State aid control of regional development policy at 60: harder and sharper, but not yet crystal clear?

placeholder (4)

Competition policy provisions for the control of State aid have been in the Treaty from the outset. These articles have remained substantially unchanged, but a large body of law and decisional practice has evolved.

The origins of regional aid control go back to the 1956 Spaak Report and ECSC experience with subsidies. The Treaty of Rome banned State aid, but allowed for three regional policy related exceptions, closely tied to the interests of France, Germany and Italy. There have been a number of key drivers in regional aid control. Over time, the European Courts have entrenched key aspects of State aid policy through case law. Theories of competitive harm have evolved: in the 1960s and 1970s, the development of regional aid control was motivated by concerns at competitive outbidding and ambitions for a Community regional policy; by the 1990s priorities shifted to competition between firms and issues of market structure and dominance. More recent developments give primacy to economic efficiency and perceived incentive effect. Accession has forced new Member States to adapt regional aid policies, but enlargement has also heavily influenced regional aid control. European regional policy and regional aid control are interlinked, but relations are increasingly complex. Wider policy agendas have been a key part of the narrative of regional aid control: the 1992 Single European Market programme, the Lisbon Strategy and Europe 2020 have all been influential in different aspects of State aid control.

The regulatory framework for regional aid falls into two phases: pre-2000 and post-2000. The pre-2000 period is characterised by an absence of legal clarity, ad hoc decision-making and significant departures from published policy positions. Since 2000, regional aid guidelines (RAG) have operated. These have been enforced as ‘appropriate measures’ so have a clear legal basis, and are characterised by fixed policy cycles determining the parameters of regional aid control timed to coincide with Structural Fund plans. SAAP and SAM aimed to focus Commission scrutiny on aid schemes or measures that raise serious competition concerns, resulting in an emphasis on ‘self-policing’ through block exemptions with fewer and fewer cases coming before the Commission. Maps approved on the basis of the RAG become an integral part of the RAG and form the basis for the exemption from notification on the basis of the GBER. Over 60 percent of regional aid expenditure is now committed on the basis of GBER exempt schemes. Fewer than 10 regional aid schemes and just five individual aid proposals have been scrutinised by the Commission since 2014.

Spatial coverage has consistently been one of the most controversial aspects of regional aid control. Approaches to spatial coverage in regional aid control fall into three distinct phases: in a ‘black box’ phase running from the early 1970s to the late 1980s, the Commission took an increasingly interventionist, but opaque approach to disciplining regional aid maps; in a second phase starting in 1988, the Commission explicitly distinguished between ‘a’ regions and ‘c’ areas, and aimed to be more transparent about the methodology it had been using. However, the methodology began to unravel due to Member State criticism and the discrepancies between national and Structural Fund assisted areas. A compromise reached with the German authorities over the post-reunification regional aid maps formed the basis for an approach based on overall population ceilings, and this was used for the remainder of the 1990s. The 2000-6 Regional Aid Guidelines marked the beginning of the third phase where EC-wide population ceilings were set, then distributed among Member States which then designated areas subject to the parameters in the guidelines. Subsequent guidelines have maintained overall population coverage discipline and fine-tuned the area designation criteria. Over time the impact of competition policy on spatial coverage and on the ‘shape’ of assisted area maps has been dramatic.

Commission perspectives on forms of aid are linked to two issues: first, the need to measure and compare different types of support – a grant equivalent; second, the desire to eliminate aid that ‘props up’ inefficient activities by supporting ongoing expenses. This led to early action against so-called operating aids in a number of countries, especially on accession. Opposition to operating aid contributed to the demise of some major instruments of regional policy, which resulted in a progressive narrowing of the forms of aid offered – by the late 1990s, the main form of regional aid was a capital grant in every Member State; operating aid is allowed in only two areas: outermost regions and sparsely-populated areas. However, the development of policy on operating aid has been tortuous – sometimes leading to curtailment of major operating aid schemes, but also extension of operating aid under the GBER, most notably enabling the compatibility of the Norwegian social security concession in a complete reversal of earlier policy.

