Call for applications: Klaus Liebscher ­
Economic Research Scholarship

The scholarship program targets Austrian and international experts with a proven research record in economics and finance, and postdoctoral research ­experience. Applicants need to be in active employment and should be interested in broadening their research experience and expanding their personal research ­networks. Given the OeNB’s strategic research focus on Central, Eastern and Southeastern Europe, the analysis of economic developments in this region will be a key field of research in this context.

The OeNB offers a stimulating and professional research environment in close proximity to the policymaking process. The selected scholarship recipients will be expected to collaborate with the OeNB’s research staff on a prespecified topic and are invited to participate actively in the department’s internal seminars and
other research activities. Their research output may be published in one of the ­department’s publication outlets or as an OeNB Working Paper. As a rule, the ­consultancy services under the scholarship will be provided over a period of two to three months. As far as possible, an adequate accommodation for the stay in ­Vienna will be provided. ¹

Applicants must provide the following documents and information:

• a letter of motivation, including an indication of the time period envisaged for the consultancy
• a detailed consultancy proposal
• a description of current research topics and activities
• an academic curriculum vitae
• an up-to-date list of publications (or an extract therefrom)
• the names of two references that the OeNB may contact to obtain further information about the applicant
• evidence of basic income during the term of the scholarship (employment ­contract with the applicant’s home institution)
• written confirmation by the home institution that the provision of consultancy services by the applicant is not in violation of the applicant’s employment ­contract with the home institution

Recent economic developments
and outlook

Developments in selected CESEE countries

War in Ukraine disrupts recovery from the pandemic and further heats up prices ^{2 , 3 , 4}

1 Regional overview

Economic activity was recovering from the pandemic’s disruptions when the war in Ukraine hit

Table 1: Real GDP growth  

	2019	2020	2021	Q3 20	Q4 20	Q1 21	Q2 21	Q3 21	Q4 21
	Period-on-period change in %
Slovakia	2.6	–4.4	3.0	9.1	0.4	–1.4	1.9	0.4	0.3
Slovenia	3.3	–4.2	8.1	11.8	–0.2	1.5	2.0	1.3	5.4
Bulgaria	4.0	–4.4	4.2	3.1	1.4	1.8	0.9	0.8	1.0
Croatia	3.5	–8.1	10.2	3.3	4.3	7.4	0.9	1.4	–0.1
Czechia	3.0	–5.8	3.3	6.7	0.8	–0.3	1.4	1.7	0.8
Hungary	4.6	–4.5	7.1	11.5	1.5	1.7	2.2	0.9	2.0
Poland	4.7	–2.5	5.7	7.6	–0.3	1.6	1.8	2.3	1.7
Romania	4.2	–3.8	6.0	4.8	3.9	1.9	1.6	0.4	–0.1
Turkey	0.9	1.8	11.0	16.4	1.2	2.2	1.7	2.8	1.5
Russia	2.2	–2.7	4.7	0.9	0.5	0.3	0.2	0.1	0.4
CESEE average1	2.5	–2.3	6.8	6.6	0.9	1.1	1.0	1.2	0.9
Euro area	1.6	–6.4	5.3	12.6	–0.3	–0.1	2.2	2.3	0.3
Source: Eurostat, national statistical offices.
1 Average weighted with GDP at PPP.

Growth was firmly based on domestic demand in the second half of 2021

Capital formation was more heterogeneous as financing conditions tightened, capital goods and construction inputs became more expensive and uncertainties around the outlook for the international economy increased. At the same time, stockbuilding contributed notably to growth in several countries as the completion and sale of semifinished industrial goods was still delayed by lingering supply chain issues.

Yet, the improving availability of key inputs had a positive impact on industrial sentiment and industrial activity in the final quarter of 2021 and in the first two months of 2022. This helped to sustain a robust export momentum. Dynamic ­domestic demand, however, strongly lifted imports so that the external sector on balance hardly contributed to growth.

Highest inflation in more than 15 years

By contrast, the momentum in energy price growth in early 2022 was ­contained by widespread government intervention to limit price increases for household ­energy (and in some cases also for food items). These measures ranged from ­compensatory payments to consumers and companies, to reductions in VAT rates and/or network fees, to direct interventions in the price structure (price cuts or price caps). As these interventions continue, they place an increasing burden on national budgets and the balance sheets of energy suppliers, and they could lead to a ­renewed price surge once they expire.

Going forward, another important determinant of price dynamics will be the extent to which rising inflation rates and expectations could lead to higher wage demands and thus trigger a price-wage spiral. In Turkey, for example, high inflation has already triggered a rapid increase in government payments for pensions and wages. In the other CESEE countries, several indicators also point to faster wage growth from mid-2021, and nominal wages advanced by some 8% on average in the final quarter of 2021. In general, however, the wage-setting process in ­CESEE is organized in a much more decentralized manner than e.g. in Austria, thereby reducing the bargaining power of employees. In the CESEE EU member states as a whole, about half of the working population are not subject to any ­collective bargaining agreements, and for another third, collective agreements are negotiated only at the company level (although there are sometimes marked ­differences between the individual countries). Moreover, the conflict in Ukraine has probably dented wage expectations in Central Europe, with the crisis adding to supply chain bottlenecks and disruptions to economic activity.

Most CESEE central banks have initiated a remarkable tightening cycle

CESEE currencies broadly stable despite growing interest rate differentials

The invasion of Ukraine: tectonic shift for CESEE

Wide-ranging economic sanctions against Russia

Market turbulences on the heels of the invasion did not persist

Following the stabilization, the Russian central bank again reduced the key ­interest rate to 17% in early April 2022, arguing that the balance of risks related to accelerated consumer price growth, a decline in economic activity and financial stability has shifted. It also highlighted that the latest weekly inflation data showed a deceleration of price growth and that capital control measures were containing financial stability risks.

Real economy seems to have handled the shock reasonably well – at least for the time being

The full impact of sanctions will only unfold in the weeks to come

Russian oil sales set to decline noticeably?

Yet, the sale of energy commodities might decline noticeably going forward. Most of the oil that has left Russia in recent weeks was bought and paid for before the war started. Worries about sanctions and bad publicity as well as logistical ­difficulties (as cautious banks cut credit, ship owners struggled to obtain insurance and freight costs soared) have prompted many Western buyers to pause purchases. Big players in the energy sector are boycotting Russian oil altogether (e.g. Shell and BP) and globally operating shippers are winding down operations in Russia (e.g. Maersk and MSC).

According to the commodity data provider Kpler, seaborne exports of crude oil from Russia averaged 4.6 million barrels per day in March, in line with export volumes in December, January and February. In the first week of April, volumes averaged 1.1 million barrels a day. This is well in line with estimates by the International Energy Agency that suggest that the supply of Russian oil will exceed ­demand by about 3 million barrels per day in April (other estimates are even higher, i.e. demand of 4.8 million barrels per day versus a total Russian oil supply of around 10 million barrels per day).

As these barrels fail to sell, Russia’s Urals crude is trading at an increasing ­discount: At the cut-off date, Urals crude sold for some USD 35 below Brent crude (see chart 5). The large discount is also a clear indication that non-Western ­countries are not yet prepared to up their purchases of Russian oil and that the ­decisive Western sanctioning is constraining the willingness and/or ability to ­actively profit from price dislocations. This applies to large Chinese energy ­companies, in particular. The reluctance on the part of China, however, might in part also be related to transport and technical issues: Whereas tanker shipment from Russia to Europe usually takes three or four days, to Asia it takes 40 days. Furthermore, most refineries are optimized to operate with certain types of crude and switching to a Russian type from a different type takes time and money.

CESEE economies will suffer setbacks too

Rising commodity prices are set to heat up inflation even further

The pass-through of spiraling world market prices on CESEE consumer price inflation will be enhanced, on the one hand, by the relatively large weight of ­energy and especially food items in CESEE consumption baskets and, on the other hand, by recent currency movements. The moderate upward trend of Central European currencies came to an end and spillovers from the war in Ukraine cost CESEE ­currencies quite a bit of their external value against the euro (and, even more so, against the US dollar) in the first days after the invasion. By mid-April, regional ­currencies had recovered some of their initial losses but generally failed to return to their pre-war levels. Geopolitical risk premiums could continue to weigh on currencies in the weeks to come.

Central banks step up their hiking cycles

War is a major deglobalization event

The trade-related fallout from the expected recession in Russia (and the ­associated reduction of import demand) alone, however, appears to be limited. The Vienna Institute for International Economic Studies (wiiw), for example, ­estimated that even in a scenario where Russian demand falls by 10%, imports by 30% and exports by 13%, the negative impact on GDP growth in the other CESEE countries via the trade channel would be –0.3% at most. This scenario, however, does not take into account potential adverse effects on growth and demand from rising costs of energy and raw materials, value chain disruptions or from ­disruptions in the supply of critical inputs ⁵ . Especially a stop of energy deliveries from Russia to the CESEE economies would impact negatively on economic activity in the ­region.

Dependence on Russian inputs above EU average in CESEE

Russia’s share in value-added in gross exports is even higher and ranged from 11.9% for Bulgarian exports to 1.2% for Croatian exports (EU-27: 1.4%). Again, Russia’s contribution to gross exports mostly came from the Russian mining ­industry (especially from energy-producing goods). The respective contributions of business services and manufacturing were somewhat less important (see chart 8).

Russia supplies several critical inputs

“Risky” products supplied by Russia include palladium, a precious metal that is embedded in engine exhausts to reduce emissions. Here, shortages have already led to price hikes and could quickly translate into supply chain issues especially for the automotive industry. Russia also accounts for large chunks of the EU’s total imports of nickel and aluminum. Disruptions in the trade flows in these areas could therefore severely impact the steel, manufacturing and construction ­industries. Furthermore, farmers across Europe rely heavily on imported ­fertilizers from Russia (and Belarus). Some 50% of the world’s semiconductor-grade neon, critical for the lasers used to make chips, came from two Ukrainian companies that have halted their operations. The stoppage casts a cloud over the worldwide output of microchips, already in short supply after the pandemic had driven up demand for cell phones, laptops and cars.

Economic sentiment remains remarkably stable so far

Against this backdrop – and somewhat astonishingly – economic sentiment in CESEE has so far deteriorated only moderately. The European Commission’s economic sentiment indicator (ESI) on average declined by some 3 points ­between February and March 2022, with the largest reductions being reported for Bulgaria, Czechia and Turkey (see chart 9). At the height of the first COVID-19 wave back in April 2020, the ESI had declined by full 37 points. Consumers are currently worrying the most. Consumer sentiment declined by an average 5.6 points, with Czechia and Hungary reporting declines in the double digits. In addition to general economic fears, consumers were increasingly concerned about a deterioration in their own finances, certainly reflecting the loss of purchasing power due to current and expected future ­inflation. According to survey data, the level of inflation expected by consumers spiked in March, reaching historically high levels in many cases.

Sentiment was more resilient in services and in the retail sector amid ­expectations of a stronger rebound after the ongoing easing of COVID-19 restrictions. The ­deterioration in the industrial sector was also rather contained. Purchasing ­managers’ indexes declined somewhat but remained in expansionary territory, at least in Czechia and Poland in March. The indicator on industrial sentiment within the ESI framework retreated by an average of 1.2 points, which is a moderate ­decline compared to the composite ESI. This was somewhat unexpected as the energy-intensive industrial sector suffers particularly from high electricity and gas prices and would have to bear the main economic burden of any gas rationing. Furthermore, industry surveys also reveal that supplier’s delivery times have again increased quite a bit in March (after several months of decline) pointing to ­recurring supply chain issues. This development, however, could also reflect renewed supply chain blockages due to China’s zero-COVID policy.

CESEE takes in millions of Ukrainian refugees within only a few weeks

2 Slovakia: brittle economic recovery amid supply shocks and lofty prices

Despite the pandemic, the situation on the labor market has improved since mid-2021 as a falling unemployment rate and slightly rising employment suggest. Nonetheless, the Slovak economy suffers from a skill mismatch and thus a lack of skilled labor. This does not only translate into rising wages but is reflected also in a record-high number of vacancies, on the one hand, and a persistently high long-term unemployment rate – one of the highest in the EU – on the other. Headline inflation rose sharply in the review period and came in at 8.3% in February 2022, the highest reading in more than 20 years. This towering inflation rate was broadly based, driven by soaring prices of almost all components, particularly services, industrial goods, processed food but, especially since the start of this year, also swelling energy prices. The Russian invasion of Ukraine and the ensuing sanctions as well es economic disturbances will further exacerbate inflation pressure. On the back of fiscal response measures to the coronavirus crisis worth 3.8% of GDP that had been budgeted for 2021, the general government deficit amounted to 6.2% of GDP. Consequently, public debt is projected to have gone up from 48.1% of GDP in 2019 to about 63.1% of GDP in 2021.

Table 2: Main economic indicators: Slovakia  

	2019	2020	2021	Q3 20	Q4 20	Q1 21	Q2 21	Q3 21	Q4 21
	Year-on-year change of the period total in %
GDP at constant prices	2.6	–4.4	3.0	–2.0	–1.8	0.2	9.6	1.3	1.4
Private consumption	2.7	–1.3	1.2	0.8	–2.3	–5.5	5.0	2.5	2.7
Public consumption	4.6	0.9	1.9	1.4	3.2	–1.7	8.1	–1.0	2.1
Gross fixed capital formation	6.7	–11.6	0.6	–8.1	–14.8	–9.3	5.6	–1.9	6.0
Exports of goods and services	0.8	–7.3	10.2	0.9	0.9	10.8	39.3	–3.0	1.6
Imports of goods and services	2.1	–8.2	11.2	–5.7	0.3	6.0	39.2	3.5	3.5
	Contribution to GDP growth in percentage points
Domestic demand	3.8	–5.2	3.8	–7.8	–2.3	–4.4	9.3	6.8	3.1
Net exports of goods and services	–1.2	0.9	–0.8	5.7	0.5	4.5	0.2	–5.3	–1.7
Exports of goods and services	0.8	–6.7	8.7	0.8	0.8	10.2	28.2	–2.5	1.5
Imports of goods and services	–2.0	7.6	–9.5	4.9	–0.3	–5.7	–28.0	–2.8	–3.2
	Year-on-year change of the period average in %
Unit labor costs in the whole economy (nominal, per person)	5.2	6.4	2.1	2.5	5.3	0.9	–1.3	4.4	4.6
Unit labor costs in manufacturing (nominal, per hour)	5.6	4.2	–4.2	–5.0	–5.3	–9.9	–19.0	8.7	5.9
Labor productivity in manufacturing (real, per hour)	1.3	1.2	10.0	7.6	9.5	10.5	23.3	1.6	7.0
Labor costs in manufacturing (nominal, per hour)	6.8	4.9	6.1	2.3	3.7	–0.5	–0.1	10.5	13.3
Producer price index (PPI) in industry	1.8	–0.5	6.8	–1.3	–1.0	–0.8	4.3	9.3	14.5
Consumer price index (here: HICP)	2.8	2.0	2.8	1.5	1.6	1.0	2.1	3.4	4.8
	Period average levels
Unemployment rate (ILO definition, %, 15–64 years)	5.8	6.8	6.9	7.3	7.0	7.2	7.0	6.8	6.6
Employment rate (%, 15–64 years)	68.4	67.5	69.5	67.5	67.8	67.9	68.8	70.3	70.8
Key interest rate per annum (%)	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
	Nominal year-on-year change in the period-end stock in %
Loans to the domestic nonbank private sector1	6.8	6.8	6.8	5.0	4.5	4.8	4.2	5.2	7.3
of which: loans to households	8.0	8.0	8.0	6.5	6.1	6.0	7.2	8.0	8.8
loans to nonbank corporations	4.4	4.4	4.4	2.1	1.4	2.5	–1.8	–0.2	4.3
	%
Share of foreign currency loans in total loans to the ­nonbank private sector	0.1	0.1	0.1	0.1	0.1	0.1	0.1	0.1	0.1
Return on assets (banking sector)	0.8	0.5	0.7	0.5	0.5	0.5	0.8	0.8	0.7
Tier 1 capital ratio (banking sector)	16.6	18.1	18.3	18.0	18.1	18.8	19.2	18.8	18.3
NPL ratio (banking sector)	2.8	2.3	1.9	2.5	2.3	2.2	2.1	1.9	1.9
	% of GDP
General government revenues	39.4	39.9	40.7	..	..	..	..	..	..
General government expenditures	40.7	45.3	46.8	..	..	..	..	..	..
General government balance	–1.3	–5.5	–6.2	..	..	..	..	..	..
Primary balance	–0.1	–4.2	–5.0	..	..	..	..	..	..
Gross public debt	48.1	59.7	63.1	..	..	..	..	..	..
	% of GDP
Debt of nonfinancial corporations (nonconsolidated)	53.8	54.5	53.2	..	..	..	..	..	..
Debt of households and NPISHs2 (nonconsolidated)	43.7	47.2	48.5	..	..	..	..	..	..
	% of GDP (based on EUR), period total
Goods balance	–1.2	1.1	–0.1	4.2	2.5	4.0	–0.3	–2.5	–0.9
Services balance	1.3	1.2	0.8	2.0	0.2	0.7	0.7	1.5	0.4
Primary income	–2.3	–1.2	–1.7	–1.2	–1.6	–0.4	–1.8	–1.6	–2.9
Secondary income	–1.1	–1.0	–1.0	–1.1	0.0	–1.9	–1.0	–0.8	–0.4
Current account balance	–3.4	0.1	–2.0	3.9	1.0	2.4	–2.4	–3.4	–3.8
Capital account balance	0.7	1.1	1.4	0.6	2.0	0.9	3.7	0.2	0.8
Foreign direct investment (net)3	–2.3	2.1	0.3	7.1	–0.7	3.0	–1.0	–1.3	0.7
	% of GDP (rolling four-quarter GDP, based on EUR), end of period
Gross external debt	112.7	120.5	137.0	121.0	120.5	119.0	117.6	118.6	137.0
Gross official reserves (excluding gold)	5.3	6.6	8.7	7.0	6.6	8.4	8.1	8.8	8.7
	Months of imports of goods and services
Gross official reserves (excluding gold)	0.7	0.9	1.1	1.0	0.9	1.2	1.1	1.2	1.1
	EUR million, period total
GDP at current prices	94,048	92,079	97,123	24,578	24,325	21,819	24,078	25,637	25,589
Source: Bloomberg, European Commission, Eurostat, national statistical offices, national central banks, wiiw, OeNB.
1 Foreign currency component at constant exchange rates.
2 Nonprofit institutions serving households.
3 + = net accumulation of assets larger than net accumulation of liabilities (net outflow of capital).
– = net accumulation of assets smaller than net accumulation of liabilities (net inflow of capital).

3 Slovenia: GDP rebound overshadowed by accelerating inflation

Slovenia’s budget deficit amounted to 5.2% of GDP in 2021, down from 7.8% in 2020. Fiscal developments benefited from the strong cyclical rebound. At the same time, expenditures in 2021 continued to be adversely affected by investment spending and government measures designed to mitigate the effects of the ­pandemic. For 2022, the government targets a reduction in the budget deficit as expenditures are expected to decline, while revenues will continue to benefit from economic growth. According to Slovenia’s independent fiscal council, however, planned expenditures pose a structural risk for public finances and violate fiscal rules, as they raise the concern of inefficient use of budget funds and open room for nontransparent spending in the election year 2022. Moreover, the fiscal council has warned that discretionary measures adopted during, but not related to, the ­pandemic will worsen public finances by around 2% of GDP per year in the future.

HICP inflation accelerated from around 2.1% in August 2021 to 7% by ­February 2022 and thus exceeded the euro area average from December 2021. ­Inflation excluding energy and unprocessed food prices rose from less than 1% to 4.7% during the reporting period. In order to mitigate the effect of rising energy prices on households and businesses, parliament adopted an aid package of around 0.4% of 2021 GDP. The package includes an energy voucher scheme for households and aid for most-affected businesses. In addition, network fees for electricity and excise duties on heating oil, petrol and natural gas were lowered for ­February-April 2022, while the government has also introduced a cap on retail fuel prices.

The banking sector was recently hit by negative news. In early February 2022, parliament passed a law which introduced a retroactive cap on exchange rate ­movements for CHF loans signed between June 28, 2004, and ­December 31, 2010, including those that have been paid off. According to ­estimates from the banking sector, the implementation of the law would cause them a loss of almost 60% of their combined after-tax profit in 2021. Following constitutional complaints, the constitutional court in mid-March 2022 suspended the implementation of the law until a final decision has been reached.

At the beginning of March 2022, following the ECB’s and the Single Resolution Board’s decision to close Sberbank Europe AG, Sberbank’s Slovene subsidiary was taken over by Slovenia’s biggest lender, NLB. The institution was the ninth-largest bank in Slovenia with a market share of nearly 4% in 2020 (in terms of total assets).

4 Bulgaria: economic growth disappointed in 2021, and strong energy dependence on Russia and high inflation pose challenges

Bulgaria went through a difficult political year in 2021: After two failed ­attempts to form a government, the country held its third round of general ­elections that year on November 14, 2021. A new four-party coalition government under Prime Minister Kiril Petkov was approved by parliament in ­December.

