Netherlands: Extended scope of the innovation box

Patrick T F Schrievers	Gert-Jan Hop

The Netherlands government is promoting engagement in research and development (R&D) activities through a preferential corporate income tax regime and specific R&D tax incentives granted to employers with regards to salaries paid to employees who carry out qualifying R&D activities and related capital expenditures. There are a number of extensions to the new innovation box legislation, which applies from January 1 2017, but has retroactive effect from July 1 2016.

Identifying which assets qualify for the innovation box is of fundamental importance. Paragraph 34 of the OECD's report on BEPS Action 5 starts by noticing that the only IP assets eligible for tax benefits are patents and other IP assets, which are functionally equivalent to patents if those IP assets are both legally protected and subject to similar approval and registration processes (where such processes are relevant). Based on this starting point, the Netherlands government has amended its innovation box legislation and has itemised various IP assets that should give access to the Netherlands' revised innovation box. In this respect, a distinction is made between small and medium-sized companies (SMEs) and other taxpayers. Qualifying intangible assets in respect of SMEs (i.e. no more than €50 million ($53 million) in global group-wide turnover and the taxpayer cannot, itself, earn more than €7.5 million per year in gross revenue from all of its IP assets using a five-year average for both calculations) are self-developed intangible assets from R&D activities in respect of which R&D wage tax certificates from the competent Netherlands governmental agency (i.e. Ministry of Economic Affairs) have been obtained. For other MNEs, the definition of qualifying intangible assets contains a limitation in comparison to SMEs. As an additional condition, such intangible assets should also qualify as one of the following:

• Patents or plant breeder's rights;

• Patents or plant breeder's rights applications;

• Software program(s);

• Market authorisation of an (orphan) medicinal or a veterinary medicinal product;

• Supplementary protection certificates for medicinal products for human use, for veterinary medicinal products and plant protection products;

• Utility models; or

• Intangible assets, related to intangible assets qualifying under any of the above.

It should also be noted that exclusive licenses to use intangible assets qualifying in a certain way, in a certain area, or for a certain period of time are also recognised. This will benefit MNEs that typically segregate the legal ownership of intangibles from the economical ownership (i.e. development, maintenance and execution of the R&D activities).

This latter point will particularly benefit knowledge institutes, research centres and other companies that actively cooperate with one another (as public-private partnerships) in R&D projects. These R&D projects require an integrated approach, pursuant to which science and industry, supported by government, share their knowledge and expertise. The IP developed through these projects is governed by project agreements. Generally, several participants have taken steps regarding the invention at issue (i.e. for patents) and have an equal undivided ownership interest in the IP developed. These joint IP owners typically agree amongst themselves on the relative allocation and related rights of the IP developed. If these R&D projects are set up as jointly owned companies that are recognised as non-transparent entities for tax purposes, the owners of these exclusive rights were not able to enjoy the benefits of the innovation box in respect of income earned via these exclusive rights under the old rules. At best, only the lead partner that has coordinated all activities regarding the IP developed and has applied for the patent in its name could (at least) try to apply the innovation box to its proportionate income share. These problems should be solved under the new provisions (although there is still some debate).

An important benefit of the new legislation is also the introduction of software to the innovation box. This confirms that copyrighted software shares the fundamental characteristics of patents, since such software is novel, non-obvious and useful and it is unlikely that core software developments will be outsourced to unrelated parties. This amendment is welcomed by most Netherlands software companies that typically had to rely on R&D declarations, but may now also and in addition rely on copyrights to support their innovation box application. Interestingly, it should be noted that the European Patent Convention, and in line therewith national patent laws, provides that software, as such, is not patentable. In practice, however, many software related inventions (that have a novel and surprising effect on a technological process) have been granted a patent. This is, however, not very common (we have seen it in a very limited number of situations) and we expect that a copyright protection is more appropriate.

Under the revised Netherlands innovation box regime, qualifying income is determined per qualifying intangible asset or per coherent group of qualifying intangible assets. It is, however, expected that the main methods used under the current innovation box regime to determine qualifying income will remain applicable to a certain extent. This includes, in particular, the commonly used profit split method. Although the new law does not shed much light on the implementation, it is expected that this will increase the administrative burden for companies, as well as the Netherlands tax authorities. It, however, also creates opportunities for companies to prepare detailed supporting documentation, which may increase benefits. In this respect we have listed a number of comments.

