The Interplay of R&D Works and Formal Incentive Effect Rules
The issue mentioned in the title appears at first sight rather uncontroversial, as it is instinctively understood that research and development (R&D) works do not form part of an investment towards which they are directed, but this brief note raises certain issues arising in that respect which have led to a prolonged dispute and litigation recently resolved by a supreme national court.
The point of departure of EU law is that State aid is as such principally prohibited and may only be allowed in specific circumstances. It is logically implied, and has been reiterated by the EU courts and the Commission throughout the decades, that even when such specific circumstances arise and State aid can be granted by a Member State, it must be strictly limited to what is necessary and proportionate to attain objectives of the aid concerned.
In that context, State aid is supposed in broad terms to be awarded to incentivize certain desirable actions of undertakings, such as launching an investment in an underprivileged region, creating jobs or researching useful innovations. Consequently, against these requirements of necessity and proportionality, aid may only be granted when it can be demonstrated that it does have a genuine potential to affect decisions of beneficiary businesses, by inducing them to pursue the goals and activities which the aid is intended to promote. This is referred to as the incentive effect requirement.
The Purpose and Shape of the Formal Incentive Effect Requirement
In practice, in many areas of EU law the incentive effect of a particular aid measure is first assessed in a strictly formal manner, through a requirement that an aid beneficiary must submit its application for aid concerning a particular project before the start of work. That requirement is based on a presumption that if prior to the application the beneficiary had started implementing the project, the aid could not be deemed as necessary by the beneficiary. This rule runs through State aid law as one of the fundamental conditions for compatibility of aid with the internal market. In certain larger or more complicated cases, mere formal incentive effect does not suffice. A substantial, in-depth test must be carried out. That being said, the focus of this brief comment is on the formal test alone.
Some of the most notorious areas where this formal test for the incentive effect is being applied is General Block Exemption Regulation (GBER).2 The GBER removes the notification and standstill obligations for a vast number of aid measures awarded by EU Member States in a variety of areas – from regional policy and local infrastructures, to environmental protection, climate action and energy. Member States can grant State aid without being required to notify it to the Commission if they ensure that (among other rules) the beneficiaries submitted their applications for aid for specific projects before they started work on these projects.3 Therefore, given the important benefits offered by GBER (and the substantial risks involved in case of its misapplication by a Member State), it is crucial that the notion of the start of works on an aided project is clearly defined in EU law and properly construed in practice.
The Definition of the Start of Works
Although it might sound surprising, a comprehensive, clear and uniform legislative definition of the “start of works” is relatively new. The 2006 version of the GBER included a basic definition of the “start of work” as:4
… either the start of construction work or the first legally binding commitment to order equipment, excluding preliminary feasibility studies, whichever is the earlier;
It made no explicit reference to commitments to making the investment irreversible and did not include special rules concerning real property acquisitions, preparatory works or take-overs.
The subsequent 2008 GBER,.5 while obviously featuring the formal incentive effect requirement (the duty to file the aid application yet before the start of works), did not explain what the start of works really meant. For guidance on the notion of the start of works, reference had to be made to the Commission’s decisional practice, as well as to various other Acts and instruments released by the Commission in the area of State aid law. In particular, useful reference could have been made to the Commission’s guidelines, such as the regional aid guidelines for the 2007 – 2013 period.6, which included a simplified and evolving version of the “start of works” definition at footnote 40, or to the FAQ issued on the 2008 GBER by the Commission.7
Still, as far as binding legislation is concerned, it is only from 2014 that the mature concept of the “start of works” has been clearly defined in article 2(23) of the 2014 GBER.8
The current definition stipulates that the start of works occurs upon construction works relating to the investment, the first legally binding commitment to order equipment, or any other commitment that makes the investment irreversible, whichever happens earlier.9 Moreover, what is most relevant for the topic under consideration, certain situations are, by way of exception, expressly deemed not to constitute a start of works; these include buying land and “preparatory works” such as obtaining permits and “conducting feasibility studies”.
Research and Development Towards Manufacturing Investment
Notwithstanding any deliberate policies of both the EU and the Member States to boost technological progress, modern industry and economy are by their nature driven by the constant and relentless search for improvement through science, technology and innovation. Technological superiority provides businesses (and countries) with a competitive edge over their competitors, and thus is a crucial element of their success.