In spite of high political uncertainty and high infection numbers in fall, GDP growth recovered to 4.2% in 2021. Not only a strong second quarter, but also a dynamic second half of the year – with private consumption as the main driver – supported the recovery. However, the recovery fell somewhat short of ­expectations, with declines in investments dragging on growth.

HICP inflation started to pick up in the fourth quarter of 2021 and continued its rise in 2022: Inflation hit a new high in February with 8.4%, predominantly driven by energy prices. In order to tackle the impact of rising energy prices, a freeze on utility prices for households and compensations for businesses were ­established in end-2021. Despite the newly emerging energy crisis triggered by the war in Ukraine, only the latter had been extended beyond March. At the same time, the Bulgarian energy and water regulatory commission has recently approved increases in the price of natural gas, following sharp price increases on the world market and a restriction on withdrawing gas from the national storage facility that has been imposed to secure reserves in preparation for a worsening energy crisis.

The war in Ukraine is strongly affecting Bulgaria. With 77% of natural gas imports coming from Russia, Bulgaria is highly dependent on Russian gas. Also, the country’s sole oil refinery, which covers more than 60% of its domestic ­demand, is owned by Russia’s Lukoil. In order to improve energy security, the state-owned gas operator launched procurement procedures to expand the underground gas storage facility Chiren. This is part of the country’s energy strategy, together with the modernization of the distribution infrastructure and the ­construction of a connection to the liquified natural gas terminal in ­Alexandroupolis (Greece), in which Bulgaria holds a 20% share. The war is likely also to weigh on the upcoming summer tourism season: with a shortfall of Russian and Ukrainian tourists and the geographical proximity to the war, a recovery of tourism is likely to be further postponed.

Because of political uncertainty until end-2021, the budget for 2022 was only approved in February 2022. A budget deficit of 4.1% of GDP is expected. The ­budget foresees sizable infrastructure investments, as well as spending increases in the areas of education, health and social protection. Pensions, minimum wages as well as social benefits have been raised by the new government, but if inflation ­remains elevated these might barely translate into real gains. Adding to the stimuli created by the expansionary budget, also the access to the NextGenerationEU funds will fuel the country’s investments. On April 7, 2022, the European ­Commission endorsed Bulgaria’s national recovery and resilience plan, unlocking a total of EUR 6.3 billion in grants.

The Bulgarian government continues to plan for euro adoption by 1 January 2024.

5 Croatia: excellent recovery momentum in 2021, but murky outlook

GDP growth was supported by the fiscal and monetary policy stance. During the second half of 2021, the government largely wound down its job preservation support, but given accelerating inflation, temporary caps on fuel retail prices were implemented at the end of 2021. The government budget deficit was 4.1% of GDP in 2021, roughly as projected. Government debt declined from 87.3% of GDP to 82.3%, as GDP growth surpassed the growth of debt. Croatia’s external debt ­decreased from 79.9% of GDP at end-2020 to 78% of GDP at end-2021. During the second half of 2021, the Croatian National Bank (HNB) did not have to ­conduct foreign exchange interventions to maintain the exchange rate of the kuna vis-à-vis the euro, and international reserves continued to grow. The precautionary swap line with the ECB expired at the end of March 2022.

The Croatian banking system’s profitability improved, with return on assets increasing from 0.6% in 2020 to 1.2% in 2021. The improvement was mostly due to lower net provisions. The tier 1 capital ratio of the banking system stood at the high level of 25.1% at end-2021. The NPL ratio declined mildly throughout 2021 and was 4.3% at end-2021; however, the share of loans with elevated credit risk (IFRS 9 “Stage 2”) remained high. The European Systemic Risk Board issued a warning regarding residential real estate dynamics at the end of 2021.

Croatia has accelerated its preparations for its targeted euro adoption on ­January 1, 2023. In late March 2022, the Croatian government announced that all commitments made under the ERM II Action Plan had been completed. If Croatia will get the green light for euro adoption in 2023 will become clear in mid-2022. Croatian HICP inflation has accelerated quickly to 6.3% in ­February 2022, ­shifting the focus of attention for the Convergence Reports to inflation.

Inflation is also an issue for Croatia’s private sector, creating uncertainty and eroding spending power. In February 2022, the Croatian government passed a large package (HRK 4.8 billion; about 1.1% of GDP) with measures to mitigate the impacts of inflation. These measures include several VAT rate cuts, energy price regulations and targeted subsidies for households and sectors.

The ongoing war in Ukraine is worsening the outlook for inflation and, ­generally, economic growth. So far, an immediate policy response has been ­necessary in financial markets, with the HNB intervening twice to stabilize the kuna, providing liquidity to banks via its regular operations and resolving ­Sberbank d.d. by sale after assessment by the Single Resolution Board (SRB) that the bank was failing or likely to fail. Further policy action may become necessary, ­depending on the extent of energy price surges or even energy shortages, supply chain ­disruptions and weaker economic growth in important trading partners.

Table 5: Main economic indicators: Croatia  

	2019	2020	2021	Q3 20	Q4 20	Q1 21	Q2 21	Q3 21	Q4 21
	Year-on-year change of the period total in %
GDP at constant prices	3.5	–8.1	10.2	–10.1	–7.4	–0.6	16.4	15.1	9.7
Private consumption	4.1	–5.3	10.0	–6.5	–3.7	–0.2	17.9	15.8	7.6
Public consumption	3.3	4.1	3.1	5.6	2.9	–5.8	8.5	–4.5	14.4
Gross fixed capital formation	9.8	–6.1	7.6	–4.2	–3.8	5.0	18.1	7.6	0.8
Exports of goods and services	6.8	–22.7	33.3	–31.1	–7.2	–1.0	43.0	48.8	31.7
Imports of goods and services	6.5	–12.3	14.7	–12.0	–5.5	–0.7	32.2	13.9	16.4
	Contribution to GDP growth in percentage points
Domestic demand	3.4	–2.8	3.4	5.5	–6.9	0.8	16.6	–6.7	4.7
Net exports of goods and services	0.1	–5.3	6.8	–17.2	–0.2	0.0	0.5	20.6	4.3
Exports of goods and services	3.4	–11.5	14.0	–22.8	–3.0	–0.3	15.2	26.9	12.9
Imports of goods and services	–3.3	6.3	–7.2	5.5	2.8	0.4	–14.7	–6.3	–8.5
	Year-on-year change of the period average in %
Unit labor costs in the whole economy (nominal, per person)	..	..	..	..	..	..	..	..	..
Unit labor costs in manufacturing (nominal, per hour)	11.4	2.5	–0.7	1.1	–2.6	–3.0	–7.4	2.9	5.2
Labor productivity in manufacturing (real, per hour)	–7.2	–2.4	4.0	–1.6	2.9	4.9	9.6	2.2	–0.4
Labor costs in manufacturing (nominal, per hour)	3.6	–0.1	3.3	–0.5	0.3	1.7	1.5	5.1	4.7
Producer price index (PPI) in industry	0.8	–3.2	11.7	–4.2	–2.9	0.9	8.0	13.1	24.6
Consumer price index (here: HICP)	0.8	0.0	2.7	–0.5	–0.2	0.7	2.2	3.1	4.6
EUR per 1 HRK, + = HRK appreciation	0.0	–1.6	0.1	–1.8	–1.6	–1.1	0.7	0.4	0.6
	Period average levels
Unemployment rate (ILO definition, %, 15–64 years)	6.7	7.6	7.6	7.5	9.2	10.0	7.9	6.3	6.3
Employment rate (%, 15–64 years)	62.1	62.0	63.4	63.0	61.5	61.4	63.6	64.6	64.1
Key interest rate per annum (%)	..	..	..	..	..	..	..	..	..
HRK per 1 EUR	7.4	7.5	7.5	7.5	7.6	7.6	7.5	7.5	7.5
	Nominal year-on-year change in the period-end stock in %
Loans to the domestic nonbank private sector1	3.4	3.4	3.4	3.0	2.8	1.3	2.4	2.8	2.4
of which: loans to households	6.7	6.7	6.7	3.2	1.6	1.9	3.7	4.5	4.1
loans to nonbank corporations	–1.3	–1.3	–1.3	2.8	4.8	0.4	0.5	0.2	–0.1
	%
Share of foreign currency loans in total loans to the ­nonbank private sector	51.5	52.0	52.2	51.1	52.0	52.1	51.8	51.5	52.2
Return on assets (banking sector)	1.4	0.6	1.2	0.7	0.6	0.9	1.1	1.1	1.2
Tier 1 capital ratio (banking sector)	24.0	25.0	25.1	24.3	25.0	24.6	25.0	25.2	25.1
NPL ratio (banking sector)	5.5	5.4	4.3	5.5	5.4	5.3	5.1	4.7	4.3
	% of GDP
General government revenues	46.3	47.2	47.2	..	..	..	..	..	..
General government expenditures	46.0	54.5	51.2	..	..	..	..	..	..
General government balance	0.3	–7.4	–4.1	..	..	..	..	..	..
Primary balance	2.5	–5.4	–2.4	..	..	..	..	..	..
Gross public debt	71.1	87.3	82.3	..	..	..	..	..	..
	% of GDP
Debt of nonfinancial corporations (nonconsolidated)	84.5	93.5	83.8	..	..	..	..	..	..
Debt of households and NPISHs2 (nonconsolidated)	33.7	37.6	34.4	..	..	..	..	..	..
	% of GDP (based on EUR), period total
Goods balance	–18.8	–17.3	–18.1	–15.9	–16.8	–20.4	–18.8	–17.2	–16.6
Services balance	18.5	10.6	17.2	26.6	5.0	3.4	9.6	41.6	8.7
Primary income	–0.1	2.3	0.3	0.3	5.3	1.1	–0.2	–0.8	1.4
Secondary income	3.4	4.4	4.0	3.9	4.7	5.9	3.7	3.5	3.4
Current account balance	3.0	–0.1	3.4	14.9	–1.8	–10.0	–5.7	27.0	–3.1
Capital account balance	1.6	2.1	2.4	1.8	2.7	2.1	2.5	2.1	2.9
Foreign direct investment (net)3	–6.1	–1.3	–5.0	–1.7	0.5	–3.5	–2.7	–7.4	–5.6
	% of GDP (rolling four-quarter GDP, based on EUR), end of period
Gross external debt	72.2	79.9	78.0	79.9	79.9	86.9	84.4	80.0	78.0
Gross official reserves (excluding gold)	33.3	37.8	43.7	35.6	37.8	42.2	41.1	44.0	43.7
	Months of imports of goods and services
Gross official reserves (excluding gold)	7.8	9.3	9.8	8.8	9.3	10.3	9.7	10.2	9.8
	EUR million, period total
GDP at current prices	55,577	50,192	57,216	13,463	12,622	12,331	14,037	16,415	14,434
Source: Bloomberg, European Commission, Eurostat, national statistical offices, national central banks, wiiw, OeNB.
1 Foreign currency component at constant exchange rates.
2 Nonprofit institutions serving households.
3 + = net accumulation of assets larger than net accumulation of liabilities (net outflow of capital).
– = net accumulation of assets smaller than net accumulation of liabilities (net inflow of capital).

6 Czechia: fragile recovery set back by a new war-induced shock

A lower surplus in goods trade and a higher primary income ­deficit on the back of revived outflow of dividends, caused a current account deficit in the second half of 2021, bringing the full-year balance into slightly ­negative territory.

The general government deficit came in at 5.9% of GDP in 2021. Despite ­expenditures significantly exceeding revenues owing to high pandemic-related ­expenses, the abolition of the “super gross wage” and relatively moderate economic growth, the deficit turned out significantly smaller than had been mandated by the parliament. Public debt increased to 41.9% of GDP in 2021. Thanks to ­government support schemes, the harm of the economic downturn remained contained in the labor market. Moreover, driven by strengthening demand for labor, the ­unemployment rate declined from its pandemic peak (3.4%) in March 2021 to 2.1% by year-end and has increased only marginally since.

Inflation has sped up dramatically since late summer and reached 10% in ­February, well above the target set by the Czech National Bank (CNB) (2% ± 1 percentage point). Both consumer and producer price inflation have reached levels unseen since the beginning of transition. The mounting inflation pressure has been broadly based as persistently high core inflation has received a further boost by soaring energy and administered prices. Core inflation has echoed a substantial discrepancy between excess demand fostered by fiscal support measures, robust wage growth and, until recently, loose monetary policy, on the one hand, and rather constrained supplies, on the other. The war in Ukraine will exacerbate price pressures but the monetary policy response has been vigorous. Since the CNB started its tightening cycle last summer it has continued its hawkish stance: In only seven steps – some of them unprecedentedly large – it raised the key interest rate from 0.25% in June 2021 to 5% effective from April 1, 2022.

Table 6: Main economic indicators: Czechia  

	2019	2020	2021	Q3 20	Q4 20	Q1 21	Q2 21	Q3 21	Q4 21
	Year-on-year change of the period total in %
GDP at constant prices	3.0	–5.8	3.3	–5.7	–5.3	–2.6	9.3	3.2	3.7
Private consumption	2.7	–6.8	4.4	–5.1	–9.8	–6.2	8.7	5.9	9.4
Public consumption	2.5	3.4	3.0	0.4	7.0	0.4	1.7	7.2	2.6
Gross fixed capital formation	5.9	–7.5	0.6	–9.0	–11.3	–4.1	4.5	0.8	0.9
Exports of goods and services	1.5	–6.9	5.1	–4.6	3.5	2.6	32.2	–3.2	–5.2
Imports of goods and services	1.5	–6.9	11.5	–6.5	–0.7	4.1	33.4	8.2	4.5
	Contribution to GDP growth in percentage points
Domestic demand	3.0	–5.3	7.1	–6.7	–8.4	–1.8	8.8	10.5	10.7
Net exports of goods and services	0.0	–0.5	–3.8	1.0	3.0	–0.8	0.6	–7.3	–7.1
Exports of goods and services	1.1	–5.1	3.6	–3.2	2.6	1.9	19.9	–2.2	–4.0
Imports of goods and services	–1.1	4.7	–7.4	4.2	0.5	–2.7	–19.4	–5.1	–3.1
	Year-on-year change of the period average in %
Unit labor costs in the whole economy (nominal, per person)	4.3	7.7	2.3	6.4	10.2	3.4	1.5	4.2	0.3
Unit labor costs in manufacturing (nominal, per hour)	9.1	2.3	–2.5	–1.7	–5.4	–5.2	–14.6	5.5	5.8
Labor productivity in manufacturing (real, per hour)	–0.8	2.7	6.7	3.8	11.5	8.1	18.4	1.5	0.9
Labor costs in manufacturing (nominal, per hour)	8.2	4.7	4.5	2.0	5.4	2.5	1.1	7.1	6.8
Producer price index (PPI) in industry	1.7	0.6	6.2	0.1	1.1	2.3	3.3	8.1	11.0
Consumer price index (here: HICP)	2.6	3.3	3.3	3.5	2.7	2.2	2.8	3.3	5.0
EUR per 1 CZK, + = CZK appreciation	–0.1	–3.0	3.2	–2.8	–4.1	–1.7	5.6	3.8	5.1
	Period average levels
Unemployment rate (ILO definition, %, 15–64 years)	2.1	2.6	2.9	2.9	3.1	3.4	3.1	2.8	2.3
Employment rate (%, 15–64 years)	75.1	74.4	74.4	74.4	74.3	73.6	73.7	75.0	75.3
Key interest rate per annum (%)	1.9	0.8	0.9	0.3	0.3	0.3	0.3	0.7	2.4
CZK per 1 EUR	25.7	26.5	25.6	26.5	26.7	26.1	25.6	25.5	25.4
	Nominal year-on-year change in the period-end stock in %
Loans to the domestic nonbank private sector1	5.0	5.0	5.0	3.5	3.0	2.6	4.4	6.3	9.7
of which: loans to households	6.1	6.1	6.1	6.3	6.5	6.9	8.1	9.1	9.9
loans to nonbank corporations	3.8	3.8	3.8	0.1	–1.3	–2.7	–0.3	2.8	9.4
	%
Share of foreign currency loans in total loans to the nonbank private sector	14.5	14.6	14.6	16.1	14.6	14.8	13.5	14.1	14.6
Return on assets (banking sector)	1.2	0.6	0.8	0.6	0.6	0.5	0.7	0.8	0.8
Tier 1 capital ratio (banking sector)	20.8	23.6	22.8	22.6	23.6	23.4	23.9	23.2	22.8
NPL ratio (banking sector)	2.4	2.6	2.3	2.2	2.6	2.6	2.6	2.5	2.3
	% of GDP
General government revenues	41.4	41.6	40.5	..	..	..	..	..	..
General government expenditures	41.1	47.3	46.4	..	..	..	..	..	..
General government balance	0.3	–5.8	–5.9	..	..	..	..	..	..
Primary balance	1.0	–4.9	–5.2	..	..	..	..	..	..
Gross public debt	30.1	37.7	41.9	..	..	..	..	..	..
	% of GDP
Debt of nonfinancial corporations (nonconsolidated)	55.1	56.4	53.2	..	..	..	..	..	..
Debt of households and NPISHs2 (nonconsolidated)	31.7	34.1	35.6	..	..	..	..	..	..
	% of GDP (based on EUR), period total
Goods balance	4.1	5.0	1.2	5.4	7.3	6.6	2.7	–2.0	–1.7
Services balance	1.8	1.8	1.8	2.1	0.6	1.7	2.1	1.9	1.6
Primary income	–5.0	–2.7	–3.3	–1.0	–6.7	–1.9	–3.6	–4.7	–2.9
Secondary income	–0.6	–0.5	–0.5	–0.7	–0.1	–1.6	0.2	–0.5	–0.2
Current account balance	0.3	3.6	–0.9	5.8	1.2	4.8	1.3	–5.3	–3.3
Capital account balance	0.4	1.3	1.6	1.3	0.8	–0.1	1.6	2.4	2.1
Foreign direct investment (net)3	–2.4	–1.3	–0.1	1.5	–4.4	2.4	–2.1	–0.7	0.3
	% of GDP (rolling four-quarter GDP, based on EUR), end of period
Gross external debt	76.5	76.1	75.3	73.6	76.1	76.7	73.7	73.8	75.3
Gross official reserves (excluding gold)	59.0	62.7	64.0	61.5	62.7	64.7	62.2	62.8	64.0
	Months of imports of goods and services
Gross official reserves (excluding gold)	10.4	11.7	11.0	11.5	11.7	12.0	11.1	11.0	11.0
	EUR million, period total
GDP at current prices	225,579	215,272	238,824	55,130	56,823	53,249	60,103	61,752	63,721
Source: Bloomberg, European Commission, Eurostat, national statistical offices, national central banks, wiiw, OeNB.
1 Foreign currency component at constant exchange rates.
2 Nonprofit institutions serving households.
3 + = net accumulation of assets larger than net accumulation of liabilities (net outflow of capital).
– = net accumulation of assets smaller than net accumulation of liabilities (net inflow of capital).

7 Hungary: economy bounced back in 2021 but inflation is soaring

Hungary’s budget deficit amounted to 6.8% of GDP in 2021, down from 7.8% in 2020. The deficit was particularly driven up during the last quarter of 2021 as, for example, extra pension payments were paid out. In mid-December 2021, the government decided to indefinitely postpone investment projects worth nearly 2% of GDP to increase fiscal reserves and lower the 2022 budget deficit target from 5.9% to 4.9% of GDP. At the same time, the 2022 budget foresees substantial ­expenditure increases and tax cuts, while rising energy prices (with household ­energy price caps having been in place for years), expected losses of the central bank (which need to be covered by the budget) and higher debt-servicing costs will add to the expected budget deficit. In mid-February 2022, the European Court of Justice confirmed the validity of the “rule of law” mechanism for the suspension of EU funds. Since no progress on this issue seems to have been achieved in ­negotiations with the EU Commission, the approval of funds to Hungary under the Recovery and Resilience Facility (RRF) is still pending.

Inflation kept accelerating and reached 8.4% in February 2022. Inflation excluding energy and unprocessed food prices also ­accelerated from 4.1% in August 2021 to 7.8% in February 2022. In response, the Hungarian central bank (MNB) continued its rate hiking cycle into 2022. Between September 2021 and March 2022, it raised the base rate in monthly steps from 1.65% to 4.4% to anchor medium-­term inflation expectations. To ­stabilize short-term financial market ­developments, starting in late November 2021, it decoupled the interest rate on its one-week deposit facility from the base rate. This “operative” policy rate reached 6.15% in early April 2022 in reaction to the forint temporarily falling to historic lows as a result of the war in Ukraine. In order to additionally tighten monetary conditions, MNB also gradually phased out its quantitative easing programs.

In late December 2021, the government fixed the interest rate on households’ mortgage loans at their end-October 2021 level for the first half of 2022. The ­measure will cost banks an estimated 3.7% of 2021 consolidated after-tax profit and may limit the contractionary effect of the central bank’s rate hikes. On the other hand, the government has introduced temporary price caps on selected fuel and basic food prices, which – together with the long-standing fixation of ­household energy prices – MNB estimates to have taken around 4.2 percentage points off the headline inflation rate in March 2022.

