Six billion dollars in a month: could the NBU have contained currency instability at a lower cost?
In late October 2026, the National Bank of Ukraine faced one of the most serious episodes of currency pressure in recent times. Over the month, the NBU sold almost $6 billion, including more than $1.8 billion in the final week alone. Yet the hryvnia continued to weaken. Does this mean the chosen strategy was ineffective? More importantly, could stability have been achieved by other, less costly methods? The KVARTA analytical system examined possible alternatives.
The dollar rises as reserves are used
October posed an uncomfortable question for Ukrainian monetary policy: if the National Bank spends billions of dollars supporting the foreign exchange market while the hryvnia continues to fall, what exactly is it defending — economic stability or a particular exchange rate level?
The NBU's foreign exchange interventions totalled around $6 billion in October. During 26–30 October, foreign currency sales reached $1.818 billion, around 52% above the previous week's level. On 28 October, the official dollar exchange rate rose to UAH 45.6008. An increase in the rate does not by itself indicate a currency crisis. Nor do record interventions mean the National Bank made a mistake. Ukraine is at war, with enormous fiscal needs, substantial imports and a structural foreign currency deficit. Under these conditions, the central bank inevitably acts as a key provider of foreign currency liquidity.
A fundamental question nevertheless remains: was the chosen combination of instruments the most economically efficient? We tried to answer it using the KVARTA analytical system.
Why selling dollars does not always solve the problem
The traditional logic of foreign exchange intervention is straightforward: when demand for dollars exceeds supply, the central bank sells foreign currency from its reserves. Additional supply restrains the dollar's appreciation and reduces short-term exchange rate fluctuations. The effectiveness of this mechanism, however, depends on the nature of the foreign currency shortage. If demand is temporary, intervention can stabilise the situation quickly. If the shortage is structural — meaning the economy consistently spends more foreign currency than it generates — interventions can offset the imbalance without addressing its causes.
There is also a behavioural dimension.
When market participants believe the hryvnia will continue to weaken, they may accelerate foreign currency purchases even amid substantial interventions. Businesses try to pay for imports earlier. Households convert savings into dollars. Exporters may postpone selling foreign currency earnings to the extent permitted by regulations. Expectations of future depreciation thus become a driver of current foreign currency demand.
The central bank risks finding that ever larger interventions are needed not only to offset an economic shortfall, but also to contain behaviour partly shaped by exchange rate expectations themselves.
This mechanism was one subject of KVARTA's preliminary research.
What KVARTA found
Within the KVARTA–AMT research programme, we examined the hypothesis of asymmetric monetary transmission. Its premise is that, in a highly dollarised economy, exchange rate expectations can affect prices and business decisions faster than the central bank's traditional interest rate instruments. This is particularly important for Ukraine, where war has weakened the credit channel of monetary policy and a substantial share of economic decisions depends directly or indirectly on the dollar exchange rate.
KVARTA's preliminary empirical tests identified some signs of this mechanism. However, the statistical advantage of extended models over baseline models was not sufficiently robust. It is therefore too early to claim definitive confirmation of the hypothesis. It nevertheless allows an important research question to be formulated.
What if currency instability is best contained not only by selling dollars, but also by simultaneously influencing hryvnia liquidity, foreign currency flows and market expectations?
We considered four possible approaches.
Four strategies: what the NBU could have done
First: continue large-scale interventions
The NBU's actual policy relied on managed exchange rate flexibility and substantial foreign currency sales. Its clear advantage is the ability to meet demand directly and restrain sharp exchange rate movements. This is particularly important in a wartime economy. Depreciation can quickly affect the cost of imported energy, equipment, medicines and other essential goods.
The drawback is a substantial need for international reserves. Although intervention does not necessarily imply an equivalent reduction in total reserves — which are also affected by external assistance and other foreign currency receipts — its cost to the system remains significant.
Second: allow the hryvnia to adjust more quickly to the market
An alternative strategy involves smaller interventions and greater exchange rate flexibility. In this case, some excess demand for foreign currency disappears as it becomes more expensive. Imports cost more, some buyers postpone transactions and exporters have an additional incentive to sell foreign currency.
In theory, this conserves international reserves. But there is a risk: rapid depreciation may increase rather than reduce demand for dollars by triggering panic expectations. Such a strategy requires particular caution during wartime.
Third: use a balanced mechanism
KVARTA identifies this approach as the most promising for further quantitative testing.
Its premise is to distribute the stabilisation burden across several instruments. The National Bank continues to intervene but does not try to offset every change in foreign currency demand using reserves alone. At the same time, it uses mechanisms to manage excess hryvnia liquidity, maintain the appeal of hryvnia savings, improve conditions for selling export earnings and coordinate large budget payments.
