6. Emergency economic and fiscal measures in Luxembourg

Faced with the health crisis, the fiscal response of OECD Member countries was swift, massive and multifaceted. In Luxembourg, the comprehensive business support package was similar to those deployed by many OECD Member countries. It included a variety of measures aimed at providing a response tailored to each company’s situation. These measures can be grouped into four main categories: i) tax deferrals; ii) measures to maintain employment; iii) direct support; and iv) State loans and guarantees. For each of these four categories, Figure ‎6.1 sets out the measures covered in this chapter. The measures analysed were selected based on their relative share of disbursements within Luxembourg’s emergency plan.

To address the economic impact of the health crisis, stabilisation and recovery plans have been put in place to support and accelerate the ongoing recovery and future growth. The recovery package will not be considered in detail in this chapter because it would be premature to conduct an impact assessment on a set of measures that are still being implemented.

An analysis of discretionary measures to tackle COVID-19 shows that the support policies adopted in Luxembourg have been commensurate with the needs of most economic sectors. The self-employed received less support at the start of the crisis; however, specific support measures were put in place during the crisis. Assistance was granted swiftly, a critical success factor in supporting companies that may run out of cash very quickly. Given the scale and speed of the intervention, the potential risk was of a windfall effect enabling companies to receive support when they did not need it. The OECD’s analysis suggests that this has not been the case: the companies hardest hit by the crisis are those that have received the most support. The analysis demonstrates that it is important to continue to monitor the financial health of businesses to ensure that future interventions can be appropriately targeted. It also seems necessary to continue to digitalise administrative procedures. The digitalisation effort has proved its value during the crisis and can help create a resilient business climate.

The rest of the chapter is structured as follows: it first assesses the overall package implemented by Luxembourg by comparing it with those adopted in other OECD member countries. It then looks at the implementation of the support measures, analysing the procedures for applying for and granting assistance, as well as the budgetary procedures used to disburse the necessary funds. The chapter then goes on to address the issue of targeting by analysing the use of support measures. It also presents the results of a microsimulation that offers some elements that will help evaluate the impact of the measures on business performance.

In Luxembourg, the first business support measures were put in place very quickly to accompany the strict lockdown implemented on 18 March 2020. Most of the measures followed the Communication from the European Commission on 19 March 2020 introducing changes to the State aid regulations (see Box ‎6.1). The objective of Luxembourg and most of the countries in the region was to protect domestic industry for a future recovery and to limit the social impact of the business shock. In Luxembourg, short-time work was the key measure in terms of the amount of total expenditure. In this respect, the choices made by Luxembourg are in line with those made by other open economies such as Ireland and Belgium. Liquidity measures, such as State loans and guarantees, have been less important in Luxembourg, limiting the risk of long-term corporate debt as the crisis ends.

As in many other European countries, the adjustment in the labour market in Luxembourg took the form of a reduction in the average working time per employee. The aim was to limit the rise in unemployment by reducing working hours through short-time work. This involved a large-scale programme in which public funding has played the most significant role in supporting businesses. In addition to the “force majeure/COVID-19” short-time work programme, which was implemented in March 2020, the employment measures included COVID-19-related family leave. This enabled employees and the self-employed to care for their loved ones following the closure of schools, childcare facilities and care homes for people with disabilities and the elderly.

At the same time, from late March 2020, the Government implemented several waves of direct, non-repayable support. This was initially aimed at businesses with fewer than ten employees (microenterprises). Self-employed people, who were also eligible for this assistance, were specifically targeted from April 2020 onwards. Companies with fewer than 20 employees received a second lump-sum payment in April 2020, and self-employed people in May 2020. From June 2020, recovery support provided all small businesses in the tourism, events, HORECA, culture and entertainment sectors, as well as the self-employed, with a monthly grant of EUR 1 250 per employee in work and EUR 250 per employee on furlough. Businesses were eligible if they experienced a reduction in turnover of more than 25% compared to the same month in 2019.1 The recovery support measures were adjusted after March 2022 as part of a gradual reduction, and came to an end in June 2022. In November 2020, the government also introduced support for “uncovered fixed costs”. This direct, monthly, non-repayable support targeted the sectors hardest hit by the COVID-19 crisis: tourism, events, HORECA, culture and entertainment. It was intended to cover a portion of the monthly expenditure not covered by businesses’ revenues. Businesses were eligible if they had experienced a reduction in turnover of at least 40% compared to the same month in 2019. Recovery support and support for uncovered fixed costs were extended until June 2022. Other specific forms of financial support for the tourism (including vouchers for overnight stays), physical retail, culture, sport and agriculture sectors were introduced in the same period.

Companies experiencing liquidity issues resulting from the pandemic were also able to request the waiver of advance payments for direct taxes (corporate income tax and municipal business tax) for the first two quarters of 2020. For businesses operating in the HORECA sector, this option was extended to the third and fourth quarters of 2020. The government also helped companies affected by the COVID-19 crisis by accelerating the reimbursement of some tax claims (indirect taxes), in particular VAT credit balances of less than EUR 10 000. Unpaid VAT credit balances of less than EUR 10 000 were automatically refunded during the week of 16 March 2020. Penalties for late submission of VAT and other returns were waived.

In addition to non-repayable assistance, the Government also awarded capital grants to businesses, in the form of repayable advances, and guarantees to facilitate access to bank financing (Box ‎6.2).

Analysis of the emergency measures implemented in Luxembourg and in six OECD Member countries (Belgium, France, Germany, Ireland, the Netherlands and Switzerland) confirms that the country took action as quickly as its neighbours (see Table ‎6.1). Other countries introduced or adapted pre-existing regular income replacement schemes for self-employed people. Luxembourg opted instead to adapt the legal framework for short-time work (which did not previously cover compensation for the self-employed). In addition to repayable advances, the government introduced a non-repayable lump-sum payment of EUR 2 500 for the self-employed on 8 April to partially offset the effects of the crisis. This payment came in addition to a first lump-sum payment granted to microenterprises, which benefitted also some self-employed people.

