Test the Grand Continent. The ideas and analyses you won’t find anywhere else. But you’ll see them everywhere soon. In your inbox from 8 euros per month
Since February 2022, a single question has repeatedly surfaced: how long can Russia still finance the war? Six months, a year? Three deadlines for exhausting its sovereign wealth fund have circulated, from mid-2026 to the end of 2027, but none has materialized so far. If the question is legitimate, it nonetheless implies a misreading of the problem. The Russian state is not a liquidity-strapped company on the brink of bankruptcy and can still sustain the immense expenditures demanded by the invasion of Ukraine. Yet the cost is rising faster than revenues and the oil rent is dwindling. Consequently, the real question now is what sacrifices the Kremlin will have to undertake to continue financing its war effort.
In this war-serving economy, the principle is that the levy finances a spending that the regime cannot reduce or admit. Line by line, the Russian public accounts, however, allow us to gauge what the communiqués seek to conceal. Our previous analyses devoted to Ukrainian strikes against Wildberries and against Yandex data centers showed how Kyiv hit Russia’s power rather than its strength. We must now examine the ultimate target of this wear-and-tear strategy, namely the state’s finances and the reserves accumulated for the war of reasserting the power Vladimir Putin seeks.
A Fund with Two Pockets
Let us begin with the object that forms the core of the “cushion” Russia prepared to weather the anticipated shock of entering the war: the National Welfare Fund (NWF). Created in 2008 by splitting the Stabilization Fund, it is chiefly fed by the surplus of oil and gas tax revenues collected when the price per barrel exceeds a threshold set by the budget law. In plain terms, if the price of a barrel is $59, as in 2026, and it trades at $103, the revenues corresponding to the $44 gap feed the fund.
What about the use of the surplus generated? The official purpose of the National Welfare Fund is twofold: to fill the deficit of the pension fund and to smooth the cycles of commodity prices. Unlike the Norwegian sovereign wealth fund, it does not hold a diversified foreign portfolio generating returns; it mainly consists of domestic assets along with some cash and precious metals. However, the de-dollarization, decided in 2021 in anticipation of a freeze on Western assets, and then the freeze itself in 2022, reduced the mobilizable share to two assets: the yuan and gold.
That is the decisive point to understand how Russia financed the early phase of its war. The fund splits into two pockets, which are not of the same nature.
- A liquid pocket gathers gold and foreign currency assets held on the fund’s accounts at the Central Bank. It is mobilizable immediately. It is the only one that can actually fund expenditures.
- Another pocket, illiquid, contains stakes in Russian companies, loans for infrastructure projects, and assets frozen since 2022. These are mere lines on the balance sheet, which cannot be sold to finance next month’s expenses nor divested without destroying their value or destabilizing the companies involved.
The mobilizable share of the fund contracts year after year, shifting toward the illiquid pocket, and its overall value thus takes on a different meaning. When a communiqué reports the fund standing at more than $150 billion, it is adding the two pockets. This is mathematically accurate, but operationally misleading.
The figures published by the French Treasury’s economic service in Moscow allow measuring the gap between the total figure and the actually mobilizable share. On February 1, 2022, on the eve of the invasion, the fund totaled nearly $175 billion, i.e., 10.2% of GDP, with a liquid portion exceeding $112 billion, representing almost two-thirds of the total and 6.6% of GDP. By September 1, 2026, total holdings stood at 13,194.9 billion rubles, i.e., $154.1 billion and 5.6% of GDP, but the liquid share stood at 3,998 billion rubles, i.e., $46.7 billion and 1.7% of GDP. A month later, by October 1, it had fallen to 3,800 billion rubles, i.e., $45 billion and 1.6% of GDP, for a total of 12,985 billion rubles ($153.8 billion).
The dollar total had fallen by only 12% while the usable pocket had been halved and more. The immobilized share thus rose from about one-third to more than 70%. It is not a reserve that is replenishing itself, but a endowment whose composition has been upended. As for the decline in its share of GDP, it also reflects nominal growth in an inflation-heavy economy.
A good analogy to understand this strategy would be a household emptying its savings book to acquire a property whose price rises. The theoretical net worth increases, but its ability to pay the following month’s charges collapses at the same time. This exactly mirrors the Kremlin’s position, which converted its liquid assets into claims whose value is far less tangible.
