Financial Resilience Perspective on COVID-19 Business Support: A Comparative Study of Four European Countries

Positioning Our Study in the Resilience Literature

Resilience is not a new concept. In his seminal paper about new public management (NPM), Hood (1991) points to a set of public sector values related to resilience, endurance, robustness, survival, and adaptivity, which allow operations to continue, even in adverse ‘worst cases’, and to adapt rapidly in a crisis (Hood, 1991, pp. 13–14). Resilience as a concept has been adopted in many different fields, such as ecological systems and physical systems. More recently, the concept received much attention in the fields of disaster management (Boin and van Eeten, 2013), organizational science, and supply chain management (see Bahmra et al., 2011).

Initially, resilience was restricted to the capacity to withstand an external shock or crisis, to react and absorb its impact in order to restore the situation (or return to the trend) before the shock, labelled as ‘bouncing back’ (Bhamra et al., 2011; Boin and Van Eeten, 2013). Later on, the concept was extended to also consider the so called ‘bouncing-forward’ capacity, which refers to setting up new post-shock configurations, called ‘transformational resilience’ (Shen et al., 2023), as well as anticipation of and quick adaptation to new challenges before they happen, called ‘evolutionary resilience’ (Saliterer et al., 2021).

Most relevant for our study are the insights derived from the financial resilience literature. Since the financial crisis of 2008 and the resulting austerity programs, the impact of shocks on public finance has been intensively studied by public policy and public sector accounting scholars. Various empirical studies have investigated financial resilience. They include: Barbera et al. (2017) in Austria and Germany; Barbera et al. (2017, 2021) in Italy; Wójtowicz and Hodži (2022) in Poland and Croatia; Padovani et al. (2021) in Portugal and Italy; Upadhaya et al. (2020) in India, Nepal, and Sri Lanka; Ahrens and Ferry (2020) in the UK; and Lee and Chen (2022) in the US.

Many studies that investigate the financial resilience of governments in a crisis context examine the relationship between hindering or stimulating factors for becoming or staying resilient. Such factors include financial aspects (e.g., indebtedness), elements of human resources (e.g., personnel capacities), and political circumstances (e.g., political stability). Financial resilience can be measured as the time for recovering from a financial setback due to a crisis, or by comparing pre- and post-crisis financial criteria, such as debt over GDP (see, for instance, Barbera et al., 2017, 2021; Padovani et al., 2020; Upadhaya et al., 2020; Lee and Chen, 2022; Wójtowicz and Hodži, 2022). Of the formerly listed studies only Upadhaya et al. (2020), Ahrens and Ferry (2020), Wójtowicz and Hodži (2022), and Padovani et al. (2021) focus on financial resilience related to the COVID-19 crisis. Our study differs from this research in two ways. First, it is dedicated to measures that specifically cope with the financial impacts of a crisis, in particular COVID-19 business support. Second, it acknowledges that the COVID-19 crisis is characterized by extreme uncertainty and turbulence, which calls for an investigation of direct consequences, such as providing business support, but also indirect consequences in terms of learning for future crises. This also underpins the importance of investigating COVID-19 coping and anticipatory responses that have never been explicitly related to business support. The next section provides a theoretical lens for the analysis of the four country studies. These country studies, including a comparative analysis are presented in the subsequent section. The following section reflects on these findings in the light of the more recent business support during the energy crisis 2022-2023. The final section comprises conclusions and reflections.

A Framework for Financial Resilience

Our study builds on the theoretical framework for financial resilience as developed by Barbera et al. (2023), the core elements of which are depicted in Figure 1. A brief clarification of this framework is given below.

Vulnerability concerns the extent to which organizations can be negatively affected by shocks and crises. Anticipatory capacities aim to identify and control vulnerabilities, for example, through the monitoring of potential threats and through risk management. Coping capacities can lead to actions or responses to manage the crisis or shock, such as buffering (by absorbing the impact of a crisis); adaptation (i.e., incremental changes of structures and processes), and transformative capacities (which point to radical changes in structures and processes). These coping responses resonate with bouncing-back and bouncing-forward strategies, which can have non-financial impacts (e.g., on health and employment) and fiscal impacts (e.g., governmental finances in terms of deficits and debts).

