1.Introduction

The Guarantee Notice was first adopted in 2000 and last revised in 2008. It sets out the Commission’s approach to State aid granted in the form of guarantees and provides Member States with guidance on when guarantees do not constitute State aid.

In August 2022, the Commission launched the process of evaluating the 2008 Guarantee Notice. The purpose of this evaluation was to check whether the Guarantee Notice remained ‘fit for purpose’, to identify any shortcomings and to verify whether there was any scope for improvement. Sources of evidence for the evaluation included public consultations, an external expert study and the Commission’s own case practice. The evaluation covers the guarantees with an aid element and guarantees without aid element that were notified to the Commission and that were introduced within the period between 2010 and 2022. The results of the evaluation are summarised in a Staff Working Document (SWD).

The SWD reflects the findings and views of the Commission’s staff and does not represent the formal position of the Commission itself. It verifies the extent to which the evaluation criteria – effectiveness, efficiency, value added, relevance and policy coherence – have been met and summarises those findings. Thus, the SWD does not make any policy proposals or prejudge the final nature or the content of any act or decision that the Commission may take.

2.Expected outcome of the initial intervention

The objective of the 2008 Guarantee Notice is to provide detailed guidance on assessing notified guarantee measures, to clarify when a guarantee constitutes State aid, and, if that is the case, to provide guidelines on how Member States should calculate the aid element (“gross grant equivalent”).

The 2008 Guarantee Notice applies to all guarantees issued by public authorities or entities under their control related to financial instruments such as loans, and counter-guarantees whereby a public guarantor undertakes to compensate an undertaking (such as a lender or a private guarantor) for payments made to the beneficiary (hereafter “public guarantees”). Thus, in the context of the Guarantee Notice, public guarantees involve public authorities, lenders (i.e. banks and other financial institutions) and borrowers.

Public guarantees cover both individual guarantees and guarantee schemes for undertakings in all economic sectors, except for export credit guarantees. Key provisions of the Guarantee Notice include criteria that, if met cumulatively, exclude aid (capping of guarantee coverage at 80%, exclusion of firms in difficulty, the need for the measure to remain “self-financing” and be market-oriented, etc.). In addition, the Guarantee Notice includes simpler approaches to assess possible State aid elements in public guarantees to small and medium-sized enterprises (“safe harbour premiums” or “single premiums”), helping to improve their access to finance.

Against this background, the 2008 Guarantee Notice aims (i) to reduce the number of notified public guarantee measures by guiding Member States in designing State aid-free schemes; (ii) to increase legal certainty; (iii) to ensure that the guarantee premiums reflect market-conform pricing thereby avoiding notification to the Commission; (iv) to serve as an appropriate tool for calculating the aid element; and, (v) to support financing of SMEs through simpler methods such as safe harbour premiums.

3.Evolution of the situation over the evaluation period

Since the 2008 revision, there have been notable developments with relevance for the functioning of the Guarantee Notice. In terms of macroeconomic developments, the 2008 global financial crisis triggered higher default rates and a prolonged phase of low interest rates. Subsequent shocks like the COVID-19 pandemic and the Russian military aggression against Ukraine further stressed the markets, causing inflation and a tightening of monetary policy starting in 2022. Credit risks fluctuated sharply in response to these crises and resulted in a tightening of funding conditions for SMEs.

Private lenders were also affected by several key regulatory developments, such as increased capital requirements through the adoption of the Capital Requirements Directive and Regulation in 2013 and the introduction of a new forward-looking approach for measuring credit risks under the IFRS 9 accounting rules, which has been effective from 2018 and onwards. As a result, lenders now follow stricter and more harmonised credit assessment practices, which may also affect the use of guarantees.

The external study estimates that in the observation period of 2010 to 2022, public authorities granted guarantees for an amount of EUR 22.1 billion under the Guarantee Notice, which is equal to 0.13% of EU GDP (2023). While Italy and Germany recorded the highest absolute amounts of public guarantees, smaller Member States like Latvia and Romania issued more guarantees relative to their respective GDP levels than these larger Member States. Overall, 34% of the total public guarantee volume reported to the Commission consisted of market-conform (i.e. without State aid) public guarantees, though this number is likely to be underreported since market-conform measures are not subject to explicit reporting obligations.

