The New Year 2026 will see negotiations on the EU’s next medium-term budget step up in gear. Many Member States have called to set a deadline of end 2026 to agree on the next Multi-annual Financial Framework (MFF). This is necessary if the new MFF programmes are to start on 1 January 2028.
That this is a very ambitious deadline is an understatement. The Danes set a cracking pace in organising the Council’s MFF discussions under their Presidency, and presented a first draft of the famous ‘negotiating box’ to the European Council meeting in December 2025. This draft, without figures, was presented under the sole responsibility of the Presidency. The accompanying progress report recognised that significant political as well as technical issues remain to be resolved. Bridging these divides under the Cypriot and Irish Presidencies will require a massive effort in the coming twelve months.
Although it is the Council that negotiates the MFF budget, the European Parliament (EP) must also give its consent. The EP has begun the process of formulating its position by preparing an interim report on the MFF proposal, intended to signal in advance what its minimum demands will be. The joint rapporteurs in the Committee on Budgets (one from the EPP and the other from the S&D political groups) published a draft Interim Report in early December. This will undergo amendment before being voted on in the Committee and presented to Parliament for decision most likely in May.
Agricultural stakeholders will be significantly affected by the outcome of these budget negotiations. In this post, I want to focus on the potential role of EU borrowing in these negotiations. As the EU Treaties lay down that the EU must be financed wholly from its own resources, its ability to borrow in limited circumstances is often overlooked, But the EU has increasingly resorted to borrowing in recent years, most recently to finance the €90 billion loan to Ukraine agreed in the European Council conclusions in December 2025. The repayment of the loan taken out to fund the Next Generation EU (NGEU) programmes in the wake of the COVID-19 pandemic in 2020 weighs heavily on the EU’s ability to increase its programme spending in the next MFF. We look at possible options to address this issue.
The MFF package also includes the Own Resources Decision (ORD) where the Parliament must be consulted but its opinion is not binding. However, it can only enter into force after ratification by all Member States in accordance with their constitutional procedures. This gives national parliaments a de facto veto over the EU’s revenue system, the size of its budget and its borrowing authority. This constitutional architecture needs to be kept in mind when discussing possible options for the next MFF.
The EU’s borrowing options
Article 310 of the Treaty on the Functioning of the European Union requires that revenue and expenditure in the EU budget must be in balance. In turn, Article 311 states that the revenue of the Union consists of its own resources, without prejudice to other revenue. The Treaties do not define what counts as own resources. Instead, these are established and updated as necessary in the ORD adopted by the Council subject to unanimous approval and national ratification. The Lisbon Treaty introduced changes to the provisions related to the system of own resources which enable the abolition of an existing category of own resources and the establishment of a new category. The ORD can also authorise borrowing if necessary, and repayment modalities, but subject to strict rules.
We can distinguish between four mechanisms that can facilitate additional EU expenditure through borrowing.
1. Providing additional fiscal flexibility. This involves amendments to the EU Stability and Growth Pact to provide greater fiscal space to Member States for specific expenditures. The SGP is a set of rules for EU countries to coordinate fiscal policies and to maintain sound public finances by limiting deficits to around 3% of GDP and government debt to around 60% of GDP. Where a country exceeds these limits, an Excessive Deficit Procedure can be activated where countries receive recommendations for corrective action and where failure to comply can lead to financial penalties or suspension of EU funds.
The ReArm Europe/Readiness 2030 package activates the national escape clauseof the Stability and Growth Pact for defence spending. This clause allows a country to temporarily deviate from the SGP limits without triggering sanctions or formal excessive deficit procedures, provided that the increased spending relates to defence. Specifically, Member States can increase defence spending by up to 1.5% of GDP per year above their previously agreed net expenditure paths without being deemed in breach of fiscal rules. This flexibility can be used for four years (2025–2028), allowing sustained build-up of defence budgets.
What is relevant to highlight here is that this mechanism involves additional national spending through borrowing but has no implications for EU borrowing per se.
