How to Use Russia’s Frozen Assets

Brad Setser is the Whitney Shepardson senior fellow at the Council on Foreign Relations.

This Ukraine Policy Brief is part of the Council Special Initiative on Securing Ukraine’s Future and the Wachenheim Program on Peace and Security.

Executive Summary

The European Union is considering proposals that would make much greater use of Russia’s immobilized reserves to finance Ukraine. A decision could come in December, when the European Council meets. All the options under serious consideration would avoid outright confiscation of Russia’s assets, while mobilizing significant additional funds to cover the financing gap Ukraine faces in 2026 and 2027. Those proposals are legally defensible, strategically vital, and pose no significant risk to the euro’s role as one of the two main global reserve currencies. The proposed mechanism ideally should be designed to make full use of all Russian frozen assets, not just the assets frozen in Belgium’s Euroclear.

Russia’s Immobilized Assets

When Russia invaded Ukraine in February 2022, the Group of Seven (G7)—at the suggestion of Italian Prime Minister and former European Central Bank (ECB) President Mario Draghi—acted decisively to freeze the reserve assets of the Russian Central Bank (CBR). Around €260 billion ($301 billion) of immobilized Russian reserves have been identified. That amount is a bit lower than pre-invasion estimates of Russian reserves in the Group of Ten (G10) currencies, as Russia’s gold and yuan holdings were never at risk of being immobilized. The bulk of those reserves—over €193 billion ($224 billion)—are held by a single Belgian custodian, Euroclear. Smaller sums are held by British, French, Japanese, Luxembourgeois, and Swiss custodians. France is thought to hold around €20 billion ($23 billion), Britain has frozen £25 billion ($33 billion)—perhaps not all of those are assets of the central bank— Japan is estimated to hold €25 billion–€30 billion ($29 billion–$35 billion), Luxembourg may have frozen over €10  ($12 billion), and Switzerland is known to hold 7.45 billion Swiss francs ( $9 billion).

At the time of the initial freeze, there was an extensive debate about what immobilizing Russian assets would do to the dollar as the leading global reserve currency. But that debate turned out to be misguided. The effect of the freeze on any individual G7 currency was always going to be modest if the G7 acted together, and, more importantly, very few of Russia’s reserves were held in dollars. The vast majority of the immobilized reserves were held in the European Union and United Kingdom and denominated in euros and British pounds. Only around $5 billion has been frozen directly by the U.S. Treasury, and just 7 percent of the Euroclear portfolio ($15 billion) is held in dollars.

As only a limited share of Russian assets is frozen in the United States, the U.S. government cannot unilaterally decide to use them to support Ukraine. The Biden administration made a concerted effort to convince more reluctant European leaders to meet Ukraine’s 2025 funding needs by pulling forward the receipt of the interest income on the Euroclear bank account, and expressed willingness to guarantee a loan based on the interest income of those frozen assets. By contrast, the Trump administration’s opposition to any further U.S. financial commitment to Ukraine has unfortunately left Washington at the margin of this debate; all the major decisions now are being taken in Berlin, Brussels, London, Paris, and Rome.

A full accounting of the immobilized CBR assets is clearly overdue. But the Euroclear account is certainly the largest pool of immobilized assets, and making greater use of that pool of funds has been the focus of recent proposals. This is in some ways unfortunate, as there are substantial Russian reserves frozen in other jurisdictions, and proposals that made full use of all the available assets would be easier for Belgium to accept. Belgian Prime Minister Bart De Wever is right to note, “The fattest chicken is in Belgium, but there are other chickens around.”

The Euroclear assets have been the focus of debate because of their size and one important peculiarity: under the terms of the custodial contract between Euroclear and the CBR, the central bank does not have a legal claim on the interest income after the original bond matures. On maturity, the bond’s proceeds are invested by Euroclear’s house bank in a deposit account at the ECB—and the interest income on that account in principle accrues to Euroclear, a private company, rather than to the frozen account of the CBR. As a result of the sanctions, Euroclear initially received a substantial windfall, and the income tax on those profits generated a related windfall for the Belgian Treasury.

But the European Council rose to the occasion and imposed regulations on central security depositories, such as Euroclear, that required separate accounting for the interest income on immobilized assets; interest income from those assets has now been redirected and is used to support Ukraine. Over time, nearly all the bonds held in Euroclear have matured—now €193 billion of the estimated €195 billion ($226 billion). But the ECB’s rate cuts have offset some of the impact of the rise in the stock of claims. Euroclear’s interest income on frozen Russian portfolio topped €6 billion ($7 billion) in 2024, with close to €5 billion ($6 billion) transferred to Ukraine. However, this flow fell by 25 percent in the first three-quarters of 2025.