The most common methodologies used to calculate the guarantee premiums during the observation period were risk-based approaches (with a value of around EUR 12 billion), used mostly in Member States like Germany, Sweden, and Italy. Risk-based approaches are bespoke, taking into account the specific circumstances of each case. Such case-specific methodologies should be distinguished from safe harbour methodologies, which were popular in other Member States (e.g., Romania, Ireland). A third set of methodologies calculate the market conform guarantee premiums and are based on available market benchmarks, such as the pricing of comparable financial products (e.g., credit default swaps) or the interest rate charged on the very loan that is subject to a public guarantee.

In its case practice, the Commission services have faced several issues during the evaluation period. For example, the safe harbour premiums have been used less regularly, most likely due to post-crisis market conditions and lower interest rates, prompting a shift toward the bespoke risk-based methodologies. Furthermore, there has been limited use of market benchmarks, due to the absence of comparable transactions, especially in smaller or less developed financial markets.

4.Evaluation findings

The evaluation assessed, how well the Guarantee Notice has met its objectives, based on five evaluation criteria: effectiveness, efficiency, coherence, EU added value, and relevance. The findings are backed by empirical data, benchmark comparisons, case practice, and stakeholder feedback.

Effectiveness

Accuracy of risk-based and safe harbour pricing methodologies

First, the evaluation’s findings suggest that existing risk-based methodologies produced lower premiums than market benchmark methodologies developed for the external study, especially for higher-risk borrowers and long maturities. Second, the safe harbour methodology produces premiums that are lower than the market benchmarks for lower risk (e.g. BBB-rated) borrowers while the opposite is true for higher risk borrowers. More generally, the existing risk-based and safe harbour methodologies exhibit lower sensitivity to market dynamics and changing risk conditions than the market benchmarks. These results imply that the popular methodologies that are used most frequently by Member States may not be adequately accurate in producing market conform fees across different risk categories and over time.

Crowding out of private investors

Crowding out occurs if publicly guaranteed loans replace private loans or prevent the development of private guarantee markets for similar loans. SMEs typically have more difficulties in accessing finance compared to larger firms. The external study therefore analyzed whether public guarantees can reduce this effect. The results suggest that public guarantees may not be sufficient in bridging the entire gap between SMEs and large firms in terms of participation in credit markets and credit rationing vis-à-vis borrowers. While the results effectively rule out full crowding out, it is also likely that some amount of crowding out takes place, especially since the impact of public guarantees on credit rationing is relatively small. The extent of crowding out remains an open question.

Unintended Selective Advantages

Public guarantees may provide unintended selective advantages to lenders if those lenders do not properly pass-through the risk-reduction benefits to the intended beneficiary, the borrower. Similarly, public guarantees may provide support for non-viable borrowers, delaying their exit from the market and undermining competition. The evaluation indicates that there may be an incomplete pass-through of the benefits of public guarantees to final borrowers, which implies that selective advantages to lenders could be present. In addition, public guarantees may reduce entry and exit rates in sectors benefiting from guaranteed loans, potentially lowering market dynamism, productivity and ultimately growth.

Coherence

The findings of the external study suggest that the Guarantee Notice is not fully coherent with other State aid legislation, such as particularly the Notice on the Notion of Aid or the Reference Rate Communication. For instance, the Notice on the Notion of Aid prioritizes the use of market benchmarks over market proxies, such as the premiums produced by the risk-based or safe harbour methodologies, which is not fully reflected in the Guarantee Notice. Additionally, the Guarantee Notice and Reference Rate Communication prescribe different fixed risk premiums – a key component of a guarantee premium – for the same firm. Finally, the Guarantee Notice does not provide any specific guidance on the ex-ante examination that the Commission should conduct for determining whether undue advantages to lenders exist, as required by the Notice on the Notion of Aid. Similarly, the Guarantee Notice does not suggest any specific safeguards to rule out indirect aid to lenders, as is done by the General Block Exemption Regulation conditions for risk finance aid.

