Designing short-time work for mass use

Abstract

In the course of the Covid-19 pandemic, short-time work as an instrument of income replacement once again proved to be an effective means of stabilising employment. However, the very concept, based on individual entitlement, led to its operational limits in respect of mass use. For example, in Germany, the complete processing of all cases in multiple stages can take years, involving corresponding strains and uncertainties for firms and labour administration. Against this background, we discuss the development of variants of job retention schemes compatible with mass use. An international comparison indicates that the legal instrument of force majeure could facilitate access with simplified criteria and procedures. We elaborate on specific proposals for the well-known German system. Going beyond simplifying existing rules, we outline a collective instrument of a wage subsidy increasing with lost revenue or hours. In this respect, drawing on results from the relevant literature, we argue that the need to limit redundancies and the precision of the instrument must be carefully balanced. Particularly in the case of mass use, qualification is indispensable, which is why the need for a concept with flexibly applicable, modular and online-based training formats, incentives and counselling services is essential. Finally, preconditions for the phasing-out of the mass use scheme are outlined. The exceptional situation would have to be officially ended – or extended – at an appropriate time with sufficient notice. Subsequent schemes may provide for transition to regular arrangements, a gradual reduction of wage subsidies, and liquidity support.

1. Introduction

Whether in the wake of the global financial crisis or more recently, since 2020 in the wake of the Covid pandemic, the concept of Kurzarbeit, hereafter referred to as short-time work, has aimed at income replacement in cases exogenous events or general economic development leading to the reduced working hours. In doing so, it subsidises the hours not worked due to these circumstances. This makes short-time work a crucial means of social security. In several countries, such as Germany, short-time allowances are a benefit of the social unemployment insurance.

Short-time work therefore serves to safeguard jobs and to avoid unemployment. Thus, it resembles, in its targeting and effect, the furlough schemes and other temporary unemployment schemes used in other countries, which focus on forced suspended labour periods (in contrast to temporary unemployment between contracts for specific fixed time periods, e.g. in platform work (Schoukens, 2020)). However, short-time work does not provide for a compulsory total reduction of hours worked (Hijzen et al., 2021). Indeed, short-time work schemes can take different forms according to the permissible range of working time reduction, that is, unrestricted, only partial or only full. Short-time allowance is an established instrument that has already proven its worth and also across the different variants in safeguarding employment (e.g. Aiyar and Dao, 2021; Balleer et al., 2016; Brey and Hertweck, 2016; Giupponi et al., 2022; Hijzen and Martin, 2013). Even during the Covid crisis, short-time schems were able to significantly limit the rise in unemployment in the face of the immense force of the economic shock.

However, in times of major crisis, such as the recent Covid-19 pandemic, it has been observed that the instrument of short-time work has reached the limits of its processability due to exceptionally high numbers of cases and fluctuating use. This is a problem that threatens to undermine the effectiveness of the otherwise successful instrument. In many countries such as Germany, short-time allowance is designed based on complex criteria and processes and on individual entitlement. While it is initially the companies that must notify the authority about short-time work, later, the settlements and claims must be checked in each individual case. Processing takes years and an enormous amount of staff. The required operating personnel data may even no longer be available at the time of review. Additionally, this entails tedious uncertainty for those affected. When the instrument was originally designed, no one even imagined millions of cases. As a result, many companies across all sectors gained extensive experience with the instrument, so that the use of short-time work is likely to remain extensive in future crises (cf. Boeri and Brücker, 2011). This is also supported by the fact that the labour market will be characterised by scarcity in the foreseeable future, which increases the willingness to retain workers (Klinger and Weber, 2020). In the event of energy rationing, there would be high demand for a short-time work instrument.

For this reason, our contribution seeks to outline provisions for the design of short-time work for mass use. We discuss the strengths and weaknesses of short-time work, especially from the viewpoint of the German scheme, and propose reform options for mass use. For this purpose, we refer to international examples. While short-time work in general is well-studied, we focus on extraordinary circumstances. Based on surveying the relevant literature and drawing international comparisons, we work out both the potential for simplifying short-time work regulations and procedures or organising it in a collective way, and the restrictions and incentives that are crucial for its effectiveness.

The first ideas for such a concept can be found in a mass influx Directive designed by the EU, which was introduced as a result of the Ukraine war. In March 2022, the EU activated the Temporary Protection Directive (2001/55/EC) for immigration from Ukraine. This is intended to allow people to obtain temporary protection status in the EU without the usual asylum review on a case-by-case basis. The Directive was developed for circumstances in which a large number of people leave their home country for the same generally recognised reason. It reduces waiting times and uncertainty for those affected, and it also cuts down on the workload for the administrations of the receiving states. The backlog in the processing of asylum applications in 2015 and 2016 is an abiding memory.

