Measuring Ukraine’s private costs during the war

There is a significant gap in the production of national-level data that captures the level of economic activity in Ukraine in the first half of 2022. As Russian aggression continues to devastate large parts of the country, the flow of official statistical data has been erratic This lack of data and the continued war have increased the challenge of assessing Ukraine’s post-war recovery needs (Becker et al. 2022).

Respondents to the State Statistical Service of Ukraine survey are allowed to suspend their reporting during periods of martial law. For example, monthly sectoral output data and enterprise payroll data have not been reported since February 2022. These sets of indicators are required to construct national output estimates under the standard aggregation models employed by economic forecasters

Business surveys are limited in their statistical value due to geographical and sectoral gaps. War created additional complications. For example, the monthly business sentiment survey of the National Bank of Ukraine was interrupted for the important months of March to May (NBU 2022) and as a result, it now points to a general trend rather than a specific and comparable level of economic activity. .

The use of other data that might provide good proxies for economic activity in normal times has become inappropriate in times of war. In particular, value-added tax (VAT) data could not be translated into sales estimates as businesses were allowed to switch to a no-VAT system in March 2022. Electricity consumption, another proxy for economic activity, has also become unavailable. Since the power load data was kept secret at the start of the war.

Researchers have explored new methods to proxy economic activity in Ukraine, for example through the use of social media data (Constantinescu et al. 2022). However, social media use can vary dramatically between seismic events like the Russian attack, making these data a noisy proxy. Also, social media is often used for non-commercial reasons, blurring the link between its use and economic activity.

We propose another method for measuring personal consumption using micro-level bank card data. The advantage of this approach is that it has a seamless flow of information and data comes from almost all Ukrainian regions, including those regions where standard data collection techniques are suspended.

We begin by describing the level of disruption across Ukraine, to give us a minimum benchmark for consumption reduction. We then report more precise estimates, acknowledging some potential bias in the estimates.

Areas affected by war

In the absence of reliable statistics or business survey data the simplest back-of-the-envelope estimation is mapping the areas most affected by war operations and overstating these affected areas and the severity of the damage with their role in national output. This estimate allows us to measure the direct war-related burden on the economy. The approach has limitations because of indirect injuries to the economy, such as a naval blockade of Ukrainian exports (Blinov and Jankov 2022a). It therefore provides a lower estimate of the negative economic consequences of war.

We use this regional approach by investigating eleven Ukrainian regions that have witnessed a significant presence of Russian military forces (Table 1). Using official government data on combat operations, we estimate capture and loss levels by assigning weights. The darker the color, the stronger the negative impact on economic activity. For example, areas coded black have most of their territory occupied by the Russian military in the corresponding month. White color means ‘no war’, as all Ukrainian regions are subject to airstrikes and some areas may see conflict (such as Dnipro or Odessa, not present in this table). Rather, it means that there is no significant troop movement on the ground.

This calculation leads us to conclude that a third of Ukraine’s economy ceases to function in March 2022, then the counter-advance of the Ukrainian army in the north halves this share to 15-18% in April, and then further reduces it to an average of May and June. 13% per month.

Table No. 1 War-Torn Areas and War Intensity (% of GDP) in Months of 2022

formula: State Statistics Service of Ukraine, authors’ estimates.

However, use also declined in geographies not directly invaded by Russian troops. Accelerating inflation, job losses, exchange rate movements all depress economic activity. This means that the above bottom-line figures represent a practical upper limit for private consumption estimates: the actual decline figures are more dramatic.

Bank card transactions

We employ bank card activity data as a proxy for personal spending. Specifically, we use monthly point of sale (POS) transaction data from Alfa-Bank, one of the largest private commercial banks in Ukraine. The data shows the exact location where a purchase is made. As of May, these data were consistent with bank card transactions reported by Privatbank, Ukraine’s largest bank (Privatbank 2022). The second provides some comfort about the robustness of our estimates for the two banks jointly accounting for more than half of the POS market.

The damaging effects of war are easily understood in the data. Bank card payments halved in March due to a war-induced drop in demand in Ukraine. April and May saw a recovery in transaction volume, and by June transaction volumes had stabilized at about 75% of their pre-war monthly levels. Adjusted for seasonality and inflation over the period, this estimate corresponds to a 40% reduction (Figure 1).

We also plot data for the pandemic-affected first half of 2020 to compare the relative depth of the war crisis in consumption with previous recessions, also due to exogenous factors. In 2020, retail sales, adjusted for seasonality and inflation, surpassed 90% of their pre-shock levels by June, reflecting a rapid recovery from the lockdown.

