The cost of waiting for Russia’s reparations

On 7 October 2026, the European Commission announced a further €1.24 billion payment to Ukraine for equipment produced by its defence industry. The money comes from the EU support loan authorised for 2026 and 2027. The announcement explains where the payment is going. Understanding its financing requires tracing the money in the opposite direction, back to the saver or investment fund buying an EU bond. ^[1]

That investor has a claim on the European Union. On the other side of the transaction, EU legislation creates a loan to Ukraine whose principal becomes repayable when specified events occur, most notably the receipt of Russian reparations. Europe is bridging a gap: it must service its own debt while the timing of its recovery remains uncertain. The budget guarantee was designed to protect precisely those European obligations to creditors. ^{[4] [5]}

This timing gap is the economic heart of the arrangement. Principal may eventually be recovered. Interest will have been paid along the way. Even a substantial Russian settlement might repay only part of the new loan, because the legislation takes other debts into account when allocating compensation. Reading those clauses identifies the exposure Europe is accepting, without treating an authorised lending ceiling as a loss already incurred. ^[4]

EU bond maturities and Ukraine’s repayment conditions

The EU borrows in its own name and lends to Ukraine in euros. It uses a diversified funding strategy: different issues and maturities support the needs of its programmes. There is no single €90 billion bond scheduled to mature on the day the war ends. The Commission issues securities across a range of maturities and manages its liquidity across its financing operations. ^{[4] [9]}

A bond buyer therefore relies on the Union’s ability to honour its own obligations. A failure to receive Russian reparations does not, by itself, suspend the coupon payable to that investor. The legislation strengthening the budget guarantee is intended to ensure that resources are available when those payments fall due. ^[5]

For Kyiv, the terms are substantially more protective than a conventional loan with fixed repayment dates. The regulation describes a limited-recourse loan: principal becomes due upon specified triggers. The design provides funding now without immediately imposing an ordinary borrower’s repayment schedule on a country at war. Its economic counterpart is the period for which Europe must carry the financing. ^[4]

The two sides of the transaction are connected, but their timetables can diverge. If an EU bond matures before the expected recovery arrives, the Union needs cash, callable budget resources or refinancing consistent with its rules. The timing exposure lies between a contractual cash outflow and a conditional cash inflow. ^{[4] [5] [9]}

Assumptions and method

Diagram of obligations required by Regulations 2026/467 and 2026/469. Arrows show contractual relationships without monetary scaling. The full signed agreement was not reviewed; additional implementation terms are not reconstructed.

A €90 billion ceiling, several routes for payment

The programme authorises up to €90 billion, with an indicative allocation of €60 billion for defence industrial capacity and €30 billion for budget support. The Council made a maximum of €45 billion accessible for 2026: €28.3 billion for defence, €8.35 billion of macro-financial assistance and €8.35 billion delivered through the Ukraine Facility. The €45 billion is part of the €90 billion, rather than a separate addition. ^[3]

The Ukraine Facility acts here as a delivery channel. It already had its own €50 billion envelope for 2024–2027, combining loans and grants. Some of the new support uses its procedures, allowing funding from different sources to appear in announcements issued under the same name. The loan regulation expressly provides for that arrangement. ^{[2] [4]}

The payment announced on 2 October illustrates the distinction. Of the €2.9 billion transferred through the Facility, €800 million came from the new loan. The difference, approximately €2.1 billion calculated from rounded figures, belongs to the rest of that payment. The €1.24 billion defence payment on 7 October also comes from the support loan. Counting all €2.9 billion as a disbursement from the €90 billion programme would overstate its execution. ^{[1] [2]}

Conditions and dates matter too. On 7 October, the Commission said €28.1 billion remained accessible for the rest of 2026, subject to the relevant requirements. The regulation distinguishes the programme’s availability through the end of 2027 from the possibility of disbursing approved tranches through the end of 2028. Neither date is a general maturity date for Ukraine’s repayment. ^{[1] [4]}

Assumptions and method

Total ceiling: 60 + 30 = €90bn, an indicative allocation for 2026–2027. Amount accessible in 2026: 28.3 + 8.35 + 8.35 = €45bn. The 2 October payment is €2.9bn, including €0.8bn from the new loan; approximately €2.1bn is a difference calculated from rounded figures. The 7 October payment is separate. Envelope and payment bars use different, explicitly identified scales.

