Svenske Shell AB imported crude oil from its UK sister company SIPC over a five-year period. Imports included the purchase and shipping of crude oil to the port of Gothenburg i Sweden from different parts of the world. The price of the oil was based on a framework agreement entered into between the parties, while the freight was calculated based on templates with no direct connection to the actual individual transport.
The tax authorities considered that the pricing in both parts was incorrect and therefore partially refused deduction of the costs of oil imports. The assessment (and the later judgement of the Supreme Administrative Court) was based on the wording of the former Swedish “arm’s length” provision dating back to 1965.
Decision of Court
The Court did not consider that a price deviation has been sufficiently established where the applied price of only a single transaction deviates from the market price. Applying such a narrow view on price comparisons in general lacks support in the preparatory work, nor can it be justified in the light of the purpose of the legislation, which is to prevent unauthorized transfers of profits abroad. Where the controlled parties have continuous business transactions with each other, it is more aligned with the purpose of the legislation to focus on the more long-term effects of the bases and methods of pricing applied during the period under review. This means that “overprices” and “underprices” that have occurred within the same tax year should normally be offset against each other. According to the Court, it is therefore often necessary to make an overall assessment of the Swedish and foreign company’s business transactions with each other.
The Court also concluded that it is sometimes possible to deviate from the principle of separate tax year’s when applying the arm’s length provisions. A pricing system that in a longer perspective is fully acceptable from an arm’s point of view can lead to overcharging in one year and undercharging in another year. It can also lead to costs in the form of premiums for a number of years leading to higher incomes or losses at a later stage. For a period of three years, Swedish Shell had applied one and the same business strategy with regard to its oil purchases. The strategy was considered justified for business reasons, which meant that the results during two of the tax years were taken into account in the assessment of the results in the third year.
The price method used in the case was the market price method (CUP). This method was chosen as no material had been presented that could form the basis for using any of the other methods. The price test and choice of method had to be made in the light of the information available on the price conditions in the crude oil and freight markets.
The Court stated that a market comparison presupposed that the contractual conditions and the market situation could be established. The sale of crude oil from SIPC to Svenska Shell had taken place on CIF terms (cost, insurance and freight) in the sense that SIPC had been responsible for shipping and insurance of the oil sold.
It could have been worthwhile, the Court held, to base the arm’s length test on a comparison between that total price and the CIF prices which occurred in similar transactions on the market between independent parties. However, there was no investigation of such CIF prices and in fact it was the case that SIPC’s and Svenska Shell’s contractual relations in significant parts had no direct equivalent on the market during the relevant time period.
Even though the sale of crude oil had taken place on CIF terms, the contractual relationship between Svenska Shell and SIPC had contained separate agreements on the pricing of crude oil and shipping. As a result of these agreements, prices of oil were more similar to FOB (free on board) prices and the prices of freight could be linked to standards that directly or indirectly reflected market prices. A separate price comparison of oil and shipping was therefore possible.
According to the Court, there were also other reasons for making such a separate assessment of the prices, since the fundamental problems that arose during the arm’s length examination in the two areas differed in significant respects.
Swedish Shell claimed that SIPC had the function of an independent trader and therefore took out a certain trading margin on crude oil sales. The Swedish Tax Agency claimed that SIPC was only a group-wide service body and therefore not entitled to any profit margin at all. However, the Swedish Tax Agency accepted a certain remuneration for the services provided by SIPC.
The Court found that SIPC had borne certain risks in its purchasing and sales activities and that these risks were such that a certain trading profit was justified. To assess the trading margin when pricing crude oil, different types of list prices were used as a comparison. Some of these list prices contained a certain trading margin that amounted to different levels. The Court found that it was not possible to determine any general levels for the size of the trading margin based on the material. An acceptable margin therefore had to be estimated in the individual case. The Court did not find it clear that the trading margin taken by SIPC on crude oil exceeded the level that would have been agreed between independent parties.
When it came to the pricing of crude oil, different price types were used based on price quotations in different markets. The Court assessed the extent to which the different price types could be used for purpose of pricing the controlled transaction.
One of the price types used was the US list prices. These were average prices that the US tax authorities produced as comparative prices for one year at a time. The list consisted of a large number of sales worldwide. The Court found that these list prices were based on a very broad and elaborate statistical data from transactions between independent parties. On that basis the Court found that the list prices provided the safest guidance in comparison to other price types for the purchase of OPEC oil and Norwegian North Sea oil.
With regard to freight pricing, one of the circumstances was that SIPC had access to a significant freight fleet with vessels in different size classes. SIPC also applied a shipping pattern that included far-reaching coordination of transports throughout the Group. The principles for pricing of the freight services that SIPC performed on behalf of Svenska Shell AB were determined in advance on a quarterly or annual basis based on expected crude oil quantities and freight routes. Svenska Shell also made recurring commitments about the tonnage volume that the company would use during a subsequent period. If Svenska Shell had failed to fulfill these commitments, it would have had financial consequences.
Quotations based on an average of shipping costs reported on the market during a certain period, formed the basis for a pricing model with predetermined shipping prices that depended on the size of the cargo and the transport route.
The model was to some extent standardized and the compensation was calculated regardless of the vessel size and route that was then used in reality. Standard surcharges were made for reserve tonnage and transports to Nigeria. If the pre-specified need for tonnage was changed, certain adjustments were also made. The Swedish Tax Agency considered that the pricing should take into account the actual size of the vessel used and the route.
The Court found that what SIPC had provided more than just the service of transporting crude oil to Gothenburg in Sweden. SIPC had been active as a carrier on behalf of several Group companies, and in order to achieve coordination gains (group synergies), a relatively complicated freight pattern was followed. This pattern had no direct equivalent on the market and according to the court, the same applied to the shipping services that were provided.
The Court found that only the market price method could be used for assessment of the freight services prices. As the freight services had no equivalent in the market, a hypothetical assessment had to be made of what an independent crude oil importer, in the same situation as Svenska Shell, would have been prepared to pay for the services provided by SIPC.
The basis for the assessment of the freight pricing was mainly quotations based on an average of freight costs reported on the market during a certain period. The calculation basis for the freight price quotations included agreements concluded several years earlier and the quotations reflected to some extent historical market situations. The Court emphasized the importance of using these data with some caution. According to the Court, however, the quotations could be used in a price comparison if the pricing system was applied consistently over a longer period of time.
The Court stated that an important principle to consider in the arm’s length examination is that the examination should not be based on hypothetical transactions but on the actual transactions chosen by the parties. It may seem that the method chosen by the lower courts, which was based on the principle of taking into account the actual size of the vessel and the route (vessel / voyage), met this requirement. However, in view of what has emerged about the shipping pattern applied by SIPC and the motives behind the applied pricing method, such a conclusion could be questioned. It could be argued that the method chosen by the lower courts deviated from the actual circumstances by not taking into account the fact that the individual transports formed part of a complicated freight pattern that included groupage etc. in the interests of several companies.
In the Court’s view, there were grounds for claiming that the standard method applied by the company, took better into account the special circumstances which existed compared to the vessel/voyage principle.
The Court approved of the pricing model but then examined how the model was structured and how the various elements in the model related to the arm’s length principle. All in all the various elements in the pricing model were considered to be compatible with the arm’s length principle. Another model was used for FY 1982, which was also considered arm’s length.