Following the ECB’s and the SRB’s decision to shut down Sberbank Europe AG, MNB revoked the operating license of the Hungarian Sberbank subsidiary, ordered its winding-up and provided the deposit insurance fund with a short-term bridge loan and a repo deal. According to data at end-2020, Sberbank accounted for around 1% of total assets and 1.2% of total deposits of the banking sector.

Table 7: Main economic indicators: Hungary  

	2019	2020	2021	Q3 20	Q4 20	Q1 21	Q2 21	Q3 21	Q4 21
	Year-on-year change of the period total in %
GDP at constant prices	4.6	–4.5	7.1	–4.1	–2.7	–1.9	17.8	6.2	7.1
Private consumption	5.0	–1.2	4.6	–1.5	–2.5	–5.5	9.5	6.5	8.0
Public consumption	4.3	–0.9	3.7	–0.3	–1.4	7.5	3.0	3.6	1.1
Gross fixed capital formation	12.8	–7.0	5.9	–13.4	1.6	–3.0	9.3	11.5	3.2
Exports of goods and services	5.4	–6.1	10.3	–3.9	3.4	5.6	36.1	2.8	2.6
Imports of goods and services	8.2	–4.0	8.7	–4.8	2.3	2.5	26.7	7.0	2.0
	Contribution to GDP growth in percentage points
Domestic demand	6.5	–2.6	5.7	–4.8	–3.7	–4.8	11.6	9.2	6.6
Net exports of goods and services	–2.0	–1.8	1.4	0.7	0.9	2.7	6.3	–3.0	0.5
Exports of goods and services	4.5	–5.0	8.1	–3.1	2.6	4.8	25.7	2.2	2.1
Imports of goods and services	–6.5	3.1	–6.7	3.8	–1.7	–2.0	–19.4	–5.3	–1.6
	Year-on-year change of the period average in %
Unit labor costs in the whole economy (nominal, per person)	3.4	6.7	4.0	5.8	5.7	9.2	–2.5	4.5	5.2
Unit labor costs in manufacturing (nominal, per hour)	7.8	7.6	–0.2	2.7	–0.1	–0.6	–15.5	7.5	10.6
Labor productivity in manufacturing (real, per hour)	4.3	–0.2	5.9	2.4	6.1	4.7	20.8	0.4	0.4
Labor costs in manufacturing (nominal, per hour)	12.5	6.8	6.4	5.2	6.0	4.0	2.1	8.0	11.1
Producer price index (PPI) in industry	2.2	4.3	13.5	4.0	6.1	8.0	10.9	14.4	20.7
Consumer price index (here: HICP)	3.4	3.4	5.2	3.8	2.9	3.3	5.3	5.0	7.1
EUR per 1 HUF, + = HUF appreciation	–2.0	–7.4	–2.0	–7.2	–7.9	–6.1	–0.8	–0.1	–1.0
	Period average levels
Unemployment rate (ILO definition, %, 15–64 years)	3.5	4.3	4.1	4.5	4.2	4.5	4.1	3.9	3.7
Employment rate (%, 15–64 years)	70.1	69.7	73.1	70.2	70.2	71.8	72.8	73.6	74.1
Key interest rate per annum (%)	0.9	0.8	1.1	0.6	0.6	0.6	0.6	1.3	2.0
HUF per 1 EUR	325.2	351.2	358.5	353.6	360.5	361.0	354.7	353.9	364.3
	Nominal year-on-year change in the period-end stock in %
Loans to the domestic nonbank private sector1	12.5	12.5	12.5	10.3	11.0	8.7	10.5	11.6	12.0
of which: loans to households	15.5	15.5	15.5	14.9	14.1	13.4	15.5	16.1	15.0
loans to nonbank corporations	10.4	10.4	10.4	7.1	8.8	5.4	6.8	8.3	9.8
	%
Share of foreign currency loans in total loans to the nonbank private sector	23.8	22.3	20.3	23.4	22.3	21.9	20.0	20.3	20.3
Return on assets (banking sector)	1.2	0.4	1.0	0.5	0.4	1.1	1.3	1.2	1.0
Tier 1 capital ratio (banking sector)	16.4	17.4	17.1	15.8	17.4	17.3	17.2	16.6	17.1
NPL ratio (banking sector)	2.6	2.4	1.7	2.8	2.4	2.3	2.2	1.8	1.7
	% of GDP
General government revenues	43.9	43.4	41.1	..	..	..	..	..	..
General government expenditures	46.0	51.2	47.9	..	..	..	..	..	..
General government balance	–2.1	–7.8	–6.8	..	..	..	..	..	..
Primary balance	0.1	–5.5	–4.5	..	..	..	..	..	..
Gross public debt	65.5	79.6	76.8	..	..	..	..	..	..
	% of GDP
Debt of nonfinancial corporations (nonconsolidated)	62.7	67.6	72.2	..	..	..	..	..	..
Debt of households and NPISHs2 (nonconsolidated)	18.2	20.0	20.1	..	..	..	..	..	..
	% of GDP (based on EUR), period total
Goods balance	–2.5	–0.9	–2.5	–0.2	0.0	3.7	–1.3	–6.0	–4.9
Services balance	4.9	3.0	3.2	4.1	2.2	2.1	3.6	4.2	2.9
Primary income	–2.5	–2.6	–3.2	–2.8	–2.5	–4.0	–2.9	–3.3	–2.9
Secondary income	–0.5	–0.6	–0.6	–0.2	–0.3	–0.8	–0.7	–0.3	–0.7
Current account balance	–0.7	–1.1	–3.1	0.9	–0.5	1.2	–1.3	–5.4	–5.7
Capital account balance	1.8	2.0	2.5	2.1	2.1	2.0	1.6	1.8	4.3
Foreign direct investment (net)3	–0.7	–1.7	–1.4	–1.7	0.5	–0.1	0.3	–1.8	–3.5
	% of GDP (rolling four-quarter GDP, based on EUR), end of period
Gross external debt	72.6	80.1	82.3	80.1	80.1	85.9	83.2	85.5	82.3
Gross official reserves (excluding gold)	18.5	23.4	21.7	22.0	23.4	20.2	18.2	22.6	21.7
	Months of imports of goods and services
Gross official reserves (excluding gold)	2.8	3.6	3.2	3.4	3.6	3.1	2.8	3.4	3.2
	EUR million, period total
GDP at current prices	145,983	137,300	154,121	35,099	38,334	32,121	38,838	39,813	43,349
Source: Bloomberg, European Commission, Eurostat, national statistical offices, national central banks, wiiw, OeNB.
1 Foreign currency component at constant exchange rates.
2 Nonprofit institutions serving households.
3 + = net accumulation of assets larger than net accumulation of liabilities (net outflow of capital).
– = net accumulation of assets smaller than net accumulation of liabilities (net inflow of capital).

8 Poland: further rate hikes to ensure credibility and avoid second-round effects

In manufacturing, nominal unit labor cost (ULC) in Poland was unchanged in the second half of 2021 against the previous year, while nominal ULC was slightly higher in the euro area. The exchange rate further enhanced Poland’s external competitiveness, as the złoty’s value in euro was lower than a year earlier in that period. Thereafter, the złoty depreciated against the euro by about 4.5% in March in the wake of Russia’s invasion of Ukraine.

According to HICP (and national CPI) definition, annual headline inflation rose from 4.1% (4.4%) in June to 8.0% (8.6%) in December and further to 10.2% (11.0%) in March. In parallel, core inflation increased from 3.1% (3.5%) in June to 7.5% (6.9%) in March, also markedly accelerating in month-on-month terms in early 2022. Services continued to be the main inflation driver within core ­inflation. The Polish central bank’s Monetary Policy Council (MPC), pursuing a CPI ­inflation target of 2.5% ± 1 percentage point, hiked its main policy rate in monthly steps from October to April, bringing it to 4.5%, having kept it at 0.1% during the pandemic. In mid-November 2021, the MPC conducted its last outright purchases of government(-guaranteed) debt securities in the secondary market for the time being. In April 2022, the MPC stated that its recent rate hike served to reduce the persisting risk of inflation running above the target in the medium term and to curb inflation expectations. Further decisions would depend on incoming ­information, including the impact of the Russian military aggression against Ukraine. The MPC will take all necessary actions to ensure macroeconomic and financial stability and it may intervene in the foreign exchange market, in ­particular to limit fluctuations of the złoty inconsistent with the direction of monetary ­policy.

Regarding fiscal policy, the general government deficit declined from 6.9% of GDP in 2020 to 1.9% in 2021, benefiting from higher nominal growth. General government debt declined from 57.1% of GDP in 2020 to 53.8% of GDP in 2021.

Table 8: Main economic indicators: Poland  

	2019	2020	2021	Q3 20	Q4 20	Q1 21	Q2 21	Q3 21	Q4 21
	Year-on-year change of the period total in %
GDP at constant prices	4.7	–2.5	5.7	–1.7	–2.6	–1.2	10.8	5.6	7.7
Private consumption	3.9	–2.9	6.0	0.3	–3.0	–0.2	13.2	4.6	7.7
Public consumption	6.5	4.9	1.1	3.6	8.2	1.1	2.7	1.4	–0.3
Gross fixed capital formation	6.1	–9.0	7.9	–7.0	–15.3	1.3	5.2	9.1	12.3
Exports of goods and services	5.2	0.1	12.0	2.2	8.2	6.8	29.0	8.6	6.7
Imports of goods and services	3.0	–1.2	17.4	0.7	8.5	10.8	34.4	14.9	13.1
	Contribution to GDP growth in percentage points
Domestic demand	3.5	–3.2	7.6	–2.6	–2.8	0.3	11.3	8.3	10.4
Net exports of goods and services	1.3	0.7	–1.9	0.8	0.3	–1.6	–0.3	–2.7	–2.7
Exports of goods and services	2.9	0.1	6.7	1.2	4.2	4.0	15.3	4.9	3.8
Imports of goods and services	–1.6	0.6	–8.6	–0.4	–3.9	–5.5	–15.6	–7.6	–6.5
	Year-on-year change of the period average in %
Unit labor costs in the whole economy (nominal, per person)	2.4	6.3	1.9	4.3	7.1	7.4	–0.9	1.5	–0.2
Unit labor costs in manufacturing (nominal, per hour)	4.2	4.9	–4.6	–1.1	–0.8	–4.1	–13.0	0.6	–0.7
Labor productivity in manufacturing (real, per hour)	2.4	1.6	12.9	4.9	7.2	10.2	23.0	9.2	10.2
Labor costs in manufacturing (nominal, per hour)	6.7	6.2	8.0	3.7	6.4	5.7	7.1	9.8	9.4
Producer price index (PPI) in industry	1.4	–0.5	8.1	–1.1	–0.1	2.6	6.6	9.6	13.6
Consumer price index (here: HICP)	2.1	3.7	5.2	3.7	3.6	3.9	4.6	5.1	7.3
EUR per 1 PLN, + = PLN appreciation	–0.9	–3.3	–2.6	–2.7	–4.9	–4.9	–0.6	–2.8	–2.4
	Period average levels
Unemployment rate (ILO definition, %, 15–64 years)	3.4	3.2	3.4	3.3	3.2	4.1	3.6	3.1	2.9
Employment rate (%, 15–64 years)	68.2	68.7	70.3	69.0	69.4	69.2	70.0	71.0	71.0
Key interest rate per annum (%)	1.5	0.5	0.3	0.1	0.1	0.1	0.1	0.1	1.1
PLN per 1 EUR	4.3	4.4	4.6	4.4	4.5	4.5	4.5	4.6	4.6
	Nominal year-on-year change in the period-end stock in %
Loans to the domestic nonbank private sector1	5.0	5.0	5.0	–0.8	–1.2	–2.2	0.3	2.6	5.1
of which: loans to households	5.6	5.6	5.6	2.1	1.6	1.3	3.0	4.0	4.3
loans to nonbank corporations	4.1	4.1	4.1	–5.6	–6.0	–8.0	–4.4	–0.1	6.5
	%
Share of foreign currency loans in total loans to the nonbank private sector	19.2	19.6	17.5	19.6	19.6	19.3	18.1	18.0	17.5
Return on assets (banking sector)	0.7	0.0	0.4	0.4	0.0	0.4	0.5	0.5	0.4
Tier 1 capital ratio (banking sector)	17.0	18.5	17.3	18.5	18.5	18.6	18.5	18.0	17.3
NPL ratio (banking sector)	6.6	7.0	5.7	7.0	7.0	6.7	6.5	6.3	5.7
	% of GDP
General government revenues	41.0	41.3	42.3	..	..	..	..	..	..
General government expenditures	41.8	48.2	44.2	..	..	..	..	..	..
General government balance	–0.7	–6.9	–1.9	..	..	..	..	..	..
Primary balance	0.6	–5.6	–0.8	..	..	..	..	..	..
Gross public debt	45.6	57.1	53.8	..	..	..	..	..	..
	% of GDP
Debt of nonfinancial corporations (nonconsolidated)	44.5	44.7	43.3	..	..	..	..	..	..
Debt of households and NPISHs2 (nonconsolidated)	34.7	33.9	32.4	..	..	..	..	..	..
	% of GDP (based on EUR), period total
Goods balance	0.3	2.4	–0.1	2.4	3.3	2.1	1.6	–1.5	–2.0
Services balance	4.5	4.4	4.6	4.3	4.0	4.9	5.0	4.6	4.1
Primary income	–4.0	–3.5	–4.5	–4.6	–4.1	–3.5	–5.1	–5.3	–3.9
Secondary income	–0.3	–0.3	–0.7	–0.2	–0.7	–1.0	–0.5	–0.2	–1.0
Current account balance	0.5	2.9	–0.6	1.9	2.5	2.5	1.0	–2.4	–2.9
Capital account balance	2.0	2.3	1.6	1.4	3.3	0.6	2.2	2.0	1.6
Foreign direct investment (net)3	–1.9	–2.1	–3.7	–1.7	–0.6	–6.0	–1.8	–5.4	–1.8
	% of GDP (rolling four-quarter GDP, based on EUR), end of period
Gross external debt	59.2	58.4	56.2	57.2	58.4	59.0	57.2	57.4	56.2
Gross official reserves (excluding gold)	19.6	21.8	23.6	20.3	21.8	23.7	22.7	24.1	23.6
	Months of imports of goods and services
Gross official reserves (excluding gold)	4.6	5.3	5.0	5.0	5.3	5.7	5.1	5.3	5.0
	EUR million, period total
GDP at current prices	533,674	523,576	569,951	131,573	144,956	129,210	136,237	141,798	162,706
Source: Bloomberg, European Commission, Eurostat, national statistical offices, national central banks, wiiw, OeNB.
1 Foreign currency component at constant exchange rates.
2 Nonprofit institutions serving households.
3 + = net accumulation of assets larger than net accumulation of liabilities (net outflow of capital).
– = net accumulation of assets smaller than net accumulation of liabilities (net inflow of capital).

9 Romania: growth weakens as inflation rises markedly, twin deficits persist

After showing strong growth in the third quarter, agricultural production ­contracted in the final quarter of 2021. Moreover, the growth contribution coming from changes in inventories turned negative in the final quarter. Private consumption, the main growth pillar since mid-2020, remained vivid in the third quarter, largely due to the release of pent-up demand and self-consumption of agricultural products. However, as households’ purchasing power was more and more eroded by rising inflation and as some COVID-19 restrictions were tightened again, ­private consumption growth lost momentum in quarter-on-quarter terms in the final quarter. Even though domestic credit growth accelerated in the second half of 2021 on the back of state guarantee programs, gross fixed capital formation ­contracted in the same period (in terms of both year-on-year and quarter-quarter growth). Though the export-oriented automotive sector was still constrained by supply chain bottlenecks, net exports delivered a slightly smaller negative contribution to growth in the second half of the year.

Nevertheless, the current account deficit did not improve and hence ended up at 7% of GDP for the full-year 2021. Compared to 2020, also the net borrowing position from the current and capital account deteriorated markedly, though the capital account surplus rose slightly. Net FDI inflows increased somewhat in 2021 and accounted for 63% of this position (compared to 43% in 2020). In contrast to the first half of 2021, unit labor costs in the manufacturing sector rose considerably in the second half of the year. At the same time, the Romanian leu weakened only slightly vis-à-vis the euro.

The general government budget deficit amounted to 6.7% of GDP in cash terms in 2021 (i.e. 7.1% of GDP in ESA terms). The general government budget plan for 2022 was based on a GDP forecast of 4.6% and envisages a deficit of 5.8% of GDP in cash terms (i.e. 6.2% of GDP in ESA terms). Public wages and compensations were frozen at the end-2021 level for all state employees except for those working in social assistance, healthcare and education. Within the framework of the excessive deficit procedure, Romania should put an end to the excessive deficit situation by 2024 at the latest and gradually reduce its deficit until then. It is worth noting that the European Commission disbursed EUR 1.8 billion to Romania in pre-financing at end-2021 (equivalent to 13% of the country’s grant allocation ­under the Recovery and Resilience Facility). Against the background of rising ­energy prices, the government introduced a support scheme (including price caps for electricity and natural gas as well as subsidies) mainly for households and small companies in November 2021. The scheme was modified earlier this year and ­extended until end-March 2023.

Inflation moved further away from the National Bank of Romania’s (NBR) ­upper bound of the inflation target variation band of 2.5% ± 1 percentage point. Headline consumer price inflation reached 7.9% in February, while core inflation went up to 5.7%. From November 2021, the NBR hiked its key policy rate to 3% − in four steps by a total of 150 basis points.

Table 9: Main economic indicators: Romania  

	2019	2020	2021	Q3 20	Q4 20	Q1 21	Q2 21	Q3 21	Q4 21
	Year-on-year change of the period total in %
GDP at constant prices	4.2	–3.8	6.0	–5.4	–1.5	–0.1	15.4	6.9	2.4
Private consumption	3.8	–4.8	7.6	–4.2	–6.1	0.9	11.7	9.0	9.4
Public consumption	8.0	1.5	5.3	–2.6	1.2	–4.3	2.1	–2.4	12.5
Gross fixed capital formation	12.7	4.4	2.5	3.2	3.5	11.3	12.9	–1.3	–6.0
Exports of goods and services	4.6	–9.3	12.8	–5.0	–1.5	1.0	41.7	7.2	7.8
Imports of goods and services	8.8	–5.9	15.0	–3.8	1.7	3.1	42.0	11.2	8.2
	Contribution to GDP growth in percentage points
Domestic demand	5.8	–2.3	7.3	–4.7	0.3	3.0	15.1	8.7	3.3
Net exports of goods and services	–1.6	–1.5	–1.4	–1.0	–1.7	–2.2	–2.2	–1.4	–0.4
Exports of goods and services	2.2	–3.8	4.7	–2.5	–1.1	0.3	13.2	2.9	3.0
Imports of goods and services	–3.9	2.3	–6.1	1.5	–0.6	–2.5	–15.3	–4.3	–3.4
	Year-on-year change of the period average in %
Unit labor costs in the whole economy (nominal, per person)	6.5	4.5	–9.1	7.1	3.3	–5.7	–13.7	–10.6	–6.2
Unit labor costs in manufacturing (nominal, per hour)	13.3	7.6	4.2	3.2	1.8	0.5	–4.9	9.3	12.6
Labor productivity in manufacturing (real, per hour)	–0.8	0.4	3.1	1.8	6.8	5.4	11.6	0.3	–4.0
Labor costs in manufacturing (nominal, per hour)	12.5	8.0	7.5	5.1	8.7	5.9	6.1	9.6	8.1
Producer price index (PPI) in industry	4.0	0.0	14.9	–0.8	–0.5	2.3	10.1	16.4	30.8
Consumer price index (here: HICP)	3.9	2.3	4.1	2.4	1.8	2.3	3.1	4.3	6.6
EUR per 1 RON, + = RON appreciation	–1.9	–1.9	–1.7	–2.3	–2.1	–1.7	–1.7	–1.8	–1.6
	Period average levels
Unemployment rate (ILO definition, %, 15–64 years)	4.0	5.2	5.6	5.4	5.4	6.1	5.1	5.3	5.9
Employment rate (%, 15–64 years)	65.8	65.6	61.9	66.0	65.8	60.8	62.4	62.3	62.1
Key interest rate per annum (%)	2.5	1.9	1.4	1.6	1.5	1.3	1.3	1.3	1.6
RON per 1 EUR	4.7	4.8	4.9	4.8	4.9	4.9	4.9	4.9	4.9
	Nominal year-on-year change in the period-end stock in %
Loans to the domestic nonbank private sector1	5.5	5.5	5.5	3.2	4.8	6.2	10.6	12.7	14.2
of which: loans to households	6.7	6.7	6.7	4.8	4.2	4.6	7.5	8.8	9.3
loans to nonbank corporations	4.2	4.2	4.2	1.4	5.5	7.9	14.3	17.3	19.8
	%
Share of foreign currency loans in total loans to the nonbank private sector	32.4	30.5	27.6	31.4	30.5	29.9	28.9	28.4	27.6
Return on assets (banking sector)	1.4	1.0	1.4	1.2	1.0	1.3	1.4	1.5	1.4
Tier 1 capital ratio (banking sector)	20.1	23.2	19.8	20.8	23.2	22.7	22.1	21.4	19.8
NPL ratio (banking sector)	4.1	3.8	3.4	4.1	3.8	3.9	3.8	3.7	3.4
	% of GDP
General government revenues	31.9	32.7	32.8	..	..	..	..	..	..
General government expenditures	36.2	42.0	39.9	..	..	..	..	..	..
General government balance	–4.3	–9.3	–7.1	..	..	..	..	..	..
Primary balance	–3.2	–7.9	–5.7	..	..	..	..	..	..
Gross public debt	35.3	47.2	48.8	..	..	..	..	..	..
	% of GDP
Debt of nonfinancial corporations (nonconsolidated)	32.2	33.2	32.4	..	..	..	..	..	..
Debt of households and NPISHs2 (nonconsolidated)	15.3	16.1	15.7	..	..	..	..	..	..
	% of GDP (based on EUR), period total
Goods balance	–8.0	–8.7	–9.6	–7.9	–8.0	–10.9	–9.3	–9.5	–9.3
Services balance	3.9	4.3	4.0	4.2	3.8	4.6	3.8	3.6	4.1
Primary income	–1.4	–1.6	–1.7	–3.5	–1.5	0.3	–3.1	–2.8	–1.1
Secondary income	0.7	0.9	0.4	0.8	1.1	0.1	0.6	0.8	0.2
Current account balance	–4.9	–5.0	–7.0	–6.4	–4.7	–5.9	–8.0	–7.8	–6.2
Capital account balance	1.3	1.9	2.2	1.0	2.4	1.1	1.4	1.5	4.2
Foreign direct investment (net)3	–2.2	–1.3	–3.0	–1.1	–1.9	–4.5	–2.8	–4.2	–1.3
	% of GDP (rolling four-quarter GDP, based on EUR), end of period
Gross external debt	49.1	58.0	56.1	54.1	58.0	56.2	56.3	56.9	56.1
Gross official reserves (excluding gold)	14.7	17.1	16.9	14.9	17.1	16.2	16.1	17.5	16.9
	Months of imports of goods and services
Gross official reserves (excluding gold)	4.0	4.9	4.3	4.3	4.9	4.7	4.4	4.7	4.3
	EUR million, period total
GDP at current prices	223,085	218,706	239,991	58,911	67,395	46,743	55,871	65,172	72,205
Source: Bloomberg, European Commission, Eurostat, national statistical offices, national central banks, wiiw, OeNB.
1 Foreign currency component at constant exchange rates.
2 Nonprofit institutions serving households.
3 + = net accumulation of assets larger than net accumulation of liabilities (net outflow of capital).
– = net accumulation of assets smaller than net accumulation of liabilities (net inflow of capital).

