Assessing fiscal multipliers in times of crisis: evidence from selected CEE countries

The ongoing public health crisis has simultaneously affected all sectors of the economy. Since its outbreak and its fast worldwide spread at the beginning of 2020, the COVID-19 pandemic has significantly changed the economic environment. Apart from its devastating effects on health and the staggering escalating death toll, the recent pandemic episode is also an unprecedented socio-economic crisis. From the social distancing measures imposed by authorities to self-induced ones, demand for certain services has dropped. Uncertainties are stemming from several sources induced by the pandemic and have the potential to lead to permanent changes in economic behaviour. Despite the introduction of vaccines against COVID-19, the pandemic threat still lingered in 2022 and is expected to weigh further on economic activity, as new and more contagious COVID-19 variants have emerged. Thus, as it might take time until the effects of vaccination campaigns become manifest, it is uncertain whether activity will return to normalcy as known before the pandemic. Even though at the beginning of 2022 the effects of the pandemic were gradually fading-off, a new threat to economic activity emerged. At the current juncture, the supply shocks induced by the Russian invasion in Ukraine have jeopardized the economic recovery of an already vulnerable global environment. The hikes in the prices of energy goods, previously triggered by the reopening of the economies, have recently been exacerbated by the international sanctions imposed to Russia. At the same time, the persistent disruptions in global production and supply chains, in tandem with novel episodes of increased risk aversion have led to high inflationary pressures as well. Given the fact that emerging economies in the geographical proximity of the war might be more exposed to the negative effects of the ongoing conflict and the necessary support measures of authorities might be relatively more consistent, the research is conducted for several Central and Eastern European (CEE) countries such as Bulgaria, Czechia, Hungary, Poland and Romania.

National authorities have tried to partially alleviate the effects of the pandemic crisis by introducing fiscal support measures. In some countries, the package of such initiatives and the variety of instruments were impressive. To these adds additional expenditure related to the public health crisis. At the same time, the European Commission has been coordinating a common European response. All these measures supported the national health systems and alleviated the socio-economic impact of the pandemic. Given the ongoing geopolitical crisis, such measures remain particularly relevant.

Economists all over the world have analysed the implications of the COVID-19 pandemic from a macroeconomic perspective. From the coronavirus impact on economic growth, labour market or certain industries, empirical evidence is gradually accumulating and bringing light on the effects of the unprecedented crisis. Even though there is no doubt the authorities’ response to the public health crisis was necessary, there is nevertheless scarce empirical evidence showing how efficient the implemented fiscal support measures proved to be.

Against this background, this paper focuses on the impact of the fiscal policies in times of crisis. As previously mentioned, in the recent period, the fiscal stance has suffered important reconfiguration. This was also true during the global financial crisis. As studies regarding the efficiency of the supportive measures related to the pandemic are scarce, this analysis concentrate particularly on the effects of fiscal measures adopted to alleviate the adverse effects of the health crisis.

The fiscal multipliers are a useful tool to quantify the effects of implemented support measures. The paper investigates the magnitude of fiscal multipliers and whether the pandemic affected the efficiency of the fiscal policies. Additionally, it provides empirical evidence for revenue and expenditure multipliers. Moreover, we consider the main sub-components of budgetary items. Furthermore, we illustrate the impact of support measures on economic activity. To this end, this analysis employs a variety of econometric tools, ranging from Bayesian Structural Vector Autoregressive (B-SVAR) models to Time-Varying Parameters Vector Autoregressive (TVP-VAR) models.

The rest of the paper is structured as follows. Section 2 provides an extensive summary of recent analysis focused on the quantification of fiscal multipliers in CEE countries and Sect. 3 presents some stylized facts regarding recent fiscal measures aimed at containing the effects of the adverse effects of the public health crisis. Section 4 describes data and methodology, while Sect. 5 summarises the main results. In a counterfactual exercise, the size of economic contractions that would have occurred in the absence of the adopted fiscal measures is also assessed. Section 6 concludes.

The novelty of the recent sanitary crisis made it difficult to assess its socio-economic effects. Under these circumstances, the COVID-19 pandemic has become a popular subject in economic literature. Since its beginning in early 2020, economists have made comparisons between the COVID pandemic and other sanitary crises or even the financial crisis. Jordá et al. (2020) compare European pandemics in order to assess their impact on the natural interest rate. As regards the comparison of the COVID pandemic with the 2008 global financial crisis, both Garrett and Gangopadhyaya (2020) and Li et al. (2021) find a higher adverse impact on the unemployment rate of the sanitary crisis. According to Gunay (2020), the effects of the financial crisis on the stock exchange are heftier than those prevailing during the pandemic. The less relevant literature is devoted to comparing the effectiveness of government support measures between the two recent crises.