2. Borrowing for back-to-back loans. Here the EU borrows on financial markets (making use of its high creditworthiness to secure favourable borrowing rates) and lends the money to Member States for specific purposes. Such borrowing was used to provide loans to Member States under the SURE programme (Support to mitigate Unemployment Risks in an Emergency) created in May 2020 during the COVID-19 pandemic. The EU also has three other loan programmes to raise funds and pay for financial assistance to countries experiencing economic difficulties (for balance of payments assistance to EU countries outside the euro zone, the European Financial Stability Mechanism for euro countries in economic difficulty, and macro financial assistance to non-EU partner countries). The most recent example is the SAFE financial instrument (Security Action for Europe) whereby the EU will provide up to €150 billion in low-cost, long-term loans to EU Member states for joint defence procurement.
These one-off borrowing operations rely on Article 122 TFEU which provides for solidarity between Member States in exceptional circumstances, and on Article 212 which provides for financial assistance to third countries. The debt is serviced by Member State repayments, and only the contingent default risk as well as administrative costs appear in the EU budget, as well as the cost of possible interest subsidies if loans are made at concessional interest rates (for example, fully covering interest rate costs for existing loans to Ukraine is estimated to cost approximately €11.5 billion in total over the next MFF, Darvas and McCaffrey, 2024). Thus, there is no repayment of principal line needed in the MFF because the loan is treated as a financial asset, not a grant. In each case, borrowing is justified as temporary, exceptional, and for a specific purpose.
In its MFF 2028-2034 proposal, the Commission proposed a new €150 billion loan support programme (dubbed Catalyst Europe) for the implementation of countries’ National and Regional Partnership Plans. The request for loan support should be justified by the higher financial needs linked to additional reforms and investments included in the NRP Plan and by a higher cost of the NRP Plan than the sum of the Union financial contribution and the national contribution. Because debt service would be covered by repayments from borrowing countries, these policy loans would remain outside the MFF ceilings. Still, this proposal for loan financing marks new ground because it is not specifically tied to an exceptional situation but rather to undertaking reforms and investments as part of ongoing EU budget expenditure.
In this proposal, as well as with the SAFE proposal, the Commission is gingerly embracing one of the recommendations of the Draghi report, which was to introduce regular and sizeable issuance by the EU of a common safe and liquid asset to enable joint investment projects among Member States and help integrate capital markets. Draghi’s focus was on the creation of a common European safe asset. Joint investment was a means to this end, although he also saw merit in increasing resources for productive investment. He recommended adopting the precedent of the NGEU model, in which issuance would remain mission and project specific. But Draghi was silent on which element of the NGEU model – loans or grants – he favoured.
3. Borrowing to provide grants for operational expenditure. This is a big ‘no no’ in the EU constitutional financial architecture and, to date, there has been only the one exception of the Next Generation EU programme (NGEU) in response to COVID. The NGEU allowed the EU to borrow up to €750 billion (in 2018 prices), of which €360 billion was to be used for back-to-back loans to EU countries and the remaining €390 billion could be used for expenditure. The borrowing again was justified by Article 122 TFEU but it also required an amendment to the ORD to temporarily raise the own resources ceiling, to explicitly authorise the Commission to borrow this amount, and to specify how repayment would be managed. Without this ORD (ratified by all Member States), borrowing under Article 122 would not have been legally viable. Article 4 of the OCR specifies that “The Union shall not use funds borrowed on capital markets for the financing of operational expenditure” before then, in Article 5, permitting such borrowing “for the sole purpose of addressing the consequences of the COVID-19 crisis”.
The key difference with back-to-back borrowing is that, for the NGEU grant element, there is no offsetting repayment stream from beneficiaries. Servicing this debt requires specific EU budget resources. The interest and principal must be paid, either from new own resources (if agreed) or by higher GNI-based contributions, and directly compete with other EU spending. The NGEU created a structural claim on future EU budgets and this is reflected in the Commission MFF proposal which sets aside €168 billion in current prices for this purpose.
The EU Commission now (since 2023) uses a unified funding approach to EU borrowing. This means that the EU does not borrow for specific programmes but under a single branded “EU Bonds”. Proceeds from these bonds go into a central pool, and are then allocated internally to different policy programs. The framework utilises long-term EU-Bonds, short-term EU-Bills, and specialised instruments like NGEU Green Bonds. By enhancing liquidity, this approach reduces borrowing costs and supports broader EU financial objectives.
