Dutch court of appeal rules on the at arm’s length interest rate of a shareholder loan

19 July 2024 – The Amsterdam court of appeal published an interesting decision on how to determine whether a shareholder loan qualifies as a so-called “uncommercial loan”.

This July, the Dutch court of appeal of Amsterdam (the ‘Court‘) issued an interesting judgment that clarifies under which circumstances a shareholder loan qualifies as a so-called  “uncommercial loan”. The Court ruled that the taxpayer could not have received financing from any third party under the terms and conditions of the shareholder loan (‘SHL’), unless at an interest rate that would have been so high that it would have become profit sharing. The Court specifically regarded the absence of a mortgage right and other security rights such as a loan to value (‘LTV’) covenant, as the relevant terms and conditions of the SHL. The Court considered that the benchmark analyses from the taxpayer and the comparison by the taxpayer with its internal rate of return (‘IRR’) not relevant. The qualification of a loan as an uncommercial loan reduces the amount of interest that can be deducted.

In this update, we provide you with a summary of the judgment, and our takeaways.

General

In 2014, the taxpayer acquired an office building for € 17 million, of which 40% was equity financed and 60% was financed through the SHL. The SHL had a 10% coupon and a maturity of 10 years. The SHL did not have a mortgage, a repayment scheme, or an LTV covenant and included the possibility of deferring interest payments in the event of liquidity problems.

The 10% interest rate on the SHL was substantiated with a transfer pricing study prepared by a tax advisory firm in 2016. The transfer pricing report assumes that the SHL consists of a financing mix of 20% senior funding, 20% junior funding and 20% mezzanine funding, because the taxpayer could not obtain this financing only from banks. Based on the Mid Year 2014 Commercial Property Lending Report from De Montfort University, and on the qualification of the acquired office building as a secondary office, the transfer pricing report concluded that the arm’s length range is between 6.9% to 10.9%.

In 2021, an additional analysis was performed, in which the expected IRR of the investment in the office building and the taxpayer’s equity IRR were calculated using Monte Carlo simulations. The taxpayer claimed that the equity IRR of 16.2% confirmed that the SHL’s interest rate was at arm’s length, because it was higher than the SHL’s interest rate of 10%. The expected unlevered IRR for the investment was 13.5%.

The tax inspector did not agree with the 10% interest rate and imposed assessments considering an interest rate of only 2.59%. Before the Court, the inspector’s primary position is that the SHL’s terms and conditions, other than its interest rate, should first be adjusted to arm’s length terms and conditions. Subsequently, the arm’s length interest rate should be determined based on the adjusted terms and conditions. As a result, the interest rate should be adjusted to the interest rate on a loan that is secured by mortgage, has repayment schedule and an LTV covenant. In this case, the highest interest rates that the inspector found in his study for such a loan range between 2.59% to 4.25%, depending on the tenor of the loan.

The inspector’s subsidiary position is that, if the SHL’s terms and conditions are not first adjusted to arm’s length terms and conditions, the loan qualifies as a so-called “uncommercial loan”. An intra-group loan qualifies as an uncommercial loan if the terms and conditions are such that an unrelated party would only have agreed to an arm’s length interest rate that is (effectively) profit sharing. If a loan qualifies as an uncommercial loan, its interest rate is reduced to the “pledge analogous interest rate”. The pledge analogous interest rate is the interest rate which would have been paid to an external lender, if the loan was provided by an external lender under a guarantee of the SHL’s lender. The tax inspector determined the pledge analogous interest rate also at 2.59%.

We note that the dispute between the taxpayer and the inspector also concerns the application of the Dutch REIT regime, but we do not address this part of the dispute in our contribution.

The Court’s considerations

The Court held that the inspector’s primary position is incorrect, because Dutch Supreme Court case law stipulates that the arm’s length character of an intra-group loan should be tested given its other terms and conditions. The inspector is therefore not allowed to adjust these terms and conditions, before testing the SHL’s pricing.

The Court agreed with the inspector’s subsidiary position and ruled that the SHL qualifies as an uncommercial loan. It based its decision on the fact that the taxpayer itself acknowledged that no third party would have agreed to provide the SHL based on its terms and conditions. According to the taxpayer, an unrelated lender would have required a mortgage, an LTV-covenant and prepayment penalties. Furthermore, it would not have provided more funding than 25-40%, at a tenor of maximum six years. This has been confirmed in a third-party investigation conducted by the inspector by an unrelated professional that was involved in the initial transaction.

