Germany’s short-time work scheme: can its past success be replicated?
By Alexandra Effenberger, Michael Koelle and Andrew Barker, OECD Economics Department
Germany has avoided a large jump in unemployment in the early stages of the COVID-19 recession, just as it did during the global financial crisis (GFC). One important factor in this success is the well-established short-time work (STW) scheme (Kurzarbeit), whereby the government subsidises wage payments for employees whose hours are cut at companies in temporary distress. Countries such as Austria, Switzerland and Italy have similar established schemes, while others such as the UK, Denmark and Latvia have just recently implemented job retention schemes. A number of questions are pertinent for countries looking to learn from the German experience: how many jobs have been saved by STW, what are the fiscal costs and how much does the scheme’s success depend on the specific institutional settings in Germany?
In its current form the scheme functions as an automatic stabiliser, as companies are generally eligible to use STW if they face a major drop in activity for economic reasons or due to extraordinary events, provided the drop is temporary and unavoidable. The application process is fairly streamlined. After notifying the labour agency of the intent and demonstrating the need to use STW, firms are flexible in the actual take-up in terms of both workers and individual work-time reductions. Similar to previous economic crises such as the GFC, exceptional measures to extend the scheme during 2020 could be readily implemented including:
- The eligibility threshold for the share of workers affected by a lack of demand within a firm has been lowered from one-third to 10%, and temporary agency workers made eligible.
- The labour agency reimburses 100% of social security contributions for lost working hours (usually covered by the employer).
- To shield workers from large earnings losses, the replacement rate of lost net earnings is increased from 60% for childless workers and 67% for workers with children to 70% and 77% from the fourth month if they have reduced their working time by at least 50%. In the seventh month, payments are increased further to 80% and 87%.
- Restrictions on taking part-time jobs while on STW have been lifted. The additional earnings are not credited against short-time working benefits as long as the total income does not exceed the previous income.
The key benefit of STW is moderation of the increase in unemployment during a downturn, with immediate payoffs through avoiding the large wellbeing and scarring costs of unemployment for workers and medium-term gains from maintaining viable job matches. Countries with large STW or job retention schemes have seen substantial take-up of these schemes alongside much smaller increases in unemployment during the current crisis (Figure 1). STW is estimated to have saved in the order of up to half a million German jobs during the GFC when at its peak around 1.4 million workers were in STW (Hijzen and Martin, 2013; Boeri and Brücker, 2011). Based on the number of notifications the labour agency estimates that 6 million individuals were in short-time work in April 2020. The flipside is that by subsidising existing positions, STW can impede labour reallocation, reduce the probability that those without a secure job find work and slow job growth during the recovery (Cahuc, 2019; Hijzen and Martin, 2013).
Will STW be an efficient instrument in the current crisis?
It remains unclear how well STW will perform in the current crisis, but it will be best-suited if there is a relatively rapid economic rebound without substantial changes to the sectoral composition of the economy (Schwellnus, Koelle and Stadler, 2020). STW was well-suited to dealing with the downturn in Germany during the financial crisis as the sectoral composition of the economy changed little. No major industry grouping gained or lost more than 1 percentage point in its share of labour between 2008 and 2011, with manufacturing experiencing the biggest decline from 19% of hours worked to 18.2% (OECD, 2020)
Certainly, the current crisis hit the economy faster and more broadly, as many stores and services were required to shut down completely. This is also visible from the concentration of STW notifications and movements into unemployment in those most affected sectors (Figure 2). However, in contrast to the GFC the decline in business activity has been determined by confinement and social distancing rather than pre-existing differences in firm performance, which is mirrored in the so far much greater reliance on STW rather than lay-offs (Weber and Gehrke, 2020). This suggests that the risk of locking workers in unviable jobs through the use of STW might be less pronounced. Sectoral shifts might be avoided again if the lifting of confinement measures continues and consumers’ demand picks up quickly. In addition, Germany entered the current crisis with very low unemployment and a high degree of labour shortages making firms reluctant to lay off valuable workers. Nevertheless, sectors such as tourism, hospitality and aviation might be subject to longer-term interruptions. A fall in demand during the downturn might accelerate structural changes in some industries, such as automotive manufacturing. As a result, reallocation of labour might become more crucial. In this case, relying on STW for an extensive period might be less efficient than implementing measures focusing more directly on job creation.
Some current additional measures increase the efficiency of the scheme. For example, expansion of STW to allow part-time working in a second job is positive as it facilitates reallocation towards sectors with a temporarily higher labour demand. On the other hand, deadweight effects might increase with the full reimbursement of social-security contributions as it makes it less costly for firms to hold on to workers even if their job is not viable and they will be laid off eventually (Crimmann, Wießner and Bellmann, 2010).
How much does it cost?
The fiscal costs from STW are generally low relative to the number of jobs saved. The costs to the labour agency of saving around half a million jobs during the GFC amounted to about EUR 4.6 billion in 2009 and EUR 3.1 billion in 2010 (Bundesagentur für Arbeit, 2011). Based on the estimate of 6 million workers in STW in April 2020, labour agency budget estimates suggest costs could have been running at almost 2% of GDP recently, with the increase in unemployment payments roughly a degree of magnitude smaller. By comparison, increases in unemployment benefits under expanded coverage were costing US governments around 5% of GDP in early May (including substantial increases in payment generosity) and the Israeli government just over 3% per cent of GDP in April.
The generosity of German STW payments is aligned with unemployment benefits, but may cost the government more upfront if it pays for reductions in hours for workers who would have been kept on anyway. Conversely, STW offers fiscal benefits because the government saves on the cost of helping those who become unemployed to find work. Compared with job-saving alternatives, STW schemes such as the German one have a lower cost per job (Cahuc, Kramarz and Nevoux, 2018). By allowing for a partial hours adjustment, the fiscal cost per worker covered is lower than job retention schemes that subsidise workers who stop work altogether. They are also better targeted than wage subsidies.