A similar mass phenomenon occurred for short-time work from the onset of the Covid pandemic. In many countries, an extremely large number of employees were forced into short-time work schemes for the same reason within a very short period of time. The OECD estimates that at the height of the crisis around 60 million jobs were protected by such measures (roughly ten times as many as during the global financial crisis), which corresponds to a 20% share of dependent employment (Hijzen et al., 2021: 108). In Germany, at the peak in 2020, the number of people in such work was around six million. In this context, and to cope with the influx of eligible claimants, the provisions for short-time work schemes were facilitated in some countries by the declaration of force majeure (act of God).

Therefore, international comparisons of the development and reforms of similar instruments at European level and beyond will serve as possible starting points for the reasoning for the simplification of short-time work schemes and the development of a collective instrument as a further option. The concept of force majeure is normally used for facilitating access to short-time work in respect of the situation of the firm. Beyond that, we propose further simplifications with regard to the process and the level of individual entitlement. Thus, the activation of force majeure would result in a collective variant of short-time work. Here, a wage subsidy would apply, calculated on the basis of total hours lost verified by lost revenue. However, layoffs would have to be excluded in order to preserve the benefits of direct job retention while simplifying matters considerably. For the sake of preserving flexibility and allowing for certain structural adjustments, a basic level of layoffs could be made possible or the subsidy could be reduced proportionately in the case of redundancies due to operational reasons. In view of the lower precision, the more far-reaching variant may be considered if there are extensive economic shortfalls in certain sectors. Finally, this variant will be distinguished from other economic assistance and weighed against potential other forms of short-time work schemes.

For the sake of clarity, the provisions for mass use presented in this contribution serve to regulate short-time work in exceptional situations. They are not intended to replace the traditional short-time allowance. Therefore, we will also outline the conditions under which a return to the regular scheme should take place.

The article is structured as follows: section 2 presents the current situation in Germany and the corresponding definition of the issue. Section 3 offers an international comparison of European countries and the development of the basic approach. Section 4 shows how the principle of force majeure could be implemented in the German context. A collective variant is discussed in section 5. Section 6 is a brief appeal for the integration of further education into the concept. Finally, section 7 outlines the exit scenario and section 8 serves as a conclusion.

2. Mass use of short-time work under complex rules

Here, we focus on German Kurzarbeit as a typical scheme of individual wage replacement in case of a reduction in working hours. The criteria for cyclical short-time work in Germany are complex. The following only roughly summarises Sections 95 et seq. Social Security Code III (SGB III). Entitlement to short-time wage replacement exists in case of a considerable reduction in working hours in connection with a loss of pay if this has been reported to the employment agency and a number of other conditions are met. The reduction in working hours must be caused by a general economic development or an unusual unavoidable event, must be temporary and unavoidable, and at least one third of the staff in the company must be affected by a loss of income of over 10% each in the respective calendar month. Except for a number of cases, holidays and working time bonuses must be taken primarily. Employment must continue after the start of the reduction in working hours or be taken up for compelling reasons or after the end of training, and the employment relationship must not have been terminated. The possibility of reducing working hours and wage replacement must be specified in a labour law agreement. The net wage replacement rate is 60% (or 67% when children live in the household). The process of short-time work processing goes through the stages of notification, application or settlement and payment, and eventually, the final review.

Hence, during the pandemic an instrument designed on an individual basis with complex criteria was being used by many. In this respect, the question that must be asked is: if a large number of people want to use the instrument of a short-time work scheme for the same generally recognised reason, could we design a collective short-time work instrument effective to this end, in the sense of the Temporary Protection Directive, on a situational and temporary basis? Should this amount to facilitating short-time work provisions, or to what extent would direct economic assistance be more appropriate as an instrument in the case of a comprehensive impact, for instance, on certain sectors?

3. International approaches

It might be worth looking abroad to find suitable starting points for the design and procedure of a directive for mass use. Similar provisions for short-time work schemes can be found in many countries. Ever since the pandemic, they have been used intensively. Information on this has been compiled for an international data set as part of the IAB project ‘The labour market effects of the design of short-time work rules’, which aims to determine the effects of access requirements, wage replacement rates and entitlement periods. Object will be general procedures or also procedures that were introduced as special provisions in special cases, such as the Covid crisis. The focus of the underlying article is not on a comprehensive comparative analysis. In the following we provide a concise overview of the European country cases to which the subsequent discussion will refer. These are: Belgium, the Czech Republic, France, the Netherlands, Italy, Spain, and Switzerland. All of these countries have in common that they already had the instrument of a short-time work scheme at their disposal prior to the pandemic, although to different degrees of implementation and use.

In March 2020 Belgium introduced an extended scheme that was much more flexible than the usual one in terms of proving economic difficulties, since all temporary unemployment due to the pandemic was automatically derived from force majeure – even if it was still possible to work on some days (Eurofound, 2020c). Nevertheless, firms were instructed to retain the relevant receipts for a possible future review – an exhaustive review procedure was not applicable (cf. Landesamt für Arbeitsbeschaffung, 2022a; Serroyen, 2021). This was mainly to ease the situation of companies whose employees were in quarantine but could not work remotely or who had difficulties organising childcare. Officials of public institutions, the self-employed and many people on temporary labour contracts were, however, excepted (cf. Serroyen, 2021).