The idea behind the G7’s $50 billion extraordinary revenue acceleration facility was to advance to Ukraine the expected future interest income on the cash balances held by Euroclear and Clearstream. The more complex issues with other pools of frozen assets were set aside, and no effort was made to increase the interest income on those other frozen assets even though a deposit account clearly left money on the table as the European yield curve normalized. In order to pull forward expected future income and provide Ukraine with the funds in early 2025, the United States provided a $20 billion loan against the expected proceeds from the interest on the Euroclear account; the European Union supplied an €18 billion ($21 billion) loan; and Canada, Japan, and the United Kingdom provided smaller loans.

That revenue acceleration loan only used the proceeds on the Euroclear account. The interest income on the €60 billion–€70 billion ($69 billion–$81 billion) held outside Euroclear could also be mobilized with a well-designed proposal. For example, the interest income on those assets, which now often flows back to the frozen estate of the CBR, could be captured through a punitive tax on interest paid on central bank assets immobilized by sanctions. Creative proposals that look to capture the full principal value of the immobilized assets similarly need not be limited to the Euroclear account.

Freezing vs. Seizing

The most direct way to use all the frozen assets of Russia’s central bank would be simply to seize the assets and then hand them over to Ukraine. There are legal arguments for such a course, namely that violations of international law such as the Russian invasion warrant international legal countermeasures.

But those arguments have not persuaded the critical European countries, as an irreversible step such as asset seizure has typically been done by direct belligerents in a conflict. Ukraine itself clearly has a claim to Russian assets to offset the damages that Russia has inflicted. But the EU claim is not as clear, and most of the frozen assets are in the European Union, not Ukraine, creating a mismatch between the holder of the strongest legal claim on the frozen Russian assets and the actual location of those assets.

However, it is now widely recognized that the funds from Russia’s frozen assets can be mobilized to help Ukraine without the direct seizure of the underlying assets. As a number of European governments have made it clear that they are not prepared to support a full asset seizure, the current debate centers around ways to use the €260 billion in immobilized assets to provide more than €45 billion ($52 billion) to cover Ukraine’s immediate military and budget needs.

Avenues for Mobilizing the Reserves

The current proposals assume that the frozen Russian reserves are effectively held in trust in anticipation of a future deal that will settle Ukraine’s unambiguous legal claim on the assets for the damages inflicted by Russia’s invasion. This allows, in theory, for the full principal value of the immobilized assets to be used to support Ukraine rather than just the windfall interest income that is accruing to the custodians that held Russia’s frozen reserves.

Two broad approaches could make better financial use of the frozen Russian assets. Both rely on a bit of financial engineering so that the Russian funds themselves are not directly lent to Ukraine, but rather are directed into financial intermediaries backed by the European Union and other participating governments. The financial intermediaries would then use the funds raised from the immobilized Russian reserves to lend to Ukraine. Such structures in theory could be reversed, so the Russian assets are still available to Russia should the sanctions ever be dropped; the financial flows are structured in a way that provides the theoretical possibility of returning the funds to Russia while providing Ukraine with grant-like financing,

The European Commission’s current proposal effectively directs Euroclear and Clearstream to loan the cash balance on the already matured Russian bonds to the European Union at a zero interest rate. This takes advantage of the fact that Euroclear and Clearstream do not owe any interest on the matured bonds to the estate of the CBR. The European Union would set up a special purpose vehicle to receive the funds and in turn would loan the money received from the immobilized Russia reserves to Ukraine, also at a zero interest rate.

For this structure to work, the European Union and its members collectively would need to guarantee Ukraine’s repayment of the loan—but with the expectation that repayment would ultimately be covered by a settlement of Ukraine’s claim for war damages that would be financed by the frozen Russian assets.

This idea has the merit of not requiring the European Union to raise any new funds for Ukraine; the funds Euroclear already holds in a deposit account at the ECB would be withdrawn and invested in the new special purpose vehicle. It does, however, create a contingent financial risk for the European Union should a decision ever be made to unfreeze the Russian reserves. And it has both the advantage and disadvantage of making direct use of the frozen assets; Belgium in particular worries about Russia challenging the requirement that its frozen funds be deposited in a structure intended solely to fund Ukraine.

It also is a structure, thanks to the use of zero-coupon loans, that is optimized around the Euroclear holdings. Concessional loans that do not require that the borrower pay any interest   are much easier to do with a pool of funds that itself does not currently receive interest income. But a facility designed specifically to use the assets frozen at Euroclear would be hard to apply to €70 billion or more in frozen assets that are not in Euroclear and that do currently pay interest back into the frozen account of the CBR.