At the current juncture, one research direction refers to the effectiveness of the measures adopted by the authorities in order to alleviate the adverse economic impact of the pandemic. Some of these initiatives have been extended given the negative economic effects stemming from the ongoing war in Ukraine, whose future evolution has become the main source of uncertainties for the outlook of the global economy. In the case of the fiscal policy, measures were adopted both on revenues and expenditure sides and implemented in a synchronized manner among European countries. Chudik et al. (2021) estimate a Global VAR model to assess the macroeconomic effects of discretionary fiscal policies in response to the COVID-19 pandemic. Their results show that fiscal policy is playing an important role in mitigating the effects of the pandemic and emerging markets are also benefiting from the synchronized fiscal actions globally. Asongu et al. (2021) conclude that there is a high heterogeneity as regards the response of economic activity to the measures adopted in the context of the sanitary crisis by analysing the economic policies mix among different economies. Croitorov et al. (2021) assess the macroeconomic impact of the COVID-19 pandemic in the euro area and highlight the heterogeneity of economic activity contraction across sectors, with a significant stronger negative impact on activities requiring human physical interaction. Weyerstrass (2021) and Prammer (2021) argue that the large-scale fiscal policy package implemented in Austria to cushion the economic impact of the containment measures contributed decisively to the stabilization of Austria’s public finances. Auerbach et al. (2021) document that the effects of U.S. government spending were stronger during the peak of the pandemic recession, but mainly in cities that were not subject to constraining “lockdown” measures. Medas et al. (2022) show that fiscal rules have been flexible during crises but have not prevented a large and persistent build-up of debt over time, with an implicitly negative impact on economic activity and highlight the need to further improve rule-based fiscal frameworks.

The impact of support measures on GDP can be inferred through the fiscal impulse (measured as the change in the primary structural balance). However, this paper does not focus only on the impact of the bulk of measures, but on the effectiveness of specific fiscal measures adopted by the authorities. Schindler et al. (2009) argue that the impact of fiscal measures on economic activity can be evaluated with the help of fiscal multipliers, defined as the ratio of the change in output in response to an exogenous change in a specific budgetary position. According to Batini et al. (2014), a fiscal multiplier measures the short-term impact of discretionary fiscal policy on output. There are various measurements of the fiscal multipliers, depending on the time horizon considered. In this context, different definitions can be found in the relevant economic literature: (i) impact multiplier or short term multiplier, (ii) multiplier at a specific time horizon, (iii) peak multiplier over any time horizon, and (iv) cumulative multiplier at a specific time horizon.

While the definition of fiscal multipliers and various time frame measurements seem to be widely accepted, there is little consensus in the literature on the size of multipliers. On the one hand, isolating the direct impact of fiscal measures on GDP is difficult given second-round effects and the twofold relationship between them. Researchers have tried to isolate exogenous fiscal shocks, namely those shocks that are not induced by the macroeconomic developments. Batini et al. (2014) note that there is no consensus regarding the methodology to identify such shocks. On the other hand, uncertainty over the size of fiscal multipliers also stems from limited data availability. The highest frequency for the budget execution according to the European System of Accounts (ESA) is quarterly data and, in case of some countries, this time series are not long enough to provide robust econometric estimations.

At the same time, empirical evidence confirms that there is no single fiscal multiplier. Hernández Hernández de Cos (2015) identify a series of factors highlighting the existence of state-dependent multipliers. For instance, given the composition of the fiscal shock, some government spending items are more likely to have higher short-run impact than government revenues. Furthermore, financial frictions, as well as nominal price and wage rigidities can lead to larger multipliers. When quantifying fiscal multipliers, the exchange rate regime (there are evidences for larger multipliers in the case of a fixed exchange rate regime) and the monetary policy reaction function (inactive monetary policy leads to higher fiscal multipliers) also play important roles. Another factor relates to the degree of openness of the economy, as a low degree of trade openness leads to a higher fiscal multiplier. Batini et al. (2014) identify similar factors influencing the size of fiscal multipliers. The determinants of the size are grouped in two categories: (i) structural country characteristics and (ii) conjunctural factors (namely, a series of temporary circumstances such as the cyclical position of the economy and the response of other policies to the fiscal shock). Within the conjunctural factors, the state of the business cycle is one of the main sources of differences among the size of fiscal multipliers. Baum and Koester (2011); Mittnik and Semmler (2012) or Auerbach and Gorodnichenko (2011) identify significant higher fiscal multipliers during recessions than in expansions. Glocker et al. (2019) also confirm that fiscal multipliers are state-dependent. Ilzetzki et al. (2011) argues that even though it seems to be widely accepted that fiscal multipliers are influenced by a series of factors and the business cycle is certainly among them, the size remains a source of controversy among economists.