The denominator itself participates in the illusion. Measured in dollars, the Russian GDP shows dramatic growth since 2022, ranking Russia among the ten largest economies worldwide according to the IMF. This growth is essentially nominal. High inflation inflates the ruble value of production, which is then converted at an exchange rate kept by capital controls and by exporters being required to surrender their currencies. The IMF, however, can no longer conduct its on-the-ground full assessments, with its Article IV consultation last in 2021. Here we find, inverted, the mechanism that flatters the sovereign fund. A weak ruble increases the ruble-denominated value of a fund denominated in gold and yuan, while an administered ruble increases the dollar-denominated value of a GDP denominated in rubles. In both cases, the currency conversion fabricates a display of wealth that the real economy does not produce.
The monthly reports from the Ministry of Finance detailing the fund’s composition line by line allow reconstructing its real trajectory and understanding the loss of value in the liquid pocket.
Two lessons emerge from these figures. The first concerns gold. Contrary to what one might expect, the stock did not decline steadily. It first rose by more than a third, from 405.7 tonnes in February 2022 to 555.0 tonnes in December of the same year. The freezing of euro, pound, and yen holdings forced Moscow to reconfigure its liquid pocket around the two assets it had left. The decline only began afterward, and the precious metal reserve has fallen by 76% since that peak.
The second is more troubling. Between January and October 2025, the fund appeared to recover by 36% in dollars, rising from $116.8 to $158.8 billion, and its liquid share from 34% to 34-?% of the total, even as the gold stock declined by 8% over the period. This increase in the total is largely due to the illiquid, ruble-denominated pocket, whose dollar equivalent rose with the ruble’s appreciation in 2025. The official dollar thus moved from 101.6797 rubles on January 1, 2025 to 82.6084 rubles on October 1, 2025, a drop of about 19% in ruble terms, or an appreciation of the ruble by about 23% in dollar terms.
It is like a building valued at 1 million euros that would be worth 1.10 then 1.35 million dollars if the euro rises from 1.10 to 1.35 dollars, without changing or rising in euros. The ruble-denominated assets within the NWF thus appear richer in dollars when the ruble strengthens, without Russia having real additional resources mobilizable. The revaluation of gold explains about half of the rise in the liquid share alone. An observer could have believed Russia was rebuilding its reserves at the same time as it sold its gold stock. The gold price rally, which climbed more than 50% from late 2024 to late October 2025, was thus a “divine surprise” for the Kremlin, over which it had no control, but which helped mitigate the war’s impact on its finances.
The Fund Is Not Spent; It Is Converted
The challenge is to understand how the mobilizable reserve has drained. Three channels coexist, and the most durable is not the most intuitive.
- The first is the direct financing of the deficit. Yuan and gold are sold, rubles are collected, and the expenditures incurred are paid. This is the channel people instinctively think of, but it is not the only one draining the liquid pocket.
- The second concerns assets immobilized by sanctions. The roughly $280–330 billion of Russian sovereign wealth assets frozen since 2022, including about €210 billion immobilized in the European Union, at Euroclear, are essentially central bank reserves rather than fund assets. The frozen share of the NWF itself is of a different order of magnitude, but the mechanism is identical in both cases. These assets appear on the books; their valuation is purely accounting, and their recovery depends on an unlikely geopolitical settlement before the war ends. They constitute a phantom reserve, present on the balance sheet but economically dead.
- The third channel is the conversion into domestic assets, and this is where the plundering, in the literal sense, occurs. For the sole month of May 2025, the Ministry of Finance reportedly drew 35.9 billion rubles to cover the deficit, but also 300 billion to fund state projects, essentially a high-speed rail line between Moscow and St. Petersburg. Comptably, nothing is lost here. The fund now holds a claim on a construction project. Operationally, that money will never return to the liquid pocket. The piggy bank does not empty; its contents are progressively replaced by government debt instruments owed to itself.
The scale of this conversion can be read in the number of lines that make up the illiquid pocket of the NWF. In February 2022, it counted eight lines, namely deposits at the VEB development bank, some foreign sovereign bonds, a few securities of Russian issuers, preferred shares of financial institutions, two subordinated deposits, as well as stakes in Sberbank and Aeroflot. By October 1, 2026, that number had grown to twenty-four.
The euro, the pound, and the yen disappeared between 2023 and 2025. The new lines bear the names of Rostec, Avtodor, GTLK, or the Russian Railways, and they finance projects whose returns are fixed by decree. The sovereign fund thus became, as the communiqués accumulate, the register of commitments the state has undertaken to itself and that it could not fully honor in a financial crisis.