An alternative structure would also require the creation of a new special purpose vehicle but would invest the proceeds of the funds from the already-matured Russian bonds into a portfolio of ten-year European government bonds and similar reserve assets. Ideally, the portfolio of European government bonds would not be a market-weighted portfolio, but rather a portfolio tilted toward French, Italian, Spanish, and common European debt. All those instruments pay a higher coupon than Dutch and German bonds. The resulting interest income would then be used to pay the interest on an EU bond that would be issued separately to finance a loan to Ukraine: €5.6 billion ($6.5 billion) would cover the interest cost on a €145 billion ($168 billion) loan, €7.5 billion  ($8.7 billion) the interest cost on a €193 billion ($223 billion) loan, and so on.

An additional income stream could be generated by imposing a special tax on the interest income of the immobilized assets held outside of Euroclear. That would assure that the interest income stream over time would not flow back to the account of the CBR, and it would generate an additional stream of tax revenue that could support emergency lending to Ukraine. However, any such action would require coordinating the actions of the British, Japanese, and potentially Swiss governments with the EU.

Because the interest income on the portfolio of bonds held by the special purpose vehicle would, by design, seek to match the interest paid on the EU bonds, the bond would have no budget cost and thus the loan to Ukraine could be provided free of interest. In other words, it is simply a different financial structure that in the end also provides Ukraine an interest-free advance against the expected Ukrainian payout from the frozen Russian assets.

The second approach maintains more distance between the actual use of the Russian foreign reserves and the loan to Ukraine. Russian assets would not be directly used to fund Ukraine but rather would be invested in a conventional bond portfolio. The income on that portfolio would then be used to pay the interest on a separate bond issued by the European Union. This could appeal to Belgium, as the Euroclear proceeds are clearly invested in a relatively standard reserve portfolio. But, as a result of this greater distance, the European Union would need to increase its market issuance, as the funds lent to Ukraine would need to be raised on the market even as the returns on the immobilized Russian assets covered the cost of the bond.

This second approach thus builds on the existing structure of the revenue acceleration facility, but rather than using future interest income to back the principal value of the loan, it uses the enhanced interest income stream to cover the interest payments on a bond—allowing the mobilization of a significantly larger sum of money. If the European Union, Japan, Switzerland, and the United Kingdom could use the full €260 billion (or a bit more) in immobilized assets, the money from the additional complexity in the financial structure would clearly be worth the effort.

Unresolved Issues

Several additional issues need to be addressed before an agreement can be reached on any facility.

One is how to protect Belgium from the risks associated with the use of the Euroclear portfolio. Belgium argues, not without justification, that it bears a disproportionate share of the risks as the bulk of Russia’s frozen reserves are held by an institution under Belgium’s legal jurisdiction. It would like to be indemnified—protected from financial losses—for the associated risks from the European Union. That is a reasonable request to guard against the contingency that Russia ever successfully challenges the legal basis of the asset freeze, or the associated use of the proceeds for the frozen funds. Belgium should also contribute the substantial funds it has withheld from the structure that used the frozen Euroclear account in the past as a form of insurance against the legal risks that it and Euroclear are facing. Similarly, Euroclear should end its current practice of withholding 10 percent of the interest income it receives on the proceeds of the frozen Russian portfolio to protect against legal risk. If future risks are assumed collectively, then Belgium and Euroclear no longer need to self-insure. A structure that only uses the proceeds of the frozen assets as a source of income to cover the interest cost on European debt issued to fund Ukraine could also reassure the Belgians, as proceeds of accounts frozen in Belgium would not be directly used to finance Ukraine.

The second issue is whether funds raised through a European guarantee for Ukraine should come with a buy European requirement on arms purchases, as the French have requested. Such conditionality is reasonable in theory, as the new structure would be backed exclusively by European guarantees and the value of assets frozen in Europe. The Trump administration is conspicuously not contributing to the structure or participating in the provision of the needed guarantees. At the same time, the overarching principle governing the use of the funds is to assure Ukraine’s capacity for self-defense. If current European production is insufficient to cover Ukraine’s needs, it would be self-defeating to insist that all the funds be used only for European weapons.

A classic European compromise should be found, one that respects the buy European principle while defining it loosely enough to allow Ukraine to buy U.S. equipment when absolutely necessary. As an example, spending a specified percentage (67 percent) of the funds on European weapons (including paying European countries for existing stocks of NATO standard kit that could be replaced by future European production) could be deemed sufficient for the buy European criteria, or an escape clause could be provided that allows open procurement for equipment needed to meet pressing military requirements that cannot be met by existing European production.