Operationalization of the Financial Resilience Framework for our Study

Because our study focuses on business support as a consequence of the COVID-19 pandemic, we will elaborate on and expand the more general framework in Figure 1 for the business support context (see Figure 2 and further clarifications below).

Perceived vulnerability is associated with a lack of capacity for adequately responding to a crisis or shock. Vulnerability can have various dimensions (see, for instance, Padovani et al., 2020), but we focus on two important ones. First, a lack of fiscal capacity, measured by government debt over GDP, can be an obstacle in providing business support during a crisis. Second, fragile governance can be a hindrance for effective business support, that is, support programs that insufficiently achieve relevant goals in financial and employment terms. Our study relies on indicator values for six governance dimensions, developed by the World Bank, which are published annually for more than 200 countries (see, for clarification, Kaufmann et al., 2010). The World Bank defines governance as the traditions and institutions by which authority in a country is exercised. Three areas of governance are identified, and each include two indicators: 1) the processes of selecting, monitoring, and replacing government (with voice and accountability, as well as political stability and absence of violence/terrorism indicators); 2) the capacity for formulating and implementing policies (with government effectiveness and regulatory quality indicators); and 3) the respect of citizens and the state for governing institutions (with rule of law and control of corruption indicators). The underlying data are mainly based on surveys and input from experts.

In general, capacities refer to a specific ability, while responses refer to the way in which reactions are shaped. Of course, coping capacities and responses are related, as are anticipatory capacities and responses.

Coping capacities for crisis-related business support may include, on the one hand, financial facilities for taking adequate measures to tackle the crisis. However, De Jong and Ho (2021; see also Chen et al., 2021) find that serious indebtedness of OECD countries did not hinder generous COVID-19-related policies, but this relationship regards fiscal measures in general, while our focus is on financial support for businesses. On the other hand, coping capacities can also refer to institutional enablers for taking action, especially the availability of governmental organizations that have the capacity to execute business support programs, which mirrors governance issues as introduced under vulnerability. Coping responses related to business support comprise specific dimensions: reaction speed; design principles (in particular, generally applicable versus tailor-made programs); and reasoning behind adaptation of programs over the course of time (e.g., new types of programs, and revisions of access conditions to existing support programs).

Anticipatory capacities concern the abilities of institutions to learn from COVID-19 business support for future support programs. A distinction can be made between first-order and second-order anticipatory responses. First-order responses refer to how governments prepare themselves for similar crises in the future, especially through monitoring the features that may impact the design of business support. Second-order responses go beyond similar crises and concern crises with different causes, scopes, and consequences; for instance, what can be learned from business support during the COVID-19 pandemic for similar support during other types of crises, such as the energy crisis (or other shocks and crises like climate catastrophes, supply-chain disturbances, or banking crises). According to Barbera et al. (2023), various measures can be part of organizational learning, that is, information sharing (so, with other organizations facing similar challenges); and monitoring (of future events and circumstances), including risk management (see also Ahrens and Ferry, 2020; Bracci et al., 2022); as well as critical thinking about existing policy-making.

The effectiveness of business support can be measured through non-financial and financial measures. The main non-financial performance dimensions of business support are concerned with avoiding employment losses and bankruptcies. The financial dimensions of business support effectiveness regard the impact on government finances in terms of debt over GDP.

Clues for a Comparative Analysis

Responses of governments to a crisis might be dependent upon country-specific characteristics (see Kuhlmann and Wollmann, 2019). Specifically, we envisage that countries with a federal structure possibly attribute responsibilities about these programs to both the federal and sub-federal layer, which could lead to either divergent or complementary approaches at both layers, resulting in additional expenses, whereas countries with a centralized governmental structure potentially rely on a more coherent (and less costly) support program. Chen et al. (2021), in an international comparative study about the first half year of the COVID-19 pandemic, confirm the thesis of a relatively larger business support in federal countries. Fiscal capacities might also contribute to a large extent of business support, but, as argued before, this relationship is contested in research about crisis-related governmental interventions. In addition, our study can provide knowledge on how coordination mechanisms and information-sharing among various governmental layers and institutions can contribute to effective business support in times of crisis, which can be part of anticipatory capacities.