In the case of Romania, there is also a lack of consensus regarding the size of fiscal multipliers. Empirical evidence differs depending on the size of the data sample and econometric tools used in estimations. Cozmanca and Voinescu (2020) find the cumulative expenditure multiplier to vary across 0.5 and 0.8 over the medium term and the revenue multiplier to be between 0.3 and 0.5. Lower revenue multiplier is found also by Voinescu (2018); Dumitrescu (2015) and Bobasu et al. (2013). Stoian (2012) estimates the revenue multiplier to pose slightly larger values than the expenditure one. At the same time, multipliers are found to be larger in booms compared to recessions and slowdowns. Nonetheless, according to Batini et al. (2014), in a comparison between fiscal multipliers in emerging market economies, their values are found to be positive and below 1. Overall, many of these studies list several caveats in their estimations.

In case of Poland, Czechia and Hungary, empirical evidence mostly finds relatively low values for fiscal multipliers, usually between 0 and 1. Szymanska (2019) estimated the 1-year cumulative spending multipliers for Czechia and Hungary at 0.4 and, respectively, 0.5. As regards Poland, empirical evidence showed a higher than 1 expenditure revenue, also confirmed by Derkacz (2020). By contrast, Batini et al. (2014) or Haug et al. (2019) estimate the spending multiplier in case of Poland to values close to 0.6.

In order to contain the spread of the SARS-CoV-2 virus, most countries introduced social distancing measures or even national lock-downs. Although the bleak picture of empty streets did not last too long, some sectors, especially those requiring high social interaction, suffered more than others and deep scars left by the pandemic can still be seen. Under these circumstances, the national authorities had a prompt response in limiting the adverse effects of the public health crisis.

Both the monetary and the fiscal policies have faced important reconfiguration. The packages of measures adopted in order to mitigate the negative effects of the crisis mainly relate to job retention, avoiding liquidity shortages and stimulating economic activity.

Countries also adopted revenue side support measures. However, those were not as popular as the expenditure related ones. Most of them had zero budgetary impact, i.e. forgone revenue associated with the COVID-19 pandemic that partially started to be paid out in 2021 at the latest. Revenue side measures, having a budgetary impact referred to: (i) social insurance contributions covered by the public budget, (ii) discount for tax payments, and (iii) reduced VAT rate for various goods and services (especially in the case of medical products and those services substantially affected by social distancing measures).

While most of the measures were synchronized among countries, the amount spent to limit the adverse effects of the coronavirus differs significantly, as shown in Fig. 1. In the case of CEE countries, namely Bulgaria, Czechia, Hungary, Poland and Romania, the most impressive package of support measures as a share of GDP was implemented in Hungary, while Romania was in the opposite side in these terms (Fig. 2). By the end of July 2021, Czechia and Hungary spent around 10% of GDP on support measures. In terms of additional expenditure in the health sector, the countries mentioned above also performed better. According to IMF estimates, the average COVID-19 health expenditure in case of the selected CEE countries is 2.1% of GDP, while the average economic support measures amounts to 4.8% of GDP.

While most countries relied more on expenditure side measures, Czechia adopted an impressive revenue package. In addition to foregone revenue, the Czech authorities also introduced lump sums to be received by self-employed and the abolition of real estate transfer taxes. Furthermore, Czechia also outperformed in terms of guarantees, another form of government support (Fig. 3). This evolution relates to the introduction of several programs, such as: (i) the COVID Plus Program (state guarantees provided by the Export Guarantee and Insurance Corporation), (ii) the COVID II Program (state guarantees up to 80% of a loan, as well as a state contribution on interest costs), (iii) the COVID III Program (state guarantees up to 30% of loan principal), and (iv) other guarantees. Romania, Poland and Hungary also offered credit guarantees and micro-loans, especially to support the activity of SMEs, but of a lower value than in the case of Czechia. No data is available for Bulgaria regarding guarantees.

In addition to the response of national authorities, international institutions also played an important role in addressing the COVID-19 pandemic. For instance, the EU is mobilizing available resources to help member states coordinate their national responses. This includes, on the one hand, providing information about the spread of the virus and effective efforts to contain it and, on the other hand, measures to repair the economic and social damage brought by the pandemic. In this regard, international authorities decided that more flexible EU fiscal rules were needed and consequently the general escape clause was activated. Since its introduction in 2011, it has never been activated until the outbreak of the COVID-19 pandemic. This has allowed EU Member States to pursue a fiscal policy that facilitates the implementation of all necessary measures to address the public health crisis. In other words, they were allowed to record budget deficits above the threshold of 3% of GDP. However, in case of Romania, the Excessive Deficit Procedure (EDP) was activated in 2020 against the background of the escalated deficit from 2019. Despite the activation of the general escape clause, the EDP is still active in 2023 for Romania (yet a longer time for fiscal consolidation has been granted).