Can Germany’s successes with STW be readily replicated?
There are some institutional settings that favour working time adjustments in Germany and may limit applicability to other countries. Stringent rules about layoffs make STW more pertinent to firms. Germany, like some other countries with extensive STW programmes such as France and Italy, ranks in the top 10 OECD countries for protection of permanent workers against dismissal (OECD, 2020). Protection includes notice periods, during which the regular salary would have to be paid, and in some cases severance payments. Total costs averaged across different tenures reach almost 22 weeks of salary, among the highest in the OECD (World Bank, 2020).
Similarly, firing and re-hiring is more costly for firms that require specific qualifications such as technology-intensive production. At the time of the GFC, the turnover costs for low-skilled workers were around EUR 7000 in Germany whereas they were almost five times that high for qualified workers (Bach and Spitznagel, 2009). OECD research shows that technology- and skill-intensity is positively related to labour-hoarding tendencies during economic downturns (OECD, 2010). In this context, the relative importance of technology- and skill-intensive manufacturing in Germany may help to explain the modest response of employment to the decline in output during crises.
In addition, internal flexibility measures such as working hour accounts and reductions in weekly hours or overtime play an important role in working time adjustments in Germany and help cushion the effects of cyclical downturns. Such internal flexibility measures are often covered through collective bargaining agreements or agreed between the social partners, which might not be easily transferred to other countries.
What’s to be learned?
In sum, the German experience shows that an established, flexible and quickly accessible STW scheme can help reduce the labour market effects of an economic crisis and save viable job matches when the downturn is short-lived, the sectoral composition of the economy is not affected, and costly firing and hiring by firms might slow the recovery. For the German case, it remains to be seen whether COVID-19 will trigger the need for substantial reallocation of labour, undermining the benefits of keeping employees in their current jobs. Moreover, any such scheme needs to be tailored to country-specific institutional settings and the eligibility for different types of workers and their share in the economy have to be carefully considered. For example, even when they are eligible for STW it might still be easier for firms to lay off workers on fixed-term contracts, which across European OECD countries on average constitute around 8% of all workers in sectors hit heavily by the COVID-19 crisis (OECD, 2020). During the GFC, take-up of STW in Germany was lower in firms with a higher share of fixed-term contracts (Boeri and Brücker, 2011). Similarly, the self-employed cannot use STW and might need different support programmes. Many countries including Germany have established special funds for the self-employed. Finally, as STW is targeted at keeping incumbent workers in employment, it needs to be recognised that potential new hires such as young graduates and those currently unemployed do not benefit and may lose out if subsidised jobs hinder their own entry into the labour market. Rates of transition from unemployment to employment during April and May 2020 were the lowest ever recorded for those months, worse even than as unemployment approached its 2009 peak during the GFC (Bundesagentur für Arbeit, 2020).
Bach, H. and E. Spitznagel (2009), “Kurzarbeit: Betriebe zahlen mit – und haben was”, IAB-Kurzbericht No. 17/2009. 
Boeri, T. and H. Brücker (2011), “Short-Time Work Benefits Revisited: Some Lessons from the Great Recession”, Economic Policy, Vol. 26/68, pp. 697-765. 
Bundesagentur für Arbeit (2020), “Der Arbeitsmarkt im April 2020”, Press release No. 27 April 30, https://www.arbeitsagentur.de/presse/2020-27-der-arbeitsmarkt-im-april-2020. 
Bundesagentur für Arbeit (2020), Monatsbericht zum Arbeits- und Ausbildungsmarkt Mai 2020. 
Bundesagentur für Arbeit (2011), Aktiv aus der Krise – Geschäftsbericht 2010. 
Cahuc, P. (2019), “Short-Time Work Compensation Schemes and Employment”, IZA World of Labor, Vol. 11/2, pp. 1-11. 
Cahuc, P., F. Kramarz and S. Nevoux (2018), “When Short-Time Work Works”, Banque de France Working Paper No. 692. 
Crimmann, A., F. Wießner and L. Bellmann (2010), The German work-sharing scheme: An instrument for the crisis. 
Hijzen, A. and S. Martin (2013), “The Role of Short-Time Work Schemes during the Global Financial Crisis and Early Recovery: A Cross-Country Analysis”, IZA Journal of Labor Policy, Vol. 2/5, pp. 1-31. 
OECD (2020), “Distributional Risks Associated with Non-Standard Work: Stylised Facts and Policy Considerations”, Tackling Coronavirus Series.  see also: https://oecdecoscope.blog/2020/06/19/policy-responses-to-covid-19-no-worker-should-be-left-behind/
OECD (2020), OECD Employment database, https://stats.oecd.org/Index.aspx?DataSetCode=EPL_OV. 
OECD (2020), OECD National Accounts database, https://stats.oecd.org/Index.aspx?DataSetCode=NAAG. 
OECD (2010), Employment Outlook 2010, OECD Publishing, Paris. 
Schwellnus, C., M. Koelle and B. Stadler (2020), “Flattening the unemployment curve? Policies to support workers’ income and promote a speedy labour market recovery”.  See also: https://oecdecoscope.blog/2020/06/17/flattening-the-unemployment-curve-policies-to-support-workers-income-and-promote-a-speedy-labour-market-recovery/
Weber, E. and B. Gehrke (2020), Kurzarbeit, Entlassungen, Neueinstellungen: Wie sich die Corona-Krise von der Finanzkrise 2009 unterscheidet. 
World Bank (2020), Doing Business 2020: Employing Workers, https://www.doingbusiness.org/en/data/exploretopics/employing-workers.