To achieve technological advances, firms consistently need to engage in R&D. This often involves costly, painstaking, time-consuming and, most of all, risky processes. It is also often the case that a firm’s decision to invest in or implement a technology requires it to be validated by a successful R&D process. R&D processes go through (often numerous) stages: early studies on the scientific and technical nature of an underlying phenomena; searches for practical uses of the outcomes; and, finally, verifications of whether and exactly how outcomes can be used at an industrial scale in an economically sound manner. Pre-investment R&D processes are a recurrent part of the industrial landscape and an inevitable phase preceding large investment in innovative manufacturing sites.
While certain R&D processes, or their parts, can possibly be carried out through study, conceptual work or computer modelling, a large portion require testing materials, expendable equipment and specific fixed assets. These assets may range from sophisticated laboratories to expensive pilot installations, or even pilot lines or plants. Certain construction works could also be necessary in the framework of the R&D process, from adjustment of buildings required to host new R&D assets and equipment, to extensive construction processes needed to erect a pilot plant. It is these assets and construction works that allow the R&D process to be put in place and, possibly, deliver solid outcomes that will encourage a company to proceed to an actual industrial-scale investment in a manufacturing facility.
Qualification of Pre-investment R&D Under Formal Incentive Effect Rules
R&D processes are not obviously indicated in the 2014 GBER definition (or in any other EU law definition) as an event constitutive of the “start of works”. Therefore, they can principally be launched and carried out yet before any aid application is filed for the subsequent investment (if any) based on that R&D.
As mentioned, the “start of works” is triggered, among others, by the first legally binding commitment to order equipment as well as construction works relating to the investment.
Still, it appears to be beyond doubt that assets (and construction works, if any) needed to carry out an R&D process preceding and conditioning the actual investment decision are not, as a matter of fact and as a matter of principle, a part of that investment. Instead, they are used by a company, prior to an investment decision, as a tool to verify whether an investment would be technically and economically viable in the first place. The R&D process intended to verify a technology to confirm its adequacy as a basis for a potential industrial-scale production facility can thus be clearly delineated from investment in such a facility.
Notwithstanding that clear distinction between the assets used for an R&D phase and the assets needed for an actual investment in a manufacturing facility, the R&D equipment and processes could fall under the notion of “feasibility studies” or under the even larger heading of “preparatory works”, which, as mentioned above, have been explicitly indicated in the current legislative definition as falling outside the scope of the “start of works”.
In fact, the Commission’s consistent decisional practice indicates that such “feasibility studies” or, in broader terms, “preparatory works” include not only typical R&D assets used exclusively for research or development (such as laboratories), but also assets and facilities which take the shape of a proper industrial infrastructure or a manufacturing site.10
This is the case for pilot plants used to carry out R&D works to validate certain technologies and their industrial use, but also subsequently being used for regular manufacturing purposes. As the Commission held in cases including Qimonda, Sovello and Normandy Ocean Energy Farms.11, pilot plants do not necessarily need to be minor installations – they can have the size and characteristics of a regular industrial installation if that is needed to test a technology intended to be used industrially at an even larger scale. Notwithstanding the fact that pilot lines or plants operating at a greater scale are necessary for industrial validation of a manufacturing technology due to their size emulating the real-life situation, they can also be condemned to commercial use simply due to their excessive cost..12 Indeed, the Commission has so far applied this approach in a uniform manner across various instruments and areas, indicating that an investment in a pilot plant is not considered as investment in commercial manufacturing assets.13
Unexpected Troubles With Pre-investment R&D Works
It is against this background that a recent Polish case was resolved. A Polish beneficiary of regional aid launched R&D work and, for that purpose, purchased specific testing assets assembled into a small-scale pilot installation envisaged to validate a technology which could potentially be used in an industrial-scale manufacturing plant. Remarkably, the beneficiary acted in line with the direct advice of a technical consultant. The consultant explicitly stated that the technology (previously examined using conceptual tools, calculations and laboratories) could not be effectively verified unless it was tested empirically. This would require assets prospectively replicating (albeit at an indefinitely smaller scale) one of core processes that would occur at the full-sized production plant. The R&D works commenced, and elements of the pilot installation were purchased by the beneficiary. All equipment procurement and construction works relating to the full-sized plant began only after the filing of the aid application.