Research Questions

Perceived Vulnerability

The vulnerability of government in the four countries comprises two dimensions, that is, lack of fiscal capacity and fragile governance. Fiscal capacity is measured by government debt over GDP, particularly in the year just before the beginning of the COVID-19 pandemic, that is, 2019. Belgium and Portugal had a relatively poor fiscal health, with a debt over GDP ratio of about 98% and 117% respectively, whereas this ratio was relatively good in Germany (around 60%) and the Netherlands (around 49%), which were close to or below the EU standard of 60%.

Governance in the four countries is measured by the World Bank worldwide governance indicators (see further Kaufmann et al., 2010). Based on the years 2015–2021, we calculated the average indicator values over the six governance dimensions as introduced in the theory section (see Table 2  and related clarifications). Values can range from –2.5 to 2.5; the higher the score, the better (source: wigidataset). This results in the following outcomes for our study: 1.3 for Belgium, 1.5 for Germany, 1.6 for the Netherlands, and 1.0 for Portugal. Consequently, all four countries score well on governance, although Portugal is somewhat lagging behind (as standard deviations are very low, average differences are significant). So, effective governance might be a bit more problematic in Portugal. It seems that the overall vulnerability was higher in Belgium and Portugal than in Germany and the Netherlands, where a stronger responsiveness could be expected.

We will explore the possible influence of these two vulnerability dimensions on expenditure for business support and impacts later in this section.

Speed, Extent, and Design of the Support

In all four countries business support measures were implemented immediately after the virus outbreak. However, over the period 2020–2021, business support as a percentage of GDP was much lower in Portugal (2.3%) than in the Netherlands (4.1%), Germany (4.3%), and Belgium (5.3%).

Some types of support were present in all four countries, especially wage supplementation and loan facilities. However, there were differences between the types of business support and target groups. Unemployment benefits were important in Belgium and Germany, short-term work allowances were applied in Germany and Portugal, and fixed-cost support was established in the Netherlands and Portugal. Business support was predominantly arranged to be generally applicable to all firms suffering from COVID-19 restrictions. The extent of income loss, in case of wage or fixed-cost supplementation, or liquidity problems due to a lack of income from sales, in case of loan facilities, were conditions for receiving financial support. Portugal also instigated specific support, that is, for the tourism and commerce sectors in particular.

The preference for generally applicable support arrangements was impacted by the need to give rapid support, which required simple and easily manageable support criteria. In a similar vein, in several countries already existing arrangements were revitalized. Germany activated automatic stabilizers (unemployment insurance) and short-time working benefits. The latter enable companies to reduce their activities while employees remain employed in their respective company but receive income compensation in the form of government funds. The instrument existed previously but due to the pandemic, an adequate application process and various other factors (such as the extension of the reference period in favour of the beneficiary), were reasons for its activation. In Belgium the legalistic framework of the measure called ‘hindrance allowance’, previously dedicated to sales losses due to road works, was reused during the COVID-19 crisis as a lump sum cash allowance for all firms suffering as a result of the pandemic.

In Belgium both federal and regional governments were engaged in specific sector support, but tax measures were restricted to the federal government. The federal state installed a temporary moratorium on company bankruptcies and granted a state guarantee for new credits for non-financial corporations, including the self-employed, and the non-profit sector. The regions were in the lead regarding the allowances (e.g., Flemish Protection Mechanism (FPM)) to compensate for lost business income, but they also funded COVID-19 loans through their public investment vehicles.