10 Turkey: unorthodox policies show strong impact on exchange rate and inflation

Official foreign currency reserves declined by 7% in EUR terms and by 12% in USD terms from end-August 2021 to end-February 2022 when they covered 2.7 months of imports and exceeded the scheduled off-balance and public sector’s on-balance net drains due in three months only moderately. These off-balance sheet liabilities consist primarily in foreign currency swaps that have been incurred to a large extent vis-à-vis domestic commercial banks. The latter have been ­prohibited to enter swaps abroad for closing their open foreign currency position since 2018. Annual headline inflation accelerated moderately until November. ­After the Turkish central bank cut the key interest rate by 1 percentage point to 18% in September, further decreases amounted to 2 percentage points in October and 1 percentage point each in November and December so that the key rate stood at 14% during the first quarter of 2022, implying a large, negative real key rate. Reserve requirements, however, were tightened in early November 2021. In ­response to this unorthodox interest rate policy and accompanying statements by the Turkish president, the lira depreciated sharply from mid-November to mid-­December by about 43% in EUR terms. This led to an immediate jump in ­inflation, resulting in annual figures of 61.1% (headline) and 51.8% (core) in March 2022. Authorities reacted by introducing measures, like budget-financed exchange rate-linked deposit schemes, to boost “liraization” and decrease both nonmonetary gold deposits and foreign exchange deposits and, hence, central bank’s contingent ­foreign exchange liabilities. In addition, since January 1, exporters have been obliged to convert 25% of their foreign exchange earnings into lira. So far, these policies have been partially successful, causing the lira to re-appreciate somewhat and stabilize at about lira 16 per euro.

Table 10: Main economic indicators: Turkey  

	2019	2020	2021	Q3 20	Q4 20	Q1 21	Q2 21	Q3 21	Q4 21
	Year-on-year change of the period total in %
GDP at constant prices	0.9	1.8	11.0	6.3	6.2	7.3	21.9	7.5	9.1
Private consumption	1.5	3.2	15.1	8.5	7.9	7.0	23.3	9.1	21.3
Public consumption	4.1	2.2	2.1	2.0	3.7	–0.1	3.2	7.9	–1.9
Gross fixed capital formation	–12.4	7.2	6.4	22.6	11.7	12.4	20.8	–1.9	–0.8
Exports of goods and services	4.6	–14.8	24.9	–21.4	0.5	3.9	60.9	25.5	20.7
Imports of goods and services	–5.4	7.6	2.0	16.4	3.0	–1.0	19.9	–8.9	2.6
	Contribution to GDP growth in percentage points
Domestic demand	–2.1	4.0	10.9	10.4	8.1	7.2	19.9	6.0	11.9
Net exports of goods and services	2.5	–5.7	5.1	–9.6	–0.7	1.2	7.1	7.4	4.3
Exports of goods and services	1.2	–4.0	5.6	–5.9	0.1	1.0	11.5	5.3	5.0
Imports of goods and services	1.3	–1.8	–0.5	–3.7	–0.8	0.2	–4.4	2.2	–0.7
	Year-on-year change of the period average in %
Unit labor costs in the whole economy (nominal, per person)	..	..	..	..	..	..	..	..	..
Unit labor costs in manufacturing (nominal, per hour)	21.9	10.0	19.1	3.6	7.0	9.0	13.2	29.2	26.7
Labor productivity in manufacturing (real, per hour)	1.7	8.3	–0.3	7.0	8.2	4.4	–6.9	–1.1	3.1
Labor costs in manufacturing (nominal, per hour)	23.8	18.9	19.0	10.8	15.8	13.8	5.4	27.8	30.6
Producer price index (PPI) in industry	17.6	12.2	43.9	11.4	22.2	28.2	38.8	44.8	60.6
Consumer price index (here: HICP)	15.2	12.3	19.6	11.8	13.5	15.6	17.1	19.2	25.9
EUR per 1 TRY, + = TRY appreciation	–10.4	–21.0	–23.2	–25.5	–31.8	–24.3	–25.2	–15.9	–26.4
	Period average levels
Unemployment rate (ILO definition, %, 15–64 years)	14.0	13.4	12.2	13.4	13.0	13.8	12.0	11.9	11.2
Employment rate (%, 15–64 years)	50.3	47.5	50.3	48.8	47.7	48.0	49.7	51.6	51.7
Key interest rate per annum (%)	20.6	10.2	17.8	8.4	12.5	17.3	19.0	18.9	15.9
TRY per 1 EUR	6.4	8.0	10.5	8.5	9.4	8.9	10.1	10.1	12.8
	Nominal year-on-year change in the period-end stock in %
Loans to the domestic nonbank private sector1	11.0	36.3	36.1	41.3	36.3	31.9	20.7	14.5	36.1
of which: loans to households	15.9	40.1	20.4	48.4	40.1	35.4	24.9	15.9	20.4
loans to nonbank corporations	9.5	35.0	41.9	39.1	35.0	31.3	20.1	14.7	41.9
	%
Share of foreign currency loans in total loans to the ­nonbank private sector	35.2	30.9	38.1	32.0	30.9	32.4	32.7	32.2	38.1
Return on assets (banking sector)	1.1	1.0	1.3	1.2	1.0	1.0	1.0	1.1	1.3
Tier 1 capital ratio (banking sector)	13.9	14.1	13.2	14.5	14.1	13.4	13.2	12.9	13.2
NPL ratio (banking sector)	5.7	4.4	3.4	4.4	4.4	4.1	3.9	3.8	3.4
	% of GDP
General government revenues	..	..	..	..	..	..	..	..	..
General government expenditures	..	..	..	..	..	..	..	..	..
General government balance	–4.4	–4.7	–3.6	..	..	..	..	..	..
Primary balance	..	..	..	..	..	..	..	..	..
Gross public debt	32.6	39.7	39.1	..	..	..	..	..	..
	% of GDP
Debt of nonfinancial corporations (nonconsolidated)	..	..	..	..	..	..	..	..	..
Debt of households and NPISHs1 (nonconsolidated)	..	..	..	..	..	..	..	..	..
	% of GDP (based on EUR), period total
Goods balance	–2.2	–5.3	–3.6	–5.7	–4.4	–3.7	–3.4	–3.2	–4.2
Services balance	4.5	1.6	3.3	2.0	1.9	1.2	1.9	5.4	4.0
Primary income	–1.7	–1.3	–1.5	–0.9	–1.2	–1.6	–1.9	–1.2	–1.3
Secondary income	0.1	0.0	0.1	0.2	0.1	0.1	0.2	0.1	0.0
Current account balance	0.7	–5.0	–1.7	–4.3	–3.7	–4.0	–3.2	1.2	–1.5
Capital account balance	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Foreign direct investment (net)2	–0.8	–0.6	–0.9	–0.5	–0.6	–0.7	–0.7	–1.5	–0.8
	% of GDP (rolling four-quarter GDP, based on EUR), end of period
Gross external debt	52.3	51.3	51.7	50.1	51.3	53.8	53.1	53.3	51.7
Gross official reserves (excluding gold)	10.3	6.5	9.4	5.0	6.5	6.5	7.5	10.7	9.4
	Months of imports of goods and services
Gross official reserves (excluding gold)	4.1	2.4	3.2	1.9	2.4	2.4	2.7	3.8	3.2
	EUR million. period total
GDP at current prices	678,772	625,264	683,223	167,165	162,128	155,985	156,492	190,618	180,128
Source: Bloomberg, European Commission, Eurostat, national statistical offices, national central banks, wiiw, OeNB.
1 Nonprofit institutions serving households.
2 + = net accumulation of assets larger than net accumulation of liabilities (net outflow of capital).
– = net accumulation of assets smaller than net accumulation of liabilities (net inflow of capital).

11 Russia: heading from strong recovery to strong recession triggered by Ukraine war and sanctions

But Russia’s invasion of Ukraine that started on February 24, 2022, and the ­unprecedented package of Western punitive sanctions that followed profoundly changed the playing field. Western sanctions include the freezing of assets of the Bank of Russia (CBR) and of some large Russian banks located in EU/G7 ­countries, the prohibition of funding in EU/G7 financial markets for the Russian sovereign, the CBR and several large Russian banks, the exclusion of some Russian banks (though not the largest, Sberbank, and the third-largest, Gazprombank) from the international payment system SWIFT, and additional export controls (on top of already existing controls) for high-tech products and aircraft parts and ­components. The G7 further decided to strip Russia of its “most favored nation” status in trade with G7 members. The USA furthermore imposed an embargo on purchases of oil, gas and gold, and the EU on purchases of coal from Russia. The freezing of almost half (about USD 300 billion) of the CBR’s international reserves (a total of USD 643 billion or 38% of GDP in mid-February 2022) – the part that had been placed in Western countries’ jurisdictions – is a particularly remarkable and ­unprecedented step.

In response, the CBR more than doubled its key rate to 20.0% on February 28 after it had already raised it in the previous six months (in four steps totaling a raise from 3% to 9.5%). Moreover, exporters were instructed to exchange 80% of their foreign currency proceeds into ruble, and some other capital controls were ­installed (e.g. retail foreign currency purchase surcharges). The CBR also ­intervened with the unfrozen part of its reserves, which, together with foreign currency refinancing and asset valuation changes, contributed to a decline of its unfrozen reserves by USD 39 billion (about 12%). The Moscow Exchange was closed for a couple of weeks, then opened again in late March, although a “­temporary” ban was imposed on foreign firms and nonresidents selling Russian assets and/or repatriating proceeds. The ruble – no longer fully convertible – lost almost half of its value against the US dollar and the euro from mid-February to mid-March, then regained most of the lost terrain. In the course of April, it came close to the level prior to the invasion. In early April, the key policy rate was ­lowered by 300 basis points to 17%.

CPI inflation, which previously had been pushed by strong domestic demand and structural bottlenecks, grew from 7.4% in September 2021 to 9.2% in ­February 2022, and further accelerated to 16.7% in March (the highest level seen since 2015). Apart from the (limited) devaluation, the sharp rise is also due to ­supply chain disruptions and consumers’ temporary hamster purchases of food and durables. The government invoked its own anti-inflation measures, including ­export restrictions on some commodities and products (e.g. sugar, grain, ­fertilizers). In mid-March, the CBR stated that the Russian economy has entered a phase of far-reaching “structural transformation” toward more self-reliance and less ­dependence on Western imports, which will also modify the domestic price ­structure. Despite economic woes, GDP is estimated (Refinitiv Datastream, OECD) to have expanded by 5% to 6% in the first quarter of 2022.

In 2021, the federal budget had produced a surplus of 0.8% of GDP, buoyed by the recovery of the oil price. Fiscal surpluses continued in January and February 2022. As of end-January 2022, the assets of the National Welfare Fund – most of which constitute Russia’s now partially frozen international reserves – came to USD 175 billion. In reaction to the sanctions, the authorities have announced stepped-up social assistance payments, pension adjustments, tax breaks and ­financial support for enterprises. That said, a substantial anticyclical fiscal stimulus is reportedly not planned. The oil price rise contributed to boosting the country’s current account surplus to 6.9% of GDP in 2021. The first quarter of 2022 ­delivered another substantial current account surplus. Despite the partial freeze of reserves, debt service has so far been upheld.

Banks’ NPL ratio slightly declined over 2021 to 15.1% at end-2021. Following the imposition of massive Western sanctions, the CBR provided extensive ­regulatory lenience for the measurement of banks’ assets and encouraged banks to grant temporary credit holidays for distressed borrowers. In mid-March, the ­authorities introduced a credit subsidy program. While fully up-to-date monthly data are not yet available, mass bank runs following the plunge of the ruble have so far not materialized, possibly due to the sharp upward key rate adjustment which pushed up deposit rates, albeit rising inflation may soon put more pressure on banks.

Outlook for selected CESEE countries and Russia

Accelerating inflation and war in Ukraine weigh on growth in the CESEE-6 region; attack on Ukraine severely hits Russia’s economy ^{8 , 9}

Russia’s war on Ukraine markedly hurts the Russian economy. In light of ­uncertainty in Russia, increased Western economic and trade sanctions as well as the country’s own countersanctions, we expect Russian GDP to contract by about 10% this year and thereafter remain at levels seen a decade ago, i.e. prior to the annexation of Crimea and initial sanctions on Russia. The ruble’s exchange rate has fallen sharply, and Russia’s imports are expected to halve to levels reminiscent of the mid-2000s. The volume of Russian exports will decline, particularly as the EU reduces its energy imports from Russia. High inflation will depress household ­consumption, and fixed investment will suffer. The risks to this forecast are ­exceptionally large and concern e.g. the war, sanctions, inflation and fixed investment. Government budget spending could grow strongly.

Table 1: OeNB-BOFIT GDP projections for 2022 to 2024 compared with the IMF forecast  

	Eurostat/ Rosstat	OeNB-BOFIT projections April 2022			IMF WEO forecast April 2022
	2021	2022	2023	2024	2022	2023	2024
	Year-on-year growth in %
CESEE-6	5.5	3.2	3.7	3.8	3.1	3.4	3.5
Bulgaria	3.8	2.9	3.5	3.2	3.2	4.5	4.2
Croatia	10.0	2.4	3.7	3.5	2.7	4.0	3.0
Czechia	3.3	0.7	3.5	3.7	2.3	4.2	3.6
Hungary	7.1	3.4	3.0	3.3	3.7	3.6	3.7
Poland	5.6	4.5	4.1	4.1	3.7	2.9	3.2
Romania	5.8	2.5	3.3	3.5	2.2	3.4	3.8
Russia	4.7	–10.0	0.0	..	–8.5	–2.3	1.5
Source: IMF World Economic Outlook (WEO) of April 2022, Rosstat, OeNB-BOFIT projections.

1 CESEE-6: inflation curbs private consumption and external demand weakens in the near term; investments strengthen in the medium term

Inflation will dampen real disposable income

Monetary policy will be challenged to rein in inflation while not supressing economic activity in the current environment. Sharply accelerating inflation ­pressures will likely affect the monetary policy stance in Czechia and Hungary, and somewhat less so in Romania, while ERM II requirements reduce the monetary policy space in Bulgaria and Croatia.

Fiscal policy tightening will be somewhat delayed for many reasons. As ­mentioned above, government support aimed at mitigating rising energy prices will burden public budgets, as will additional expenses for refugees. In most ­countries, spending on pensions will rise (Bulgaria, Czechia, Hungary and ­Romania). Also, the green transition will cause fiscal costs in the near term; for instance, Czechia plans to increase subsidies for renewable resources. Further additional ­social spending is envisaged in Bulgaria and Croatia (i.a. maternity and unemployment benefits).

Ukrainian refugees will not immediately support labor supply

Nominal wage growth will be strong, but given even stronger inflation ­dynamics, real wages are likely to remain flat or decrease. Minimum wages are being ­increased in all countries, bringing them closer to average wages. Bulgaria, ­Czechia and Hungary will also see strong public wage growth, while Romania has imposed a wage freeze in most public sector areas, except for moderate increases in education and health.

Private consumption will remain a stable, albeit smaller, growth factor

As mentioned before, public spending will increase so as to mitigate price ­effects for households and corporates, to cushion the effects from the war and ­related sanctions and to provide assistance for refugees from Ukraine. In 2022, annual public consumption growth will be at 3% in all countries except for ­Romania, where fiscal space is limited by the ongoing excessive deficit procedure. For the CESEE-6 region in total, the GDP growth contribution of public ­consumption growth will triple to almost 1 percentage point in the current year, before falling back to around 0.3 percentage points in the subsequent years. Apart from Hungary, where some pre-election spending continued until April 2022, no further elections are scheduled in the CESEE-6 this year.

Investment growth will accelerate in 2023

External demand is set to weaken

What is more important, foreign demand will recede in general, as the growth projections for all export markets, in particular in the euro area, have been revised downward. We revised export growth for the CESEE-6 region down by 3.3 percentage points to 5.3% year on year in 2022. At the same time, import growth will decelerate by 3.7 percentage points to 5.8% year on year, due to the strong ­import-export nexus, rising uncertainty and lower real income (thus also reducing import demand for final goods). Hence, the strong negative growth ­contribution from net exports will narrow to remain at –0.4 percentage points throughout the forecast horizon. Both export and import dynamics should marginally increase again in 2023 and 2024, as the geopolitical situation should stabilize and delivery bottlenecks and hence supply shortages should soften again.

Risks to CESEE-6 projections are strongly tilted to the downside; uncertainty is exceptionally high

A rapid and lasting peace in Ukraine seems unrealistic at the current juncture; hence, the severe and unprecedented sanctions against Russia are more likely to stay in place (as assumed in our baseline) than to be relaxed or lifted. On the ­contrary, it cannot be ruled out that the sanctions against Russia are intensified or further countersanctions are imposed by Russia, also restricting oil and gas ­deliveries to the rest of Europe. This would constitute a major downside risk to our GDP forecasts, as the dependency on energy from Russia is particularly high in the CESEE-6 countries.

In the same vein, supply disruptions are likely to prevail longer than assumed in our baseline. In particular, the war has extended the range of products affected by supply shortages to more inputs, including neon from Ukraine, palladium from Russia and, most importantly, steel with a strong impact on almost all industrial sectors, including construction. Building on the flexibility observed for global value chains during previous crisis events (such as the earthquake in Japan in 2011), we assume that shortfalls of inputs from Russia and Ukraine will, to a large extent, be substituted with other sources. Yet, this process may take longer than expected or remain incomplete, given its magnitude and suddenness, which would pose a serious threat to industrial production, not least in the important automotive ­sector.

While we do not expect widespread restrictive measures to be re-imposed in the region in response to new surges in COVID-19 infection rates, renewed supply chain disruptions arising from the pandemic cannot be ruled out. In particular, China’s zero-COVID strategy represents a risk, as closures of production sites and ports in China have the potential to shake supply chains worldwide – especially given concurrent interruptions on land routes from China due to the war in Ukraine.

While financial market reactions to the military attack have been contained so far, negative confidence effects could intensify and worsen risk sentiment further, leading to higher financial market volatility beyond the current level and to a strong rise in risk premia, which would match badly with burdened public budgets in the region.