Communication plays a special role. If the central bank convinces the market that it will defend a specific exchange rate, participants may start testing that boundary. By contrast, a predictable policy aimed at containing excessive fluctuations rather than guaranteeing a particular dollar price could potentially reduce one-sided exchange rate expectations.
A balanced strategy is not cost-free, however. Absorbing liquidity has a cost of its own, while postponing budget payments can harm the economy. The effectiveness of each instrument must therefore be calculated separately.
Fourth: tighten monetary policy sharply
Another option is to raise interest rates and constrain hryvnia liquidity more actively. This policy may support demand for hryvnia financial instruments. But it also raises borrowing costs, affects government debt servicing and may restrain economic activity. In Ukraine, where interest rate transmission is weak, the effectiveness of such a move is not obvious.
KVARTA finds insufficient grounds to regard a sharp policy rate increase as the optimal response to a short-term currency shock.
How many dollars could have been saved?
To illustrate the potential scale, consider the final week of October. Actual interventions amounted to around $1.818 billion. If an alternative strategy had reduced them by 10%, the NBU would have sold $182 million less.
A 15% reduction would correspond to $273 million; a 25% reduction to around $455 million. These figures look striking, but an important qualification is necessary. They are not proven reserve savings, but arithmetic scenarios illustrating the scale of a possible effect.
We cannot yet reliably establish how the exchange rate would have changed with smaller interventions, how importers and exporters would have responded, or whether the NBU would have had to spend the saved funds the following week. Potential inflationary effects must also be considered. Saving $400 million in international reserves is not necessarily beneficial if accompanied by a substantial rise in prices and deteriorating economic expectations.
KVARTA therefore proposes assessing each strategy's total economic cost, rather than simply the amount of reserves used.
The central problem is the structure of the economy, not the exchange rate alone
Debate over foreign exchange policy often boils down to the question: what should the dollar exchange rate be? We believe this approach is too narrow. The exchange rate is just one indicator within a much more complex system. The hryvnia's resilience depends on public finances, export volumes, import needs, international assistance, the behaviour of banks and households, and confidence in economic policy.
If the government undertakes large-scale hryvnia spending without the economy generating a corresponding supply of foreign currency, additional pressure arises in the foreign exchange market. If international assistance finances a substantial share of imports, international reserves effectively become an important part of the mechanism financing the wartime economy.
It is therefore inappropriate to assess the NBU's effectiveness in isolation from the government's fiscal policy.
Exchange rate stability is not solely the central bank's responsibility. It results from coordination between monetary, fiscal and external economic policy.
In our view, this is where the greatest scope lies for improving the effectiveness of economic management.
What we still do not know
We deliberately distinguish three categories of results: confirmed facts, economically grounded hypotheses and quantitatively validated causal effects. The actual scale of NBU foreign exchange interventions is known. The mechanisms through which exchange rates, liquidity and expectations operate are reasonably well described by economic theory.
KVARTA's preliminary research has helped formulate alternative monetary policy scenarios, but a definitive conclusion that one scenario outperforms another requires a full counterfactual test.
Such a test must use only information available to the NBU at the time of each decision and assess outcomes over horizons of 5, 20 and 60 trading days. It must compare not only exchange rate movements and reserve use, but also inflationary effects, the cost of monetary instruments, fiscal effects and the risk of a renewed currency shock. Until this test is complete, it would be wrong to claim that KVARTA has already proved the NBU's actual policy mistaken.
Even now, however, there are sufficient grounds to question whether exchange rate stability should be assessed solely by the level of the exchange rate.
Conclusion: Ukraine needs the most efficient stability, not the most expensive
October's events showed how costly it can be to maintain foreign exchange equilibrium in an economy dependent on external financing and imports. Almost $6 billion in foreign exchange interventions in a single month is a compelling reason to analyse monetary policy efficiency, but record interventions alone prove neither that the National Bank's decisions were right nor that they were wrong.
An alternative to large-scale foreign currency sales may be a balanced system that simultaneously influences the exchange rate, hryvnia liquidity, foreign currency flows and economic expectations. KVARTA's preliminary analysis identifies this approach as the most promising for further research.
Its actual advantage still needs to be demonstrated, but the central conclusion can already be stated:
Ukraine needs monetary policy that measures success not by the number of dollars sold, or even by the stability of a particular exchange rate, but by the economic stability achieved at the lowest total cost.
Otherwise, we risk spending billions for years to alleviate symptoms without addressing the causes of currency instability.
KVARTA — an independent system for integrated economic analysis, forecasting and evaluation of management decisions.
Research: KVARTA–NBU / Monetary Policy Counterfactual Analysis.