The measures adopted in Luxembourg with the largest budgetary impact were:

  • short-time work, with a total of EUR 1 208 million granted;

  • family leave, amounting to EUR 288 million in 2020 and 2021;

  • the Recovery and Solidarity Fund, with a total of EUR 271 million granted (EUR 110 million for recovery support and EUR 161 million for support for uncovered fixed costs);

  • the State guarantee scheme, with a total of EUR 194 million guaranteed for loans taken out in 2020 and 2021;

  • repayable advances granted to businesses amounting to a total of EUR 180 million.

To better assess the relative importance of the measures implemented in Luxembourg, the budgetary effort has been compared with the efforts of other OECD Member countries (Belgium, France, Germany, Ireland, the Netherlands and Switzerland) along the four main types of business support measures: i) tax and social security contribution waivers and deferrals; ii) short-time work and leave; iii) direct support; and iv) State loans and guarantees.

Estimating the total value of emergency schemes in OECD Member countries, as well as the breakdown of expenditure across different areas, is a difficult task. Measures are not categorised the same way in different countries. The availability and frequency of data updates also varies. Nevertheless, this exercise is useful because it enables a better assessment of the choices made in a given country compared with the efforts of other OECD Member countries.

Overall, a degree of heterogeneity can be seen in the relative scale of national schemes.1 In 2020-21, of the EUR 3 570 billion allocated to tackle the COVID-19 crisis, Luxembourg allocated EUR 2 850 billion in support to businesses and households. This represents 3.9% of its annual GDP in 2021. It is an effort comparable to that of Belgium, which devoted 4.7% of GDP, and Ireland, which devoted 3.3% of GDP. These economies were relatively less affected by the crisis and were therefore able to bounce back more quickly. In other countries, the effort was greater: it amounted to 12.6% of GDP in the Netherlands, 11.9% in France and 5.6% in Germany.

Support measures have been divided into two categories: i) fiscal measures (short-time work and leave, direct support and tax and social security contribution waivers); and ii) liquidity and guarantee measures (State guarantees, tax and social security deferrals and State loans). This categorisation enables a distinction to be made between expenditure with an immediate and definitive effect on public finances, and temporary disbursements and guarantees that are expected to have only a limited impact on public finances in future years. In general, all of the countries considered in the analysis relied mainly on immediate fiscal measures, with the exception of the Netherlands and France, which favoured liquidity and guarantee measures within their overall packages.

Across the board, the main fiscal measures are comparable, ranging from 3.0% (Ireland) to 4.3% (Netherlands) of GDP. In terms of liquidity and guarantee measures, France, with measures worth around 8.6% of GDP, and the Netherlands, with measures worth 8.3% of GDP, stand out, followed by Switzerland (2.3%), Germany (2.1%), Luxembourg (1.1%), Belgium (0.8%) and Ireland (0.3%) (see Figure ‎6.2).

A comparative analysis of emergency measures shows that Luxembourg (along with Ireland, Belgium and Switzerland) stands out for the predominant focus on employment support measures, mainly in the form of short-time work (see Figure ‎6.3). Luxembourg’s decision to focus on maintaining employment from the early phases of the crisis seems appropriate given the country’s heavy reliance on cross-border workers and its need to retain as much of this workforce as possible. This decision may also have facilitated the economic recovery following the rebound in demand in the second half of 2020 (OECD, 2021[1]; IMF, 2021[2]).

In all the countries considered in this comparative analysis, public financing of short-time work has played a significant role in the schemes supporting businesses. In Luxembourg, short-time work accounted for 60% of the funds spent to support businesses, or 1.6% of GDP. The coverage rate of this scheme is among the highest, at 80% of gross wages, rising to 100% for wages at the level of the Social Minimum Wage. In Ireland, short-time work accounted for 70% of spending to support businesses, or 2.1% of GDP. In contrast, in Germany, which has one of the lowest replacement rates (60% of the benchmark net hourly wage), short-time work accounted for only 20% of spending to support businesses, or 0.7% of GDP (see Figure ‎6.4 and Figure ‎6.5).

Direct support includes subsidies or similar granted to companies and self-employed people. These measures represent recurring and important budgetary support mechanisms for businesses. They made up the largest share of support in Germany and France, accounting for 50% of spending allocated to help businesses in Germany (1.7% of GDP) and 44% in France (1.4% of GDP). In Luxembourg, this kind of support accounted for a smaller share of spending (16%, or 0.6% of GDP), as the employment component covered the initial shock of the crisis. However, as the second part of this chapter will show, more direct support was used in 2021 (see Figure ‎6.5).

In Luxembourg, tax deferrals accounted for approximately the same amount of spending as direct support (with the difference that deferrals represent temporary expenditure for the State). These measures include the waiver of quarterly tax advances, the deferral of tax due dates, the deferral of social security contributions and the refund of VAT credit balances.

With respect to liquidity and guarantee measures, State-guaranteed loans were the most widely used measure in most of the countries surveyed. In Luxembourg, however, such loans accounted for just 23% of total liquidity spending, or 0.3% of GDP. This type of measure accounted for almost all of the liquidity and guarantee provision in Switzerland, representing 2.3% of GDP. It accounted for more than 60% of spending in all countries surveyed, except for Luxembourg and Ireland (see Figure ‎6.6 and Figure ‎6.7).