Security over Ukraine’s reparations claim

References to “Russian assets” can suggest that a vault has already been assigned to EU bondholders. The loan arrangement is more specific. The regulation requires Ukraine to provide the Union with a security interest over its claim against Russia for reparations, equal in value to funds disbursed. The security attaches to a claim for compensation. Separately, the regulation reserves the right to use Russian assets immobilised in the EU in accordance with European and international law. ^[4]

This distinction has practical consequences. The amount legally claimed, its expected economic value and the cash ultimately recoverable can differ. Paying a coupon requires resources available on the relevant date. A reparations claim can retain legal and political significance while the route from that claim to a payment remains unresolved.

Article 20 specifies the repayment events. The receipt of cash as Russian reparations, indemnities or a financial settlement triggers an obligation to repay within the prescribed thirty-day period. Non-monetary assets are also contemplated, with an independent valuation, a trigger after ninety days and a possible extension where strictly justified. The general thirty-day repayment period then applies. Territory is expressly excluded from that category. ^[4]

The text also contains protections extending beyond reparations. A breach of the fundamental condition covering democratic mechanisms, the rule of law and human rights makes the entire outstanding loan repayable. Established fraud, corruption or another illegal activity harming the EU’s financial interests in the management of the loan requires repayment of the amount concerned. Contractual breaches can also interrupt disbursements; breaches of repayment terms can make some or all of the balance due. ^[4]

The political shorthand that Kyiv will repay when Moscow pays describes the central structure but leaves these protections out. Equally, a delayed reform milestone can affect a tranche without automatically making €90 billion immediately payable. The authorised envelope, the disbursed balance and the relevant contractual event remain distinct. ^[4]

Sharing reparations among creditors

One of the most important clauses is also among the least intuitive. When reparations trigger repayment, the share allocated to the new EU loan depends on its outstanding balance relative to a larger pool: that balance, other outstanding G7 reparations loans and outstanding ERA liabilities, under the Extraordinary Revenue Acceleration arrangement. Amounts already repaid should no longer appear in the same balance-sheet snapshot. ^[4]

Consider an illustration, rather than an estimate of the amounts actually owed in October. The new EU loan stands at €90 billion; ERA liabilities total €40 billion and other reparations loans €20 billion. The reference pool is €150 billion. The EU loan accounts for 60%. If Ukraine receives €30 billion in cash reparations, the formula allocates €18 billion to that loan, reducing its outstanding balance to €72 billion. The remaining €12 billion falls outside this particular payment; its use depends on other applicable arrangements.

This pro-rata rule changes how a future diplomatic announcement should be read. “Thirty billion received” and “thirty billion recovered on the new EU loan” can describe very different outcomes. The allocation will depend on the debts still outstanding at the relevant time. In our model, if the other balances had already disappeared, the same reparations payment would reduce the EU loan by the full €30 billion.

The regulation leaves amounts not covered by a repayment in place until further triggers occur. A first compensation payment therefore need not close the financing arrangement. Its size, form and timing, together with the presence of other creditors, all shape the period for which European financing remains outstanding. ^[4]

Assumptions and method

Hypothetical €bn example, not an estimate of actual balances: loan share = 90 / (90 + 40 + 20) = 60%. Payment to the new loan = 30 × 60% = 18; remaining balance = 90 − 18 = 72. This calculation does not allocate the other €12bn. With no other outstanding balances, the share is 100%. Formula: Article 20(2)(k) of Regulation 2026/467.

The EU budget carries the wait

The guarantee relies on the room available below the EU’s own-resources ceiling, usually called headroom. This is a legal capacity to mobilise resources, linked to the gap between authorised ceilings and budget requirements. It is not a bank account holding €90 billion in advance. ^[6]

The budget regulation adopted in April allows necessary amounts to be mobilised above the expenditure ceilings of the multiannual financial framework while remaining within the own-resources ceiling. It therefore protects European creditors beyond the ordinary spending envelope. That protection has a framework and a limit; it does not create unrestricted funding. ^[5]

Interest follows a different, recurring route. Article 22 allows the Union to meet funding, liquidity-management and certain administrative costs that Ukraine would otherwise bear. Kyiv may request this subsidy annually. An award depends on available resources and budget appropriations. The special instrument added to the budget framework provides a route to finance debt-service costs after examining the other funding options set out in the rules. ^{[4] [5]}

The distinction matters for public finances. Subsidised interest is expenditure accruing over time. The guarantee is a capacity to intervene when European payment obligations require it. The regulation does not establish advance funded provisioning for the loan. That describes how the arrangement is resourced; it does not establish that its economic risk is zero or provide an accounting measurement of expected losses. ^[4]