The Czech Republic set up a completely new programme tailored to the situation in the pandemic. The so-called ‘Antivirus programme’ started as a temporary short-time work scheme, open to all employees, reimbursing employers a large part of their costs when employees were not able to work due to isolation, quarantine or other restrictions put in place by the authorities. The same applied if the employer faced economic difficulties for the above-mentioned reasons. In July 2021 this transitioned into a permanent scheme. After a tripartite consultation the government can now launch a short-time work period that supports employment threatened by natural disasters, cyberattacks, an epidemic or an economic crisis (cf. Drahokoupil, 2021).

Initially, France opted to introduce specific emergency measures to its already-existing ‘Activité Partielle’ (AP) scheme, such as the retrospective application of the short-time work regulation for up to 30 days from the first reduction in hours, an extension of the maximum duration from six to 12 months, and a significant reduction in the notification period from 15 days to two days (following which applications for the AP would be automatically accepted) (cf. Ministère du Travail, du Plein emploi et de l'Insertion, 2020; OECD, 2020). In July 2020 a further instrument was introduced (cf. Vincent, 2021). The Activité partielle de longue durée (APLD) aims to support firms with a prolonged reduction in activities. With the help of this, working time may be reduced even further (up to 40%, and in exceptional circumstances up to 50%) on condition that the employment level is maintained. Its maximum duration is 24 months, which may be split up into periods of six months. After frequent modifications to the job retention scheme almost all categories of employees in France were eligible during the pandemic, no matter if they worked part-time or full-time, in private sectors or in state-owned enterprises.

The Netherlands chose a quite different path. Instead of modifying the existing short-time work scheme it opted for a temporary wage subsidy that did not have the reduction in working hours as its object (cf. OECD, 2020). Hence, employers had to continue to pay their employees the full usual wage, after which they would receive a subsidy proportional to the reduction in sales, up to a maximum of 90%. All firms that suffered a certain revenue loss over a given period were eligible. Employees on permanent or flexible contracts (even on zero-hour or on-call contracts) could benefit from the scheme. Layoffs were forbidden during the period covered by the allowance (cf. Eurofound, 2020a).

Italy extended its pre-existing STW scheme to such a degree that all firms of any size from all sectors and all regions could apply and all employees could benefit. Firms had only to state that they have been negatively affected by the Covid crisis within four months of the start of their reduced activity, and no further evidence had to be provided (cf. OECD, 2020). Employers still had to inform and consult the unions, but they were released from co-financing obligations for the hours not worked. For this, the short-time work was funded entirely by the state (cf. Faioli and Bologna, 2021).

In Spain, different types of extraordinary Expediente Temporal de Regulación de Empleo (ERTE), also fully funded by the state, were implemented in order to tackle the economic and social impact of the Covid pandemic. All firms of every sector severely affected by the crisis, or firms directly subjected to administrative regulations implemented to fight the virus, such as shutdowns, were eligible. No categories of employees were excluded. For these firms, administrative procedures were simplified (cf. Cruces Aguilera, 2021). Instead of the usual 15-day mandatory consultation period with worker representatives there was only a notification period of five days. The labour authority simply inspected briefly whether the presented case fulfilled the requirements of force majeure (cf. Eurofound, 2020b) – this principle will become crucial in this paper in a moment.

In the course of the pandemic, Switzerland also facilitated the application for and processing of short-time work. Prior notification and waiting periods were abolished, the maximum period of entitlement was increased to 18 and 24 months respectively, and the categories of eligible employees was also extended to atypical employees, such as fixed-term employees, apprentices and on-call workers (cf. Bundesrat, 2020a, 2020b, 2021a, 2021b). Even before the pandemic, a minimum reduction in working hours of 10% for all eligible employees in a firm (or part of it) was a prerequisite for access (cf. SECO, 2022). However, the Covid regulations also allowed for other more collective approaches. Not only did individual unreduced overtime and interim employment not have to be factored out for each individual employee (cf. SECO, 2020), but a summary accounting procedure was introduced (cf. Bundesrat, 2021c). In this simplified procedure, the employer only had to inform the unemployment insurance fund of the total wage amounts of all eligible employees with the percentage of work lost. However, this information, as well as the information on the scheduled hours and the hours lost for economic reasons, had to be verifiable.

Faioli and Bologna (2021) are right to point out, in their article, that the state funding was supported by the European instrument for temporary Support to mitigate Unemployment Risks in an Emergency (SURE). This instrument is available to any EU Member State in order to finance short-time work schemes and other similar measures of job retention due to the Covid crisis. SURE is financed via social bonds and up to EUR100 billion may be provided to Member States. According to its website, ¹ the Council has approved a total of EUR98.4 billion in support, already disbursed to Member States. The figures for the countries discussed here, provided that SURE was used, are as follows: Belgium, EUR8.197 billion; Spain, EUR21.324 billion; Italy, EUR27.438 billion; and the Czech Republic, EUR4.5 billion.