While these measures have been widely used in all of the countries studied, the implementation mechanisms differed. Most countries, like Luxembourg, have guaranteed loans for a maximum of six years. Some countries have allowed a ten-year period in special circumstances (for example, small businesses experiencing serious cash flow problems in France). The interest rate varies according to the duration of the loan. In Luxembourg, it is 0.25% for a one-year maturity, 0.5% for three years and 1% for six years (see Figure ‎6.6).

Finally, most countries have put in place tax deferrals for businesses. This type of measure accounts for 60% of the liquidity and guarantee provision in Luxembourg, compared with 42% in Ireland. Deferrals can take the form of deferring direct tax payments (as in the Netherlands, Germany and Ireland), deferring social security contributions for businesses and self-employed people, or deferring VAT payments (as in Belgium) (see Figure ‎6.6).

In Luxembourg, support through advances for short-time work and ex-post controls helped provide a rapid response to businesses’ demand and needs, maintain people in employment, while limiting public expenditure. The digital tool used to process direct support for businesses also made it possible to respond quickly to applications and to grant assistance within a relatively short time. Speed of delivery has been recognised as a key factor in protecting the liquidity of businesses (Schivardi, 2020[3]). The fiscal procedures were flexible enough to allocate the necessary funds quickly, while keeping parliament informed.

In consultation with social partners, the Government introduced the “force majeure/COVID-19” short-time work programme on 18 March 2020. Unlike the system already in place, this special programme was based on advances paid to businesses in order to support them quickly. The existence of the Economic Committee (Comité de conjoncture) and its consultative framework for approving short-time work measures (see Box ‎6.3) facilitated consultation with social partners and a rapid response to applications from businesses.

Once a business’s application was accepted by the Economic Committee, the National Employment Agency (ADEM)2 paid the special advance. This meant that businesses were able to make use of this advance within ten days of the ADEM receiving their application. An ex-post accounting process made it possible to verify that the advances paid matched the actual short-time hours declared via the businesses’ accounts. A total of 20 000 applications were audited. The differences between the amounts actually owed to businesses and the advances received by them have been accounted for in benefit payments for subsequent financial years.

The use of this advance payment system with ex-post verification seems to have been particularly effective in responding very quickly to business needs while reducing the risk of wasteful use of resources. Similar systems were also implemented in other OECD member countries (OECD, 2022[4]). In Luxembourg, the risk of fraud seems to have been relatively contained: in 2021, the Ministry of Labour reported just six cases of fraud, either because the beneficiary business had gone bankrupt or because the beneficiary business received the advances while keeping the employees at work (Chamber of Deputies, 2021[5]; Jacquemot, 2021[6]).

Direct support took longer to be implemented than short-time work due to the lack of a pre-existing legal framework. The first lump-sum payment of EUR 5 000 to microenterprises on 25 March 2020 benefited only those businesses closed due to government decisions. This excluded about 6 000 businesses. Some of these businesses were not considered to be microenterprises as they fell below the qualifying thresholds. Others were affected by the sanitary restrictions, but they were not obliged to close. The Ministry of Economy’s Directorate for Small and Medium-Sized Enterprises (DGCM) decided to open up the subsequent programmes also to these companies.

At the beginning of the crisis, business support was not always appropriately targeted. Repayable advances and short-time work targeted sectors defined by the Statistical Classification of Economic Activities in the European Community (NACE), which is an economic activity classification system based on the business’s principal activity code (APE code). However, this classification appeared to be unsuitable for allocating support. By using this classification to determine which businesses should receive support, the public authorities excluded businesses that were highly dependant on a NACE sector directly affected by the crisis, even though they did not belong to it. This was particularly the case for business working on the organisation of events (sound engineers, communication managers, online corporate event organisers, and so on), which are not classified as belonging to the events sector and did not therefore receive any support. Yet the decline in activity in the sector probably had a similar impact on them. There was also the risk of excluding businesses whose APE code had not been updated but had in fact an activity affected by the crisis. The DGCM subsequently sought to avoid these pitfalls by asking businesses to describe the main activity in their applications for support. This approach helped improving targeting. The ministry’s teams then had to analyse these activity descriptions on a case-by-case basis.

Some self-employed people (around one in six) benefited from the first lump-sum payment of EUR 5 000 and were eligible for the repayable advances introduced on 3 April 2020. The second support package introduced by the DGCM on 8 April specifically targeted the self-employed, whether or not they had been required by the authorities to cease their activity. This allowed some self-employed people to receive EUR 2 500 directly into their bank accounts, which to some extent mitigated their ineligibility for short-time work. However, most self-employed people did not receive this support until May or even June 2020 (Figure ‎6.8). For the first time, all applications for this support to the self-employed were submitted entirely in digital format, thus requiring some time for the ministry’s teams to adapt to the digital tool. As a result, the DGCM was only able to start processing applications from 16 April. It is also possible that some eligible self-employed people were not immediately aware of this support. The fact that the number of applications was higher for the second and third lump-sum payments for the self-employed is consistent with this possibility. Some self-employed people may also have thought that they were ineligible for the first payment of EUR 5 000 (because they were not required by the authorities to cease their activity).

The analysis shows that providing support to the self-employed proved more difficult. This seems to be partly due to their special status, which excluded them from short-time work. In addition, short-time work benefited from an existing consultative framework with social partners, namely the Economic Committee. This was not the case for the implementation of direct support, including for the self-employed, for which consultation and dialogue took place in a more informal way. To remedy this situation, Luxembourg should, in the future, assess the needs of the self-employed, take their specific needs into account and consider strengthening recurrent support that is provided in the early stages of a crisis. It would also be important to establish a more formal framework for dialogue, including with representatives of the self-employed, when putting in place measures aimed at supporting business, in order to better target direct assistance and monitor the implementation of the measures.