The Court dismissed the taxpayer’s transfer pricing studies. According to the Court, the financing mix used in the 2016 study is not comparable, because the SHL does not comprise separate types of financing. The 2021 study is dismissed, because the IRR on equity does not substantiate the arm’s length character of this interest rate, nor can it be used to assess whether independent third parties would be willing to provide the loan at the same remuneration.

The Court confirmed the methodology applied by the inspector for determining the pledge analogous interest rate, but adjusted this interest rate “in good justice” to 3.09%, because the tenor of the SHL is longer than the tenor used by the inspector. Furthermore, the Court held that, insofar the SHL is provided by private individuals (15% of the total loan), the interest rate is not adjusted, because the inspector did not successfully argue that these individuals were sufficiently related to the taxpayer.

Takeaways

  • The Court rejected a traditional benchmark study because of insufficient comparability between the controlled transactions and the benchmark data. This shows that traditional benchmarks can easily be challenged by the tax authorities. We recommend performing economic analysis (in addition to benchmarks) to substantiate an arm’s length interest rate.
  • Although the taxpayer seemingly attempted to overcome comparability issues by performing an economic sanity check using Monte Carlo simulations, the Court found that a comparison of the coupon with the expected equity IRR in and of itself is not sufficient to substantiate the arm’s length character of an interest rate. We agree with that. Nevertheless, the expected IRRs on the various instruments can be used to substantiate the loan size and interest rate, if properly used, as they have a very strong relationship with the arm’s length returns on these investments:
    • The unlevered IRR is an indicator of the debt capacity, because, if the acquisition price is equal to the fair value of the property, a high unlevered IRR indicates high risk and vice versa. The higher the risks, the earlier a loan becomes effectively profit sharing.
    • If the equity IRR is higher than the cost of equity, the net present value of the equity will be positive. To assess whether it is beneficial for a borrower to obtain additional funding, one should therefore test the loan’s effect on the equity’s net present value.
    • The cost of debt at the SHL’s leverage level should be proportionate to the risk-free rate and the expected unlevered return. Under normal circumstances, risk increases exponentially, in which case the cost of debt would also increase exponentially from the risk-free return once the instrument is no longer free of risks. The cost of debt is capped at the expected unlevered return if the acquisition price is equal to the fair value. Based on this notion, a cost of debt of 10% at a leverage level of 60% appears very high compared to the unlevered return of 13.5%.

We therefore disagree with the Court’s statement that IRRs are not in line with the arm’s length principle.

  • By arguing that no unrelated party was willing to provide the loan at such “covenant-lite” conditions (and without mortgage), the taxpayer hoped to substantiate the relatively high interest rate of the loan. However, by doing so, the taxpayer confirmed the inspector’s argumentation that the loan is uncommercial, because no third-party was willing to provide the loan under these conditions. This demonstrates that it is crucial to be thorough, careful and complete with argumentation before court.  
  • The Court argues that the fact that the SHL does not comprise multiple tranches means that the blended rate of multiple tranches found by the taxpayer is not comparable. We disagree with this statement, because it means that a blend of financing can be priced differently from one financial instrument with economically the same characteristics. This statement therefore violates the generally accepted economic principle known as “the law of one price”, and implies that arbitration is possible. For clearness’ sake, we note that the above does not mean that the benchmark analysis of the taxpayer is comparable, but we believe that the Court has erred with his argumentation, or at least has not motivated it sufficiently.
  • The Court held that the taxpayer’s Monte Carlo analysis is not in line with the methodologies that are prescribed in the OECD transfer pricing guidelines. Although we agree with the Court that the added value of the taxpayer’s Monte Carlo analysis is not significant, we point out that especially with respect to financial transactions, economic modelling is included as an approved methodology for the substantiation of the arm’s length character in Chapter X of the OECD transfer pricing guidelines. In our view, the use of economic models can add superior value to the substantiation of the arm’s length character of financial transactions. This is supported by the fact that many banks use Monte Carlo analysis to test their equity buffer for Basel regulation purposes.

 

Contacts

Photo Quirijn Knab

Quirijn Knab

+31 20 247 03 03
+31 6 10 89 76 10
quirijn.knab@huygenstax.com

Photo Mark van Casteren

Mark van Casteren

+31 20 247 03 04
+31 6 10 89 88 38
mark.van.casteren@huygenstax.com