Figure 1 Retail Sales Trends for 2020 and 2022 (January = 100)

formula: State Statistics Service of Ukraine, authors’ estimates based on micro-level bank transaction data.

During war the pattern is different. When the trend estimates are adjusted for seasonality and price, it appears that consumer demand in the second quarter of 2022 was 3.3 times higher than during the initial COVID-19 lockdown in 2020. Household consumption fell 9.3% year-on-year. Second quarter of 2020. However, applying POS-derived coefficients, personal consumption declined by around 30% in the second quarter of 2022, plateauing at around 70-74% of the previous year’s levels by June 2022.

The trend has been uneven for different bundles of consumption. Although total nominal POS sales in June remained slightly above 75% of their pre-war level (Figure 1), some of their contributors have already exceeded their pre-war points, while others are still less than half of their pre-war levels. Battle standard (Figure 2). Food purchases in supermarkets have remained fairly stable. At the same time, sales of clothing and shoes showed rapid recovery, while sales of entertainment services (cinemas, bars, etc.) and dining remained depressed.

Figure 2 Alfa-Bank’s POS Sales by Key Sector in 2022 (January = 100, Unadjusted)

formula: Author’s estimates based on micro-level bank transaction data.
Comment: No season and price adjustments.

There are limitations to the use of bank card transaction data as a proxy For example, Ukrainian households withdraw more cash during a crisis and then make payments with that cash. This consideration makes POS-based cost estimates conservative, especially for March and April 2022, when a lot of cash was withdrawn and then returned to the bank through various vendors. It is also worth noting that POS as a payment method has different penetration levels in big cities and small towns, with spending in the latter being more cash-driven. These two biases serve to make the estimate more conservative and thus serve as an upper bound on the decline in private consumption.

In summary, this aggregated micro data suggests that private consumption in Ukraine fell to half of its pre-war level in March 2022. In the second quarter, it recovered to 70%, while the monthly breakdown indicated that personal consumption plateaued at 70% by June 2022. -74% of the previous year’s level.

These micro-data estimates of private consumption are significantly affected by the negative effects of war due to large refugee waves following the outbreak of hostilities (Blinov and Jankov 2022b). Estimating the cost of the more than five million Ukrainians who fled the country and have yet to return presents a statistical challenge. They are still Ukrainian consumers, even though their consumption counts as imports for national accounting purposes. The repatriation of these Ukrainian citizens will likely increase consumption and push it away from the current plateau.


Micro-data methods for estimating private spending during the war in Ukraine show a dramatic decline of 70-74% from the previous year’s level and then some recovery. Because of the obvious differences in the nature of the two crises, the path to recovery after the COVID-19 pandemic has some similarities, albeit from a much deeper trough and at a much slower pace. In particular, the difference between the two scenarios is the large refugee wave, which has not yet subsided and creates significant uncertainty about future recovery trends. These early estimates indicate that Ukraine may indeed be poised for a slow recovery path (Blinov and Djankov 2022c).

The use of micro data can also be warranted in calculating other components of national accounts, for example net exports and public investment. For the former, bank card data from Ukrainian refugee accounts can improve import estimates, while data from public procurement contracts can shed light on public investment (Bosio et al. 2022). These estimates are only imperfect substitutes for standard government statistics, which will hopefully return to normal production in the near future.


Alfa-Bank (2022), “Private consumption recovery slows in June”, 5 July (in Ukrainian).

Becker, T, B Eichengreen, Y Gorodnichenko, S Guriev, S Johnson, T Mylovanov, K Rogoff and B Weder di Mauro (2022), “A blueprint for restructuring Ukraine”,, 7 April.

Blinov, O and S Djankov (2022a), “Ukraine’s agricultural exports resume,”, 10 June.

Blinov, O and S Jankov (2022b), “Ukraine’s Population Depth Challenges,”, 28 June.

Blinov, O and S Djankov (2022c), “Ukraine’s recovery challenges,”, 31 May.

Bosio, E, S Djankov, E Glaeser, and A Shleifer (2022), “Public Procurement in Law and Practice.” American Economic Review 112 (4): 1091-1117.

Constantinescu M, K Kappner, N Szumilo (2022), “Estimating the short-term impact of war on economic activity in Ukraine,”, 21 June.

NBU (2022), Monthly Business Outlook Survey, National Bank of Ukraine, June.

Privatbank (2022), “Consumer market recovery: In May, Ukrainians bought more food and appliances, visited cafes and restaurants and paid with cards”, 16 June (in Ukrainian).

Leave a Reply

Your email address will not be published.