The change in the number of businesses accessing support reflects the different waves of infection and the associated sanitary restrictions. The first wave corresponded to a very high use of support measures for a three-month period, followed by a lower level of use from July to October 2020 and then a slight upward trend from November 2020 to June 2021 (see Figure ‎6.9, Panel b). There were significant differences in the amounts spent during the first and second waves. These differences can be explained both by the changing number of beneficiary businesses and by the redefinition of some measures, in particular the Recovery and Solidarity Fund, which was stepped up in the second phase of the pandemic, responding to the need to revive the economy as the crisis was ending. The Recovery and Solidarity Fund also focused on those businesses that were hardest hit by the COVID-19 pandemic.

For short-time work, the funds paid out amounted to EUR 482 million for the first wave, and just EUR 81.5 million in the subsequent wave, before rising again in the third wave to EUR 280 million (see Figure ‎6.9, Panel a). A total of 33 032 businesses received these funds in the first wave, and 10 868 in the second wave. This decrease was primarily due to the difference in the intensity of the shock experienced by business during the first and second lockdown waves (see Figure ‎6.9, Panel b). According to STATEC data, following a 7.3% decline in activity in the second quarter of 2020, at the height of the crisis, Luxembourg was able to return to pre-crisis activity levels in the third quarter of 2020.

Conversely, the funds allocated to direct support almost doubled between the first and third periods, rising from EUR 60.9 million to EUR 125 million. This increase was not due to more beneficiary businesses (the number actually fell), but to a greater diversity of direct business support and an increased concentration of support for the hardest-hit sectors.

Digitalisation played a very important role in implementing direct support to business. The DGCM provided back-office management for processing applications using a software package known as MMAET. This software was operational less than a week after the first support measure was launched (25 March 2020) and its development was accelerated during the crisis.

The DGCM quickly went “all digital” for processing support applications. From the launch of the lump-sum payment for the self-employed (8 April 2020), applicants were required to use the professional area on the MyGuichet.lu website.3 This caused an initial delay in implementation, as applications could not be processed before 16 April. However, this approach delivered significant time savings for businesses and gains in effectiveness for the government.

In parallel, the MMAET software has been continuously improved since the beginning of the crisis, providing information about the number of applications submitted and the status of applications (in process, in payment, rejected, etc.). These figures are available in real time.

To respond to the many requests for information, a hotline and a dedicated e-mail address were set up. The DGCM updated a frequently asked questions (FAQ) document listing eligible and ineligible types of activity. This FAQ was updated and sent to employers’ associations daily. An e-mail address dedicated to the support for uncovered fixed costs and recovery support programmes was created. A hotline was provided by staff at the House of Entrepreneurship, a dedicated point of contact for entrepreneurship launched in 2016 by the Chamber of Commerce and the Ministry of Economy (House of Entrepreneurship, n.d.[7]). Getting the House of Entrepreneurship involved in responding to questions and providing more information about support measures not only offered answers which better met the needs of entrepreneurs but also allowed the DGCM to focus on processing applications.

Finally, the DGCM put in place a verification process to check the information supplied by businesses and avoid the risk of fraud. An internal audit was also carried out using sample checks.

Analysis of indicators relating to the applications received highlights the positive impact of these initiatives on application processing. For the initial EUR 5 000 support payment for microenterprises, of 14 145 applications, only 6 220 enterprises were approved (see Table ‎6.2). This large discrepancy is due to non-compliance with the eligibility criteria, mainly because of a misunderstanding of the criteria relating to the size of the business. The number of full-time equivalent employees used to determine eligibility was that of the group and not of the individual organisation or business. The rejection rate improved significantly following the second support payment of EUR 12 500, with 8 022 of 8 944 applications being accepted. In total, business support payments amounted to about EUR 76 million.

On average, companies waited 15 days to receive the second lump-sum payment of EUR 5 000 and EUR 12 500. To receive the amount requested, businesses had to wait an average of 15 days for recovery support and 30 days for support for uncovered fixed costs. Under the repayable advance programme, the average time taken to receive the support was 29 days (see Figure ‎6.10).

Over time, an increasing proportion of the applications received were approved. As regards lump-sum payments and repayable advances, two weeks after their introduction, almost all applications were being processed within seven days (see Figure ‎6.11). Support for uncovered fixed costs had the largest mean (and median) delay between application and receipt of payment. It is also the programme for which the DGCM took the longest time to deal with all applications. This is partly due to the number of steps required to validate an application and grant such a large amount of support (support under this measure was capped at EUR 300 000). In addition, businesses were required to send a number of documents and other items to receive the support. This resulted in a significant number of incomplete applications.

The improvement in processing times is also the result of the decrease in the number of applications received as the crisis eased (see Figure ‎6.12).

Crisis management has demonstrated the advantages of simplifying and digitalising administrative procedures so that support can be awarded quickly, protecting business liquidity and jobs. This process should continue.

  • Recommendation: the Ministry of Economy and the DGCM should conduct an in-depth analysis of the administrative procedures for businesses, especially small and medium-sized businesses, and seek to simplify and digitalise them. This process can build on the tools put in place during the crisis.

Luxembourgish businesses affected by the pandemic took advantage of the guaranteed loan programme, as soon as it was implemented. In the first three months of the programme’s implementation, 222 loans were granted, totalling EUR 95.6 million. As of 31 December 2021, the State Treasury had identified a total of 415 loans granted, totalling EUR 194 million. The amount guaranteed by the State is equivalent to some EUR 164.9 million (85% of the nominal amount of the loans). As the recovery progressed, the number of applications gradually declined, holding steady at relatively low levels during the first three quarters of 2021, underlining the resilience of Luxembourg’s economy to the pandemic (see Figure ‎6.13).