After this short excursion we will draw conclusions from the presented cases for our analysis. Some European countries, for instance, France, Italy, Belgium, Spain and the Czech Republic, were united in their approach to the pandemic as a trigger for a state of emergency. The pandemic was considered a force majeure (OECD, 2020), that is, an act of nature beyond control that affects the economy and the labour market to be tackled with short-time work schemes. On the one hand, as has become clear in the overview, this led to an expansion or simplification of already existing statutory options, such as in the case of France, Italy or Belgium; and on the other hand, as already described, specific regulations were launched which, as in the case of the Czech Republic, became the blueprint for a new generally applicable instrument.

In all cases, this state of emergency of force majeure is defined and declared by the government, partly in consultation with the social partners, and thus defined as a framework for the subsequent possibilities of applying for short-time work. For this to be possible, an attempt is made beforehand to define the nature of such an exceptional situation and how it can be recognised. In summary, these are particularly serious situations of economic crisis or situations beyond human agency that affect economic activity, such as environmental disasters. In France, force majeure is described as an epidemic, disaster or storm of an exceptional nature (Ministère de l'Économie, des Finances et de la Souveraineté industrielle et numerique, 2020). In the Czech Republic, force majeure, which the general short-time work scheme implemented in the course of the pandemic is based on, is defined as a threat to employment in connection with a natural disaster, a cyberattack, an epidemic or an economic crisis (cf. Drahokoupil, 2021). The Belgian provision, on the other hand, defines force majeure somewhat more universally as the occurrence of a sudden and unforeseeable event that was not brought about deliberately, which makes it impossible to carry out work (cf. Landesamt für Arbeitsbeschaffung, 2022b). This covers a broad spectrum from company destruction due to fire and power cuts to a pandemic and company closures, as well as staff quarantine periods.

Force majeure – and this is key to a mass use directive – comes with an absolute validity claim, provides room for action in times of great challenge, and yet does not simply undermine existing law and customary procedures without a sense of proportion. This effectiveness of the principle becomes clear when one briefly recalls the case studies of the countries under consideration. The example of France shows that force majeure facilitates access to an already-implemented instrument of short-time work and rearranges the process of notification and application. A similar principle also applies Spain, where access to assistance is additionally facilitated by the accompanying definition of sectors that are particularly affected and, therefore, worthy of support (cf. Cruces Aguilera, 2021). The actual admission examination is facilitated as well. The approach typically pursued in France and Italy to legitimise short-time work through collective agreements instead of examining individual cases seems promising. More evidence of the crisis situation is not necessary in combination with force majeure and its specified case of validity, which is why the formal fulfilment of the collective process alone and the principles and declarations recorded in the contracts are examined (cf. Faioli and Bologna, 2021; Ministère du Travail, du Plein emploi et de l'Insertion, 2020; OECD, 2020). In Belgium, one does not have to verify the actual economic difficulties of the company under force majeure, since the pandemic was generally considered, as stated in our overview, to be the cause of short-time work in that event.

Regarding the question of how a firm's losses to be compensated are defined, the European cases examined with the principle of force majeure show a tendency towards extending the possibility of access to short-time work with simultaneous settlement per affected employee or individual reduction of hours (cf. Eurofound, 2022, entries on Italy, Spain, Belgium and France). This becomes particularly clear and understandable in the case of France. The onset of force majeure in the form of the pandemic allowed an application of short-time work to atypical employment relationships as well. However, this was associated with a change from the payment of a lump sum per employee to an amount proportional to the remuneration of employees on short-time work (Eurofound, 2022).

So far, results of evaluations of the impact of the instruments during the Covid crisis are only available for some cases. For Spain, Congregado et al. (2022) draw a positive conclusion to the short-term effects on employment. Albertini et al. (2022) note that French short-time work schemes have stabilised employment and consumption, however, with significant bandwagon effects. The latter should be considered in light of minimal preconditions combined with a high level of performance (the benefits for workers amounted to 100% for those on minimum wages or less. For other wage levels they were at times up to 70% of the original gross wage; subsidies for employers have been up to 4.5 times the minimum wage; cf. Vincent 2021). Windfall effects already significantly increased the cost per job saved of the French short-time work scheme in the Great Recession (Cahuc et al., 2021). Deadweight effects were also found for Germany (Boeri and Brücker, 2011). The issue of bandwagon effects is particularly evident in all countries where the costs of hours not worked are not co-financed by firms, including Belgium and Italy (OECD, 2022). This design of the instrument fuels the danger that more employees benefit from the instrument of short-time work than necessary. Permanently unprofitable jobs, in particular, may thus be secured at high financial cost, which could undermine the anticipated upswing.