Inflation may stay high for a longer period of time than currently anticipated by policymakers. This would bring central banks increasingly under pressure and could – if not sufficiently addressed in time – also trigger turbulences on financial markets, worsen financial conditions beyond expectations and negatively impact on financial market stability.

Apart from geopolitics, political uncertainty in the region itself prevails. It ­remains to be seen how the dispute with the EU about the conformity of ­Hungarian and Polish national regulations with EU law develops against the background of a more united EU against an external aggressor. Thus, uncertainty about the ­disbursement of EU funds, including NGEU funds, to Hungary and Poland ­remains high for the moment.

Finally, climate risks are no longer relevant in the long term only. Recent ­extreme events, such as floodings and droughts, have exposed the vulnerability to natural disasters of economies in moderate climate zones. In combination with imminent food shortages caused by the war raging in Ukraine, such events would yield stronger effects on food supplies and prices than under normal circumstances. More generally, the green transition coupled with currently high levels of ­uncertainty and depressed consumer sentiment may also affect consumer behavior and curb demand for automobiles in the near term and in the longer term alike, with a detrimental impact on this dominant sector in many CESEE-6 economies, unless structural reforms reduce dependency on the sector.

2 Projections for Bulgaria, Croatia, Czechia, Hungary, Poland and Romania

Bulgaria: consumption depressed by inflation; investment will lead growth in the medium term

We expect that domestic demand will be the key contributor to growth over the next years. Private consumption, entering with a high base from 2021, will support growth over the forecast horizon, although only to a limited extent in 2022. The government adopted increases in minimum wages, pensions and social benefits, but these are likely to barely offset the increases in price levels that are expected to continue through 2022. Labor market participation rates and employment rates have been rising recently, but they are not yet back to pre-pandemic levels. The uncertainty related to the war in Ukraine, including its consequences for tourism and trade, can hinder a swift progression and constitute an obstacle for strong consumption growth.

Gross fixed capital formation is expected to be a key driver of Bulgaria’s growth over the next years. This is in part due to a low base effect resulting from a sharp decline in investment in 2021 which was related to ongoing political and ­pandemic-induced uncertainty throughout the year. More substantially, we expect a boost to investments from the accommodative budget, effective as of April 1, 2022. The budget focuses on increasing public investment to strengthen the ­recovery after the pandemic. Apart from raising social spending, the government plans to bolster the education and health sector and make sizable capital ­investments, in particular to update infrastructure. The budget will be revised in mid-2022, and the revised budget should provide buffers to address accelerating inflation and spending related to refugees from Ukraine.

The political instability in 2021 led to a postponed submission of the country’s Recovery and Resilience Plan, the approval of which is a precondition for accessing NGEU funds. Bulgaria submitted its plan in mid-October 2021, requesting a total of EUR 6.6 billion in grants under the Recovery and Resilience Facility. At the time of writing, Bulgaria’s Recovery and Resilience Plan is still under review by the European Commission. Funds are not expected to flow before the second half of 2022, but we expect them to promote further investment growth in 2023.

The number of international tourists remained well below pre-pandemic levels in 2021. A shortfall in Russian and Ukrainian tourists and the geographical ­proximity to the war raging in Ukraine combined with the very low and stagnating level of COVID-19 vaccination rates are likely to severely hamper a recovery of the tourism industry in 2022. Subdued growth in other EU countries, the ongoing disruptions in global value chains and the high degree of uncertainty, now fueled by the war in Ukraine, will be a further drag on exports. For 2022, we expect a moderately negative contribution of net exports to growth, due to moderate growth in imports. For 2023, however, we expect imports to pick up, triggered by investment activities. Overall, net exports are expected to slow down GDP growth over the forecast horizon.

Croatia: accelerating inflation and geopolitical crisis weigh on outlook

We expect GDP growth to be mostly driven by domestic demand, which is however difficult to predict, given the multitude of growth-retarding and -­enhancing factors currently at work. Uncertainties relating to the pandemic and the geopolitical crisis, elevated and accelerating inflation and potentially also some monetary and macroprudential tightening ¹² will likely dampen domestic demand. However, the Croatian government has already undertaken measures to limit the impact of higher inflation on the private sector: It passed a substantial package that includes VAT cuts, energy price regulations and subsidies for vulnerable ­households and sectors. Private consumption will also be supported by another increase in minimum wages and a renewal of the housing subsidy program. On the investment side, the absorption of funds and start of projects related to the EU’s MFF and the Recovery and Resilience Facility are expected to outweigh the negative effects on private investments from the factors mentioned above. Anticipating growth-­retarding factors to abate over time, we see growth of private consumption ­accelerating from 2.8% year on year in 2022 to 3.8% year on year in 2023, and investment growth rising from 3.2% year on year in 2022 to 5.7% year on year in 2023.

Public consumption growth will make mildly positive and declining contributions to growth over the forecast horizon. Croatia must balance the need for fiscal support with the need for fiscal prudence, given high public debt levels.

We expect another successful tourist season in 2022, as the geographic ­proximity and perceived predictability and safety of a holiday in Croatia will likely continue to be important factors drawing (mostly) Central European tourists to the Croatian coast. This will not only support growth of exports but also feed back positively to domestic demand, given the high share of the tourism sector in gross value added and employment. We are more pessimistic about goods exports in 2022, given likely disruptions to supply chains and fluctuations of prices of raw materials. Export growth is expected to accelerate over the forecast horizon. Net exports make a negative contribution to growth in 2022 and 2023.

While there is some upside risk to our forecast, on the whole, we still consider that the risks are mostly on the downside. Upside risks would arise from a ­quicker-than-expected softening of inflation, an even better-than-expected tourist season and stronger investments. Downside risks largely stem from spillovers from the external environment related e.g. to the war and the pandemic, and are ­exacerbated by high domestic public and private debt levels.

Czechia: pandemic- and war-induced frictions turn recovery into stagflation

Apart from the impact of demand related to a significant number of Ukrainian refugees, private consumption growth in 2022 will benefit from a low base in the first half of 2021 and pent-up savings. However, the buoyant nominal disposable income growth on the back of a tight labor market, rising wages and pensions will be more than offset by very high and rising inflation. Moreover, owing to the war-induced supply disruptions and other factors, including domestic ones, ­inflation rates will not only peak at higher levels than previously expected but will also stay high noticeably longer. Hence, in the months ahead, real wages are set to experience a plunge unseen since the beginning of transition. This will put a ­significant drag on household consumption growth in 2022 and beyond as is also echoed in still subdued consumer confidence. Even before the war, frictions in raw material and component supplies, particularly the shortage of semi-conductors, were expected to continue – despite some easing – in the short to medium term. Large stocks of unfinished products waiting for completion upon arrival of missing components have been driven down significantly. Nonetheless, supply chain ­bottlenecks translated, i.a., into reduced production in the first quarter of this year. The war in Ukraine has exacerbated the situation, as the supplies of some key components from Russia and Ukraine, particularly in the automotive industry, have been interrupted and will lead to outright production reductions at least until provisions can be recovered or substituted. As a result, supply chain disturbances and dearth of inputs in combination with impaired foreign demand not only in the EU but also in Russia – which is, for instance, the third biggest sales market for the car producer Škoda – will significantly hamper net exports and fixed investment.

To contain domestic inflationary forces and prevent inflation expectations from unanchoring, the Czech National Bank has repeatedly signaled that it will continue its monetary policy tightening, and with more vigor in light of the impact of the war on prices. Dearer credits and stricter lending conditions will thus pose an additional damper on private consumption and especially on fixed investment in the short to medium term before strengthening. While the fiscal stance was ­expected to be broadly neutral over the forecast horizon, with unexpected expenditures triggered by the war and the massive migration wave, it is likely to turn significantly accommodative.

Hungary: GDP bounced back in 2021, but growth outlook deteriorates

As for 2022, we expect negative repercussions both for private consumption and business investments to dominate. Consumer sentiment, export expectations and overall business sentiment will likely be dented by the war, while supply chains will likely be disrupted both upstream (e.g. deliveries from Russia, shortage of raw materials and semi-products) and downstream (e.g. production disturbances in major export markets/industries, export sanctions, loss of Russian tourists). ­Rising prices for raw materials, transport costs and potential supply shortages will burden both producers and consumers with higher costs.

As far as not consumed by higher inflation, households will likely use additional income from tax rebates/cuts, wage and pension hikes to increase their savings rate, but in 2022 private consumption will still be supported by various price and interest rate caps. Housing investments will likely remain supported by home ­construction subsidies, but the substantial rise in financing costs will deter credit financing.

We expect government consumption to accelerate in 2022 on the back of ­increased transfer payments, public sector wage hikes and operating costs (e.g. ­energy prices, election costs, costs of care for refugees). Thereafter, we expect government consumption to reverse owing to the need to tighten the fiscal stance (less favorable financing conditions, need to cover the expected annual losses of the central bank).

The outlook for investments remains good but less favorable than at the time of the last forecast. On the one hand, interest rates on corporate loans have started to rise in reaction to higher central bank interest rates and the end to the central bank’s preferential facilities. Also, in late 2021, the government postponed investment projects worth nearly 2% of GDP, and public investments will increasingly hinge on the inflow of EU funds, notwithstanding unresolved issues concerning the release of Recovery and Resilience Facility funds by the EU. Business investments will likely be put on hold, given major uncertainties about short- to ­medium-term domestic and global demand conditions, deteriorating corporate earnings and high financing costs. On the other hand, the corporate sector’s net financial asset position improved during 2021. Also, the economic recovery pushed capacity utilization in industry again beyond its long-term average.

We expect the contribution of net real exports to worsen in 2022, as the ­deteriorated global environment hampers the outlook for exports more than for imports, which will be supported by domestic demand. Overall, net exports should have a slightly negative or neutral impact on the GDP growth rate over the forecast horizon.

Poland: growth slows in 2022 as most demand components weaken

As a result, in both 2022 and 2023, the slowdown of foreign and domestic ­demand growth will decelerate import growth sharply, after imports ballooned by more than 17% annually in 2021. Likely, supply chain problems will further ­contribute to the weakening of real import growth, while higher import prices may somewhat uphold imports in nominal terms. In both years, real import growth is forecast to exceed real export growth, albeit to a lesser extent than in 2021. ­Despite starting from a sizable external surplus, the growth differential will be sufficiently large to continue rendering a negative contribution of net exports to GDP growth, but a much smaller one than in the previous year.

Private consumption is expected to grow year on year at about 5% in 2022 and 4.2% in 2023, after growing by 6.1% in 2021 when it benefited from the partially favorable base effect and from anti-COVID-19 crisis measures. Post-pandemic pent-up demand on the back of accumulated extra savings together with hikes in minimum wages will constitute supportive factors. Moreover, refugees fleeing from the war in Ukraine will provide additional private consumption demand, ­using both own private savings and transfers received in Poland. At the same time, the rise of inflation and the expectation of a somewhat prolonged period of higher inflation (given supply bottlenecks) will dampen private consumption, as wage ­increases are unlikely to reflect this acceleration of inflation to the full extent. Moreover, employment growth will slow in the wake of shrinking foreign ­demand, and consumer sentiment is bound to deteriorate. Public consumption growth will accelerate, given measures related to infrastructure for incoming refugees in 2022 and likely also in view of upcoming elections in 2023.

Fixed investment is expected to grow at about 5% annually in 2022, after growing by about 7% in 2021. After all, both public and corporate sector fixed investment benefit from the new MFF funding cycle and the post-pandemic ­National Reconstruction Plan, and corporate investment receives further support from substantial corporate savings. Yet, supply chain bottlenecks will be a partially limiting factor. Moreover, weaker foreign demand and war-related uncertainty will prevent fixed investment growth from accelerating at high rates. For 2023, no substantial acceleration is expected either in view of these factors. Besides, the forecast incorporates a moderate growth contribution from inventory increases in 2022, albeit a far smaller one than in 2021 when increases contributed almost 3 percentage points to annual GDP growth.

Romania: moderate growth expected with EU fund inflows playing a key role

Gross fixed capital formation will be an important growth driver over the ­forecast horizon due to sizable EU fund inflows (from the EU’s multiannual budget frameworks and the NGEU recovery instrument). Yet, there is quite some ­uncertainty about effective EU fund absorption and the implementation of the ­national Recovery and Resilience Plan as a requirement for disbursements from the Recovery and Resilience Facility. Robust domestic credit growth (largely on the back of state guarantee programs) will support private investments. While Russia’s war on Ukraine will entail negative confidence effects, Romania could benefit from foreign direct investments related to production reallocations from Ukraine and Russia.

We expect rather moderate private consumption growth figures in the coming years. The lifting of COVID-19 restrictions should underpin private consumption particularly in the first half of 2022. The minimum wage hike as well as the ­increase of pensions and other social security benefits effective from the beginning of 2022 will also support private consumption. High inflation rates will erode real disposable income, but high inflation expectations will likely also result in higher renegotiated wages in sectors facing labor shortages. Moreover, the extension of the energy price cap until end-March 2023 increases the leeway for consumption spending. Yet, the continued public sector wage freeze (with exceptions for state employees working in social assistance, healthcare and education) will clearly have a dampening effect on private consumption growth.

Based on the assumption of continued positive growth in the euro area, ­Romanian exports are projected to rise over the forecast horizon. However, their growth will be restrained by supply chain disruptions in particular in the auto­motive industry. Moreover, unit labor costs in the manufacturing sector started to rise markedly again in the second half of 2021, while the nominal exchange rate vis-à-vis the euro weakened only a little. Domestic demand growth will keep up import growth so that the contribution of net exports will remain negative.

3 Attack on Ukraine severely hits Russia’s economy

We expect the Russian economy to contract in 2022 even with higher prices for oil and other key export commodities produced in Russia. Markets expect ­energy prices to fall especially next year, and international sanctions and Russia’s countersanctions to remain in place throughout the forecast period. The EU’s fresh energy goal means that EU energy imports from Russia are to decline steadily. Moreover, the pandemic is expected to subside during this year. Facing very large uncertainty, we estimate that Russian GDP will contract by about 10% this year, to the level seen in 2011−2012. The economy is not seen to turn to recovery next year, being strained by instability, sanctions and Russia’s own restrictive measures. The long-term growth outlook, tepid already earlier at around 1%−1.5% a year, has now weakened further, as it is increasingly unlikely that Russia moves ahead with needed economic reforms. Russian imports are foreseen to decline by about half this year, and to remain at that lower level for the remaining forecast period. It may be possible to substitute some imports with domestic products, but unlike in earlier recessions, Russia’s domestic production will, this time, suffer from ­import disruptions due to export bans imposed by foreign countries.

Private consumption should fall this year to levels seen a decade ago, as ­inflation, already relatively high at 9% before the invasion, erodes purchasing power and depresses real household income. With the ruble’s collapse, inflation has accelerated and is anticipated to accelerate further with supply and production disruptions as well as higher world market prices. Russian inflation could reach as much as 20% or more this year. Distress in corporations will, like in earlier ­recessions in Russia, probably lead to larger cuts in real wages than in employment. Nevertheless, unemployment is expected to rise. Real wages in the government sector are also foreseen to decrease. Under the Russian constitution, pensions need to be adjusted to keep up with inflation, while other social supports have already been scheduled to increase. Consumer credit turned to rise after the 2020 ­recession, but is now likely to come to a halt.

Inflation will also depress public consumption. In order to mitigate harms in the current downturn, the country’s leadership may also focus on easily implementable transfers of funds, i.e. social supports that will feed into private consumption and corporate subsidies that keep the economy going.

Fixed investment should contract to levels last seen a decade and a half ago. The business environment for private firms has become exceptionally unstable and lacks perspective. The decline in foreign corporate investment to a necessary ­minimum will be felt, as investment by foreign-owned firms has represented ­almost 15% of total investment. The leadership will likely seek to stave off a deeper collapse in fixed investment by increasing orders from oligarch and other private companies and commanding state-owned enterprises. While government sector investment may increase, the effects are likely to be relatively limited, as it ­represents no more than about 20% of total fixed investment. We expect to see an exceptionally large drop in inventories that significantly exacerbates the declines in GDP and imports.

The paradigm that growth of the world economy fuels Russian exports is changing, as the EU pursues its goal of reducing energy imports from Russia and some non-EU countries and large corporations have decided to stop importing Russian oil. While Russia may be able to shift a tangible part of its oil exports to other markets, a particularly large part of its gas exports is tied to pipeline transmission. Russia’s own export bans will reduce exports further. In the worst ­previous recessions, the drops in GDP and the ruble’s exchange rate have knocked down Russian imports by 30%. In addition to these two factors, imports will now also be reduced by bans on exports, payment restrictions and voluntary decisions by foreign firms to cease exporting to Russia. The expectation that imports fall by half this year means a decline to a level last seen in the mid-2000s.

The ruble’s exchange rate is expected to remain at its current level during the forecast period. The central bank is unable to support the exchange rate through purchasing rubles on the market, as the EU and the US have prohibited trans­actions with the central bank. On the other hand, the fresh requirement for ­Russian firms to convert 80% of their foreign exchange export earnings to rubles will provide coverage for payments going abroad, especially as imports have fallen sharply. High export prices and falling imports will push up Russia’s current account surplus, which already reached nearly 7% of GDP last year. Nevertheless, the surplus should shrink considerably during our forecast period, as export prices are ­expected to come down and EU countries cut back on their energy imports from Russia.

Russia’s shrinking economy will undermine the fiscal balance following an ­episode in which a rapid expansion of budget revenues after the 2020 recession had turned the deficit into surplus last year. Although the economy is now shrinking, budget revenues stand to increase markedly faster than inflation in 2021, but not beyond, as dollar-based proceeds from exports by and production taxes on the oil and gas sectors rise sharply in rubles due to the collapse of the ruble’s exchange rate. Markets also expect oil and gas prices this year to remain higher on average than in 2021. Other budget revenues will shrink this year due to the severe ­economic slump. In the next few years, total government revenues will decline in real terms, as changes in oil prices and the ruble’s exchange rate no longer support revenues, while inflation probably remains rather brisk. Government spending ­increases scheduled so far do not look very large, so that high government revenues this year should keep the government budget in surplus. Russia could afford ­further increases in government spending, as revenues are up this year, but not over a ­longer term. The National Welfare Fund, the state’s reserve fund, appears to be largely blocked due to the central bank transaction ban imposed by the EU and the US, unless the state and/or the central bank handle the conversion of the Fund’s foreign exchange assets into rubles in a way that does not violate the ban.

Soon after Russia’s invasion, the central bank raised its key rate to 20%, which staved off a run on household bank deposits. At the same time, the central bank increased the liquidity supply to banks as much as needed and granted a variety of regulatory relaxations. State banks dominate Russia’s banking sector, and various funds are available to keep other banks afloat. Nevertheless, banks may face some difficulties, as falling household and corporate income hurts the deposit intake and the weak economic outlook depresses bank lending.

Table 2: Russian GDP growth and oil prices  

	2019	2020	2021	2022	2023
	Year-on-year growth in %
GDP	2.2	–2.7	4.7	–10	0
Private and public consumption	3.4	–4.9	7.1	–14	..
Fixed investments	1.0	–4.4	7.0	–20	..
Exports	0.7	–4.1	3.2	–2	..
Imports	3.1	–12.1	16.7	–50	..
Brent oil price (bbl) in USD	64.0	42.0	70.4	100	89
Source: BOFIT.
Note: Realized data for 2019 to 2021, projections for 2022 and 2023. Forecast oil prices are based on the average for oil futures contracts for the ten days preceding March 25, 2022.

Forecast risks are exceptionally large

Government budget spending could increase significantly during the forecast period, including the possibility of reliance on central bank funding to cover spending.

Studies

Mitigating the impact of the pandemic on personal finances in CESEE: descriptive evidence for 2020

Thomas Scheiber, Melanie Koch ¹³

This study describes the economic consequences the COVID-19 pandemic had on people in Central, Eastern and Southeastern Europe until October 2020. We use data from an annual survey of individuals in ten different countries. Specifically, we employ a special module from the OeNB Euro Survey in 2020 to assess what kind of measures individuals took to mitigate ­negative effects of the pandemic and how this relates to income shocks. Reducing expenditure was by far the most common measure, followed by reducing savings, informal support and borrowing against future income. Only very few respondents stated that they had been forced to move. Descriptive results seem to suggest that experiencing income shocks and being ­financially vulnerable are related to taking significantly more mitigating measures and, hence, that the mere number of different measures taken can be a proxy for how severely an ­individual is affected by the pandemic. However, taking more measures can also be related to having more options to actually smooth out negative effects. Therefore, classifying those who report a larger number of different mitigating measures as more vulnerable without taking other socioeconomic characteristics into account can be misleading.

