On November 11, 2022 the Dutch Supreme Court rendered judgment in an important case concerning the VAT recovery right of a bank. The bank in question wanted to determine the VAT recovery right on its mixed costs on the basis of a financial analysis of the profit and loss (‘P&L’) per product. It believed that this VAT recovery method constituted an ‘actual use method’ permitted for determining the VAT recovery right. The Court of Appeals in Den Bosch had ruled in favor of the taxpayer, but in the cassation proceedings the Supreme Court ruled that the Den Bosch Court of Appeals had interpreted the actual use method too liberally. The Supreme Court also ruled that the taxpayer should have applied the actual use method to all the mixed costs.
This case is not only relevant for financial institutions, but also for other taxpayers performing VAT-taxed and VAT-exempt services, such as parties in the public sector, education and healthcare.
1. Background and points of law
The taxpayer in the case at hand is a bank that performs VAT-exempt and VAT-taxed services. All the costs incurred by the bank can be regarded as mixed costs. Its turnover consisted for the most part of interest income and commission income. The majority of the interest income was exempt, while a large part of the commission income was subject to VAT. The bank had transferred part of its mortgage receivables to separate securitization companies. It had passed on the interest received on these receivables to the aforementioned companies.
The taxpayer prepared a financial analysis of the ‘P&L per product. To this end, the costs were apportioned to the various product groups by means of three interval-based allocation formulas (based on time registration, actual products purchased and proportional distribution). This resulted in the bank’s mixed costs being allocated to the various product categories.
In its VAT return, the taxpayer recovered VAT on the mixed costs in accordance with the turnover pro rata method. Under this standard method, the VAT recovery is calculated in accordance with the ratio: turnover that allows VAT recovery (in particular VAT-taxed turnover) to total turnover (including VAT-exempt turnover). In doing so, no account is taken of (i) the interest expenses paid, (ii) the interest the bank received on its notes in the securitization companies, and (iii) the interest passed on to the securitization companies. In essence, the taxpayer wanted to use this procedure to apply the recovery of VAT in accordance with actual use.
The Court of Appeals in Den Bosch had ruled that the VAT recovery calculation based on the P&L per product results in a more accurate determination of the VAT recovery than the turnover pro rata method and is based on (sufficient) objectively and accurately established data. The taxpayer therefore was allowed to determine its VAT recovery on mixed costs on the basis of this actual use method. The Deputy Minister of Finance initiated cassation proceedings. The taxpayer initiated a cross-appeal in cassation and argued that the interest expenses paid can be deducted from the interest income received. It also claimed that the interest that was passed on to the securitization companies could be excluded from its turnover.
2. Supreme Court judgment
The Supreme Court upheld the appeal in cassation, quashed the judgment by the Den Bosch Court of Appeals and confirmed the judgment by the District Court of Zeeland-West-Brabant. In its judgment, the Supreme Court examined the actual use.
Firstly, it addressed the test that must be performed when applying the actual use method. The Supreme Court ruled that the Den Bosch Court of Appeals erred in law by ruling that it was sufficient that an actual use method led to a more accurate result than the turnover pro rata method. The Supreme Court reiterated its previous conclusion that an actual use method must be based on objectively and accurately ascertainable data, on the basis of which the actual use of mixed costs can be objectively and accurately established.
The Supreme Court then addressed the tax inspector’s assertion that all the mixed costs must be taken into account when deciding on an actual use method. It ruled that the actual use method can only be applied if it plausible that the actual use of the mixed costs as a whole is not in line with the turnover pro rata method. The ‘as a whole’ implies that all the mixed goods and services used by the VAT taxable person must be taken into account. In the case of a VAT group, all the mixed costs of all the members of that VAT group must be taken into account. If a VAT group consists of several separate banks and the taxpayer makes use of the approval in the Bank Decree (Decree by the Deputy Minister of Finance dated November 9, 1982, no. 282/15703) to calculate the VAT recovery for each separate bank, then the mixed costs of all the business units of that separate bank must be taken into account. The taxpayer had made use of the approval to calculate the VAT recovery for each bank but wanted to use the actual use method to calculate the VAT recovery for one of the two banks. However, the Supreme Court inferred from the court documents that the tax inspector had argued before the Court of Appeals − an argument that went unchallenged − that the taxpayer had not taken all the costs of all the business units of that bank into account when calculating the VAT recovery on the basis of actual use. It is unclear whether the Court of Appeals took this unchallenged argument into account in reaching judgment in this case. The Supreme Court ruled that the Court of Appeals either assumed an error of law or wrongly failed to take the tax inspector’s argument into account in its ruling. The Supreme Court ruled that the tax inspector’s unchallenged argument must be the starting point for the cassation proceedings, which means that in the cassation proceedings it must be assumed that the taxpayer did not apply the actual use method to all the bank’s mixed costs. The taxpayer thus lost its case and the actual use method cannot be applied.
In the case at hand the taxpayer had also contended that in determining the pro rata the interest expenses paid should be deducted from the interest income. The taxpayer also wanted to exclude the interest that had been passed on to the securitization companies from the calculation of the turnover. These arguments were rejected without any further substantiation.
3. Practical aspects to consider
This was an eagerly-awaited Supreme Court judgment. However, the outcome was not as hoped. The Supreme Court has not completely rejected the actual use method but has left the door ajar. Despite the fact that the EU Court of Justice has repeatedly ruled that an actual use method may be applied if this leads to a more accurate recovery calculation, the Supreme Court has persisted with its previous position that an actual use method must be based on objectively and accurately ascertainable data, on the basis of which the actual use of mixed costs can be objectively and accurately established. The Supreme Court neither further defined the terms ‘objective’ and ‘accurate’, nor did it rule on the method applied by the taxpayer. Proving that the actual use method leads to an objective and accurate result and is applied to all costs, remains difficult given that no guidance has been provided.
We can imagine that the taxpayer had done that in the case at hand. The taxpayer had a registered controller carry out an analysis for which assumptions and presumptions were used for financial reasons to arrive at a P&L for each product. As far as we are concerned the latter is an objective and accurately prepared calculation, or at least justifies asking the tribunal of fact to assess whether this is the case. The Supreme Court did not address this in its judgment. It would have been very welcome for practical purposes if the Supreme Court had provided guidance on how to establish whether data is objective and accurate.
If you would like to discuss this judgment, feel free to contact the advisors of KPMG Meijburg & Co’s Indirect Tax Financial Services Group or your usual advisor.