The State Treasury manages the day-to-day registry of guaranteed loans, including the approval and rejection of applications submitted by banks. The arrangements and practical implementation are formalised in an agreement signed between the State and the banks participating in the guaranteed loan programme (BCEE, BIL, Banque de Luxembourg, Banque Raiffeisen, BGL BNP Paribas, ING, Bank of China, Banque BCP).

In Luxembourg, beneficiary businesses have six years to repay the loans, as is the case in most countries (Anderson J., 2020[8]). As of 30 June 2022, less than 7% of loans were in default. Of these, only 28 loans were classified as "in default" for a total amount of EUR 12 781 800. It should be noted that “a loan in default" does not necessarily mean that bankruptcy proceedings have been initiated and that the State may have to cover all or part of funds loaned. So far, no State guarantee has been activated (Government of Luxembourg, 2022[9]).

Given the scale of the economic measures implemented in Luxembourg, the Government had to ensure that there was sufficient liquidity to disburse the funds quickly, and that it also had the authority to do so. A mechanism for proper monitoring of the commitment and disbursement of funds was also required.

To this end, from the beginning of the health crisis, Luxembourg ensured that it had sufficient liquidity to be able to disburse the sums linked to the exceptional measures adopted, and to absorb the drop in revenue associated with the crisis (for example, income tax and VAT revenue fell).

During the first two months of the crisis, Luxembourg was able to benefit from surplus liquidity thanks to a EUR 1.7 billion loan taken out in November 2019 to repay a loan maturing in May 2020. This situation gave the State Treasury valuable time amid great uncertainty and increased volatility on the markets due to the rise in international financing rates. This period of uncertainty did not last long in Europe, thanks to the intervention of the European Central Bank (ECB), which introduced the pandemic emergency purchase programme on 18 March 2020. The act passed on 18 April 2020 also allowed the Government to borrow an additional EUR 2.5 billion on 28 April 2020. All these circumstances meant that Luxembourg did not experience any liquidity problems.

In addition to having the money available to fund the measures, the Government also needed the authority to commit and disburse the funds. In principle, the funds that a Government can use are limited to the amount specified in the budget passed by Parliament. Any increase in these funds therefore requires Parliament to make changes to the budget. There is, however, one important exception to this rule in Luxembourg: items may be given an "open-ended" limit in exceptional cases. For these funds, the Minister of the Budget may grant an overrun at the request of the authorising minister. Each year, the budget contains several items that have open-ended limits for each ministry, with the aim of adding some flexibility to the annual budget in case of force majeure or an unforeseen economic downturn. In the case of the COVID-19 crisis, this exception allowed the Government of Luxembourg to commit and disburse the majority of the necessary funds in this flexible way, without having to go through Parliament. More than 80% of the additional COVID-related spending in 2020 was related to items with open-ended limits that were already included in the original 2020 budget (that Parliament passed in late 2019).

Nevertheless, new budget items still had to be adopted for some measures. An act on 24 July 2020 thus created a Recovery and Solidarity Fund for businesses, and another on 15 December 2020 established a legal basis for introducing a new open-ended item to fund COVID-19-associated health insurance expenditure. Most of the crisis-related expenditure already had a legal basis in the initial budget, however, and this facilitated a quick Government response. Unlike France or Germany, for example, Luxembourg did not have to introduce an amended budget act in Parliament.

This flexibility was accompanied by frequent communication between the Minister of Finance and Parliament. The Minister of Finance provided regular updates on changes in the State’s financial position at joint meetings of the Chamber of Deputies Finance and Budget Committee and Budget Oversight Committee. At these meetings, the Minister reported on the amounts disbursed in response to the COVID-19 pandemic. However, the information was limited to amounts. No details were provided on how the funds were allocated by measure or by type of beneficiary. For 2022, on the other hand, the Ministry of Finance took note of the recommendation issued by the Court of Auditors and produced a document featuring detailed information for each measure introduced in 2020. This document is currently in the process of being published. This process of discussion and providing information should continue. In the future, the Government of Luxembourg should continue such regular exchanges with the Chamber of Deputies on the beneficiaries, targeting and impact of the measures put in place, even in crisis situations.

Monitoring by the Ministry of Finance and the Inspectorate of Finance focused on the fiscal and financial aspects to ensure continuity of the fiscal system and sufficient liquidity. Since Luxembourg has not (yet) introduced performance budgeting, the fiscal authorities were not able to monitor the efficiency or effectiveness of the measures adopted. While such an analysis was not necessarily a priority in the acute phase of the crisis, a budgetary perspective that included performance data would have been useful in the context of a long crisis to determine whether certain measures should have been extended or better targeted. A performance-based approach could also have provided additional information to Parliament, thereby enhancing transparency. In general, performance budgeting would help improve the budget process in Luxembourg. The approach used by France can serve as an example in this area: annual performance reports attached to the 2020 budget accounts with objectives and performance indicators.

Cross-analysis of data on the financial position of businesses in Luxembourg and data on support to business demonstrates that the support measures in Luxembourg met the needs of most businesses (see Box ‎6.4). The businesses hit hardest by the crisis received the most support. Conversely, measures addressed to all businesses might have benefited businesses that could have weathered the crisis without the need for support, creating a windfall effect. A broad support package would have also heightened the risk of protecting businesses that would have gone bankrupt without the crisis. The available data suggest that this did not happen in Luxembourg. Public support benefited businesses that were in good shape in 2019 and were particularly hard hit by the crisis. The analysis of the impact of the support on businesses’ performance also shows that the financial position of most businesses has been protected.