Overall, however, a positive summary can once again made with regard to the use of short-time work. For example, a Europe-wide evaluation by the University of Konstanz in 2016, which evaluated short-time work provisions during the financial crisis between 2007 and 2009, shows that this instrument is most effective when corresponding provisions exist in a country and GDP growth is strongly negative. Hence, short-time work is most effective at the beginning of a recession and when it can serve as a quick-acting automatic stabiliser (cf. Brey and Hertweck, 2016). In light of the heavy use of short-time work schemes, especially in the early stages of the pandemic, this was obviously also demonstrated in the recent crisis. As a matter of fact, job retention instruments were used by all European countries regardless of the nature of their welfare state (cf. Ebbinghaus and Lehner, 2022).

4. Force majeure as the basis of a mass use provision for short-time work

Insofar as a special provision for mass influx is desired, the conclusions could be as follows. Force majeure could be defined along the lines of the international examples discussed. In accordance with force majeure, the federal government would have to specify a generally recognised incident with exceptional and serious effects to activate the mass use directive. This would remove the requirement for the standard case-by-case review as a guarantee of appropriateness of use. Therefore, the bar would have to be set high. Normal, limited recessions (cf. Hausner and Weber, 2017) would not come into consideration. It could therefore be argued that the effects would need to go well beyond such a recession, at least for parts of the economy. Among other things, current leading indicators and forecasts could be used for this purpose. The involvement of consulting social partners and science representatives would be conceivable. For instance, a straightforward application could be presented in the event of a possible energy rationing.

One approach to setting the bar high could be by involving common European decisions. Force majeure could then be activated with a sufficiently broad consensus on the exceptional significance of an event. Clear and efficient rules of procedure would be necessary in order to guarantee fast decision-making. A positive side effect would be that a European alignment would manifest a common basis on which support measures could be taken in line with EU competition regulations.

Another argument in favour of such an approach is that an extension of short-time work provisions only has an additional positive effect in particularly severe crises (Gehrke and Hochmuth, 2021), and bandwagon effects can be limited by focusing on companies with heavy losses (Cahuc et al., 2021). Moreover, as in Spain or Italy, applications could be limited to certain industries and, if necessary, fields of activity (or, if relevant, regions or enterprises directly affected by ordered measures such as business closures) if this is in the nature of the situation. In the remaining sectors, the established short-time work scheme would continue to apply. This is because blanket measures increase the risk of bandwagon effects, are less cost-effective and can prevent productivity-enhancing reallocation (Di Nola et al., 2022). The German labour market and those of many other states will probably be characterised by shortages for the foreseeable future. Consequently, instruments that finance the loss of labour should be handled in a restrictive manner.

In the event of energy shortages, an obvious delimitation would be limited to those establishments directly rationed on the amount of gas used, for instance. While this criterion has the advantage of clarity, its scope in the economy would be very limited. In addition, this could include sectors that are highly dependent on the directly affected energy-intensive industries in supply chains. This may involve procuring energy-intensive products, that is, supply chain disruptions, or suppliers or service providers losing business as a result of energy rationing. In principle, these criteria could also be applied to individual establishments, but a definition according to industries would create clarity in advance and avoid individual review efforts. The full crisis effects would also go beyond this delimitation. Thus, an energy crisis would lead to further sharp increases in energy prices, which would additionally drive inflation. Labour-intensive service industries, such as the hospitality industry, would also be affected by the corresponding loss of purchasing power (cf. Wolter et al., 2022).

In existing applications, the concept of force majeure usually facilitates eligibility at the firm level by alleviating the need for an economic justification of short-time work. We adopt this principle and furthermore propose comprehensive simplifications of the process of application, processing and verification. Furthermore, in the subsequent section, in an additional step we apply the force majeure principle also at the level of individual entitlement, developing a collective variant of short-time work.

In the case of an activation of force majeure, the fact that the establishment is affected by a temporary unavoidable loss of working hours for economic reasons or as a result of an unavoidable event would no longer have to be examined on a case-by-case basis. Nevertheless, bandwagon effects must be limited as effectively as possible. It is conceivable (as in Belgium) that enterprises would be required to keep documents for proof of appropriate use, but these may only need to be checked on an ad hoc basis (e.g. in the case of irregularities or previous complaints) or randomly. Plausibility could already be estimated to a very large extent in a standardised way using revenue data, which are also available promptly in respect of value-added tax (VAT). ² In obvious and verified cases – for example, a government decision such as business closures in a pandemic or rationing in a gas supply freeze – the short-time work notification could be omitted, because the business would not need any further confirmation of access to short-time allowance in advance. Below the threshold of legal changes, a more detailed examination of the notification could at least be dispensed with in such cases. If a negative decision is not received by a certain deadline, this could count as confirmation. In addition, a positive initial application, which already includes the necessary business documents, could (except in cases of suspected benefit fraud) already be considered as a final confirmation of payment. The criterion of a minimum level of exposure of the enterprise (one third of the staff with a loss of 10%) could be omitted, since, on the basis of assumptions, only very severely affected sectors would be considered anyway. Holidays and working time bonuses could be disregarded, as was temporarily decided in the Covid crisis by amendment. A refund of social security contributions and an increase in the wage replacement rate may be available as options. Remanence costs basically limit bandwagon effects; however, expansions that reduce remanence costs can have positive employment effects, especially in severe crises, when it is also particularly a matter of securing liquidity (Gehrke and Hochmuth, 2021).