Efficiency

Efficiency refers to the determination of how successful the Guarantee Notice has been in producing methodologies that are cost-effective. The general unavailability of reliable data on administrative costs (which form part of the price of public guarantee together with the premium, i.e. the renumeration) has been a key impediment for the independent experts to address empirically the questions relating to efficiency. Nevertheless, the experts did use the limited available data to the fullest extent and supplemented this with voluntary interviews to draw some qualitative results.

Overall, the interviewed stakeholders flag that the Guarantee Notice is too complex, requiring a significant amount of resources from Member States, especially in the design and setting up stages. The qualitative data also highlights that the lengthy pre-notification, notification and adoption stages relating to individual State aid decisions on public guarantees result in uncertainty and costly staff re-allocations over time. In turn, the quantitative analysis revealed that the guarantee premium collections by the State may not cover total costs. Furthermore, larger schemes benefit from economies of scale, resulting in lower unit administrative costs. In turn, no-aid measures typically incur higher administrative expenses, which could be explained by the fact that in addition to remaining self-financing, Member States may have to devote further resources to ensure that the measure does not become too costly for borrowers, resulting in a healthy take up.

EU Added Value

The evidence gathered in the evaluation suggests that the Guarantee Notice has a clear EU added value compared to a situation without any guidance. In particular, the Guarantee Notice has provided a framework within which the Commission can approve different methodologies for determining the value of public guarantees developed by Member States. The Notice also sets out directly applicable safe harbour premiums. But, on the other hand, the Guarantee Notice does not reflect the rules applicable when State guarantees are complemented with a support stemming from Union Programmes. Overall, the Notice has increased predictability and legal certainty, ensuring a level playing field between Member States.

Relevance

The increasing number of pre-notifications and notifications for the approval of guarantee methodologies and a continued use of the safe harbour premiums shows that the Guarantee Notice remains highly relevant. From a purely practical point of view, this relevance is likely explained by the fact that the calculation of a market conform premium and the measurement of the aid element are quite distinct for guarantees than for other support measures, such as loan subsidies and grants. Nevertheless, due to the changing market conditions and regulatory environment since 2008, the Guarantee Notice would benefit from a revision to increase its relevance.

5.Conclusions and Lessons Learned

Overall, the Guarantee Notice has been widely used by Member States, especially due to its flexibility in allowing the use of both risk-based methodologies and safe harbour premiums for providing public guarantees. This has contributed to maintaining a level playing field within the EU and enabling the Commission to approve guarantee methodologies developed by Member States.

The evaluation highlights several areas for improvement of the Guarantee Notice. First, the risk-based and safe harbour methodologies in most cases tend to underestimate guarantee premiums compared to market benchmarks.

Furthermore, while there is evidence that public guarantees do result in lower interest rates than what would otherwise be possible without the guarantees in place, this finding is not sufficient to rule out the possibility that some of the benefit of the aid is retained by the lenders. The Guarantee Notice also seems overly complex, potentially resulting in high administrative burdens for the Member States. Finally, there appears to be a certain degree of incoherence between the Guarantee Notice and other State aid rules, such as the Reference Rate Communication and the Notice on the Notion of Aid.

The evaluation has also highlighted a number of lessons learnt.

First, since existing safe harbour premiums are much higher than comparable benchmarks for higher risk SMEs, many Member States may be opting for developing bespoke risk-based (and more complex) methodologies for those beneficiaries.

Second, the evaluation demonstrates that the accuracy of risk-based methodologies may be undermined if the parameters used are not aligned with market reality.

Third, there may be potential unintended aid to lenders which only has been partially addressed by recently approved methodologies to tackle this issue.

Fourth, there is a lack of a framework for voluntary information sharing between Member States. Such framework for Member States wishing to cooperate both among themselves and with lenders could be instrumental to improve access to data and state-of-the-art risk assessment approaches. In addition, the current reporting requirements only lead to limited information being available to properly monitor the implementation of the Guarantee Notice.

Fifth, there are some common interpretation and implementation issues that emerged from the evaluation. This for instance concerns the assessment of portfolio guarantees (i.e. the State guarantee covers losses up to a portfolio-level rather than a loan-level cap), the meaning of the so-called “self-financing” condition (i.e. the requirement that the collected premiums should be sufficient to cover all relevant costs), and criteria and conditions for other types of guarantees than loan guarantees (e.g., performance guarantees).