The crisis affected almost all sectors of the economy (excluding the financial sector) and the sample data used in this study confirm that they all received support in 2020, i.e. at the peak of the crisis (see Figure ‎6.15 and Figure ‎6.16). The variation in the amounts disbursed in 2021 by sector reflects the evolution of the crisis. Aggregate amounts increased in 2021 for the HORECA sector and remained virtually stable for arts and entertainment (see Figure ‎6.15). This is due to the sectoral conditions for granting short-time work and the main direct support measures in 2021, as these sectors were considered the "vulnerable" sectors. While the amounts allocated to these two sectors did not decrease in 2021, the number of beneficiary businesses did (see Figure ‎6.16). The measures thus increasingly targeted vulnerable sectors on the one hand, and the hardest hit businesses within these sectors on the other.

In 2020, employment support measures, direct support and loans mainly benefited three service sectors, which together accounted for 57% of employment support, 70% of direct support and 64% of total loans: accommodation and food services (EUR 130 million in short-time work and leave, EUR 39 million in direct support and EUR 37 million in loans), construction (EUR 102 million, EUR 42 million and EUR 10 million), and trade (EUR 91 million, EUR 61 million and EUR 17 million).

After the sanitary restrictions were relaxed in 2021, the use of these measures became more targeted; the accommodation and food services sector accounted for 48% of short-time work and leave payments and 75% of direct support payments. The trade, construction, and HORECA sectors accounted for 64% of newly allocated State-guaranteed loan funds in 2020, rising to 73% in 2021 (see Figure ‎6.15 and Figure ‎6.16).

The impact of the pandemic and the support measures is reflected in the bankruptcy rate for each sector. Figure ‎6.17 shows the trend in bankruptcies since 2016, based on aggregated public data from STATEC. In 2020, the bankruptcy rate decreased in three sectors (out of seven available). However, HORECA is the only sector in which the bankruptcy rate fell for two consecutive years. This seems to confirm that this sector, which was particularly affected by the pandemic, was particularly targeted by the support measures.

The aim of the measures, as regulated by the European Union, was to target businesses that were healthy before the crisis but suffered a sharp decline in activity in 2020. According to the data collected, the businesses hardest hit by the COVID-19 crisis benefited more from the support measures. This suggests that the support was relatively well targeted. Regardless of the measure concerned, the rate of use increased with the scale of impact on the activity. This relationship becomes even clearer in 2021 (see Figure ‎6.18).

The first four deciles of hardest-hit businesses in terms of reduced turnover in 2020 received 54% of the total support paid out over the period considered. This is in line with the objective of supporting the sectors most exposed to the crisis, and was reinforced in 2021, when the top two deciles received 48% of the total funds allocated (see Figure ‎6.18).

The data indicates a threshold effect for direct support, some of which was awarded on condition of a reduction in turnover of at least 25% compared to the same month of the previous year.4 Between the third and fourth decile, the total amount of direct support disbursed decreases more than the reduction in turnover. Adjustments to support measures on the basis of a progressive reduction in turnover could have avoided this effect. This approach would have helped businesses that were less affected because they adapted better than others. To remedy this situation, all new direct support measures could include a subsidy proportionate to the reduction in turnover to help target businesses more precisely.

In two years, businesses with fewer than 50 employees received 74% of the total amount granted under all support measures (see Figure ‎6.19). Nevertheless, different beneficiary profiles can be distinguished according to the measures examined. With respect to employment support, the amount of support was divided equally across businesses. Conversely, more than half of direct support went to microenterprises (fewer than ten employees) in 2020. These smaller firms were generally harder hit during the crisis in OECD Member countries (OECD, 2020[10]). The Government of Luxembourg introduced several lump-sum payments specifically for very small businesses, which explains this concentration in 2020. In contrast, direct support was no longer so focused on microenterprises the following year (36%). In 2021, recovery support and support for uncovered fixed costs, which were not conditional on business size, accounted for most of the direct support.

Linking data from the Central Balance Sheet Office with administrative data on support measures provides information on the pre-crisis financial position of beneficiary businesses. This information is important in determining whether support measures helped businesses that were already failing before the crisis. The amount of support received by companies in 2020 and 2021 can be broken down by decile of profitability and corporate debt. The upper deciles of the profitability indicators reflect a better financial position, while the upper deciles of the debt indicator include businesses with high financial costs.

In 2020, the rate of use of loans, direct support and employment support by the least profitable firms (first decile of the profitability indicator) and the most profitable firms (last decile of the profitability indicator) was below 7%. Businesses in the fourth decile received 23% of repayable advances and State-guaranteed loans, while those in the fifth decile received 18%. By 2021, the distribution of recipients was more even, although the deciles at either extreme continued to receive less support. Nevertheless, the use of loans shifted in favour of businesses in the sixth decile, which received 20% of loans compared with 11% in 2020 (see Figure ‎6.20).

In terms of the debt indicator, businesses in the sixth and seventh deciles made the most use of loans. The use of the different support measures according to debt level once again reveals a bell curve, with businesses in very good or poor financial health making less use of support measures (see Figure ‎6.21).

Overall, disbursements were concentrated among businesses whose financial health before the crisis was average. Businesses in poor or excellent health received less support overall. This remains true to varying degrees regardless of the type of support analysed and regardless of the financial health indicator used, suggesting that firms with very high or very low debt levels received less support overall, and that the support measures therefore largely protected healthy businesses that needed assistance to survive the COVID-19 crisis.

The health restrictions put in place in Luxembourg from March 2020 to curb the rise in COVID-19 infections hit businesses hard. In some sectors, businesses were forced to deal with a sharp decline in their activity. Some businesses had to reduce their costs and increase their debt levels to meet their payments. This had a significant impact on their financial position. One measure of the impact of the support masures is therefore to estimate the extent to which the support received by businesses actually protected their financial position.