We thus understand why the earlier predictions of depletion mentioned above have so far been debunked. Three mechanisms have delayed the deadline. First, the price of gold rose sharply since the start of the war, which mechanically revalues the remaining stock. Second, the ruble’s depreciation inflated the ruble-denominated value of foreign-currency assets. The very event that widens the deficit inflates the fund’s display, and one can draw from it year after year without the amount, published in rubles, diminishing proportionally. Finally, Moscow shifted toward domestic borrowing rather than exhausting the fund.
That is what misled forecasters, who divided the balance by the annual withdrawal pace, assuming the first would shrink by that amount. This offset cannot be renewed indefinitely, and it does not erase the major fact. Since February 2022, the liquid pocket has fallen from 112.7 to 45 billion dollars, and its weight in GDP has dropped from 6.6% to 1.6%. Between September 2025 and September 2026, this mobilizable pocket declined by 6% in rubles and 5.4% in dollars. By contrast, in August 2026 alone, it rose by 8.3% in rubles, and only 1% in dollars. In September, conversely, it fell by about 5% in rubles. The ruble’s depreciation in 2026 thus creates, month to month, the illusion of a replenishing reserve, even as it continues to erode in reality.
November 2025: A Decisive Threshold
Yet a qualitative signal largely went unnoticed. Until autumn 2025, gold sales from the fund were a matter of internal bookkeeping. The state sold to the central bank, and the metal did not leave Russia’s sovereignty perimeter; only the accounting line changed ownership. On November 19, 2025, the Interfax agency announced that the Central Bank had begun selling ingots on the domestic market to contribute to the budget financing, the first such announcement since 2022. In the first four months of 2026, the Central Bank’s stock of gold thus posted its largest decline since 2002.
Under these conditions, the central bank’s gold stock, which had remained roughly stable around 2,300 tonnes since the eve of the invasion (73.9 million troy ounces as of February 1, 2022), fell to about 2,270 tonnes as of September 1, 2026 (73.0 million ounces), below its pre-war level. As for the gold held by the fund itself, it declined from 406 tonnes in 2022 to 131.7 tonnes by October 2026. At the same date, yuan reserves stood at 184.4 billion, down 25 billion since October 2025. They had peaked at 309.7 billion on December 1, 2022, and were at 226.8 billion as of February 1, 2022. They had thus lost 40% since that peak.
These figures do not signal a bankruptcy, but the end of an accounting loophole. The real problem now lies in the deficit and the Kremlin’s ability to offset losses by making political choices that, ideally, would not affect its capacity to wage war and would not provoke domestic resistance. It remains to be seen whether such choices still exist.
Constraints on Flows Rather than on Stocks
National defense spending rose from about 4,700 billion rubles in 2022 to 13,490 billion in the 2025 budget law, and the 2027 budget proposal places them at 17,100 billion. When we add internal security and maintaining order, the military and security apparatus now absorbs more than 40% of federal spending. Revenues do not follow this pace. In the first eight months of 2026, expenditures rose by 14.7% year over year, compared with 9.2% for revenues. Public procurement, which makes up most arms purchases, rose by 36.2%.
As a result, the federal budget deficit widened sharply. For 2025, it reached 5,600 billion rubles, nearly five times the initially planned figure. The 2026 deficit was projected at 3,786 billion rubles, i.e., 1.6% of GDP. In the first eight months of the year, it already stood at 5,795 billion, or 2.5% of GDP and 53% higher than the target for the full year. After a September surplus of 609 billion, the nine-month deficit stood at 5,186 billion rubles, i.e., 2.2% of GDP and still 37% above the annual target, compared with 3,471 billion for the first nine months of 2025.
Petroleum and gas revenues for 2025 fell to their lowest level since the pandemic, with the Urals trading at an average of $44.90 in November, and below $35 in December, while the 2026 budget had been built on a price of $59.29. They fell by 17.2% year over year in the first nine months of 2026, despite the surge in prices caused by the Ormuz crisis, due to both a stronger-than-expected ruble and the Ukrainian campaign, whose cost is now seen in the compensations paid to oil companies.
On September 28, 2026, the Ministry of Finance acknowledged this deviation. The 2026 deficit is now estimated at 7,300 billion rubles, or 3.2% of GDP, nearly double the target and the highest level since the war began. Expenditures would reach 48,600 billion rubles, i.e., 10.2% higher than planned, while petroleum and gas revenues would amount to 7,600 billion rubles, i.e., 1,300 billion rubles less. The gap would be covered by 459 billion rubles drawn from the sovereign fund, i.e., 11-12% of its liquid portion, and by net borrowing of 5,000 billion rubles. The structural primary deficit would reach 1.7% of GDP, whereas the budgeting rule called for a zero primary balance.