JEL classification: D14, G50

Keywords: household finance, COVID-19, survey data, Central, Eastern and Southeastern Europe

Nevertheless, early evidence suggests that the economic impact within ­countries was felt rather unevenly (e.g. Alstadsæter et al., 2020, for Norway; ­Adams-Prassl et al., 2020, for Germany, the UK and the US; Bundervoet et al., 2022, for 34 countries). The crisis has affected different people in different ways and, therefore, the impact on different groups cannot be assessed based on macroeconomic figures (Basselier and Minne, 2021; Bundervoet et al., 2022). However, a better understanding of how different groups of people have been mitigating the adverse effects of the COVID-19 pandemic is important from a policy perspective as it could shape the recovery of private consumption and therefore GDP. Moreover, such an understanding makes it possible to identify more financially vulnerable groups of people, i.e. those people who suffer severe consequences when hit by an income shock, hence informing the policy debate. It is therefore crucial to use more ­granular microdata to complement the general picture. In this sense, the COVID-19 pandemic is also unprecedented in terms of the production and use of microdata, in particular, survey data – probably another lesson learned from the global ­financial crisis. Aspects that deserve further exploration are an improved understanding of individuals’ adjustment behavior during the pandemic and, related to this, finding ways to measure how strongly an individual is affected by the ­pandemic in overall economic terms.

This study provides additional microlevel evidence shedding some light on these two aspects based on data from a special module of the OeNB Euro Survey conducted in October 2020. With our paper, we contribute to understanding how the COVID-19 pandemic has economically affected individuals in different regions of the world. Our sampled economies mostly consist of middle-income countries in CESEE which are often overlooked in the literature and for which survey data are less frequently assessed. The study is descriptive in nature and briefly addresses the following questions: Which common measures did individuals take to mitigate the effects of the pandemic? How do these measures relate to income shocks? And does the mere number of different mitigating measures an individual took tell us something about how severely a person was affected by the pandemic? We present comprehensive descriptive evidence to establish some stylized facts for the early phase of the pandemic in CESEE.

The special module of the OeNB Euro Survey collected information on varying mitigating measures, like reducing consumption, dissaving and borrowing, and on which respondents were actually hit by an income shock. A caveat ­regarding the module is that it covers only the extensive, not the intensive margin of measures taken. Still, we find that for every single measure elicited, those who were hit by an income shock were more likely to have taken this measure. Moreover, ­individuals who experienced an income shock took more different measures at once than those who did not experience a shock. The measure reported most often in both groups (income shock, no income shock) was reducing consumption, which was followed by reducing savings and informal support, and eventually by borrowing against future income and, apparently as a last resort, by moving. With the help of ­generalized ordered logit regressions with partial proportional odds, we analyze if the mere number of measures taken is a good indicator for how severely a person was ­affected by the pandemic. We find that financially vulnerable people who experienced an income shock are significantly more likely to take more measures, which speaks in favor of this interpretation. However, we also see that more affluent people are more likely to take measures if hit by a shock. This could be driven by the ability and desire to optimally distribute the negative impact across several measure ­categories. Finally, we find some evidence that individuals taking measures even if not hit by an income shock might be driven by restricted consumption ­opportunities due to lockdowns but also by precautionary motives. Summarized, considering merely the number of measures an individual took to counteract the negative ­effects of the pandemic is an imperfect proxy for how severely that person was ­affected. Socioeconomic characteristics should also be considered. These ­characteristics might lead to differing motives for how many mitigating measures are taken.

The differing impact of the pandemic, at least in terms of income shocks, was already documented in several studies. For example, based on real-time unemployment register data from Norway, Alstadsæter et al. (2020) find that layoffs hit ­financially vulnerable populations and had a high socioeconomic gradient. Moreover, layoffs were more common in less productive and financially weaker firms so that employment loss may cause an overestimation of total output loss. ­Adams-Prassl et al. (2020) employ real-time survey data for Germany, the UK and the US. They show that institutional factors and implemented policies explain a lot of the ­variation in labor market impacts. Within countries, the impacts are very unequal and aggravate existing inequalities. Moreover, Bundervoet et al. (2022) conclude that the crisis induced dynamic risks cementing inequality of opportunity and ­undermining social mobility. They use survey data from 34 countries and show that the pandemic disproportionally affected vulnerable segments of the ­population, i.e. women, lower-skilled workers and children. Similar to the aforementioned study, we go a step further, not only looking at the prevalence of income shocks but what potentially happens beyond that shock and what this might tell us about the general impact the pandemic has had.

Several central banks in Europe have produced and employ survey data to gauge the impact of the pandemic on household balance sheets, preferences and ­sentiments, or consumption behavior (e.g. Goldfayn-Frank et al., 2020; Bernard et al., 2020). ¹⁴ The European Central Bank launched the pilot of its future monthly consumer ­expectations survey in January 2020, an online household panel covering six euro area countries. Using the data elicited by this pilot survey, Christelis et al. (2020) find that the adverse effects of the COVID-19 pandemic on consumption expenditure mainly came from households’ perceptions of financial repercussions of the shock and not from their concerns about potential health implications. Moreover, controlling for socioeconomic factors, financial concerns due to the COVID-19 pandemic amplify the negative consumption effect of a negative income shock, while consumption adjustment due to a positive income shock is rather insensitive to COVID-19-related financial concerns. Our study tries to add to the existing literature by providing descriptive evidence on more detailed individual economic responses to mitigate the impact of the pandemic, particularly for countries for which, usually, much less information is available.

This paper is structured as follows: In section 1, we describe the data and ­variables we use. Then, section 2 presents the descriptive results and an in-depth analysis of the measures taken to mitigate the impact of the COVID-19 pandemic and how they are related to income shocks and other socioeconomic ­characteristics. Section 3 concludes.

1 Data and variables

1.1 The economic impact of the COVID-19 pandemic on individuals

Thus, one certain way how the pandemic initially affected people’s finances is through income shocks. According to economic theory, individuals can react in several ways when experiencing an income shock. They may adjust expenditure, and hence the consumption of durable and nondurable goods, or adjust their ­savings behavior. In case of a negative shock, they may moreover take out a loan or delay payments to the future, meaning borrowing against future income. In the OeNB Euro Survey 2020, respondents were asked how the pandemic affected their ­economic situation and, indirectly, about the actions they had taken to mitigate a potential negative income shock. The question was phrased very broadly (see ­below).

Ad hoc question for the OeNB Euro Survey 2020: How did respondents try to mitigate the impact of the COVID-19 pandemic on their personal finances?

[ASK ALL] If you think about your economic and financial situation, since the outbreak of the Corona crisis have you been affected in any of the following ways? Please name all that apply.

1. I had to reduce the amount spent on everyday expenses

2. I had to reduce or postpone larger expenditures

3. I had to reduce money set aside for savings

4. I had to utilize savings or sold possessions

5. I had to reduce help to friends or relatives whom I helped before

6. I had to delay payment of loan installments

7. I had to delay payment of rent

8. I had to delay payment of other bills

9. I had to take out a loan from a bank

10. I had to over-draft my bank account

11. I received financial help from family or friends

12. I had to borrow money from another source

13. I received social benefits or other financial aid from the state

14. I had to decrease work hours and received a reduced salary

15. I was laid off from a job / lost a job

16. I was forced to move

For each item: Yes / No / Don’t know / No answer

End of Box 1

This is because there are several other ways in which people could be ­financially affected by the pandemic besides an income shock. They could suffer from other shocks like confidence and health shocks. They could be affected because ­consumption opportunities have been limited since the start of the pandemic or because new kinds of expenses, e.g. health-related expenses, have become ­necessary. The question tries to capture all these aspects at once. ¹⁷ Overall, it ­elicits information on two different kinds of negative income shocks and 14 possible ­measures to counteract a drop in income that are, however, not exclusively related to income shocks. Moreover, strictly speaking, these measures are not always in the hands of an individual because, for example, item 13 captures whether people received social benefits from their governments. The wording for most items ­deliberately implies necessity instead of preference (“I had to...”) to highlight the crisis character of the pandemic. We still refer to the items as mitigating measures rather than as “economic affectedness” because we think there is often still an ­element of choice in which measures to take and how many. We are also aware that the list does not include all potential financial consequences of the pandemic, ­especially not consequences of positive income shocks and a higher preference for precautionary savings. For example, aggregated bank deposits of private households increased in CESEE, indicating that at least some persons may have increased rather than reduced their savings. Still, we will briefly discuss the role of ­precautionary savings in subsection 2.4.2.

Since negative income shocks are directly linked to financial vulnerability, we want to focus our analysis on how the presence of income shocks relates to the other ways in which an individual was affected. Using items 14 and 15, we ­construct a COVID-19-related income shock dummy variable that is measured on the ­individual level whereas the other items are often added up together. An important point is that individuals are not only affected by income shocks that happen to themselves but also by those that happen to other persons within their household. For this reason, we also use another income shock variable derived from a question that asks respondents more generally whether their household has experienced any significant reduction of income over the last 12 months. The caveat regarding this question is that it likely includes some income shocks that are not related to COVID-19. Our main results will focus on all shocks together, but we will also present results for COVID-19-related income shocks separately.

1.2 Descriptive statistics on sociodemographics

2 Descriptive analysis

2.1 Incidence of income shocks during the first half year of COVID-19

2.2 Number of measures taken to mitigate negative effects of the pandemic

Before we look at the prevalence of certain measures in detail, we first describe the number of measures taken, meaning how many measures at once individuals took. In total, around two-thirds of the individuals at least took one mitigating measure. ¹⁹ Given that only one-third were hit by any income shock, this already tells us that some individuals seem to have taken a measure without having experienced an income shock. Thus, as expected, negative income shocks are not the only way in which respondents may be economically affected by the pandemic. Separating the sample into those who have experienced any shock and those who did not, we still clearly see that mostly those who were hit by an income shock are those who took more than five different measures to cope with the pandemic (see chart 3). Only 10% of those with an income shock did not take any measure at all, most took two or three measures (20% each). The distributions of mitigating ­actions of those who were or were not hit by a shock are significantly different from each other. It is striking that still more than 55% of those who did not ­experience an income shock chose at least one mitigating measure.

The number of measures varies across countries, not only because the ­prevalence of shocks is different. Chart 4 shows that the variations are also conditional on ­either having experienced an income shock (left panel) or not (right panel). ­Especially for the first group, Bulgaria, Croatia, North Macedonia and Serbia stand out on the “negative” side. They have relatively low shares of shock-affected respondents with no mitigating measures (around 5% to 11%) and comparatively large shares with more than five reported items (32% to 46%). In Poland, the share of respondents reporting more than five items is also around 37%, but the share of people taking no action at all is somewhat higher than in the other four countries. Albania has the lowest share of individuals who suffered from a shock and did not take a mitigating measure (around 4%); most of the respondents there reported one or two ­measures (around 55%). Czechia, Hungary and, interestingly, Romania stand out on the rather “positive” side, with the highest shares of shock-affected respondents who took no action in response to the pandemic and the lowest shares of respondents who took three or more measures. The picture is similarly diverse for those ­respondents who were not hit by an income shock. Strikingly, in every country except Czechia even some of those who did not report a shock took more than five mitigating measures. Czechia and Poland are the only two countries in which the majority of those without reported shocks did not take any action. In subsection 2.4, we will discuss what the number of measures taken can tell us about the overall impact the pandemic has had on an individual. ²⁰

2.3 Type of measures taken to counteract negative effects of the pandemic

Chart 5 shows which share of individuals took the respective measures, again broken down into those who were and were not hit by any income shock. ²¹ The mitigating measures that were mentioned by the largest shares of respondents in both groups were consumption-related items. More than 70% of those who ­experienced a shock reduced daily and/or durable consumption. That more than one-third of those with no reported shock did so as well could be a sign of ­restricted consumption opportunities or precautionary motives; this will be ­discussed in ­subsection 2.4.2. Another rather common measure was to reduce the amount saved, reported by roughly 46% (shock) and 20% (no shock) of ­respondents, ­respectively. Utilizing savings or selling possessions was less commonly mentioned by both groups but still more often than the remaining items. Thus, the first resort after an income shock seems to be reduced consumption, followed by dissaving, relying on informal or formal support ²² and delaying payments. Borrowing comes in fifth and seems to be rather focused on short-term liquidity as using overdraft was the most-cited measure in this category. As expected, moving is only the last resort.

2.4 Regression analysis: Do more mitigating measures imply that a respondent has been more severely affected?

Summarized, it is not clear a priori how the number of mitigating measures taken relates to being affected by the pandemic. Looking at how certain personal characteristics are empirically correlated to the number (and type) of measures taken could yield some insights as to whether this number is a good proxy for how severely an individual was affected by the pandemic. In the following, we use ­generalized ordered logit (gologit) regressions with partial proportional odds to analyze which factors are associated with the number of measures taken to better understand what this number might tell us about the degree to which individuals have been affected by the pandemic. Against this background, we control for the need to adjust using information whether the individual has been hit by an income shock over the last 12 months, whether related to the COVID-19 pandemic or not. More specifically, we split the sample into those with income shocks and those without. In each specification, we additionally control whether the respondent is or was employed in an industry class that was severely hit by lockdowns and other containment measures. These industry classes comprise transportation, trade, personal services, tourism and food services, as well as art, entertainment and ­recreation. Concerning the capacity to smooth out shocks, we include ­socioeconomic factors both at the individual and at the household level. The individual factors are age, gender and employment status; the household factors are household net ­income, remittances, household size and whether there are children in the household. Concerning wealth and liabilities, we include information whether ­respondents (personally or together with their partners) have any loans, savings or secondary residences and whether the dwelling is in excellent, good or poor condition. The latter variable is a proxy for wealth reported by the interviewer. ²³

To derive the dependent variable, we recode the number of mitigating ­measures into an ordinal variable with four categories: no measures, 1 to 2 measures, 3 to 4 measures and 5+ measures (analogous to chart 4), balancing the need for a ­sufficiently high number of observations per category and imposing arbitrary ­restrictions through categorization. Economically, the difference between taking no measure and some measures and between 1 to 2 measures and 3 to 4 measures might not be equidistant, which favors an ordinal interpretation. ²⁴ When using models for ordinal dependent variables, we need to test whether the ­proportionality assumption (parallel lines assumption) holds. Since the Brant test (Brant, 1990) and the Wolfe-Gould test (Wolfe and Gould, 1998) rejected the null hypothesis of proportional odds, we explore whether a more generalized specification with ­variable parameters for selected explanatory factors may be a better fit. In our analysis, following the procedure of Williams (2006, 2016), we detect some ­explanatory variables for which variable parameters could potentially increase the goodness-of-fit. Thus, the gologit regressions take the following form (Peterson and Harrel, 1990; Williams 2016):

where m is an ordered response category, X1 and X2 are vectors of independent variables, αm is a cut point, β is the vector of logit coefficients which are fixed across cut point equations, and γ is a vector of logit coefficients which vary across cut point equations, i.e. for those variables that empirically violate the proportional odds assumption.

We pool our sample over all ten OeNB Euro Survey countries due to ­sample size restrictions. Regressions are estimated with country dummies using sampling weights and robust standard errors which are clustered at the primary sampling unit level. For each country, we create income tercile categories and ­another ­category for nonresponse. For the same reason, we add a dummy variable ­capturing item nonresponse for accumulated savings. There is valid concern that country-­specific characteristics like economic structure, unemployment benefits, health and social security systems, containment measures or furlough schemes might ­affect how severely individuals are hit by potential health, income and confidence shocks and how they respond to such shocks. This cannot be adequately addressed with country dummies. A closer microlevel inspection of how different countries are weathering the pandemic is an interesting avenue for future research. ­However, figures A1 and A2 in the annex show that there is considerable regional variation in the prevalence of income shocks, meaning that within-country differences might be even larger than cross-country differences. This might reflect differences in ­income levels and the presence of different industrial sectors in different regions, which we hope to catch with our control variables as well.

2.4.1 Main results

A close inspection of the average marginal effects in specification 1, i.e. the sample without the income shock, shows that there are two types of factors. First, there are factors that significantly decrease the likelihood of being in category 1 (no measures taken) and increase the likelihood mainly of being in category 3 (3 to 4 measures taken), namely working in an industry class that was potentially ­affected by containment measures, belonging to a household with children, and having a loan. Having savings weakly increases the likelihood for being in the ­highest category (5+ measures taken). Second, there are factors that produce the opposite pattern of marginal effects, increasing the likelihood of having taken no measures and decreasing particularly the likelihood of being in category 3, namely being an individual from a household of the third net income tercile (that is, more affluent) or who refused to report the household net income, living in a relatively better maintained dwelling and being below 25 years (only weakly significant).

Specification 2, i.e. the sample with the income shock, exhibits the two ­patterns for the marginal effects too, yet size and significance level of the effects are even higher. Additionally to the aforementioned four factors, belonging to a household of the first net income tercile (less affluent) and owning a secondary residence ­significantly decrease the likelihood of having taken only a few or no measures at all and increase the likelihood of having taken more measures, in particular 5+ measures. As before, high or nonreported income, living in a relatively better maintained dwelling and being below 25 years increase the likelihood of being in the first and second category while they decrease the likelihood of being in the third and fourth category. Another significant factor turns out to be whether the individual’s household receives remittances. Average marginal effects exhibit a u-shaped pattern, pushing observations either into the no-measures category or into the 5+ measures category. Regressions by country show that Albanians drive the significant effect for the no-measures category while remittances in Bosnia and Herzegovina, North Macedonia, Romania and Serbia are significantly associated with the 5+ measures category. ²⁸

What do the regression results tell us about the relationship between the ­number of measures taken and how severely people were affected by the ­pandemic? Financially vulnerable groups, i.e. indebted individuals, individuals from low-­income households or individuals working in an industry class that was potentially affected by containment measures, who have experienced an income shock have a significantly higher likelihood of having taken 5+ measures. This speaks in favor of the number of mitigating measures being an indicator of the severity of the ­pandemic impact on these persons. On the other hand, individuals who ­experienced some kind of income shock but who own a secondary residence or have savings, i.e. economically more affluent persons, also report significantly more measures than the average person. It seems that this group prefers to distribute the painful impact of an income shock by taking several measures and is able to do so.

Summarized, the number of mitigating measures alone is an insufficient proxy for how severely an individual was affected by the pandemic, as completely ­opposing groups of the population took more measures than the “average” person. Depending on socioeconomic characteristics, taking more measures is more likely related to the need to adjust or the capacity to adjust. Among the group of ­individuals who did not experience an income shock but still, on average, report a higher number of mitigating measures are individuals working in an industry that was potentially affected by containment measures, individuals from households with children and indebted individuals. In these cases, precautionary motives or limited consumption possibilities (forced savings) might be the driving forces ­behind the number of measures taken.

Table 1: Generalized ordered logit regression: number of mitigating measures by category  

Type of income shock	Any income shock		Income shock related to COVID-19
Shock experience in 2020: yes/no	No (1)	Yes (2)	No (3)	Yes (4)
Average marginal effects by outcome
Job in a lockdown industry (dummy)
No measures	–0.066***	–0.018**	–0.062***	–0.023
	(0.016)	(0.007)	(0.013)	(0.017)
1 to 2 measures	0.015***	–0.031***	0.002	–0.069***
	(0.004)	(0.012)	(0.001)	(0.024)
3 to 4 measures	0.035***	0.007**	0.036***	0.080***
	(0.008)	(0.003)	(0.007)	(0.030)
5+ measures	0.016***	0.041***	0.024***	0.011
	(0.004)	(0.016)	(0.005)	(0.028)
Respondent aged under 25 years
No measures	0.047*	0.054***	0.074***	–0.007
	(0.028)	(0.017)	(0.022)	(0.013)
1 to 2 measures	–0.011*	–0.009	–0.003	–0.014
	(0.006)	(0.028)	(0.002)	(0.029)
3 to 4 measures	–0.025*	0.045	–0.043***	–0.003
	(0.015)	(0.038)	(0.013)	(0.007)
5+ measures	–0.011*	–0.089**	–0.029***	0.024
	(0.007)	(0.037)	(0.009)	(0.049)
Household net income: 1st tercile
No measures	0.028	–0.016*	0.014	–0.013
	(0.018)	(0.009)	(0.015)	(0.010)
1 to 2 measures	–0.007	–0.027*	–0.001	–0.028
	(0.004)	(0.015)	(0.001)	(0.020)
3 to 4 measures	–0.015	0.006*	–0.008	–0.006
	(0.010)	(0.004)	(0.009)	(0.004)
5+ measures	–0.007	0.036*	–0.005	0.048
	(0.004)	(0.020)	(0.006)	(0.034)
Household net income: 3rd tercile
No measures	0.047***	0.017**	0.051***	0.007
	(0.018)	(0.008)	(0.014)	(0.009)
1 to 2 measures	–0.011***	0.030**	–0.002	0.014
	(0.004)	(0.014)	(0.001)	(0.019)
3 to 4 measures	–0.025***	–0.007**	–0.030***	0.003
	(0.009)	(0.003)	(0.008)	(0.004)
5+ measures	–0.011***	–0.040**	–0.020***	–0.024
	(0.004)	(0.018)	(0.006)	(0.032)
Household net income: not reported
No measures	0.095***	0.018**	0.076***	0.010
	(0.020)	(0.009)	(0.016)	(0.011)
1 to 2 measures	–0.022***	0.032**	–0.003	0.022
	(0.005)	(0.016)	(0.002)	(0.025)
3 to 4 measures	–0.051***	–0.008*	–0.044***	0.005
	(0.010)	(0.004)	(0.009)	(0.006)
5+ measures	–0.022***	–0.043**	–0.029***	–0.036
	(0.005)	(0.021)	(0.006)	(0.041)
Household receives remittances
No measures	0.005	0.028*	0.037	0.054***
	(0.024)	(0.017)	(0.024)	(0.020)
1 to 2 measures	–0.001	–0.045	–0.058***	0.001
	(0.006)	(0.029)	(0.020)	(0.037)
3 to 4 measures	–0.003	–0.053*	–0.015	–0.045
	(0.013)	(0.032)	(0.018)	(0.042)
5+ measures	–0.001	0.070**	0.036***	–0.011
	(0.006)	(0.029)	(0.012)	(0.047)
Children (dummy)
No measures	–0.051***	–0.014*	–0.038***	–0.009
	(0.017)	(0.008)	(0.014)	(0.009)
1 to 2 measures	0.012***	–0.024*	0.001	–0.020
	(0.004)	(0.014)	(0.001)	(0.020)
3 to 4 measures	0.027***	0.006*	0.022***	–0.004
	(0.009)	(0.003)	(0.008)	(0.005)
5+ measures	0.012***	0.032*	0.015***	0.034
	(0.004)	(0.018)	(0.005)	(0.034)
Source: OeNB Euro Survey 2020.
Note: Dependent variable with four ordered categories: (1) no measures, (2) 1 to 2 measures, (3) 3 to 4 measures and (4) 5+ measures. Average marginal effects from a generalized ­ordered logit estimation with country dummies using sampling weights; robust standard errors are adjusted for clustering at the primary sampling unit (PSU) level and reported in parentheses. ***, **, * denote that the coefficient is statistically different from zero at the 1%, 5% and 10% level, respectively. For definition of explanatory variables with ­proportional parameters (=betas) see annex table A2 for specifications 1 and 2 and A3 for specifications 3 and 4. For a definition of variables, see annex table A1. Base categories are: ­respondent aged 35 to 44 years; 2nd income tercile; unemployed, inactive, retired or student; dwelling in poor condition, needs major repair; accumulated no savings; and Czech resident. The sample comprises all ten OeNB Euro Survey countries.