A final issue is the relationship between the new facility and the existing €45 billion revenue acceleration facility. The current proposals aim to raise €150 billion ($174 billion)—the difference between the accounts frozen at Euroclear and the €45 billion existing facility, which draws on the same asset pool. There is no reason to cap the total provision of funds at €193 billion; the full €260 billion that has been frozen by the G7 and Switzerland should be used, not just the plurality held at Euroclear.

The cleanest solution is to use a portion of the proceeds of the new facility to retire the existing, smaller facility. However, this removes the United States from the structure, as it has guaranteed $20 billion of the current €45 billion facility. An alternative, which would require the consent of the Trump administration, would roll the U.S. guarantee over to the new facility. This would reduce the scale of the European guarantee by $20 billion—incidentally, matching the dollar portfolio in the frozen Euroclear accounts and the funds directly frozen in the United States. Additional guarantees from countries with substantial reserves that are committed to support Ukraine, such as Norway, could similarly reduce the size of the needed European guarantee.

Impact of the Euro’s Global Status

Thankfully, none of the discussed proposals pose a significant risk to the euro’s role as the world’s second reserve currency.

The euro portfolio of the frozen CBR accounts was not just larger absolutely than the dollar portion, but also a much larger fraction of the total global stock of euro-denominated central bank reserves. Russia’s €155 billion ($180 billion) in identified euro reserves are just under 7 percent of all euro reserves held globally, while the $20 billion in dollars is less than 0.5 percent of the world’s roughly $7 trillion in dollar reserves.

There is no evidence that freezing Russia’s reserves back in 2021 had a significant impact on the euro’s global role.

Moreover, the risk that taking additional action on the CBR reserves will lead to a new shift away from the G7 and G10 currencies is mitigated by two important facts.

One, China is incapable of holding its reserve assets in its own currency, and thus it has few options for holding its reserves other than in one of the G10 currencies (with smaller allocations to Korea and Brazil).

Two, most of the remaining three quarters of the world’s reserves are held by either the G7 countries themselves or by countries that are much more closely aligned with the G7 than with either China or Russia. Japan is the world’s second-largest holder of reserves, and it is extremely unlikely to diversify its reserves out the other G7 currencies. Korea and Taiwan are aligned with the United States, not with China. India, although formally nonaligned, has an unresolved border conflict with China. Norway, Poland, Sweden, and other European countries similarly are unlikely to diversify out of the G7 into the currency of a country that is aligned with Russia. Only the Arab Gulf countries potentially would be able to reduce their exposure to sanctions by diversifying their reserves out of the G7. Federal Reserve Economist Colin Weiss has made this point [PDF] forcefully; there simply are not large reserve holdings in countries that realistically could opt to diversify out of the G10 currencies.

Russia had gone to much greater lengths than any other country to diversify away from the dollar: it reduced the dollar’s share of its reported reserves to 15 percent and, in the summer of 2019, moved almost all its dollars out of U.S. custodians. Russia likely further reduced its dollar holdings in the months immediately preceding the invasion of Ukraine. Yet those precautions did not protect the bulk of its reserves from being frozen. Russia was able to safeguard its gold reserves—but those reserves only have value if a country is willing to barter goods for gold or if the gold is swapped for a currency like the Chinese yuan. The sanctions limit Russia’s ability to exchange the gold into any G7 currency. So far, Russia’s gold has not been used to help finance its war; it sits in the Russian central bank’s vaults in Moscow, unused.

The simple reality is that most countries are not at risk of any future freeze from the euro, limiting the pool of funds that could leave the euro if the European Union takes additional action to mobilize the frozen CBR funds. The freeze itself was a much more significant shock, and there simply is not good evidence that it led to any subsequent reduction in euro holdings by other reserve managers. If anything, the euro’s share of global reserves has actually increased marginally since Russia’s reserves were frozen, while the share of global reserves held in Chinese yuan is down a bit.

Conclusion

Realistic, legally viable mechanisms exist that would respect the lack of consensus in the European Union around full seizure of Russia’s frozen reserves. Those mechanisms would be consistent with a strong commitment to the rule of law and have no additional impact on the euro’s role as the world’s second reserve currency. The incentive for many countries to move reserves to the euro to avoid the policy uncertainty that President Donald Trump has introduced into the dollar would remain, and global holdings of euro reserves would likely rise over the next five years.

Agreeing on the needed financial structure would allow Ukraine to cover its 2026 and 2027 financing needs without an immediate budget cost to the European Union—and demonstrate how Europe can use its financial strength to support its strategic autonomy. Decisive action is needed, and possible.

This work represents the views and opinions solely of the author. The Council on Foreign Relations is an independent, nonpartisan membership organization, think tank, and publisher, and takes no institutional positions on matters of policy.