In the absence of sufficient liquid reserves to sustain the deficit, the Kremlin executed a genuine fiscal pivot, the scale of which is measurable. As the hydrocarbon rent contracts, VAT revenues rise by 26.3% year on year. Some of this increase stems from the VAT rate rising from 20% to 22% on January 1, 2026. But the bulk of this gap comes from widening the tax base, the threshold for small businesses on the simplified regime being reduced from 60 million rubles in 2025 to 20 million in 2026.
Thus the state taxes more on each purchase and has brought into tax revenue enterprises that previously escaped. The burden weighs especially on small and medium-sized enterprises, precisely those hit by the war against warehouses. The authorities recognize the political cost, having frozen the threshold at 20 million rubles through 2029 in July 2026, even though it would have had to be lowered to 15 million in 2027 and to 10 million from 2028 onward.
At the same time, the share of hydrocarbons in federal revenue has almost been halved since the start of the war, and non-hydrocarbon revenue now accounts for nearly four-fifths of the total.
Thus, the Russian state funds a growing portion of its war effort through domestic consumption rather than exports. It is no longer the barrel that supports the budget, but the consumer. This helps explain why the Ukrainian campaign has expanded from refineries to warehouses. The impact of strikes on refining is now measurable in volumes processed, pump prices, and subsidies paid, but it was not visible in petroleum revenues.
The Ministry of Finance partly attributed this downturn to the ruble’s appreciation relative to the budget assumption. The 2026 finance law assumed an average rate of 92.2 rubles per dollar, whereas the ruble traded around 84 rubles in early October. Each dollar of petroleum revenue thus converted to fewer rubles. The strikes, however, remained invisible in these revenues for a simple reason. The crude that is no longer refined is exported. Yet their effect is now felt through another channel, namely the compensation paid to refiners, whose rise transforms each strike into a public expenditure.
The campaign of “long-range sanctions” did indeed scale up as it became clear Moscow would benefit from a sharp rise in energy prices. By late August, the volume of refined crude in Russia reached its lowest level in 21 years. Estimates indicate that between 30% and 40% of primary refining capacity was out of service. In July, 78 of 83 regions reported fuel shortages, 38 imposed sales restrictions, and four regions, including Crimea, declared a state of emergency.
A Modest Yet Very Costly Debt
It is frequently argued that Russia’s public debt is very low, at 17.1% of GDP as of July 1, 2026, probably 19.9% by year’s end according to the Ministry of Finance, versus 115.6% for France at the end of 2025. But this comparison is misleading for two reasons.
The first is the price of this debt. Cut off from international markets, Russia borrows domestically, at rates aligned with a policy rate of 14%. It thus spends roughly 3,900 billion rubles on interest in 2026, about 1.7% of GDP and nearly 9% of federal outlays, and plans to devote 4,600 billion in 2027, whereas France spends 2.2% of GDP on interest on a debt six times larger. The average rate borne by the state is around 8.5%, versus under 2% for France. It is not the debt level that constrains Moscow, it is its cost. And the authorities themselves set the 20% of GDP threshold as the prudent limit, a threshold they will approach by the end of the year.
The second reason is that this figure measures only part of the war debt. On February 25, 2022, the Law No. 29-FZ required banks authorized to lend to army suppliers on favorable terms, while exempting them from remunerating advances the state deposits on their accounts. This is not new; since 2010, the Kremlin had financed part of rearmament through loans to manufacturers, later having to restructure bad debts in 2016 and 2017, and again in 2019 and 2020. Legally, these debts belong to the companies; economically, they are those of the state, which has already repaid them twice.
Craig Kennedy, former vice president of Bank of America and Merrill Lynch and a researcher at Harvard’s Davis Center, has reconstructed the scale. Between mid-2022 and the end of 2024, corporate Russia debt rose by 71%, about $415 billion, and he estimates that half to 60% of this growth, i.e., $207–$249 billion, funds war-related activities. When related to 2026 GDP, these loans represent 8 to 10 percentage points. If we include this off-budget debt, Russia’s real public debt would lie between 25% and 30% of GDP.