These provisions facilitate the criteria. Importantly, they can also streamline and speed up the procedure as one or two stages in the three-step process could be saved. While these provisions lead to a simplification, the loss of working hours of individual employees would still be considered. An even more far-reaching option could be based on flat-rate variables such as revenue loss, as in the Netherlands or Australia, for example, or collective hours loss, as already illustrated for Switzerland. This would make it more similar to direct economic assistance. When thinking in such a direction, however, one should first understand why short-time work schemes are so effective.

5. Variant of a collective instrument: securing employment and accuracy are essential

One important reason for the effectiveness of short-time allowances is that by supporting temporary working time reductions they provide incentives to avoid layoffs and thus draw on a flow variable. That directly influences decisions relevant to the labour market and the economy (e.g. Faia et al., 2013). The fact that a loss of working hours and certain remanence costs (Bach and Spitznagel, 2009) have to be accepted to avail of short-time allowances also strongly limits bandwagon effects. These advantages are indicative of the basic principle of short-time work schemes as compared to direct economic assistance. The same applies in respect of the fact that the establishment can plan with short-time allowance schemes, other than with discretionary economic assistance. This increases the willingness to hire even before crises (Balleer et al., 2016). Additionally, workers are less able to diversify risks than the capital side and are therefore more risk-averse. That is why stabilising employment has greater welfare effects than stabilising revenue. Consequently, aggregate demand is supported particularly effectively by short-time allowance schemes, as the unemployed cut back their consumption to a greater extent (Aiyar and Dao, 2021, Dengler and Gehrke, 2021). Similarly, the risks of consolidated unemployment (Klinger and Weber, 2016) and of taking up low-value jobs when under pressure (‘sullying’) (Barlevy, 2002) do not arise in the first place if job losses are deliberately avoided. This preserves the perceived value of the established matches of person and workplace. Costs and delays in dismissals and rehiring are avoided.

Therefore, a more far-reaching option of the mass use directive should at least retain the direct avoidance of redundancies as a principle. A collective (i.e. firm-based) subsidy would be conceivable. Thus, the activation of force majeure would also apply at the level of individual entitlement, resulting in a collective variant of short-time work. The subsidy should be granted as a percentage of wage costs, because a direct revenue loss refund would actually increase the incentives for redundancies (e.g. Hentze, 2020). After all, lost revenue would be replaced even if wage costs are saved through layoffs. When countries introduced wage subsidies in the pandemic, they mostly relied on proportionate capped support (ILO, 2020). Such subsidies are more efficient at targeting support than flat-rate schemes, avoid paying more than a recipient's normal wage and ensure that workers retain a level of salary commensurate with their productivity (Cassells and Duncan, 2020). The subsidy could be granted from a certain level of revenue losses, and the percentage would increase with the size of the loss. Thus, the share of total wage costs subsidised would be directly proportional to the share of lost revenues. In fact, this would be equivalent to subsidising a percentage of the wage costs connected to the lost share of business. Indeed, apart from inventory changes, taxable revenues and hours are likely to be highly correlated (e.g. in Germany, the correlation coefficient for all sectors is 0.99 from 2017 until 2020). By default, the percentage would be equal to the replacement rate of existing short-time work (e.g. 60% in Germany).

While the subsidy would operate at the firm level, the firm would compensate individual hours lost. For bilateral regulation of working time reductions and wage replacement between employer and employees, one would resort to established labour law agreements. Thus, just as in standard short-time work schemes, based on collective bargaining or employment contracts, the employees concerned would receive wage compensation in the same amount for their hours lost. This guarantees legitimacy of the adjustments just as acceptance and codetermination.

In this variant, the revenue loss would serve as the final criterion for the subsidy. This would have the advantage of very high simplicity because only revenue losses and total wage costs would have to be verified. Instead, in a modification, a collective loss of hours could be considered as the relevant measure. While conventional short-time work benefits are often calculated via a replacement rate for the individual net wage, here one could refer to the overall gross wage and deduct taxes and social security contributions at a flat rate. The loss of hours initially stated could be verified against the revenue loss as an efficient plausibility check once the data are available.

Automated data reconciliation with the tax authorities would be possible, as with Covid economic aid. Accordingly, the labour administration and the ministries of finance and economics could cooperate on implementation. Employment status and earnings could be determined from continuously available data (as in Australia; Hamilton, 2020) from social security payments or payroll taxes.

