main risks can almost automatically in part or in whole be shifted from the distributor to the foreign principal: 1) Bad debt risk; 2) Currency risk 3) Inventory risk What results, in a typical transfer pricing analysis, is a sales entity or distributor that gets rewarded for its functionality on the basis of the so-called ‘resale minus’ method: the entity gets a percentage of the sales price for each product or service sold. This leads to substantially lower pre-tax margins than the ones one will find if a sales or service entity assumes all of the basic economical risks itself. For the MNE as a whole, it really does not make much difference which group company bears a given risk: all will end up at the bottom line of the group’s parent, in the end. On further analysis, before the large international tax adviser attached their OK, in the form of a positive tax opinion, to commissionaire structures, they already invented the ‘limited risk distributor’ as an alternative for a commissionaire: stripping business risks out of a sales entity in the group could also be done by contractual arrangements between members of an MNE, other than concluding a commissionaire agreement: the production entity or group parent company could also outside a commissionaire environment reduce a number of business risks for its sales entities by contractually assuming these risks itself. Currency risks can obviously be avoided by the group selling its products and/or services to the distributing entity in the currency of sale rather than in the currency of production. And inventory could either be delivered to the distributor on a consignment basis (so the distributor only becomes the owner of the item once he has found a customer for it) or by giving the sales entity the contractual right to ship unsold and obsolete product back to the producing group entity for its historical acquisition price. For dealing with bad debts, similar contractual arrangements could be made. In this way, the distributor operates not only in its own name, but also for its own account, albeit that it does not assume certain risks or has contractually arranged that those are born by the production entities in the group. Such a ‘limited risk’ sales entity, in no way, acts as an agent for a principal elsewhere in the group, so the risk of being seen as a dependent agent in the sense of the OECD comments to its Model Tax Treaty, applicable to genuine bilateral tax treaties is much more remote than via commissionaire structures. In the last several years, commissionaire structures have been tested before the supreme tax courts of a fair number of countries and in the cases I have seen, the tax judges have taken the (expected) legal approach: a commissionaire is not a dependent agent of his principal, even though, economically, he operates to a large extent for the risk and account of his principal. Conclusions If the OECD now sees a benefit in rewriting the 'dependent agent' rules in its comments to its Model Tax Treaty, it may become the victim of an optical illusion: even if this could be done on short notice and even if it could be made to work also in existing situations, ie. the new interpretation could be launched as a refined interpretation of an existing phenomenon and not as a new interpretation, which is a big question all by itself, then all the OECD will likely find is that MNE’s will have changed their intercompany relationships between the producing entities of the group and the distributors, from commissionaires into limited risk distributors. MNE’s should, of course, take a close look at any existing commissionaire agreements in place today, to determine how those should be cancelled and replaced by the agreements that fit a limited risk distributor relationship, without additional tax cost. One important element here will be timing: contracts if they are to be ‘at arm’s length’ cannot normally be cancelled and replaced by something else on very short notice. Independent parties would normally include penalty clauses for early termination of distribution agreements. So it might be better not to wait to see how the OECD plans to live up to (this aspect) of its anti-BEPS report and to start the contractual investigations now, to make sure that one is ready if needed. One of the action points that has been announced, is to bring so-called commissionaire structures under the working of the permanent establishment article of tax treaties