These facts led Polish authorities (the Minister of Development and Technology, then acting as the Minister of Enterprise and Technology) to challenge the aid on the basis of a claim that the purchase of the R&D assets should be qualified as the first legally binding commitment to order equipment and, accordingly, as the start of works on the full-sized production plant. The Polish authorities did not refer to any of the exceptions to the start of works definition, although
the case could clearly be assessed in the context of both preparatory works in general and, more specifically, feasibility studies. The Polish authorities did not to comment on the Commission’s decisional practice mentioned above. In light of the Commission’s approach in cases cited above, a small-scale pilot installation could hardly be considered part of an investment in the actual full-scale manufacturing plant which the R&D works were intended to validate.
It is fortunate that a Polish administrative court ruling in the first instance dismissed the action of the Polish government.14 and ruled that the regional aid awarded to the beneficiary could not be revoked. This judgment was upheld in the elaborate decision of the Supreme Administrative Court of 22 September 2022 (ref. no. II GSK 823/19), which fully concurred with the approach of the first-tier court.
To that effect, the Supreme Administrative Court effectively held that the purchase of the R&D assets for a pilot plant testing aptitude of a mass-scale production taking place in a potential manufacturing plant could not be qualified as the start of works on that manufacturing plant. Rather, this was held to be feasibility studies (along the lines of the Qimonda decision) and manifestly did not breach the formal incentive effect rules with respect to the aid supporting that plant.
The principal takeaway from the Polish case is perhaps that even relatively straightforward concepts of State aid law, such as the start of works in the context of the formal incentive requirement, can potentially be distorted in everyday application and disconnected from their genuine content and rationale. More importantly, however, on the substance, the verdict of the Supreme Administrative Court also sends a welcome signal to industries that whenever research, development and innovation is needed before a decision can be made whether to invest in commercialization of a novel technology, under EU State aid law prior R&D processes can be carried out, including on larger assets such as pilot installations or plants, without undermining the State aid available for the actual manufacturing facilities.
- Dr Michał Bernat is an attorney-at-law, chartered tax adviser, managing counsel with Dentons, Warsaw and an assistant professor at the Warsaw University.
- See currently the Commission Regulation (EC) 651/2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty  OJ L 187 1-138.
- This being said, the CJEU has also recently held that in individual proceedings the Commission may find an aid measure to have an incentive effect even if the aid application was filed after the start of works on a project (see C-470/20 Veejaam and Espo  EU:C:2022:981).
- See Commission Regulation (EC) 1628/2006 on the application of Articles 87 and 88 of the Treaty to national regional investment aid  OJ L 302 29, art (2), s 1(j).
- Commission Regulation (EC) 800/2008 declaring certain categories of aid compatible with the common market in application of Articles 87 and 88 of the Treaty  OJ L 214, as amended.
- The Guidelines on national regional aid for 2007-2013, OJ C 54 13.
- General Block Exemption Regulation (GBER) – Frequently Asked Questions (Commission Regulation (EC) No 800/2008 of 6 August 2008 declaring certain categories of aid compatible with the common market in application of Articles 87 and 88 of the Treaty), as available at <https://ec.europa.eu/competition/state_aid/legislation/gber_practical_faq_en.pdf>.
- GBER (n 2) art 2(23).
- As a special rule, in case of take-overs the start of works is considered to occur upon acquisition of the assets directly linked to the acquired establishment.
- See, for instance, the Commission’s decision of 20 March 2020 in case State aid SA.39078 – 2019/C (ex 2014/N) which Denmark implemented for Femern A/S.
- See: Commission decision N 872/2006 Individual aid to Qimonda  OJ C 170; Commission Decision on the State aid C 21/08 (ex N 864/06) which Germany is planning to implement in favour of Sovello AG (formerly EverQ GmbH)  OJ L 237; Commission Decision State Aid SA.46874 (2017/N) France – Project for an ocean energy farm in Normandy  OJ C 14
- That aspect of pilot plants is expressly highlighted in art 2(86) of the 2014 GBER, which admits that R&D (and more specifically, the experimental development phase of the R&D process) may include “the development of a commercially usable prototype or pilot which is necessarily the final commercial product and which is too expensive to produce for it to be used only for demonstration and validation purposes”.
- See, for instance, the Communication from the Commission of 2021 on the “Criteria for the analysis of the compatibility with the internal market of State aid to promote the execution of important projects of common European interest”, OJ C 528 10.
- See the judgment of the Regional Administrative Court in Warsaw of 28 November 2018, ref. no. VI SA/Wa 1642/18.