Twenty-four member states participate in the enhanced cooperation. Czechia, Hungary and Slovakia receive an adjustment shielding them from the relevant operational costs, including debt service and guarantee calls; the institutions’ administrative expenditure is treated separately. Calculating a definitive French contribution would therefore require actual costs, the relevant financial year, the resources mobilised and the adjustments. Dividing by twenty-seven, or applying a rough contribution share, would create spurious precision. ^{[4] [7]}

Assumptions and method

The diagram distinguishes the ceilings and measures neither available headroom nor a national contribution. Headroom is calculated relative to the own resources needed for the budget, rather than simply the distance between the MFF expenditure ceiling and the own-resources ceiling. The guarantee remains subject to the own-resources ceiling. The cost subsidy depends on an annual request and available appropriations. Adjustments excluding three states concern operational expenditure; institutional administrative costs follow a separate rule.

The cost of elapsed time

Time creates a cost even when the principal is eventually recovered in full. In February, the European Parliament reported Commission estimates of approximately €1 billion in debt-service costs in 2027, followed by €3 billion annually from 2028. These were projections published at that time, not bills already paid or a permanently fixed financing rate for the full €90 billion. ^[8]

To isolate the mechanism, take a hypothetical €10 billion financing requirement, carried at 3% a year. Interest is paid annually and is not added to principal. Five years cost €1.5 billion; ten years cost €3 billion; twenty years cost €6 billion. Without a repayment, the original €10 billion financing need remains after all that interest has been paid.

Now assume the initial funding is refinanced after five years at 5%, rather than remaining at 3%. Cumulative interest over twenty years reaches €9 billion. The €3 billion difference comes from more expensive refinancing over the following fifteen years. This is neither an assumption about future ECB policy nor a forecast of the Union’s borrowing terms.

Refinancing also requires careful arithmetic. Replacing a maturing €10 billion borrowing with a new borrowing of the same amount leaves €10 billion to carry. The new issue increases gross financing undertaken over time; it does not automatically double net debt. The exposure is to the new rate and to the ability to find buyers when the old borrowing falls due. The investigation into the price of time in French debt explains this trade-off between maturity and refinancing.

Earlier receipt of reparations can reduce the financing need or prepare repayment of a maturity. The practical result depends on treasury management and securities already issued: incoming cash does not instantly cancel their coupons. Our model measures the simplified carrying cost of the funding requirement at par, excluding management fees, hedging and early-repayment costs. It does not reconstruct the Commission’s actual bond portfolio.

Assumptions and method

Illustration in current €bn, annual interest paid, with neither capitalisation nor discounting. In year t: cumulative cost at 3% = 10 × 0.03 × t. After refinancing at 5% following five years: 10 × (0.03 × min(t, 5) + 0.05 × max(t − 5, 0)). At twenty years: €6bn and €9bn. Principal to finance remains €10bn. Hypothetical rates, no forecasts or probabilities; fees, hedging and early-repayment costs excluded.

Immobilised principal, extraordinary income and reparations

The first resource is the immobilised principal: the Commission identifies more than €210 billion of Russian central-bank assets blocked in the Union. The second is the extraordinary income generated by cash accumulating at central securities depositories because of immobilisation. The third is Ukraine’s reparations claim. Their ownership, legal bases and timetables differ. ^{[12] [14] [4]}

The G7’s ERA, or Extraordinary Revenue Acceleration, arrangement, adopted in 2024, uses future extraordinary revenues to support loan repayment. It advances money today against a resource expected over several years. The EU contribution ultimately disbursed under that programme was €18.1 billion in 2025, separate from the new €90 billion ceiling. The larger European ceiling considered when ERA was launched should not be confused with the amount actually deployed. ^{[11] [12]}

Under the allocation currently described by the Council, 95% of the relevant extraordinary revenues transferred to the Union goes to the Ukraine Loan Cooperation Mechanism, supporting EU–G7 loan servicing, and 5% to the European Peace Facility. The percentages apply to the specified revenue flow after the prescribed steps, never to the entire immobilised Russian principal. ^[14]

Euroclear’s reporting usefully distinguishes income, a provision and a payment. In its 17 July publication covering the first half of 2026, it reported €2.3 billion of interest linked to Russian sanctioned assets and a €1.5 billion provision for the windfall contribution. The first is interest income; the second is a provisioned charge. Multiplying €2.3 billion by 95% would not reconstruct cash delivered to Ukraine. The figures come from the company’s unaudited financial presentation. ^[13]