In Germany, the 16 states (Länder) were involved in the implementation of the response packages to ensure a rapid outflow of funds, supplementing the federal aid only in some cases, for example, if certain sectors in the Länder were particularly affected (e.g., tourism). The program design and the provision of funds have nevertheless been carried out mainly by the federal government. An exception is ‘aid for hardship cases’ financed equally by the federal and state governments.

In the Netherlands as a decentralized unitary state, COVID-19 business support was only provided by the central government. Two business support measures were by far the largest: wage supplementation due to income loss for businesses and non-profit agencies (NOW) and fixed costs supplementation (TVL/TOGS; details are in van Helden et al., 2022). The latter type of support started later (in the Summer of 2020) than the former (March 2020).

Portugal is a centralized and unitary country with a market dominated by SMEs that were most affected by the COVID-19 shocks (Ebeke et al., 2021). It approved the first support measures (simplified lay-off) quickly and in a relatively simple way (Gomes, 2021) to mitigate, contain, and manage the spread of COVID-19. The support for income losses and fixed costs increased at a later stage (APOIAR program). The target sectors of these support measures were tourism, manufacturing industries, trade (non-food), automotive, and transport equipment. Companies with up to 10 workers represented around 80% of the businesses supported in 2020.

In all four countries, the business support packages were aimed to avoid bankruptcy and employment losses in all industries, in order to restore the situation before the pandemic outbreak. We therefore consider these support measures to be a bouncing-back response. Governments were not prepared for providing business support on short notice after the virus outbreak. So, they were forced to develop and implement simple support arrangements, and did not appear to have the time to consider types of support that would create new post-shock configurations, called bouncing-forward responses.

Dynamics in the Extent and Design of the Support

After the initial and quick implementation of the support in all four countries, large differences in the dynamics of the extent and design of the support among the countries are observed. While in Germany and Portugal support increased by 0.5% of GDP from 2020 to 2021, the opposite was true for the Netherlands (decrease of 0.5%) and Belgium (decrease of 1.5%).

In the Netherlands and Portugal access conditions for support were relaxed and support became more generous, thereby responding to the increasingly longer duration of the pandemic and its consequences for survival of specific categories of firms.

In the Netherlands, wage support was implemented as a simple measure in order to provide rapid support. However, this simplicity came at the expense of leaving little room for customization. Whereas the percentage of wage costs compensated originally was expected to diminish over time, it largely remained unchanged over the period. Fixed-cost support was initially meant to be temporary for a period of three months, but eventually it was continued until the third quarter of 2022. The first tranche was targeted at SMEs in branches that were hit most by the pandemic, but in the second tranche this group was extended to all SMEs and from the third quarter of 2020 onwards, all business entities were included. Despite the more generous access conditions for wage supplementation applications in 2021 in comparison to 2020, total support diminished over these years due to a substantial lower number of applications as a consequence of less severe COVID-19 constraints.

In Portugal, programs that were initiated at the beginning of the pandemic were extended and expanded. The short-term work scheme approved until the end of August 2020 was adjusted and enlarged in line with the evolution of the pandemic. Government decided to extend the first support measures allowing the companies affected by the new economic restrictions to keep or join the ‘simplified lay-off’ or the ‘support for progressive recovery’ until December 2020. The support for progressive recovery was revised from October to December 2020. In addition, in the last trimester of 2020, a financial incentive scheme was installed for covering additional fixed costs and tax deferral. This package was adjusted over time and aligned with the evolution of the pandemic and the worsening of the pandemic situation in January 2021. Support decisions to companies are dependent on analyses and validations by certified accountants and tax advisors.

At the onset of the crisis, the Belgian government responded quickly by installing generic, simple, lump sum types of support that could be accessed by all sectors (Dhyne and Duprez, 2021). Yet, over time, measures became less generous and more sector-specific, conditional, and restricted to healthy and tax-compliant businesses (e.g., Flemish Protection Mechanism).

