Multi-year margin averaging: Administrative Court of Zug decision
A decision by the Administrative Court of Zug on December 5 2024 (Case No. A 2023, 1) involved a taxpayer that reported a negative margin of -21.8% in 2018 but argued that its result should be accepted because its three-year average (2016–18) fell within the interquartile range (IQR) of comparable companies. The taxpayer had only arrived at that outcome through a year-end adjustment in 2018, to align the three-year average with the benchmarking results; i.e., an average operational margin of 1.2% over the three-year period 2016–18 since the margins were too high in 2016 and 2017.
This ‘margin smoothing’ approach — using multi-year averages to justify the tested party’s own results — was rejected by the court. The judgment emphasises that under Swiss tax law, based on the principle of periodicity, each tax year must stand on its own. The taxpayer’s negative result for 2018 could not be ‘rescued’ by averaging, absent any proven extraordinary external cause.
The OECD TP Guidelines: use of multi-year data for comparability analysis
The OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (the OECD TP Guidelines), particularly paragraphs 3.75 to 3.79, recognise that multi-year data can be a useful tool in transfer pricing analysis. The rationale is that such data can offer broader context, especially when analysing volatile industries, cyclical businesses, or markets subject to unpredictable economic shocks.
If a potential comparable company shows a temporary loss due to a one-off event — such as a product recall, litigation, or a market downturn — examining its financials over several years may reveal a more stable and representative picture. In this sense, multi-year data enhances comparability, allowing the practitioner to make better-informed selections of comparables.
Multi-year analysis can also help to identify trends in a taxpayer’s business, assess recurring losses versus isolated events, and better understand the economic rationale behind fluctuations in margins. It may assist in situations where transfer pricing adjustments are delayed due to the nature of the business cycle or regulatory timing, especially in industries such as pharmaceuticals, high-tech, or commodities.
However — and this is the most important element of the issue — the OECD TP Guidelines do not endorse multi-year averaging of the tested party’s results as a default method for determining whether the arm’s-length principle is met. Paragraph 3.79 clearly states that multi-year averaging is acceptable only when it improves reliability in the context of comparability — not to justify results that fall outside the arm’s-length range in any particular year.
In January 2024, the Swiss Tax Conference (STC), in collaboration with the Swiss Federal Tax Administration (SFTA), published a comprehensive commentary on transfer pricing, marking the most detailed official guidance to date. This was complemented by the SFTA’s updated questions and answers on transfer pricing released around the same time. The 2024 guidance from the STC and SFTA confirmed that Switzerland aligns with OECD TP Guidelines methodology in using multi-year data for comparables to improve reliability and identify trends, stating that comparability analyses should focus on the year under review and preceding years only, and generally reject data from periods too distant from the year under review. Thus, Switzerland’s recent guidance mirrors the OECD TP Guidelines in encouraging multi-year analysis for context and comparables.
The periodicity principle according to Swiss tax law
Swiss tax law takes a firmly annual view on income determination, whereas it accepts and supports the OECD TP Guidelines principles of using multi-year data for the comparability analysis.
The concept of periodicity is embedded in Article 58 of the Swiss Federal Direct Tax Act (FDTA), which mandates that income and expenses be assessed for each tax year independently. This means that each tax year must be assessed on a standalone basis.
The aforementioned court cited the transfer pricing expert of the SFTA, who was consulted on the case at hand: “The application of transfer pricing methods does not grant taxpayers the right to make arbitrary adjustments to their taxable profit in individual tax periods, in disregard of the periodicity principle, in order to subsequently compensate for over-taxation claimed in other tax periods”.
Key facts of the case
The court case involved a legal entity based in the Canton of Zug, acting as a limited-risk distributor (LRD) for an international pharmaceutical group. For the 2018 tax year, the company reported an operating margin of –21.8% — a significant loss that triggered scrutiny by the cantonal tax authorities.
Notably, the taxpayer had initially applied standard transfer prices during 2018 but later adjusted them retroactively as part of the year-end transfer pricing review. The goal was not to reflect new economic data or unforeseen events but to lower the 2018 margin so that the three-year average across 2016–18 would fall within the lowest quartile of the IQR of comparable companies.
The tax administration rejected this result for 2018 as non-arm’s length. Referring to the taxpayer’s benchmarking study, it imposed a margin of 1.1%, which corresponded to the lower end of the IQR of comparable companies. The resulting transfer pricing adjustment added approximately CHF9 million to the taxpayer’s taxable income.
The taxpayer objected, citing the OECD TP Guidelines and arguing that averaging was appropriate, particularly considering price lags and timing effects common in pharmaceutical distribution. The Administrative Court of Zug, however, also considering the expert opinions of the SFTA, dismissed this position.
The court found that the taxpayer had failed to demonstrate any extraordinary factors beyond its control to justify the deviation in 2018. There was no evidence of a market disruption, no regulatory delay, and no force majeure event that could explain its significantly negative result and justify it as arm’s length. Notably, the court emphasised that there is no legal basis in Swiss tax law allowing for multi-year smoothing, citing Article 58 of the FDTA as the legal anchor.
Economic substance and transfer pricing adjustments
Beyond the technical transfer pricing argument, the court also raised concerns regarding the economic substance of the intercompany transactions. It noted that the taxpayer’s retroactive adjustments were not sufficiently documented and appeared to be aimed primarily at achieving target margins, rather than reflecting real economic performance.
This undermined the credibility of the taxpayer’s entire position — particularly given the lack of contemporaneous evidence explaining the magnitude of the 2018 loss. The judgment reinforces the importance of aligning pricing policy with actual business activity and ensuring that any adjustments are defensible and economically grounded.
Key takeaways
While multi-year data remains useful in evaluating the consistency of comparables or identifying anomalies, it cannot be used as a shield for low-performing years — especially where pricing has been adjusted retroactively to achieve a target average. For practitioners, this creates a clear division:
Acceptable use – multi-year data to support the selection of third-party comparables (e.g., explaining volatility or filtering out one-off results); and
Unacceptable use – averaging the tested party’s own results over multiple years to justify an outlier year, or – as in the Zug case – to retroactively adjust transfer pricing outcomes to bring the average within the desired range.
In other words, while multi-year data may support the range, it does not rescue the tested party from being outside of it in a specific year – unless extraordinary justification is provided.
Given this evolving Swiss jurisprudence in transfer pricing matters, taxpayers with operations in Switzerland should take several steps to ensure compliance:
Annual documentation – ensure that each year’s transfer pricing documentation justifies the tested party’s result as arm’s length on a standalone basis, referencing the relevant benchmarking study and economic circumstances of the year.
Economic justification for deviations – if a year shows a negative or an unusually low margin (for example, for an LRD, as in the case at hand), the company should immediately document any extraordinary events – such as supply chain disruptions, product issues, regulatory delays, or economic downturns – and quantify their impact.
Cautious use of multi-year data – limit the use of multi-year data to benchmarking comparables and not to justify the taxpayer’s own results. Multi-year averages may offer useful insight into market trends and comparability, but they do not override the legal obligation to justify transfer pricing results on an annual basis.
Substance over form – ensure that intercompany transactions have economic substance, that pricing is based on real market behaviour, and that retroactive adjustments are defensible and well documented.
This case is emblematic of a broader shift in Swiss tax practice and greater scrutiny of transfer prices. Cantonal tax administrations are increasingly well equipped, drawing on in-house transfer pricing experts or transfer pricing specialists from the SFTA.