This exercise presents many difficulties. First, data on the financial position of businesses combine the impact of the crisis and the impact of the support received by companies. Furthermore, it is not possible to estimate the impact of the measures by comparing over time (before and after the COVID--19 crisis) the financial position of businesses that received support with similar businesses that did not receive support (for example by using a difference-in-difference estimation), since most businesses in each sector (excluding the financial sector) benefited from support, as shown above (see Figure ‎6.15).

To overcome this problem, a model has been used to simulate the income statement and balance sheet of companies under three scenarios: 1) no health crisis; 2) a health crisis without financial support from the government; and 3) a health crisis with financial support from the government (see Box ‎6.5).

The use of this analytical tool makes it possible to draw some conclusions about the impact of support measures. In general, they have helped most beneficiary businesses to reduce their losses and diminish the risk of bankruptcy. In some sectors, such as HORECA, the microsimulation seems to indicate a particularly significant impact, confirming that this sector has benefited from substantial support.

It is important to keep in mind, however, that this is a simplified accounting framework, which does not take into account the interactions between businesses; inter-firm credit, for example, may have been significantly affected when activity declined (Hadjibeyli, Roulleau and Bauer, 2021[11]). The results are based on assumptions about the businesses’ cost structures and how these changed during the crisis, as described in Box ‎6.5. In addition, further assumptions have been made due to the lack of a clear definition of how businesses may have accounted for support in their balance sheets during the COVID-19 crisis, and these assumptions may affect the results (see Box ‎6.5).

The contribution of employment support programmes, business subsidies and State-guaranteed loans to earnings before interest, taxes, depreciation and amortisation (EBITDA) varies by sector. The support measures have helped to mitigate the decline in EBITDA. With the support measures, the arts and entertainment, other services, and accommodation and food services sectors benefited from an increase in EBITDA of more than 60% compared to a scenario without any support (see Figure ‎6.22). This suggests that these sectors would have experienced the most significant decline in EBITDA without support.

The microsimulation suggests that the crisis and the support measures had a particularly strong impact on sectors affected by the health restrictions, such as the HORECA or events sectors. Businesses directly impacted by the health crisis benefited from unprecedented public support, which made it possible to sustain their activity by improving their liquidity and solvency. The Government's financial support measures therefore limited cash flow losses and the accumulation of debt by businesses (see Figure ‎6.23).

Comparing the growth rates of median income by sector in the different COVID-19 scenarios with the “no COVID” scenario also confirms the effectiveness of the support for all beneficiary sectors. In line with the previous results, the HORECA and other services sectors (associations, personal services, repair of computers and personal and household goods) benefited the most from the Government's intervention, with support measures having a positive effect on income compared with a scenario in which there was no crisis or intervention.

Without drawing any causal conclusion, the results produced by the linear probability model show that some businesses, classified according to pre-crisis variables, are more likely to become vulnerable due to the crisis. These businesses include those in the trade, HORECA and administrative and support service activities sectors, as well as medium and small businesses, which are at greater risk of illiquidity or insolvency compared to large firms (Table ‎6.3, model 1).

As expected, businesses with more assets or liquidity, or lower pre-crisis debt levels, are less likely to experience difficulties during the crisis. However, businesses with higher productivity are more affected by the crisis. This finding is consistent with the results of a study of the impact of guaranteed loan programmes on business productivity in 14 OECD Member countries (Demmou et al., 2021[13]).

These results are confirmed when a COVID-19 shock is added to the model as a control variable (Table ‎6.3, model 1), with the exception of the coefficient for the HORECA and administrative and support service activities sectors, which becomes statistically non-significant. This suggests that these sectors were the hardest hit by the crisis, experiencing a sharp decline in turnover and therefore seems to confirm the priority given to these sectors and firms’ sizes in the support measures.

When the same model is estimated using data recorded for the pre-crisis period (2017-2018-2019), the results are similar. However, it is interesting to note that in this case, unlike the results estimated by models 1 and 2, more productive firms have a lower probability of becoming insolvent or illiquid. This finding suggests that during the COVID-19 crisis, healthier companies also faced the risk of becoming insolvent or illiquid.

It will be necessary to continue to monitor certain sectors, such as HORECA, and the small businesses that were hardest hit during the COVID crisis, as new crises are emerging, and some sectors will be affected by structural changes due to the spread of remote working and videoconferencing. The risks of illiquidity and insolvency remain, and so targeted support measures will be required. In the future, Luxembourg should set up a monitoring unit at the DGCM to track the financial position of businesses and put in place targeted support in consultation with social partners. The unit should also monitor the impact of the crisis on the self-employed.

Existing databases will also need to be better integrated and their use facilitated to support monitoring of sectors experiencing economic difficulties. While ensuring that data remains confidential, the economic ministries should be given the ability to access business performance data, allowing them to target support measures if necessary. It would therefore be appropriate to involve STATEC in establishing the DGCM monitoring unit and facilitate the use of business data to target support measures.

  • Take better account of the self-employed: it will be important to assess the needs of the self-employed and take their needs into account, including by creating a recurrent support measure at the very onset of a crisis, or by including the self-employed under the short-time work scheme.

  • Monitor the hardest-hit sectors in the short term to ensure their resilience to crises: this could be done by establishing a monitoring unit at the DGCM to track businesses’ financial position and put in place targeted support measures in consultation with social partners. The unit should also monitor the impact of the crisis on the self-employed.

  • Better integrate and facilitate access to businesses’ administrative and balance sheet data: to support monitoring of sectors experiencing difficulties, existing databases will need to be better integrated and their use facilitated. Involving STATEC in establishing the DGCM monitoring unit will be important to facilitate the use of business data to target measures.