In view of bandwagon effects, full cost reimbursement should be avoided, that is, both employers and employees would bear limited shares of the loss (cf. e.g. Albertini et al., 2022). Also, the support should be conditional on the companies refraining from making redundancies during this time and for some time afterwards (with the exception of dismissals for personal and behavioural reasons according to Section 1 Para. 2 German Employment Protection Act). Such protection against dismissal during individual short-time work exists in France, Italy and Spain, for example, while in the US during the pandemic, Paycheck Protection Program loans were forgiven in proportion with the number of jobs retained. In Germany, according to directives, when short-time work is used overall employment in the enterprise must be predominantly maintained. A simultaneous use of established individual short-time work would be excluded if the subsidies were used. This variant would result in a radical simplification, but the fundamental advantages of an instrument for employment stabilisation and the prevention of unemployment would be maintained. It would guarantee high flexibility for management decisions in a crisis and provide stronger incentives for firms to keep hours worked up and to increase them quickly when conditions improve (OECD, 2020).

Obviously, there would also be disadvantages. For instance, it would lead to a loss of flexibility if the use of short-time work ruled out dismissals for operational reasons throughout the establishment. Options would be to allow a certain basic level of dismissals or to reduce the subsidy proportionally in case of dismissals for operational reasons. This would maintain some flexibility when it is no longer reasonable and necessary, respectively, to maintain certain matches because of the consequences of the crisis. The instrument would lose accuracy if all employees of a company were treated summarily, because reductions in working hours can differ significantly depending on the type of job. Short-time work schemes based on the lost working hours of individual employees allow establishments to self-select by considering only employees whose jobs would be at risk, and also only the relevant working hours (Cahuc et al., 2021). This highlights that a more far-reaching variant should only be considered if there is a widespread and profound breakdown of economic activity in specific industries. Differentiating between employees would become less important in a situation like that. That said, we stress again the point that under a collective variant, reductions in working hours would also be determined bilaterally between employer and employees. That is, accuracy and flexibility would not be affected in this respect, but a collective regulation would considerably simplify the instrument of public support.

If the subsidy was treated as operating revenue, part of it would be returned to the state via taxes if a profit is made despite the drop in revenue. In net terms, therefore, there is automatically a differentiation according to profit or loss – especially if a higher tax rate is applied to the subsidy. An early separate determination and inclusion of data on the profit situation would not be necessary.

Economic assistance independent of the principle of short-time work or job retention can additionally take effect if further measures beyond the replacement of labour costs are required to safeguard liquidity in a crisis. One example is the interim aid to offset fixed costs introduced in Germany during the Covid crisis. Simultaneous use of short-time work and possible other aids that cover personnel costs would be excluded. Targeted economic assistance is also the more appropriate instrument when it comes to avoiding rather than absorbing production losses (Weber, 2022). This is the case, for example, when costs skyrocket in an energy crisis. Here, continuing energy-intensive production would be possible but may no longer be economically viable. However, a shutdown would entail shortfalls in supply at subsequent economic stages and would risk irreversible effects. It is feasible to already create a conceptual framework for action for such cases, even if no actual standard instrument is created. Finally, economic assistance may be preferred to activating a mass use rule for short-time work in situations that require serious adjustments and make securing employment in the affected sectors no longer a sensible aim (cf. Giupponi et al., 2022). Admittedly, such an assessment is unlikely to be possible in advance (but may, as discussed below, play a role when exiting a crisis). Even in the contact-intensive service industries, which experienced extensive and lengthy downtimes during the pandemic, the demand for labour is now so high again that bottlenecks are occurring, and that despite the fact that layoffs have already been successfully limited (cf. Weber and Röttger, 2021). Furthermore, short-time work rates during the Covid crisis have declined rapidly after their peaks in the lockdowns. After the Great Recession, Aiyar and Dao (2021) have found no evidence of unusual increases in misallocation for Germany. All in all, the risks of foregoing an instrument of employment protection would be immense in view of the main advantages of short-time work programmes previously discussed.

6. Combining short-time work and qualification

The Achilles heel of short-time work is that it finances stagnation, while recessions, in particular, are times of upheaval. This can be counteracted with a sensible combination of qualification and short-time work, which explains, for example, the special efforts of France in the field of adult learning during short-time work schemes (cf. Hijzen et al., 2021). This would become even more important in the case of collective use of short-time work.

The previously rare use of periods of short-time work schemes for training is mainly explained by uncertainty and organisational difficulties of fitting into the continuously adjusted work schedule (Bellmann et al., 2020). Therefore, a concept with flexibly applicable, modular and online training formats, incentives and counselling services is needed (OECD, 2020, Weber, 2021). Such qualification programmes should be developed and provided with the support of the federal employment agency together with providers of further education. Similarly, conditions for public funding, such as minimum periods for training measures etc., must be constructed in a sufficiently flexible way for the event of a crisis. The positive wage and employment effects of qualifications can return a considerable part of the costs of short-time work into public budgets via tax and contribution revenues (Kruppe et al., 2020).