In Germany, initially, leakages were accepted for offering quick and lump-sum aid. Due to the need to act quickly, few clear guidelines were formulated, which led to extensive use of some programs. The later programs were increasingly tailored to individual sectors and formulated stricter access conditions, and companies had to provide more evidence to receive support. In the course of the pandemic, several extensions were gradually provided until 2022, when special regulations were created for particularly hard-hit sectors, such as the travel, culture, and event industry, and also for start-ups. In contrast to ‘immediate aid’, later aid could only be provided with the help of a tax advisor, lawyer, auditor, or certified public accountant.

As we found that most adaptations were not designed to set up new post-shock configurations, we consider these to be bouncing-back responses. However, by restricting the support to healthy and tax-compliant businesses, Belgium also shows some indications of a bouncing-forward approach. It may come as a surprise that governments in the four countries did not consider more specific types of business support that would respond to the post-pandemic weaknesses of particular sectors, such as a lack of IT integration or a poor competitive profile. Although time pressures became less prominent in the course of the pandemic and bouncing-forward responses would become more likely, the simplicity of bouncing-back responses was seemingly preferred.

Understanding the Impacts of the Support

Figure 3 shows the impacts of the support. It indicates that the fiscal impact differed considerably among the four countries. Whereas debt over GDP was already high before the start of the pandemic in Belgium (101%) and Portugal (120%), these percentages increased further to 113% and 133%, respectively, at the end of 2021. In the other two countries, these percentages also rose, but from a lower base and to a lesser extent: in Germany from 60% to 70% and in the Netherlands from 49% to 53%, in 2019 and 2021 respectively.

In addition, Figure 3 shows the number of bankruptcies and the unemployment rate before and during the COVID-19 period, as crucial non-financial indicators of the business support. In general, the number of bankruptcies remained low during the pandemic, and after a rise in unemployment, a quick recovery in employment figures was noticeable.

In Belgium, firms seemed to have coped relatively well thanks to government support (NBB, 2022). Compared to pre-crisis years, the percentage of firms that ran losses and the percentage of overleveraged firms decreased. The number of bankruptcies was lower than before the pandemic. Zegel et al. (2021) evaluated the impact of the Flemish support measures from the start of the pandemic up to the third round of the Flemish Protection Mechanism. More than 50% of the support was directed towards sectors that were hit the most (SMEs in retail and catering). The Flemish regional product was estimated to have decreased by at least 8.5% in the absence of the support measures. Graydon's (2021) analysis for 2020 reveals that without the support measures, the Flemish pre-COVID-19 extremely healthy SMEs that were not crisis resistant would have increased by 16%. The percentage of Flemish zombie firms (heavily indebted firms that face the risk of going bankrupt) was estimated at 2.4% of Flemish SMEs.

In Germany, the impact of the support packages was generally considered to have been positive. Severe negative effects on the labour market were prevented and unemployment already reached its pre-crisis level in late 2021. Numbers of bankruptcies were lower than before the crisis and did not increase noticeably even after the expiration of the special regulations for corporate insolvencies in summer 2021. From the companies’ point of view, non-repayable aid measures and relaxations in profit taxation were rated as very helpful for companies while loans or guarantees were seen as less beneficial (DIHKT, 2020). Short-time work, bridging aid, and the economic stabilization fund were also generally seen as positive.

Although unemployment increased at the beginning of the COVID-19 period, later on it decreased to a lower level than before the outbreak in the Netherlands. Furthermore, the number of bankruptcies in 2021 was about half the amount in 2019. The measures helped to avoid bankruptcies and allowed employees to keep their jobs. In the absence of business support, the unemployment rate would have been between 0.7 and 2.0 percentage points higher. So, extensive support in 2020 has limited the economic damage (CPB, 2021). However, a substantial part of the support was given to firms with either a lack of economic potential, or firms with enough equity to survive without governmental support. This implies that, due to the generic type of support, part of it was wasted (CPB, 2021).