  • Consider some degree of progressivity for access support to better target any future intervention: eligibility criteria for direct support created a threshold effect. Adding some degree of progressivity would have helped to improve targeting. If necessary, all new direct support measures could include support proportionate to the reduction in turnover to help target businesses more precisely.

  • Consider setting up a consultation structure with social partners for all future interventions: it will be important to set up a more formal framework facilitating dialogue with social partners for the implementation of support measures. This framework could help improve the targeting of direct support and monitor its implementation. Such a framework could build on the monitoring unit recommended above and also include representatives of the self-employed.

  • Continue to simplify and digitalise administrative procedures: the process of digitalising administrative practices proved valuable during the crisis and should continue. The Ministry of Economy and the DGCM could conduct an in-depth analysis of the administrative procedures concerning businesses, especially small and medium-sized businesses, to simplify and digitalise them. This effort could build on the tools put in place during the crisis.

  • Continue to support Parliament’s information-sharing and control role: this will require informing the Chamber of Deputies about the beneficiaries, targeting and impact of the measures implemented, even in crisis situations, and planning to provide this type of information for all future interventions.

  • Continue a regular exchange with the Chamber of Deputies on the beneficiaries, targeting and impact of the measures implemented, even in crisis situations, and plan to provide this type of information for any future interventions.

  • Continue monitoring and evaluating business support measures: more evaluations of the impact of the support measures, including on business performance, could be conducted. For this purpose, it will be important to involve all public authorities to evaluate the relevance of each of the measures for investment, the sustainability of businesses and job creation, considering all support measures put in place.


[8] Anderson J., P. (2020), Government-guaranteed bank lending in Europe: Beyond the headline numbers, https://www.piie.com/blogs/realtime-economic-issues-watch/government-guaranteed-bank-lending-europe-beyond-headline.

[5] Chamber of Deputies (2021), Question parlementaires QP3766: Fraude au chômage partiel (02-04-2021), https://chd.lu/wps/portal/public/Accueil/TravailALaChambre/Recherche/RoleDesAffaires?action=doQuestpaDetails&id=20894.

[13] Demmou, L. et al. (2021), “Insolvency and debt overhang following the COVID-19 outbreak: Assessment of risks and policy responses”, OECD Economics Department Working Papers, No. 1651, OECD Publishing, Paris, https://doi.org/10.1787/747a8226-en.

[14] Gourinchas, P. et al. (2020), “Estimating SME Failures in Real Time: An Application to the COVID-19 Crisis”, NBER Working Paper, Vol. 27877, https://doi.org/10.3386/w27877.

[9] Government of Luxembourg (2022), Rapport sur l’application du régime de garantie en faveur de l’économie luxembourgeoise dans le cadre de la pandémie Covid-19. Situation au 30 juin 2022, https://te.public.lu/content/dam/tresorerie/fr/garanties/analyse-regime-de-prets-garantis-covid-30-juin-2022.pdf.

[15] Guerini, M. et al. (2020), “Dynamique des défaillances d’entreprises en France et crise de la Covid-19”, OFCE Policy Brief, Vol. 76, https://www.ofce.sciences-po.fr/pdf/pbrief/2020/OFCEpbrief73.pdf.

[11] Hadjibeyli, B., G. Roulleau and A. Bauer (2021), “L’impact de la pandémie de COVID-19 sur les entreprises françaises”, Direction générale du Trésor -Trésor-Éco, Vol. 282.

[7] House of Entrepreneurship (n.d.), Nos missions, https://www.houseofentrepreneurship.lu/nos-missions/ (accessed on 13 June 202).

[2] IMF (2021), Luxembourg: Staff Report for the 2021 Article IV Consultation: IMF Country Report, No. 21/93, International Monetary Fund, https://www.imf.org/en/Publications/CR/Issues/2021/05/25/Luxembourg-2021-Article-IV-Consultation-Press-Release-Staff-Report-and-Statement-by-the-50188.

[6] Jacquemot, P. (2021), Six plaintes pour fraude au chômage partiel, https://www.wort.lu/fr/luxembourg/six-plaintes-pour-fraude-au-chomage-partiel-6092749cde135b923657e963.

[4] OECD (2022), “First lessons from government evaluations of COVID-19 responses: A synthesis”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/483507d6-en.

[1] OECD (2021), OECD Economic Outlook, Volume 2021 Issue 2, OECD Publishing, Paris, https://doi.org/10.1787/09bf9e01-fr.

[10] OECD (2020), “Coronavirus (COVID-19): SME policy responses”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/04440101-en.

[3] Schivardi, F. (2020), A simple method to estimate firms liquidity needs during the Covid-19 crisis, Presentation at a OECD GFP Seminar, 5 June 2020, https://www.oecd.org/global-forum-productivity/webinars/SchivardiLiquidityOECD.pdf.

[12] STATEC (2021), “La situation économique au Luxembourg Évolution récente et perspectives”, Note de conjoncture Nr 2-2021, https://statistiques.public.lu/dam-assets/catalogue-publications/note-conjoncture/2021/ndc-02-21.pdf.


← 1. An exception to this rule was made during the period of sanitary restrictions in December 2021 and January 2022, when shops and personal care businesses received the support automatically.

← 2. The ADEM operates under the supervision of the Ministry of Labour, Employment and the Social and Solidarity Economy. Its remit includes registering unemployed people as jobseekers, determining their rights to benefit payments, paying their benefits and helping them access the measures it puts in place to support their return to the labour market.

← 3. The online form available on the website was developed by the Ministry of Economy with support from the Government IT Centre.

← 4. To receive support from the Recovery Fund, businesses had to register a reduction of at least 25% in turnover compared to the previous year. For support for uncovered fixed costs, turnover generally had to have fallen by at least 40% compared to the previous year.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2022

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.