4. Permanent EU Treasury bonds. This final option for EU borrowing is purely speculative at the moment but deserves to be treated seriously. The EU has long looked enviously at the United States’ ability to issue Treasury securities, which are seen globally as a safe, liquid, and risk-free benchmark for pricing other financial assets. This status provides the US with lower borrowing costs and makes the dollar a central vehicle for international finance. For the EU, issuing comparable “EU Treasury bonds” could strengthen the international role of the euro, provide a deep, liquid asset for investors, and create a reliable, long-term source of funding for EU priorities.
However, such a development would constitute a qualitative regime change: under current Treaty rules, the EU cannot borrow on its own account for general expenditure. Any large-scale debt issuance would also require a legally guaranteed repayment source. In practice, this means the creation of EU Treasury bonds would necessitate an extension of own resources, ensuring that debt service could be met without depending on the political discretion or approval of Member States. Such a step goes far beyond the temporary, programme-specific borrowing instruments currently permitted.
Borrowing to respond to severe crises
The Commission has learned from the succession of crises since COVID-19 in 2020 of the need to be prepared. It is therefore proposing, as part of the Own Resources Decision accompanying its MFF 2028-2034 proposal, “to establish a new limited, extraordinary and targeted tool to respond solely to severe crises, severe hardship or serious threat thereof. This extraordinary crisis tool should allocate the budgetary resources for the granting of loans solely in the period of the upcoming MFF 2028-2034”.
Articles 6 through 8 of the proposed Decision mirror in many respects Articles 5 and 6 in the current ORD which enabled the joint borrowing to establish the NGEU instrument. The one important difference is that the Commission proposes only to use joint borrowing for the purpose of loans to Member States, while the NGEU instrument combined both loans and grants. As we have seen, such back-to-back lending is specifically allowed in the EU’s financial architecture.
Specifically, the new instrument would allow the EU to engage in extraordinary borrowing to address a severe crisis, severe hardship or serious threat thereof. The advantage of including this provision in the ORD is that it would allow the Commission to activate this instrument, should the need arise, using the simpler legislative procedure set out in Article 311(4) TFEU. This allows the Council to act in accordance with a special legislative procedure, requiring only the consent of the Parliament. It would avoid the requirement, should a serious crisis arise, of the need to seek the ratification by national parliaments as they would already have given this umbrella permission by ratifying the OCR itself.
There would be a budgetary implication of such borrowing as the EU would have to cover the contingent liabilities resulting from the borrowing empowerment. The Commission recognises there is a need for certainty about the Union’s liability. It proposes that the ceiling for appropriations for payments and the ceiling for appropriations for commitments should each be increased by 0.25 percentage points, with the sole purpose to cover all liabilities of the Union resulting from its borrowing for loans to address the consequences of such events. The sum of outstanding principal amounts which the Commission may be authorised to borrow on capital markets would be limited to the amount, which, in view of the forecast multiannual evolution of contingent liabilities resulting from this borrowing, remains within these 0.25 percentage point ceilings. I have seen an estimate that this might allow for up to €400 billion in borrowing, but I have not confirmed this estimate myself.
Borrowing for EU support for Ukraine
At the European Council on 18–19 December 2025, EU leaders endorsed a €90 billion financial support package for Ukraine covering 2026–2027, to be provided in the form of a loan financed by EU borrowing on capital markets and backed by the EU budget “headroom” (reserves and guarantees). This loan provides an interesting case study of how it fits into our borrowing typology.
First, the loan is justified on the basis of Article 212 TFEU. Second, and most important, the loan is structured as repayable by Ukraine (eventually through reparations) rather than as a grant financed by the EU budget. The European Council envisages repayment only when future reparations by Russia for the damage caused to Ukraine due to the invasion are paid, and until then the EU budget headroom is a legal backstop. It thus is deemed to be a back-to-back loan which would be agreed through a co-decision process rather than requiring, as for the NGEU loan, a specific amendment to the OCR. This option was chosen because of unyielding resistance by Belgium to use frozen Russian assets to assist Ukraine.
If the EU budget ends up servicing the debt (e.g., because reparations do not materialise), then at that point debt service would need to be reflected in future MFFs (as occurred with NGEU). But as it stands, the €90 billion support is structured to avoid automatic inclusion in the MFF’s mandatory repayment lines until the repayment situation crystallises.
A further noteworthy aspect is the use of the enhanced cooperation procedure (Article 20 TEU) with respect of this loan instrument based on Article 212 TFEU. Any mobilisation of resources of the Union’s budget as a guarantee for this loan will not have an impact on the financial obligations of Czechia, Hungary and Slovakia.