Euroclear also attributed a 13% year-on-year decline in that interest income to lower rates. This is an important sensitivity: lower short-term rates can reduce the revenue expected for ERA while the fixed coupons already promised by the Union continue under their contracts. The effect of a rate cut therefore depends on the timetable and the particular financing arrangement. It cannot be represented as a uniform saving across every part of Ukraine support. On the Russian side, financing the 2027 budget involves a separate circuit through domestic debt and banks. ^{[13] [9]}

Assumptions and method

These categories differ in ownership and legal status. The 95% / 5% allocation applies to extraordinary income transferred to the EU after the required regulatory steps, rather than blocked assets. Euroclear reports €2.3bn interest and a €1.5bn provision for January–June 2026 in an unaudited presentation; neither figure measures a payment to Ukraine.

Recovery still has a legal dimension

The regulation reserves the right to use immobilised Russian assets in accordance with EU and international law. Moving from immobilisation to final application of the assets therefore remains a legal step to address. In July, the Official Journal published notice of a Bank of Russia action filed on 22 May, Case T-331/26, challenging in particular the provision concerning that use of assets. The court register consulted lists the case as pending. ^{[4] [15] [16]}

The action establishes that the measure is contested. It does not determine the validity of the arrangement or the eventual prospect of recovery. For financial analysis, its significance is that future recovery should not be treated as a cash-flow date already secured. Even a defensible legal position may take time to produce money.

The detailed agreement between the Union and Ukraine is another relevant document. Ukraine’s finance ministry reports that it was signed on 27 May and ratified by parliament on 28 May. The complete contract was not reviewed for this investigation. The clauses described here are those required by the regulations; implementation details not located remain outside the verified scope. ^{[19] [4]}

Defence orders and their fiscal effects

The programme also pursues a defence-industrial objective: directing orders towards Ukrainian and European capacity under its eligibility rules. Certain third countries can gain access to procurement by contributing to financing costs. In July, the Council announced political agreement on UK participation following a contribution agreement, with associated rights and obligations. ^{[18] [17]}

The economic argument is coherent: funded, more predictable demand can support production and investment in capacity. But a manufacturer’s revenue belongs to that business. The wages, margins and taxes arising from it do not constitute a payment stream automatically assigned to EU bond redemption. Claims of a budget return would require estimates of genuinely additional public revenue, associated expenditure and their distribution across countries.

The strongest argument for the arrangement lies elsewhere. It finances the continuity of a state under attack and its capacity for defence while avoiding the immediate full burden of a commercial loan. That is the purpose set out by the institutions. Alternatives could involve more grants, a different distribution of assistance or less support, each with different security and economic consequences. Comparing the loan only with the financing cost of doing nothing would omit those consequences. ^{[3] [4]}

Tracking disbursements, interest and recoveries

Analysis should track actual disbursements, funding costs borne by the budget, forthcoming European maturities and sums genuinely recovered. Recovery analysis should also identify the other outstanding liabilities entering the pro-rata calculation. An announced reparations amount becomes much more informative when readers know how much is allocated to this loan and when. ^{[4] [5]}

The principal exposure is the period for which European resources remain committed, followed by the share of principal that may ultimately not return. Interest is already a separate category of cost. The guarantee makes it possible to carry the timing gap; the security over reparations preserves a possible route to recovery. Together, they explain the financing structure. ^{[4] [5]}

At this investigation’s cut-off, the legislation establishes who promises to pay investors and how Ukraine’s repayment can be triggered. It supplies neither a certain date nor a certain amount for a future Russian settlement. Democratic scrutiny requires visibility over payments to Ukraine, the interest incurred by the Union and the conditions governing recovery of principal.

Sources and method

Research cut-off: 11 October 2026. Amounts are in current euros; stocks, ceilings, payments and projections are identified separately. The pro-rata and carrying-cost examples explain the rules. They reconstruct neither the actual outstanding balances of all G7 loans nor the precise issuance schedule supporting this assistance. Assumptions and formulas are included in the detailed figure notes. Simulated costs are nominal cumulative amounts, without discounting.

The core sources are Regulations 2026/467 and 2026/469, disbursement announcements, EU budget and funding documents and Euroclear’s reporting. Institutional announcements establish decisions and their intended purposes; they do not provide an independent evaluation of effectiveness. The evidence reviewed does not support a definitive national contribution, a single realised rate on the full €90 billion, or a reliable market valuation of the reparations claim. ^{[4] [5] [13]}