Table 1 continued: Generalized ordered logit regression: number of mitigating measures by category  

Type of income shock	Any income shock		Income shock related to COVID-19
Shock experience in 2020: yes/no	No (1)	Yes (2)	No (3)	Yes (4)
Average marginal effects by outcome
Secondary residence (dummy)
No measures	0.019	–0.019**	0.015	–0.024***
	(0.017)	(0.008)	(0.014)	(0.010)
1 to 2 measures	–0.005	–0.034**	–0.001	–0.054***
	(0.004)	(0.014)	(0.001)	(0.020)
3 to 4 measures	–0.010	0.008**	–0.008	–0.012**
	(0.009)	(0.003)	(0.008)	(0.006)
5+ measures	–0.005	0.045**	–0.006	0.090***
	(0.004)	(0.019)	(0.005)	(0.033)
Dwelling is excellent and well maintained
No measures	0.091***	0.046***	0.097***	0.044***
	(0.023)	(0.012)	(0.020)	(0.014)
1 to 2 measures	–0.021***	0.080***	–0.004	0.096***
	(0.006)	(0.020)	(0.002)	(0.028)
3 to 4 measures	–0.049***	–0.019***	–0.056***	0.021**
	(0.012)	(0.005)	(0.011)	(0.009)
5+ measures	–0.022***	–0.107***	–0.037***	–0.160***
	(0.006)	(0.027)	(0.008)	(0.047)
Dwelling in good condition
No measures	0.051**	0.028***	0.046***	0.034***
	(0.022)	(0.010)	(0.018)	(0.013)
1 to 2 measures	–0.012**	0.049***	–0.002	0.075***
	(0.005)	(0.017)	(0.001)	(0.026)
3 to 4 measures	–0.027**	–0.012***	–0.027***	0.016**
	(0.011)	(0.004)	(0.010)	(0.007)
5+ measures	–0.012**	–0.065***	–0.018***	–0.125***
	(0.005)	(0.024)	(0.007)	(0.043)
Respondent has accumulated savings
No measures	–0.014	–0.012	–0.006	–0.013*
	(0.017)	(0.012)	(0.015)	(0.008)
1 to 2 measures	–0.020	–0.078***	–0.039***	–0.029*
	(0.015)	(0.017)	(0.014)	(0.017)
3 to 4 measures	0.019	0.042**	0.022*	–0.006
	(0.012)	(0.019)	(0.012)	(0.004)
5+ measures	0.016**	0.048**	0.023***	0.048*
	(0.007)	(0.019)	(0.008)	(0.028)
Refused to reveal the extent of savings
No measures	–0.077**	–0.007	–0.047	–0.011
	(0.039)	(0.026)	(0.034)	(0.031)
1 to 2 measures	0.018*	–0.013	0.002	–0.025
	(0.009)	(0.046)	(0.002)	(0.068)
3 to 4 measures	0.041**	0.003	0.027	–0.005
	(0.021)	(0.011)	(0.020)	(0.015)
5+ measures	0.018*	0.017	0.018	0.042
	(0.009)	(0.062)	(0.013)	(0.113)
Respondent has a loan
No measures	–0.125***	–0.055***	–0.116***	–0.047***
	(0.017)	(0.008)	(0.015)	(0.010)
1 to 2 measures	0.018	–0.097***	–0.007	–0.103***
	(0.015)	(0.011)	(0.014)	(0.015)
3 to 4 measures	0.060***	0.023***	0.046***	–0.023***
	(0.012)	(0.005)	(0.011)	(0.007)
5+ measures	0.047***	0.129***	0.077***	0.173***
	(0.007)	(0.015)	(0.008)	(0.025)
Country dummies	Yes	Yes	Yes	Yes
Other control variables (insignificant)	Yes	Yes	Yes	Yes
Log-likelihood	–7,243.6	–3,777.0	–10,187.8	–1,344.4
Probability > Chi squared (df_m)	407.91 (51)	441.16 (43)	511.29 (43)	221.17 (41)
Pseudo R-squared (McFadden)	0.049	0.074	0.045	0.095
BIC	14,959.5	7,924.5	20,790.5	3,000.6
Number of observations	6,300	3,152	8,253	1,199
Unconditional mean of dependent variable	0.88	1.82	1.07	2.03
Source: OeNB Euro Survey 2020.
Note: Dependent variable with four ordered categories: (1) no measures, (2) 1 to 2 measures, (3) 3 to 4 measures and (4) 5+ measures. Average marginal effects from a generalized ­ordered logit estimation with country dummies using sampling weights; robust standard errors are adjusted for clustering at the primary sampling unit (PSU) level and reported in parentheses. ***, **, * denote that the coefficient is statistically different from zero at the 1%, 5% and 10% level, respectively. For definition of explanatory variables with ­proportional parameters (=betas) see annex table A2 for specifications 1 and 2 and A3 for specifications 3 and 4. For a definition of variables, see annex table A1. Base categories are: ­respondent aged 35 to 44 years; 2nd income tercile; unemployed, inactive, retired or student; ­dwelling in poor condition, needs major repair; accumulated no savings; and Czech resident. The sample comprises all ten OeNB Euro Survey countries.

2.4.2 Restricted consumption vs precautionary savings?

We hypothesize that individuals with a rather pessimistic outlook are more ­inclined to increase precautionary savings and, thus, reduce consumption (e.g. Dees and Brinca, 2013, for the US and the euro area; Soric, 2018, for CESEE; Binder, 2020; Christelis et al., 2020, for the euro area during the pandemic). Therefore, we try to approximate pessimistic outlooks with some additional ­control variables. We include three indicators of economic sentiments in the ­regression, looking at those individuals who did not experience any income shock and/or who did not experience a COVID-19-related income shock. The economic indicators comprise respondents’ agreement on a 7-point Likert scale to the ­following two statements: “Over the next five years, the economic situation of [MY COUNTRY] will improve” and “Over the next 12 months, I expect the ­financial situation of my household to get better.” The third indicator captures ­inflation expectations over the next 12 months. Inflation expectations were ­recoded into a variable with three categories (i.e. prices will stay the same or ­decrease a little, prices will increase a little, prices will increase a lot). Thus, we look directly at future expectations. Moreover, we already control indirectly for a more pessimistic outlook by having included the dummy for working in a key ­lockdown industry. For persons working in such industries, income expectations should be less optimistic, and they should be more likely to expect a negative ­income shock. ²⁹ Table 2 reports the average marginal effects from gologit regressions with partial proportional odds, in which the dependent variable is an ordinal variable with three categories for no, one and two consumption-reducing measures (items 1 and 2). Again, we include country dummies, sampling weights and robust standard errors adjusted for clustering at the primary sampling unit level and ­follow the procedure of Williams (2006, 2016) to determine which explanatory variable will have proportional parameters. ³⁰ The results on economic sentiments show that individuals holding more pessimistic views about the economic situation of their country tend to take more consumption-reducing measures. However, ­results are only significant for the group of individuals who did not experience a COVID-19-related income shock (specification 2). Furthermore, individuals ­expecting rising inflation over the next 12 months are associated with reducing consumption significantly stronger in both specifications. Binder (2020) shows that many consumers associate bad events with high inflation. Accordingly, greater concern about the COVID-19 pandemic is associated with expectations of higher inflation and more pessimistic unemployment expectations, which should give rise to higher precautionary savings (see Bernard et al., 2020, for Germany). ³¹ Looking at socioeconomic background, we find a higher likelihood for reducing consumption for individuals working in key lockdown industries, indebted individuals and individuals living in larger households. By contrast, significantly fewer measures to reduce consumption are associated with higher income, being self-­employed, ­having accumulated savings, living in a dwelling in excellent condition and owning a ­secondary residence. But also individuals below 25 years and ­individuals from ­income-poor households report significantly more outcomes of no measures.

Macroeconomic literature on the business cycle stresses the role of durable goods for intertemporal substitution and as an important signal regarding households’ expectations and the strength of the recovery (Beraja and Wolf, 2021). Hence, we rerun the regressions of table 2 with the binary dependent variable ­capturing whether the respondent had to reduce or postpone larger expenditures. The estimations for the group of individuals who did not experience any shock qualitatively yield the same results. Pessimistic economic sentiments, higher ­inflation expectations, working in key lockdown industries and indebtedness ­increase the likelihood of reducing or postponing the consumption of durable goods, while more affluent individuals exhibit a lower likelihood of doing so during the first six month of the pandemic. Thus, we find some support in our sample for the existence of precautionary savings and for the hypothesis that a pessimistic ­outlook is related to taking more consumption measures. This is also in line with what Ercolani et al. (2021) argue for Italy. They find that the amounts people think their household should set aside for unexpected events, i.e. precautionary savings, are positively related to perceived income uncertainty and health risks. Using ­almost the same microdata, Guglielmetti and Rondinelli (2021) conclude that not only economic conditions but also fear of infections and uncertainty about the ­future explain the observed drop in consumption during the pandemic; for some parts of the population, like the self-employed, the latter motives are even more important than the first.

Still, we cannot rule out forced savings completely. For example, reduced ­consumption opportunities could lead to both more savings and expecting ­stronger price increases in the future, meaning the regressions suffer from omitted-variable bias. The same might be true for the lockdown industry dummy as people working in such industries might be living in areas in which more lockdowns were imposed. Nevertheless, it is harder to explain why individuals with a loan should suffer more from suppressed consumption. Furthermore, that very affluent individuals who own secondary real estate and live in “high-class” dwellings are less likely not more likely to report reduced consumption, suggests that survey respondents were less inclined to report forced savings through restricted consumption opportunities as a mitigating measure.

Table 2: Generalized ordered logit regression: number of consumption-reducing measures  

Type of income shock	Any income shock	Income shock related to ­COVID-19
Shock experience in 2020: yes/no	No (1)	No (2)
Average marginal effects by outcome
Job in a lockdown industry (dummy)
No measures	–0.048**	–0.050***
	(0.019)	(0.016)
1 measure	0.030*	0.035**
	(0.016)	(0.014)
2 measures	0.018	0.015
	(0.015)	(0.015)
Economic situation will improve over the next 5 years (7-point Likert scale)
No measures	0.005	0.008**
	(0.004)	(0.004)
1 measure	–0.001	–0.001**
	(0.001)	(0.000)
2 measures	–0.004	–0.008**
	(0.003)	(0.004)
Expected inflation over the next 12 months (3 categories)
No measures	–0.083***	–0.095***
	(0.012)	(0.011)
1 measure	0.019***	0.008***
	(0.003)	(0.002)
2 measures	0.065***	0.088***
	(0.009)	(0.010)
Respondent aged under 25 years
No measures	0.063**	0.065**
	(0.032)	(0.027)
1 measure	–0.014**	0.036*
	(0.007)	(0.019)
2 measures	–0.049**	–0.101***
	(0.025)	(0.028)
Self-employed (dummy)
No measures	0.050*	0.040*
	(0.028)	(0.024)
1 measure	–0.011*	–0.003
	(0.006)	(0.002)
2 measures	–0.039*	–0.037*
	(0.022)	(0.022)
Household net income: 1st tercile
No measures	0.053**	0.024
	(0.022)	(0.017)
1 measure	–0.035**	–0.002
	(0.015)	(0.001)
2 measures	–0.018	–0.022
	(0.018)	(0.016)
Household net income: 3rd tercile
No measures	0.041**	0.038**
	(0.019)	(0.016)
1 measure	–0.009**	–0.003**
	(0.004)	(0.001)
2 measures	–0.032**	–0.035**
	(0.015)	(0.015)
Household net income: not reported
No measures	0.075***	0.052***
	(0.023)	(0.018)
1 measure	–0.039**	–0.004**
	(0.016)	(0.002)
2 measures	–0.035*	–0.048***
	(0.019)	(0.017)
Source: OeNB Euro Survey 2020.
Note: Dependent variable with three ordered categories: (1) no measures, (2) 1 measure and (3) 2 measures. Average marginal effects from a ­generalized ordered logit estimation with country dummies using sampling weights; robust standard errors are adjusted for clustering at the primary sampling unit (PSU) level and reported in parentheses. ***, **, * denote that the coefficient is statistically different from zero at the 1%, 5% and 10% level, respectively. For definition of explanatory variables with proportional parameters (=betas), see annex table A4. For a definition of variables, see annex table A1. Base categories are: respondent aged 35 to 44 years; 2nd income tercile; unemployed, inactive, ­retired or student; dwelling in poor condition, needs major repair; accumulated no savings; and Czech resident. The sample comprises all ten OeNB Euro Survey countries.

Table 2 continued: Generalized ordered logit regression: number of consumption-reducing measures  

Type of income shock	Any income shock	Income shock related to COVID-19
Shock experience in 2020: yes/no	No (1)	No (2)
Average marginal effects by outcome
Household receives remittances
No measures	0.064**	0.065**
	(0.027)	(0.026)
1 measure	–0.014**	–0.005**
	(0.006)	(0.002)
2 measures	–0.049**	–0.060**
	(0.021)	(0.024)
Household size (number of persons)
No measures	–0.008	–0.018***
	(0.008)	(0.007)
1 measure	0.002	0.001**
	(0.002)	(0.001)
2 measures	0.006	0.017***
	(0.006)	(0.006)
Secondary residence (dummy)
No measures	0.031*	0.050***
	(0.018)	(0.015)
1 measure	–0.007*	–0.004***
	(0.004)	(0.001)
2 measures	–0.024*	–0.046***
	(0.014)	(0.014)
Dwelling is excellent and well maintained
No measures	0.077***	0.080***
	(0.026)	(0.023)
1 measure	–0.017***	–0.025**
	(0.006)	(0.012)
2 measures	–0.060***	–0.055**
	(0.020)	(0.021)
Respondent has accumulated savings
No measures	0.031*	0.035**
	(0.018)	(0.016)
1 measure	0.019	0.017
	(0.014)	(0.012)
2 measures	–0.051***	–0.051***
	(0.016)	(0.015)
Respondent has a loan
No measures	–0.086***	–0.089***
	(0.017)	(0.014)
1 measure	0.019***	0.007***
	(0.004)	(0.002)
2 measures	0.067***	0.082***
	(0.013)	(0.013)
Country dummies	Yes	Yes
Other control variables (insignificant)	Yes	Yes
Log-likelihood	–5,806.0	–7,775.5
Probability > Chi squared (df_m)	284.55 (39)	441.28 (40)
Pseudo R-squared (McFadden)	0.040	0.054
BIC	11,967.9	15,927
Number of observations	5,891	7,739
Unconditional mean of dependent variable	0.74	0.92
Source: OeNB Euro Survey 2020.
Note: Dependent variable with three ordered categories: (1) no measures, (2) 1 measure and (3) 2 measures. Average marginal effects from a ­generalized ordered logit estimation with country dummies using sampling weights; robust standard errors are adjusted for clustering at the primary sampling unit (PSU) level and reported in parentheses. ***, **, * denote that the coefficient is statistically different from zero at the 1%, 5% and 10% level, respectively. For definition of explanatory variables with proportional parameters (=betas), see annex table A4. For a definition of variables, see annex table A1. Base categories are: respondent aged 35 to 44 years; 2nd income tercile; unemployed, inactive, ­retired or student; dwelling in poor condition, needs major repair; accumulated no savings; and Czech resident. The sample comprises all ten OeNB Euro Survey countries.

3 Conclusion

In this study, we use data from the OeNB Euro Survey, an annual survey of individuals conducted in ten different countries in CESEE, to analyze which ­factors potentially determine how individuals have responded to the crisis in financial terms. With a special module in 2020, the survey assessed what kind of measures individuals have taken so far to mitigate negative effects of the pandemic. We ­compile descriptive evidence on the prevalence of income shocks in CESEE and how individual shocks relate to the number of mitigating measures taken. Then, we use regressions to establish correlations between the number of measures taken and personal characteristics, including if individuals were hit by an income shock, to understand if the total number of mitigating measures taken is a good proxy for how severely an individual was economically affected by the pandemic.

We find that income shocks in 2020 (income shocks in general and purely COVID-19-related shocks) are related to taking significantly more measures to mitigate the adverse effects of the pandemic, even when controlling for other ­personal characteristics. Moreover, for financially vulnerable individuals, the ­number of measures taken might indeed reveal the severity of the effect the ­pandemic had. For the less financially vulnerable, it might be different, however. For them, several measures might merely indicate that they are in a position to distribute the negative impact of the pandemic across different measure categories. They might even take some measures for precautionary reasons. Thus, the number of mitigating measures alone is not a good indicator for how severely individuals were affected by the first half-year of the COVID-19 pandemic. At least, socioeconomic characteristics should be considered. Additionally, asking people directly how severely they have been affected by the pandemic could serve as an anchor for what the number of measures taken actually indicates.

Still, as the pandemic goes on, more and more measures will be needed to counteract the impact of lasting negative income shocks, savings might become exhausted, reducing consumption might not be enough and borrowing against ­future income might increase. Inequality will continue to widen due to rich ­households increasing their wealth while poor households have to increase their debt. Beyond unexpected income shocks, differences in educational outcomes among children and labor force participation by women in poor and rich ­households will add to long-run inequality if not addressed. The policy implications are neither new nor innovative: policies should restore confidence, support vulnerable people and dampen the effects the pandemic has on income inequality. To identify ­vulnerable individuals, it is important to continue collecting microdata and exploring how data can actually reveal vulnerability. Simply counting the variety of mitigating measures people had to take can only serve as an insufficient proxy.

References

Adams-Prassl, A., T. Boneva, M. Golin and C. Rauh. 2020. Inequality in the Impact of the Coronavirus Shock: Evidence from Real Time Surveys. In: Journal of Public Economics 189. 104245.

Alstadsæter, A., B. Bratsberg, G. Eielsen, W. Kopczuk, S. Markussen, O. Raaum and K. Røed. 2020. The first weeks of the Coronavirus crisis: Who got hit, when and why? ­Evidence from Norway. In: Covid Economics 15. 63–87.

Baker, S. R., R. A. Farrokhnia, S. Meyer, M. Pagel and C. Yannelis. 2020. How does household spending respond to an epidemic? Consumption during the 2020 COVID-19 ­pandemic. In: The Review of Asset Pricing Studies 10(4). 834–862.

Bank of Albania. 2020. Monetary Policy Report 2020/1.

Basselier, R. and G. Minne. 2021. Household savings during and after the COVID-19 crisis: ­lessons from surveys. In: NBB Economic Review 3. 60–78.

Beraja, M. and C. Wolf. 2021. Demand composition and the strength of recoveries. NBER Working Paper No. 29304.

Bernard, R., P. Tzamourani and M. Weber 2020. How are households’ consumption plans affected by the COVID-19 pandemic? Bundesbank Research Brief 35.

Binder, C. 2020. Coronavirus fears and macroeconomic expectations. In: The Review of Economics and Statistics 102(4). 721–730.