The linking of aid intensities to the severity of the regional problem has been a core component of regional aid control since the outset. Initially, there was no direct calibration between the two – the Commission simply took the prevailing rates in the Member States as a starting point. Until the 2000s, Commission ceilings did not bite: 75% NGE was typical in many ‘a’ regions, but not necessarily affordable; generally, aid schemes became more discretionary and policymakers more value-conscious. Since 2000 aid intensities have been lowered across the board both in nominal terms and by using gross grant equivalent in place of NGE. This was partly motivated by competition concerns, as well as the affordability of high intensities for less prosperous countries. Aid intensities are as low as 10% GGE in some ‘c’ areas, raising questions about the ‘incentive effect’ of aid.

The Commission was concerned from the outset that large scale regional aid could have undesirable effects, but Member States resisted Commission attempts at case-by-case scrutiny. The evolution of policy reflects the tensions between the need for transparency and mechanisms that capture aids with the most serious effects, and the resource implications of case-by-case scrutiny. An important step was the 1997 motor vehicle framework which subjected all regional aid over a given threshold to Commission scrutiny. However, Commission ambition was for a multisectoral approach to limit capital-intensive aid and balance benefits to regional development against competition effects. Its first attempt, the 1998 multisectoral framework (MSF), did not perform as intended. This was abandoned in favour of a further multisectoral framework in 2002, which imposed reduction formula on aid rate calculations and a notification threshold for exceptionally large projects or awards. This has now been integrated into the Regional Aid Guidelines, with some changes to the basis for Commission assessment. This comprises a strong incentive to remain under the notification threshold as it is increasingly likely that notification will trigger a full investigative procedure. For ‘c’ areas, large investment projects and large awards are no longer very relevant since RAG 2014-20 mainly excludes aid to large firms in ‘c’ areas except green-field projects.

In the early years, competition policy control of regional aid was itself perceived as an instrument of Community regional, now Cohesion policy, though bespoke European regional policy funding did not emerge until after UK accession in 1972. The implementation of the ERDF in the 1970s was uncontroversial since it mainly cofinanced nationally selected projects. The 1988 reform of the Structural Funds raised major issues in State aid relations, primarily relating to spatial coverage, requiring compromise between the DGs for Competition and Regional policies. These tensions ultimately dissipated through the extension of Cohesion policy to all regions in 2007-13. Cohesion policy State aid relations are now affected by a range of issues that go beyond regional aid due to two major underlying trends. First, the notion of State aid has expanded along with the concept of economic activities. Second, ESIF interventions now extend beyond basic infrastructure and grants to firms, which in the past either did not entail aid or could be accommodated within the regional aid rules. State aid has an increasingly high profile in Cohesion policy due to the State aid modernisation initiatives (SAAP and SAM), the introduction and expansion of the GBER and the ‘upgrading’ of State aid considerations in the 2014-20 Common Provisions Regulation. RAG 2014-20 and GBER ‘nod’ to some Cohesion policy particularities, but a number of issues have been or are problematic, including: definitional issues, financial instruments and the treatment of simplified cost options. Although many Managing Authorities have bespoke systems for handling State aid, the error rate is high. The possibility of audit results in a risk-averse approach to State aid compliance, with borderline projects more likely to be funded from domestic sources alone.

Competition policy concerns with the effectiveness of State aid are comparatively new. They can be traced back to the SAAP emphasis on economic analysis and SAM concerns with the efficiency of public spending. RAG 2014-20 and GBER 2014 enable the Commission to authorise and limit the validity of large aid schemes, conditional on the outcomes of evaluation. Several major aid schemes are undergoing evaluation; however, the results of these studies and their impact on compatibility are as yet unknown.

Consideration of post-2020 State aid reforms seems to be at an early stage; there is currently no ‘high level’ State aid agenda like SAAP or SAM to provide a leitmotif for possible change. The Commission is taking soundings about the future of RAG and GBER, but there are no concrete outputs so far. Reapplication of the RAG 2014-20 post-2020 is subject to a large number of variables, including population coverage and thresholds, and other factors including data availability and NUTS boundary changes, which make outcomes difficult to predict. However, the current eligibility criteria suggests some changes to ‘a’ regions, with a reduction in some eastern European states, in part induced by a ‘Brexit effect’ on thresholds, with some southern European regions ‘regaining’ ‘a’ region status as a consequence of economic decline.