The initial stage in performing a profit allocation can be complex. In addition, the method of allocating the profits and any assumptions made in doing so need to be documented. Internal data may be helpful where the profit allocation factor is based on a cost accounting system, e.g. headcounts involved or time spent by a certain group of R&D employees on certain tasks. Typically, however, the profit allocation should be based on a supply chain or functional analysis. This should determine where and how value is created in the business and also determine the level of integration between products (which may determine the level at which profits or revenues should be allocated) and the economically relevant contributions (which may determine the factors to use to allocation of profits).

Interestingly, we often see that additional arguments related to certain products or product groups that are associated with unique intangibles are visualised and captured in this kind of analysis (i.e. first mover advantages, unique R&D contributions, protection due to barriers to entry to potential competitors, among others). This often is an unexpected benefit of any supply chain or functional analysis and supports reasons why additional products should be attributed to R&D based on the arm's-length principle and the applicable nexus approach (OECD BEPS Action 5).

Subsequently, the financial data needs to be segregated and allocations need to be made so that the profits solely relating to the R&D activities can be identified. Any company will need to segregate its financial data in a way that reflects the revenues, costs, and profits in relation to R&D investments and R&D capital employed. The choice of the measure of profits to be allocated will depend on the nature of the products and the extent to which these products are integrated.

For example, company Q produces two products, A and B. Both products are the result of innovative, complex R&D activities performed by the company. Company Q shares the production of A and B and the risks associated with the success of the products. However, the selling and other expenses are largely unintegrated. Product A needs significant selling support and product B does only require limited after sales service. Using a profit allocation based on combined operating profits (including production and selling expenses) would have the potential result of sharing selling expenses that are unrelated. In such a case, an allocation of gross profits (before selling expenses) may be more appropriate and reliable for A and B since this level of profit captures the outcomes of R&D investments, market and production activities that the products share together with the associated risks.

It should be noted that we expect that in a number of day-to-day cases asset-based or capital-based factors can be used to allocate profits where there is a strong correlation between assets or capital employed and creation of value. Identification of R&D assets can sometimes be difficult because not all valuable intangibles are legally protected and registered and not all valuable intangibles are recorded in the accounts. In this case, an essential part of the analysis is to identify what intangibles are contributing to the production of each product or product group. A significant issue for the reliability of any allocation factor is the determination of the relevant period of time in relation to the appropriate allocation key (e.g. assets, costs, or others). A difficulty arises because there can be a time lag between the time when expenses are incurred (especially R&D expenses) and the time when value is created.

For example, in the case of a cost-based factor, using the R&D expenditure on a single-year basis may be suitable for some ICT-related cases, while in some other cases it may be more suitable to use accumulated R&D expenditure (net of depreciation or amortisation) incurred in the previous as well as future years. Depending on the facts and circumstances of the case, the determination of the appropriate allocation factor may have a significant effect on the allocation of profits between the products/product groups and ultimately the net tax benefits.

Taxpayers should take into account that any advance pricing agreement (APAs) concluded with the Dutch tax administration, such as innovation box agreements, will be subject to legislation governing the exchange of information. The Netherlands tax administration will exchange information on unilateral APAs (e.g. innovation box rulings) that have an impact on taxation in the other countries (which may be affected in terms of taxation). In this respect, the Netherlands tax administration will focus on transfer pricing principles when discussing and concluding innovation box agreements to avoid transfer pricing discussions with other countries following disclosure of the innovation box ruling. Subsequently, taxpayers may also want to consider not applying for APAs governing the application of the innovation box. Although this might seem theoretical or even far-flung, observing that almost all innovation box rulings are APAs, from a theoretical perspective it might benefit both taxpayers and tax administrations. In the end, taxpayers based on an interpretation of the facts and circumstances can apply the innovation box legislation without discussing upfront the implication with the tax administration. However, any position taken needs to be supported with proper documentation and preferably should be available upon request to the Netherlands tax administration.

Patrick T F Schrievers (patrick.schrievers@noviotax.com) and Gert-Jan Hop (gert-jan.hop@noviotax.com), Nijmegen

NovioTax

Tel: +31(0) 610 246 140 and +31(0) 627 328 647

Website: www.novio.tax