In Portugal, in the period 2020–2021, almost one third of the national workforce benefited from the ‘simplified lay-off’, which is the type of support with the highest fiscal impact. Furthermore, the ‘APOIAR’ program caused the greatest impact in 2021 among the various business support measures (1.1 billion euros to help about 100,000 businesses). Businesses that benefited from this kind of support were obliged to retain employment and not pay out profits or other funds to partners. Unemployment rose at the start of the pandemic, from 6.7% in 2019 to 7.1% in 2020, but unemployment fell to 6.3% at the end of 2021. In addition, the bankruptcies index became lower in 2022 than in 2019 and GDP increased to higher levels than in 2019.

Table  2 summarizes our evidence on the impacts of COVID-19 business support on fiscal matters and employment in Belgium, Germany, the Netherlands, and Portugal, including possibly relevant antecedents for these impacts, as derived from our theoretical framework about financial resilience in Figure 2.

Table  2 reveals that the fiscal impacts have a similar pattern in all four countries: after a rise of indebtedness during the COVID-19 years 2020–2021, a recovery is noted after these years. The impacts on unemployment are different for two pairs of countries: while in Belgium and Germany an initial rise in unemployment is followed by a decrease, in the Netherlands and Portugal unemployment decreases between 2019 and 2022. However, differences between both pairs of countries are small. This analysis suggests that a main reason for these differences might be the government structure: a unitary rather than a federal state with relatively low coordination burdens may enhance support effectiveness. It is noteworthy that support is far less generous in Portugal than in the other countries, which can be explained by Portugal's high level of indebtedness, while the governance score is only moderate. Portugal therefore performs well given the limited resources spent: so, it can be labelled as a ‘cost-effective’ provider of business support.

Inspired by the financial resilience framework in Figure 2, the above explorations give rise to the following reflections on the influence of antecedents on the impacts of business support. Government structure—either federal or unitary—seems to matter. Belgium and Germany as federal countries are spending more on business support than the Netherlands and Portugal as unitary states, while their performance is not significantly better. A strong governance does not have a systematic influence on business support effectivity. Germany and the Netherlands with higher governance scores than Belgium and Portugal do not perform better than the latter two. Moreover, the influence of fiscal opportunities on delivering support is mixed. Whereas the high level of indebtedness may have prevented Portugal from providing generous support, almost equally indebted Belgium provided the most generous support. The generosity of support in the less indebted countries, Germany and the Netherlands, is also high.

Prudence is required regarding the interpretation of the figures in Table  2. The table can only present the possible impacts of business support in the short term, while longer-term impacts can be different. Moreover, causal links between antecedents and impacts may be disputable. This is especially the case with regard to debt over GDP, where many other factors can be influential, for instance in Belgium the government had to spend billions of euros to cope with flood damages. In addition, government governance is assumed to be equally rated for business support institutions as for government as a whole, for which indicator values were used. Hence, the figures in Table  2 need to be understood in combination with the detailed evidence on each of the countries, as presented earlier in this subsection.

Consultation and Information-sharing

All four countries were overwhelmed by the COVID-19 pandemic in March 2020. Without a quick intervention from their governments, many employees would have lost their jobs. Providing business support on short notice with limited monitoring and surveillance was therefore the highest priority. This meant that in these first months of the pandemic, room for information-sharing and critical thinking within the governmental system was widely lacking. But after this first episode, governments became more prone to benefiting from expertise and experiences of other stakeholders in government.

While in Germany, the Netherlands, and Portugal governments were not prepared for providing business support, evidence about this issue in Belgium seems inconclusive. While the Belgian Court of Auditors signals an absence of preparedness, the agency for delivering support in Flanders claims that it was well-prepared, given its long-lasting experience in this domain that dates back to 2002.