Given that Ukraine was likely to go bankrupt without this loan, the use of Article 212 to justify joint borrowing in itself is not unusual, although the scale of financing dwarfs what previously has been provided as macro-economic financial assistance to third countries. What is unusual is the ‘collateral’ accepted by the 24 Member States that agreed jointly to backstop this loan. Accepting that Russia will, at some point in the future, be willing to pay reparations to Ukraine at the end of the war is a brave bet on the future. However, the European Council conclusions make clear that Russian assets will remain frozen and could ultimately be used to repay the loan should Moscow refuse to pay reparations to Ukraine once the war ends.
Conclusions
This summary of issues around EU borrowing and the role it can play in the next MFF raises several issues for agricultural stakeholders.
- Whether and how the NGEU borrowing should be paid back.
- Whether the EU should be endowed with permanent borrowing powers.
- How to expand the EU’s own resources.
- Whether agriculture and rural development might benefit from the proposed Catalyst Europe loan programme in the next MFF.
Repayment of NGEU borrowing. The political agreement reached in 2020 envisaged that the repayment of both interest and principal on the grant component of NGEU borrowing would be financed by new own resources. This was intended to avoid the ‘cuckoo in the nest’ problem where debt servicing would crowd out expenditures on other programmes. The Commission tabled proposals to this end in 2021 and June 2023 and, more recently, in its proposal for an Own Resources Decision in July 2025. However, in the absence of a decision on new own resources, effective spending power at the EU level is mechanically reduced by interest and principal payments.
The Parliament’s Budgets Committee draft interim report on the MFF calls for the 2028-2034 MFF to be set at 1.27% of EU gross national income (GNI), excluding NGEU repayments. It considers that NGEU debt servicing, representing an additional 0.11% of GNI, should be treated separately from funding for EU programmes within the future MFF architecture so as to ensure that available resources for these programmes remain unaffected, bringing the total to 1.38% of EU GNI. But wherever the NGEU repayment is placed in the MFF budget, in the absence of an increase in own resources it automatically competes with resources for spending programmes including agriculture.
The alternative proposed is to simply roll-over the NGEU debt, as would normally happen with sovereign debt. This option was proposed in the Draghi report, who noted that Member States could consider increasing the resources available to the Commission by deferring the repayment of NGEU. Apart from the fact that it would ease the pressure on spending programmes, there is a solid financial market argument for this course of action. Under current plans, the Commission will cease borrowing this year (2026) and then gradually buy back the stock of outstanding bonds until 2058. Other things equal, this would mean that the supply of EU bonds would gradually fall, reducing liquidity in the market and potentially increasing borrowing costs.
In practice, this would mean moving from Options 2 and 3 in my typology to Option 4 where the EU is endowed with permanent borrowing rights. This would certainly require a Treaty change. It would also require a significant expansion in the EU’s own resources as a permanent EU borrowing capacity must rest on a firm fiscal foundation. Without dedicated revenue streams, the burden of repayment will continue to fall predominantly on national contributions, undermining both the legitimacy and resilience of the common fiscal architecture. At least in the short-run, these conditions are not likely to be present.
Could agriculture and rural areas benefit from the proposed new policy lending instrument? We have seen that the Commission has proposed a Catalyst Europe lending instrument with €150 billion in funding. The request for loan support should be justified by the higher financial needs linked to additional reforms and investments included in the NRP Plan. At first sight, this seems of limited interest for agricultural spending which mainly consists of income transfers rather than investment supports. Still, there are elements of agricultural and rural spending of an investment nature, such as building up AKIS capacity, rural infrastructure, or support for the installation of younger farmers. A recent paper by the IEEP and Concito (Muro et al, 2025) explored the role of the European Investment Bank in financing the green transition in EU agriculture. While this refers to opportunities for the EIB to fund private borrowing by farmers, it does suggest there are investment opportunities in agriculture that can also be supported by public funding.
An important constraint will be the fact that, in spite of the likely favourable interest rate on these policy loans, making use of EU borrowing instruments (whether SAFE or the proposed policy lending instrument) requires countries that are often already heavily indebted to take on more debt. Still, in the debate on future funding for the CAP, this possibility to draw down additional funding should not be overlooked.
This post was written by Alan Matthews.
Photo credit: Money Images in Flickr used under a CC by 2.0 licence.