7. Exit rules

Not only are regulations for activation important for such an instrument suitable for mass influx, but also those for exiting the crisis. The risk of delaying necessary labour market adjustments must be taken into account. The employment balance of short-time work can deteriorate sharply if it continues beyond the crisis (Hijzen and Martin, 2013). However, stopping too early can endanger the retained jobs and liquidity of the companies involved (Hamilton, 2020). The exceptional situation would have to be terminated officially at an appropriate time. If activation is accurately targeted, a differentiation by sector, for example, should not be necessary when exiting. However, this should remain an option, also because crises can be asymmetrical for different sectors. For instance, in France, more generous short-time work provisions for particularly affected industries remained in force longer during the Covid crisis. Decisions on the expiry or extension of schemes should be made with sufficient advance notice to ensure plannability. Objective criteria as a guideline can reduce uncertainty while ensuring that the exceptional situation applies for as long as necessary but is not overstretched. This could be based on the fact that the revenue development in the affected sectors is sufficiently similar to that of the pre-crisis level (e.g. over 90%); also, leading indicators, such as orders, could be included. It would also make sense to assume that by this time, the situation has normalised to such an extent that the use of the mass use directive has sufficiently decreased. When it comes to government orders such as business closures or rationing, their revocation could also play a role.

Irrespective of such criteria, the application would have to be phased out in a regulated manner when the crisis has a lasting impact so that the pre-crisis level is no longer a relevant benchmark for the affected sectors. Of course, this cannot be proven definitively in the ongoing process without long-term observations. With one to two years of experience, it should be possible to make an assessment based on information such as bankruptcies or closures, redundancies, development of demand and market changes. However, these are likely to be exceptional cases. The export economy, for instance, recovered after the severe slump in 2009, and the Covid pandemic shows a similar pattern, too. While the crisis dragged on for a long time due to the many waves of infections, the rapid recovery of the hospitality industry, for example, after the lifting of respective lockdowns indicated that the business model would fundamentally meet with strong demand again.

New short-time work measures would only be possible under normal provisions in the event of expiry of the mass use directive. A follow-up provision should be made for the remaining existing cases. In the case of wage subsidies, a gradual reduction should be considered (e.g. Hijzen et al., 2021). In the variant of individual lost working hours, at least the use of short-time work schemes according to normal rules would have to be made possible over a period remaining until the expiry of the maximum period of eligibility in total. Slovakia, for example, switched to a German-style short-time work system as of March 2022, having previously had a wage subsidy model in place during the Covid crisis when sales slumped. The provisions should be coordinated in such a way that transitions are as smooth as possible if the additional requirements of short-time work are fulfilled. A transitional phase in which existing cases can still be continued according to mass influx rules would be conceivable. In addition, economic assistance could be used to facilitate the exit from short-time work schemes for firms with temporary but still persisting financial difficulties. Access to a credit programme, for instance, would help to prevent establishments from remaining in short-time work schemes for too long due to liquidity bottlenecks.

8. Conclusion

In principle, it would be conceivable to introduce such a simplified variant of a short-time work scheme, which is geared to large numbers of cases, only when a crisis situation relevant to it occurs. However, complex conceptual decisions would have to be made under significant time pressure, then pass through the legislative process, and the new measures would have to be implemented immediately. One strength of the short-time allowance measure is, however, that it is prepared, immediately available and known in advance. Economic assistance during the Covid crisis, on the other hand, which was introduced ad hoc, was naturally subject to delays, ambiguities, and adjustments. Also, other countries, while being able to scale up a new job retention scheme at the beginning of the crisis, had to compromise on its design and could not build on existing political consensus (Hijzen et al., 2021). This speaks in favour of preparing the provisions in advance. It also clarifies the important role of a force majeure regulation, which defines the conditions for a mass use regime transparently in advance.

However, it is important to limit activation of a mass use directive to exceptional events so that even in situations of political pressure, no use is made for less serious situations. In France and Belgium, for instance, the Ukraine war and the associated bottlenecks have already been explicitly defined as another case of force majeure. In view of the stable labour market situation in Germany, at least so far, that alone would not be a sufficient trigger in the sense of the instrument discussed here. However, provisions should already be made for the event of a serious aggravation of the energy crisis.

A mass use directive for short-time work could simplify existing rules and procedures or install a fully collective instrument. These two general lines of thought may be relevant for the systems of many countries, and we elaborated on both variants. The two directions span a space of reform possibilities and thus are not mutually exclusive. For example, a collective instrument could also benefit from simplifications proposed before.

In serious crisis situations, the use of a mass use directive could considerably relieve the burden on all parties involved and greatly reduce waiting times and uncertainty for those affected. It is important to design simplifications in such a way that they safeguard employment and avoid abuse and excessive bandwagon effects as far as possible. The examples and considerations in this article are intended to provide indications for such a design of labour market policy provisions.