Bounie, D., Y. Camara, E. Fize, J. Galbraith, C. Landais, C. Lavest, T. Pazem and B. Savatier. 2020. Consumption Dynamics in the COVID Crisis: Real Time Insights from French Transaction & Bank Data. CEPR Discussion Paper No. 15474.

Brant, R. 1990. Assessing proportionality in the proportional odds model for ordinal logistic ­regression. In: Biometrics 46. 1171–1178.

Bundervoet, T., M. Dávalos and N. Garcia. 2022. The short-term impacts of COVID-19 on households in developing countries: An overview based on a harmonized dataset of high-­frequency surveys. In: World Development 153. 105844.

Carvalho, V. M., J. R. Garcia, S. Hansen, Á. Ortiz, T. Rodrigo, J. Rodríguez Mora and P. Ruiz. 2021. Tracking the COVID-19 crisis with high-resolution transaction data. In: Royal Society Open Science 8(8).

Chen, H., W. Qian and Q. Wen. 2021. The impact of the COVID-19 pandemic on ­consumption: learning from high-frequency transaction data. In: AEA Papers and Proceedings 111. 307–311.

Chetty, R., J. N. Friedman, N. Hendren, M. Stepner and the Opportunity Insights Team. 2020. The economic impacts of COVID-19: evidence from a new public database built using private sector data. NBER Working Paper No. 27431.

Christelis, D., D. Georgarakos, T. Jappelli, G. Kenny. 2020. The Covid-19 crisis and ­consumption: survey evidence from six EU countries. ECB Working Paper No. 2507.

Dees, S. and P. Brinca. 2013. Consumer confidence as a predictor of consumption spending: Evidence for the United States and the euro area. In: International Economics 134. 1–14.

Delle Monache, D., S. Emiliozzi and A. Nobili. 2020. Tracking economic growth during the COVID-19: a weekly indicator for Italy. Note Covid-19. January. Banca d’Italia.

Enzinger, M., M. Koch and A. Riedl. 2021. Die Auswirkung von Covid-19 auf die finanzielle Vulnerabilität verschuldeter Haushalte in CESEE. In: Konjunktur Aktuell 6/2021. OeNB. 60–64.

Ercolani, V., E. Guglielminetti and C. Rondinelli. 2021. Fears for the future: saving ­dynamics after the Covid-19 outbreak. Note Covid-19. June. Banca d’Italia.

Goldfayn-Frank, O., G. Kocharkov and M. Weber. 2020. Households’ Expectations and Unintended Consequences of Policy Announcements. Bundesbank Research Brief 34.

Grieveson, R., O. Pindyuk, V. Astrov, A. Bykova, R. Dobrinsky, S. Duraković, D. Hanzl-­Weiss, G. Hunya, B. Jovanovic, N. Korpar, S. Leitner, I. Mara, L. Podkaminer, S. Richter, B. Ströhm and M. Tverdostup. 2021. Darkest before the dawn? Economic ­Analysis and Outlook for Central, East and Southeast Europe. Forecast Report. The Vienna ­Institute for International Economic Studies. Spring 2021.

Guglielminetti, E. and C. Rondinelli. 2021. Consumption and saving patterns in Italy during Covid-19. Banca d’Italia. Occasional Paper No. 620.

Koch, M. and T. Scheiber. 2022. Household savings in CESEE: expectations, experiences and common predictors. In: Focus on European Economic Integration Q1/22. OeNB. 29–54.

Peterson, B. and F. E. Harrel. 1990. Partial proportional odds model for ordinal response variables. In: Journal of the Royal Statistical Society. Series C (Applied Statistics) 39(2). 205–217.

Soric, P. 2018. Consumer confidence as a GDP determinant in New EU Member States: a view from a time-varying perspective. In: Empirica 45(2). 261–282.

Williams, R. 2006. Generalized ordered logit/partial proportional odds models for ordinal ­dependent variables. In: The Stata Journal 6(1). 58–82.

Williams, R. 2016. Understanding and interpreting generalized ordered logit models. In: Journal of Mathematical Sociology 40(1). 7–20.

Wolfe, R. and W. Gould. 1998. An approximate likelihood-ratio test for ordinal response ­models. In: Stata Technical Bulletin 7(42). 24–27.

Annex

Table A1: Control variables used in the regression analysis  

Label	Description
Any income shock (dummy)	The respondent has experienced one or more of the following income shocks: Lost job due to ­COVID-19, ­reduced salary due to COVID-19 and/or the household of the respondent had a significant reduction of income in the last 12 months.
Income shock related to COVID-19 (dummy)	The respondent has experienced one or more of the following income shocks: Lost job due to ­COVID-19 and/or reduced salary due to COVID-19.
Job in a lockdown industry (dummy)	Dummy variable that equals 1 if the respondent has worked or works in one or more of the following ­industries: transportation; trade; personal services; tourism and food services; art, entertainment and recreation.
Respondent aged [...]	7 variables for the age of the respondent: age brackets with a span of (mostly) 10 years, starting from 25 years to 69 years, one open age bracket for 24 years or younger, and one open age bracket for 75 years or older.
Male (dummy)	Dummy variable that equals 1 if the respondent is male and zero if the respondent is female.
Employed (dummy)	Dummy variable that equals 1 if the respondent is employed and zero otherwise (retired, unemployed, inactive, student).
Self-employed (dummy)	Dummy variable that equals 1 if the respondent is self-employed and zero otherwise (including ­nonworking).
Household net income […]	Household income per month (after taxes) divided into three percentiles per country: low, middle and high. Nonresponse to income is coded as fourth category.
Household receives remittances	Dummy variable that equals 1 if the respondent has received remittances in the previous year and zero otherwise.
Household size (number of persons)	Number of people living in the household of the respondent (including the respondent).
Children (dummy)	Dummy variable that equals 1 if the respondent lives with children below 18 years and zero otherwise.
Secondary residence (dummy)	Dummy variable that equals 1 if the respondent’s household owns a secondary residence or other real estate and zero otherwise.
Dwelling is excellent and well maintained	Dummy variable that equals 1 if the interviewer assesses the respondent’s dwelling as being in an ­“excellent condition and well maintained” and zero otherwise.
Dwelling in good condition	Dummy variable that equals 1 if the interviewer assesses the respondent’s dwelling as being in a “good condition” and zero otherwise.
Respondent has accumulated savings	Dummy variable that equals 1 if the respondent has savings and zero otherwise.
Refused to reveal the extent of savings	Dummy variable that equals 1 if the respondent refused to report if there are savings and zero ­otherwise.
Respondent has a loan	Dummy variable that equals 1 if the respondent has a loan and zero otherwise.
Economic situation will improve over the next 5 years (7-point Likert scale)	Answer to “Over the next five years, the economic situation of [MY COUNTRY] will improve” on a scale from 1 – “strongly agree” to 6 – “strongly disagree.” Recoded to a 7-point Likert scale with “don’t know” as middle category.
Better financial situation of the household over the next 12 months (7-point Likert scale)	Answer to “Over the next 12 months, I expect the financial situation of my household to get better” on a scale from 1 – “strongly agree” to 6 – “strongly disagree.” Recoded to a 7-point Likert scale with “don’t know” as middle category.
Expected inflation over the next 12 months (3 categories)	Answer to “Over the next year, prices will strongly increase in [MY COUNTRY]” categorized into 1 – “prices will stay the same or decrease a little,” 2 – “prices will increase a little” and 3 – “prices will increase a lot.”
Source: OeNB Euro Survey 2020.

Table A2: Generalized ordered logit regression: number of mitigating measures by category  

Sample split by income shock	Any income shock: no (1)			Any income shock: yes (2)
	Betas	Gammas (variable parameters)		Betas	Gammas (variable parameters)
Dependent variable base category	No measures	1 to 2 measures	3 to 4 measures	No measures	1 to 2 measures	3 to 4 measures
	Coefficients			Coefficients
Job in a lockdown industry (dummy)	0.295***			0.231***
	(0.070)			(0.087)
Respondent aged under 25 years	–0.210*			–0.704***	–0.213	–0.502**
	(0.125)			(0.224)	(0.169)	(0.209)
Respondent aged 25 to 34 years	–0.050			–0.088
	(0.088)			(0.110)
Respondent aged 45 to 54 years	0.030			–0.030
	(0.086)			(0.108)
Respondent aged 55 to 64 years	0.012			–0.050
	(0.097)			(0.122)
Respondent aged 65 to 74 years	0.085			–0.294
	(0.128)			(0.184)
Respondent aged 75 years and above	0.075			–0.631
	(0.158)			(0.402)
Male (dummy)	0.043			–0.064
	(0.051)			(0.070)
Employed (dummy)	–0.098			–0.153
	(0.082)			(0.097)
Self-employed (dummy)	–0.152			–0.021
	(0.112)			(0.137)
Household net income: 1st tercile	–0.127			0.204*
	(0.082)			(0.111)
Household net income: 3rd tercile	–0.210***			–0.226**
	(0.080)			(0.102)
Household net income: not reported	–0.426***			–0.241**
	(0.088)			(0.118)
Household receives remittances	–0.021			–0.373*	0.081	0.394**
	(0.110)			(0.222)	(0.167)	(0.165)
Household size (number of persons)	0.009			–0.038
	(0.032)			(0.043)
Children (dummy)	0.229***			0.178*
	(0.075)			(0.102)
Secondary residence (dummy)	–0.087			0.251**
	(0.075)			(0.105)
Dwelling is excellent and well maintained	–0.410***			–0.599***
	(0.105)			(0.148)
Dwelling in good condition	–0.228**			–0.367***
	(0.097)			(0.131)
Respondent has accumulated savings	0.065	0.200**	0.297**	0.156	0.426***	0.269**
	(0.074)	(0.080)	(0.126)	(0.157)	(0.098)	(0.109)
Refused to reveal the extent of savings	0.346**			0.097
	(0.175)			(0.347)
Respondent has a loan	0.561***	0.624***	0.898***	0.722***
	(0.079)	(0.083)	(0.127)	(0.088)
Alphas: cut point parameters	–0.132	–2.113***	–4.381***	2.286***	0.453*	–1.136***
	(0.228)	(0.221)	(0.291)	(0.293)	(0.275)	(0.275)
Country dummies	Yes			Yes
Log-likelihood	–7,243.6			–3,777.0
Probability > Chi squared (df_m)	407.91 (51)			441.16 (43)
Pseudo R-squared (McFadden)	0.049			0.074
BIC	14,959.5			7,924.5
Number of observations	6,300			3,152
Unconditional mean of dependent variable	0.88			1.82
Source: OeNB Euro Survey 2020.
Note: Dependent variable with four ordered categories: (1) no measures, (2) 1 to 2 measures, (3) 3 to 4 measures and (4) 5+ measures. Coefficients from a generalized ordered logit ­estimation with country dummies using sampling weights; robust standard errors are adjusted for clustering at the primary sampling unit (PSU) level and reported in parentheses. ***, **, * denote that the coefficient is statistically different from zero at the 1%, 5% and 10% level, respectively. Variable parameters are assumed for two and three explanatory variables ­(=gammas), respectively, proportional parameters for all other explantory variables (=betas) as well as for all country dummies except for Bulgaria in specification 1 and except for ­Albania, Bosnia and Herzegovina and Poland in specification 2. For a definition of variables, see annex table A1. Base categories are: respondent aged 35 to 44 years; 2nd income tercile; unemployed, inactive, retired or student; dwelling in poor condition, needs major repair; accumulated no savings; and Czech resident. The sample comprises all ten OeNB Euro Survey countries.

Table A3: Generalized ordered logit regression: number of mitigating measures by category  

Sample split by income shock related to COVID-19	Income shock related to COVID-19: no (1)			Income shock related to COVID-19: yes (2)
	Betas	Gammas (variable parameters)		Betas	Gammas (variable parameters)
Dependent variable base category	No measures	1 to 2 measures	3 to 4 measures	No measures	1 to 2 measures	3 to 4 measures
	Coefficients			Coefficients
Job in a lockdown industry (dummy)	0.295***			0.412	0.521***	0.055
	(0.060)			(0.303)	(0.163)	(0.139)
Respondent aged under 25 years	–0.353***			0.120
	(0.107)			(0.244)
Respondent aged 25 to 34 years	–0.107			0.142
	(0.076)			(0.167)
Respondent aged 45 to 54 years	0.034			0.160
	(0.076)			(0.164)
Respondent aged 55 to 64 years	0.046			–0.072
	(0.081)			(0.213)
Respondent aged 65 to 74 years	–0.140			–0.137
	(0.107)			(0.642)
Respondent aged 75 years and above	–0.274**			–0.836
	(0.137)			(0.639)
Male (dummy)	–0.004			–0.082
	(0.044)			(0.113)
Employed (dummy)	–0.157**			–0.546***
	(0.068)			(0.175)
Self-employed (dummy)	–0.039			0.073
	(0.101)			(0.221)
Household net income: 1st tercile	–0.066			0.236
	(0.072)			(0.168)
Household net income: 3rd tercile	–0.244***			–0.120
	(0.068)			(0.158)
Household net income: not reported	–0.361***			–0.180
	(0.076)			(0.205)
Household receives remittances	–0.176	0.102	0.446***	–0.984***	–0.315	–0.053
	(0.116)	(0.109)	(0.143)	(0.340)	(0.229)	(0.235)
Household size (number of persons)	0.049*			0.009
	(0.028)			(0.069)
Children (dummy)	0.179***			0.168
	(0.065)			(0.166)
Secondary residence (dummy)	–0.070			0.446***
	(0.065)			(0.165)
Dwelling is excellent and well maintained	–0.462***			–0.796***
	(0.094)			(0.234)
Dwelling in good condition	–0.221***			–0.621***
	(0.084)			(0.217)
Respondent has accumulated savings	0.031	0.222***	0.286***	0.239*
	(0.069)	(0.070)	(0.098)	(0.139)
Refused to reveal the extent of savings	0.226			0.207
	(0.162)			(0.563)
Respondent has a loan	0.553***	0.608***	0.947***	0.860***
	(0.072)	(0.070)	(0.091)	(0.131)
Alphas: cut point parameters	0.318	–1.302***	–3.009***	3.104***	1.045**	–0.585
	(0.202)	(0.200)	(0.211)	(0.577)	(0.464)	(0.448)
Country dummies	Yes			Yes
Log-likelihood	–10,187.8			–1,344.4
Probability > Chi squared (df_m)	511.29 (43)			221.17 (41)
Pseudo R-squared (McFadden)	0.045			0.095
BIC	20,790.5			3,000.6
Number of observations	8,253			1,199
Unconditional mean of dependent variable	1.07			2.03
Source: OeNB Euro Survey 2020.
Note: Dependent variable with four ordered categories: (1) no measures, (2) 1 to 2 measures, (3) 3 to 4 measures and (4) 5+ measures. Coefficients from a generalized ordered logit ­estimation with country dummies using sampling weights; robust standard errors are adjusted for clustering at the primary sampling unit (PSU) level and reported in parentheses. ***, **, * denote that the coefficient is statistically different from zero at the 1%, 5% and 10% level, respectively. Variable parameters are assumed for three and two explanatory variables ­(=gammas), respectively, proportional parameters for all other explantory variables (=betas) as well as for all country dummies except for Albania, Bosnia and ­Herzegovina and North Macedonia in specification 1 and except for Albania, Poland and Romania in specification 2. For a definition of variables, see annex table A1. Base categories are: ­respondent aged 35 to 44 years; 2nd income tercile; unemployed, inactive, retired or student; dwelling in poor condition, needs major repair; accumulated no savings; and Czech resident. The sample comprises all ten OeNB Euro Survey countries.

Table A4: Generalized ordered logit regression: number of consumption-reducing measures  

Sample split by type of income shock	Any income shock: no (1)		Income shock related COVID-19: no (2)
	Betas	Gammas (variable parameters)	Betas	Gammas (variable parameters)
Dependent variable base category	No measures	1 measure	No measures	1 measure
	Coefficients		Coefficients
Job in a lockdown industry (dummy)	0.204**	0.099	0.220***	0.072
	(0.080)	(0.085)	(0.072)	(0.071)
Economic situation will improve over the next 5 years (7-point Likert scale)	–0.022		–0.036**
	(0.019)		(0.017)
Better financial situation of the household over the next 12 months (7-point Likert scale)	–0.013		–0.003
	(0.021)		(0.019)
Expected inflation over the next 12 months (3 categories)	0.356***		0.420***
	(0.053)		(0.049)
Respondent aged under 25 years	–0.271**		–0.286**	–0.484***
	(0.135)		(0.118)	(0.134)
Respondent aged 25 to 34 years	–0.124		–0.121
	(0.097)		(0.084)
Respondent aged 45 to 54 years	0.081		0.105
	(0.092)		(0.080)
Respondent aged 55 to 64 years	0.044		0.098
	(0.103)		(0.085)
Respondent aged 65 to 74 years	0.008		–0.138
	(0.133)		(0.111)
Respondent aged 75 years and above	0.087		–0.234
	(0.168)		(0.151)
Male (dummy)	0.003		–0.028
	(0.055)		(0.048)
Employed (dummy)	–0.031		–0.066
	(0.086)		(0.072)
Self-employed (dummy)	–0.215*		–0.175*
	(0.120)		(0.106)
Household net income: 1st tercile	–0.226**	–0.101	–0.104
	(0.093)	(0.097)	(0.077)
Household net income: 3rd tercile	–0.174**		–0.167**
	(0.083)		(0.070)
Household net income: not reported	–0.318***	–0.194*	–0.229***
	(0.098)	(0.104)	(0.082)
Household receives remittances	–0.271**		–0.290**
	(0.116)		(0.115)
Household size (number of persons)	0.035		0.080***
	(0.035)		(0.030)
Children (dummy)	0.122		0.060
	(0.086)		(0.072)
Secondary residence (dummy)	–0.133*		–0.220***
	(0.078)		(0.068)
Dwelling is excellent and well maintained	–0.330***		–0.353***	–0.263**
	(0.111)		(0.103)	(0.103)
Dwelling in good condition	–0.090		–0.042
	(0.101)		(0.090)
Source: OeNB Euro Survey 2020.
Note: Dependent variable with three ordered categories: (1) no measures, (2) 1 measure and (3) 2 measures. Coefficients from a generalized ordered logit estimation with country ­dummies using sampling weights; robust standard errors are adjusted for clustering at the primary sampling unit (PSU) level and reported in parentheses. ***, **, * denote that the coefficient is statistically different from zero at the 1%, 5% and 10% level, respectively. Variable parameters are assumed for four explanatory variables (=gammas), proportional ­parameters for all other explantory variables (=betas) as well as for all country dummies except for North Macedonia in specification 1 and except for Albania and North ­Macedonia in specification 2. For a definition of variables, see annex table A1. Base categories are: respondent aged 35 to 44 years; 2nd income tercile; unemployed, inactive, retired or student; dwelling in poor condition, needs major repair; accumulated no savings; and Czech resident. The sample comprises all ten OeNB Euro Survey countries.

Table A4 continued: Generalized ordered logit regression: number of consumption-reducing measures  

Sample split by type of income shock	Any income shock: no (1)		Income shock related COVID-19: no (2)
	Betas	Gammas (variable parameters)	Betas	Gammas (variable parameters)
Dependent variable base category	No measures	1 measure	No measures	1 measure
	Coefficients		Coefficients
Respondent has accumulated savings	–0.134*	–0.278***	–0.153**	–0.246***
	(0.078)	(0.087)	(0.070)	(0.072)
Refused to reveal the extent of savings	0.064		0.033
	(0.168)		(0.160)
Respondent has a loan	0.367***		0.394***
	(0.074)		(0.064)
Alphas: cut point parameters	–0.564**	–1.630***	–0.273	–1.248***
	(0.257)	(0.259)	(0.230)	(0.228)
Country dummies	Yes		Yes
Log-likelihood	–5,806.0		–7,775.5
Probability > Chi squared (df_m)	284.55 (39)		441.28 (40)
Pseudo R-squared (McFadden)	0.040		0.054
BIC	11,967.9		15,927
Number of observations	5,891		7,739
Unconditional mean of dependent variable	0.74		0.92
Source: OeNB Euro Survey 2020.
Note: Dependent variable with three ordered categories: (1) no measures, (2) 1 measure and (3) 2 measures. Coefficients from a generalized ordered logit estimation with country ­dummies using sampling weights; robust standard errors are adjusted for clustering at the primary sampling unit (PSU) level and reported in parentheses. ***, **, * denote that the coefficient is statistically different from zero at the 1%, 5% and 10% level, respectively. Variable parameters are assumed for four explanatory variables (=gammas), proportional ­parameters for all other explantory variables (=betas) as well as for all country dummies except for North Macedonia in specification 1 and except for Albania and North ­Macedonia in specification 2. For a definition of variables, see annex table A1. Base categories are: respondent aged 35 to 44 years; 2nd income tercile; unemployed, inactive, retired or student; dwelling in poor condition, needs major repair; accumulated no savings; and Czech resident. The sample comprises all ten OeNB Euro Survey countries.