According to Eurostat data, the budget deficit in European countries, including those in Central and Eastern Europe considered in this paper, has escalated rapidly amid both shrinking revenue and higher expenditure to help contain the virus and support economic activity (Fig. 4). In 2020, Romania recorded the highest ESA deficit, followed closely by Hungary. However, in the case of Romania, the expansionary fiscal policy stance before the outbreak of the coronavirus pandemic contributed to a deterioration of the public budget balance resulting in the activation of the EDP in 2020. In that year, spending related to COVID-19 and the contraction in public revenues led to a significant deficit. The deterioration of the balance compared to 2019 data is not the largest in this case, but for Poland, closely followed by Bulgaria.

The activation of the general escape clause was necessary given the outbreak of the novel coronavirus. The deterioration in the budget balance is expected to be only temporary, as support measures will fade and revenues could return to pre-pandemic levels. Given the protracted public health crisis, the EU’s fiscal rules also remain suspended in 2021 and 2022. Furthermore, the outbreak and extension of the war in Ukraine, which triggered an acute energy crisis and led to persistent bottlenecks in global value chains, determined the suspension of the general escape clause in 2023 as well. The risks to the budget balance outlook are rather downside, given the difficult socio-economic context at the current juncture. In this context, the implementation of new support measures to partly alleviate the effects of generalized high inflationary pressures is not excluded.

The COVID-19 pandemic proved to be an unprecedented socio-economic crisis, prompting swift and decisive action from both national and international authorities to mitigate its economic impacts. In response, the European Commission activated the general escape clause within the Stability and Growth Pact for the first time, allowing European countries to run high deficits starting with 2020. The measure was also extended for the year 2023 in the context of the energy crisis accentuated by Russia’s invasion of Ukraine. Under these circumstances, impressive fiscal support packages were adopted. Our research reveals that, while the magnitude of these measures varied significantly across CEE countries, the instruments for alleviating adverse effects were largely similar, and have proven effective in containing the economic contraction.

In selected Central and Eastern European (CEE) countries - Bulgaria, Czechia, Hungary, Poland, and Romania - the magnitude of fiscal stimulus packages adopted in response to the COVID-19 pandemic between its outbreak and 2021 ranged from approximately 3.1% of GDP in Romania to 10.5% of GDP in Hungary. Apart from Czechia, the selected countries largely implemented expenditure-side measures, such as subsidies for furloughed workers and reduced-time employment schemes.

In this analysis, the size of fiscal multipliers is evaluated in order to assess the efficiency of individual measures. Empirical evidence suggests a greater GDP response to expenditure shocks, with estimated fiscal multipliers within a quarter ranging from 0.03 in Czechia to 0.61 in Poland, values that approach or exceed one on medium-term in all countries except Bulgaria. Conversely, results indicate that multipliers for revenues are lower than those for government expenditure.

Empirical evidence suggests that there are difficulties associated with the measurement of revenue multipliers. A sign-restriction identification scheme was implemented in order to identify the automatic response of revenue variables to output fluctuations and yielded negative multipliers, as expected. These results suggest that further research should be conducted to explore alternative methods for the identification of a pure fiscal shock. For instance, a narrative approach could be employed to construct a new revenue series based on specific policy instruments.

In the related economic literature, the size of fiscal multipliers is a matter of debate. However, the sensitivity of fiscal multipliers to economic cycles is an accepted characteristic. To assess the impact of the COVID-19 pandemic on fiscal multipliers, a time-varying B-SVAR model was employed. The empirical results indicate an intensification of both expenditure and revenue multipliers during the economic crisis.

In light of our results and knowing the composition of the fiscal support packages, the impact of such measures on economic growth was estimated. The prevailing use of expenditure side measures in tandem with further amplifications in the size of fiscal measures is assessed to contribute to a lower contraction in 2020 and a speedier recovery in 2021 by almost 2 percentage points, on average at the level of the 5 analysed CEE countries. Empirical evidence suggests that this emphasis on expenditure-side measures over revenue-side ones was a beneficial choice, as expenditure multipliers tend to be higher. In the upcoming period, economic recovery should further rely on similar measures.

To sum up, the measures implemented by authorities proved to be highly effective. Nevertheless, had they been aware of fiscal multipliers of certain expenditure and revenue items, their positive impact could have been maximized. Taking into account the unstable socio-economic situation in CEE region given the ongoing war in Ukraine, new support fiscal measures could be implemented. The empirical evidence in this paper can be used to evaluate their efficiency. Although this results provide useful information for policymakers, we are aware of their limitations. As previously noted in the literature review section, the size of fiscal multipliers is influenced by a series of structural and conjunctural factors. A more granular approach could be a subject of further research. For instance, investment expenditure could be disaggregated by the source of financing (domestic or foreign) or by specific sectors. Moreover, with the implementation of the recovery instrument Next Generation EU already in progress, more evidence regarding the impact of EU funds could be an interesting research topic.