The Belgian Court of Auditors has encouraged business support agencies to develop more formal information systems for underpinning support decisions, and it has also warned against a lack of structural consultation among support-providing agencies, which could be harmful for providing justified support. Despite the advice from the Court of Auditors, no formal consultation structure of various actors in the provision of business support could be realized. The German Federal Court of Auditors criticized the lack of a legal framework for delivering support, including confusion about responsibilities at the federal and Länder level. The German Ministries for Economic Affairs at the federal and Länder level confirm that they had to improvise in the early stages of the crisis, while there was a good information flow. The Dutch agency for wage support delivery became involved in information-sharing and consultation with other agencies with similar tasks, but also with the tax office (about recovery of unjustified payments), and the Chambers of Commerce (about the most vulnerable branches). A representative of the Ministry of Economic Affairs in Portugal highlights the involvement of the Association of Certified Accountants for assuring information provision for proper business support. In addition, the Court of Auditors highlighted the need for a clear definition of target groups and specific objectives in the design of future support programs.

Risk Management

In Belgium no risk management systems were in place. There were initiatives for developing such systems, but the energy crisis hindered the realization of these initiatives because priority had to be given to the operations of giving business support.

In Germany risk management was criticized by the Federal Court of Auditors because the federal government was the only funder of the support. In addition, it was acknowledged that a revision of the support would have been desirable when the crisis turned out to be less threatening than initially expected.

The agency that executed wage support in the Netherlands indicates that risk management systems were in place, and there was also intensive coordination with the audit and enforcement departments of the ministry of Social Affairs and Employment that is politically responsible for this type of support. Wage support is divided into subsequent stages of three to six months and at the beginning of each stage a risk assessment was made, for example about improper data supply by companies, or improper use of regulations.

Risk management systems were also developed by the Portuguese central government in the form of control and auditing systems, sometimes based upon requests by the European Union. In this respect the collaboration with the Tax Office and the auditing profession is seen as a guarantee for avoidance of risks in the application and payment of support. The administrative simplification and automation of the application process became more efficient and transparent.

Critical Thinking and Change

Whilst information-sharing and consultation refer to an anticipatory capacity of openness towards inputs from other stakeholders, critical thinking and related action might have stronger consequences. It is concerned with an inclination to be prone to revising existing practices. Our evidence from the interviews reveals that critical thinking was seen as infeasible during the first months of the pandemic, due to the pressures for delivering fast support, but it gradually took shape.

The Belgium government made the decision to make business support more sector-specific in the course of the pandemic. In addition, data systems were improved to mitigate the risks of ex-post controls and repayment claims.

In Germany, it was determined that unnecessary funding and double funding, at both the federal and Länder level, should be excluded from the outset in the future. In addition, the various support measures must be more specific and more clearly demarcated from each other, although support should not be sector-specific. The Federal Court of Auditors clearly preferred the bouncing-back strategies in business support. While the auditors have criticized the insufficient ex-post performance review, the ministries of economics at the two government layers note that the learning effects have already been used for more recent crisis management.

The agency for providing wage support in the Netherlands argues that this support was gradually enhanced through so-called client trajectories, in which various actors coordinated their contributions, while clients are only faced with one service point. Although generally applicable wage support was reconsidered in this country, the alternative option of sector-specific support was ultimately rejected due to a lack of reliable data for handling applications. The Court of Auditors signalled deficiencies in the control systems for fixed-cost support, which were partly attributed to the application of automated devices for decisions regarding applications.

In Portugal a system for spotting vulnerable sectors was developed in order to effectively provide sector-specific support. The Court of Auditors promoted learning by advocating openness of different institutions to cooperate in the future.

Overall Assessment

This section about lessons learned from COVID-19 business support indicates that the need to provide quick support after the outbreak of the pandemic gave little room for consultation and information-sharing among actors within the governmental system. But, in the course of the pandemic these actors became more prone to benefiting from the experiences of others. The evidence about the development of risk management systems and critical thinking about existing support practices suggests less progress. Intentions for installing risk management systems were not realized in some countries due to, among other factors, pressures for delivering support during the energy crisis. Neither critical thinking nor related actions, in which alternative options for more sector-specific business support could have been considered, were developed to a significant extent. In sum, a quick response to the virus outbreak by providing business support was a symptom of well-functioning crisis management, but there was a lack of ex-post monitoring and critical reflection.