The weakness of Russia’s public debt is thus less about state solidity than about how it finances the war, by lodging part of its cost in the banking sector’s balance sheets where it cannot be seen.
The budget proposal for 2027, submitted to the Duma in late September 2026, confirms the enduring anchoring of the war economy at the expense of civilian priorities. National defense receives 17,100 billion rubles (6.9% of GDP), i.e., 2.6 times more than social policy, which is reduced to 6,600 billion (2.7% of GDP). With internal security and maintaining order (4,200), the military-security bloc absorbs 21,300 billion rubles, i.e., 8.6% of GDP and almost 44% of federal spending. For every ruble spent on social benefits, the state plans more than three on arms and maintaining order.
This does not mean Moscow will be unable to finance the deficit (5,443 billion rubles planned for 2027, then 5,120 and 5,245 billion in 2028 and 2029, i.e., around 2% of GDP each year), but that such financing is becoming more expensive and more visible to the population. We are approaching the end of cycles of 12 or 24 months during which the Kremlin sacrificed its own public policies and welfare state to sustain its war effort. The first trillion redirected is still easy to justify and comparatively painless; the tenth is much less so.
The cushion is now thin. As of October 1, 2026, the liquid portion of the National Welfare Fund represented only 3,800 billion rubles, i.e., less than three months of defense spending at the 2027 pace, and barely more than two-thirds of the deficit projected for 2027. The year 2027 will undoubtedly be decisive for Russian finances, given that defense expenditures are set to rise by more than 30% again.
What the Accounts Say, and What They Conceal
If the sovereign fund of Russia runs dry, it does not do so in a linear fashion, as a private bank account would. Since 2022, the federal state behaves more like a loss-making conglomerate, distributing losses among its subsidiaries to present a balance sheet that looks more favorable than reality. Its liquid assets become receivables on state-owned enterprises, its gold leaves its perimeter, and part of its war debt is recorded on the books of public banks rather than on its own. None of these movements constitutes a bankruptcy, and that is precisely what makes them convenient for the regime. Each move shifts the cost of the war to a place where it is not yet visible, but where it accumulates.
This reconfiguration has a limit, which the 2027 figures may eventually reveal. The liquid reserve no longer covers a substantial portion of the announced deficit, and every additional ruble will have to come from taxes, inflation or household savings—levers that people will feel directly. Strikes against refineries caused shortages in nearly all Russian regions in the summer of 2026 and increased the cost of compensations paid to refiners.
The campaign targeting Wildberries and Ozon warehouses weakened two flagship sectors of the Russian economy and ruined some of their sellers, whom the state had to support through tax deferrals and by VTB’s entry into Wildberries’ capital. Strikes against the data centers of Yandex, Sberbank, and later, at the end of October, against a Finance Ministry center housing state registries and tax data, finally strike the digital backbone on which payments, commerce and public services depend. Taken in isolation, none of these attacks threatens the Russian financial balance. Collectively, they convert an increasing share of destruction into public expenditures or implicit state liabilities.
Long-planned preparation through reserve accumulation and by building bypass channels allowed the Kremlin to fund the war without making its cost visible to the population. But for nearly five years, the war has devoured resources that could have modernized the Russian economy and its infrastructure, and the reserves that once made this sacrifice invisible are now running out. This delayed cost could catch up with the leadership at a moment when they would want to intensify pressure on Kyiv, just as the offensive shows signs of fatigue. In fact, during the first ten months of 2026, the Russian army gained only about 117 square kilometers of net territory, according to the Institute for the Study of War, and it even lost ground in September and October.
On October 29, 2026, Vladimir Putin asserted that the West had failed to inflict a strategic defeat on Russia, before warning that a war between Russia and Europe would have “no winner.” The statement was meant to project strength, but it betrays a change in rhetoric. After an invasion expected to last a few days, the Kremlin now presents not being defeated as a success. It may be because financially, Russia is already in a difficult position. The crisis is not yet inevitable, but it would no longer require a major shock to erupt. And it would not stay confined to the Federation’s accounts. By distributing losses among public banks, state-owned enterprises, and the sovereign fund, the attempt to obscure costs has, in effect, prefigured contagion pathways. The crisis could thus spread through the entire economy, then the political order, and into the war effort in Ukraine itself. The final act of this war remains to be written. It could play out less on the front than in the accounts, unless the Russian leader judges it safer to moderate demands and embark on substantive negotiations, to “extract gains” before the situation becomes intractable. There is no such indication at present.