The spread of the coronavirus has plunged the German economy into the worst recession in its post-war history. The corona crisis has impacted the German economy and labour market much more heavily than the financial crisis of 2009: Significantly more sectors are affected today than in 2009. Nevertheless, the number of layoffs has so far been limited compared to the number of workers on short-time work. The fact that, to date, all sectors have relied much more on short-time work than on layoffs shows that the majority of companies have – thus far – been willing to keep their staff.

What is the short-time work allowance?

The short-time work allowance (Kurzarbeit) is intended to support companies in retaining their employees during economically weak times. The German Federal Employment Agency pays a wage replacement for lost working time. Currently, the compensation is 70 (77) percent from the fourth month onward and 80 (87) percent from the seventh month onward. Short-time work can be applied for by companies in the event of a temporary, substantial and unavoidable loss of work – provided that the employees or their representatives have given their consent. In the ongoing crisis, the use of short-time work has been prolonged for up to 24 months and the additional measures that have been taken to make the use of short-time work more attractive were extended until 2021.

Germany was seen as a pioneer in the use of short-time work during the great recession of 2009, which was triggered by the financial crisis. At its peak, there were over 1.4 million short-time workers. Various studies have suggested that short-time work was instrumental in mitigating the recession’s contribution to rising unemployment: Overall, the negative impact on the German labour market was indeed much more moderate than in other countries.

In view of this positive experience, the European Union has now launched a programme (“Support to mitigate Unemployment Risks in an Emergency”) to expand short-time work in other member states during the corona crisis. For more on the use of short-time work in a broader European context, see also “Short-time work in Europe: Rescue in the coronavirus crisis? An interview with IAB researcher Regina Konle-Seidl”.

In the 2009 financial crisis, short-time work was mainly used in the manufacturing sector

Due to a global drop in demand, the global financial crisis in Germany primarily affected the export industry. Short-time work was therefore also concentrated in this sector. This time, in addition to manufacturing, many other sectors are affected by the corona crisis, especially public services. Hitherto, there had been little experience with short-time work in the non-manufacturing sectors; the extent to which it is now utilised in these sectors will be a decisive factor in determining to what extent employment levels in Germany drop substantially.

That is why the IAB has analysed which industries have registered for short-time work since the beginning of the corona crisis, and for how many employees, and compared the results with the situation during the 2009 financial crisis.

Currently, short-time work is being registered much more frequently than during the 2009 financial crisis

Figure 1 shows the number of employees for whom short-time work was registered beginning with the first lockdown, and also for whom it was actually implemented during both recessions, as far as data were available as of May 2020. It is clear that the number of registrations in March and April this year shot up much more rapidly than at the beginning of the financial crisis. In these two months alone, short-time work was reported for around 10.1 million employees. This is more than three times the reported level throughout 2009 (3.3 million people). Note: The monthly number of people on short-time work usually exceeds the monthly number of registrations after a certain period of time. This is mainly due to the fact that a notification only has to be made once, even if a person is on short-time work for several months subsequently.

In 2009, due to the relatively rapid economic recovery in Germany, an average of 40.3 percent of short-time workers were on the scheme for less than three months; 68.3 percent for less than six months. But even in 2009, the number of short-time workers, which stood at around 1.3 million in spring, only fell below the million threshold towards the end of the year – despite the rapid economic recovery. The number of people on short time work during the corona crisis and the duration of its use will depend heavily on the length of the recession.

An industry-specific comparison of the two recessions is also enlightening (see Figure 2). It is striking that short-time working has already been registered for significantly more employees in every industry during the corona crisis than in the first half of 2009.

The maximum value can be identified in the hospitality industry (accommodation and gastronomy), where already 93 percent of all employees have been registered for short-time work. This sector is particularly affected by the state-prescribed containment measures. The same applies to the “other services; art, entertainment and recreation; private household services”, where nearly 43 percent of all employees have been registered for short-time work.

On the other hand, the manufacturing sector and “other economic service providers” (including temporary workers) are generally affected more indirectly by the corona-containment measures. Rather, supply difficulties and (worldwide) loss of demand have had a strong impact on these sectors. Traditionally, these industries react quickly to economic downturns, for example by cutting their numbers of temporary workers. An exception are travel agencies and tour operators, which are also part of the “other economic service providers”, but, like the hospitality industry, were directly affected by the containment measures. As a result, short-time work notifications have already been received for around 40 percent of those employed in the manufacturing sector and 38 percent of those employed by “other economic service providers”.

The “transport and storage” sectors are also severely affected with 32.2 percent of workers on short-time work; construction with 30.1 percent, real estate and “freelance, scientific and technical services ”with 30.0 percent and trade with 28.7 percent. These sectors contain numerous professions for which working from home is difficult or barely possible.

During the 2009 recession, which mainly affected the export-oriented industry, workers in the manufacturing sector were by far the most likely to be on short-time work. Within this sector, registrations for short-time work were made for 36.6 percent of employees in the first half of 2009. However, this value is still below the value that was already achieved in March and April of this year – i.e. within two months – for the manufacturing sector. In 2009 it was 10.7 percent in traffic and storage, 10.6 percent in “other economic service providers” and 8.8 percent in the construction industry.

Overall, it can be concluded that during the corona crisis, those industries that are directly affected by the containment measures are making intensive use of short-time work, even if these industries have relatively little prior experience of the instrument.

More layoffs, fewer new hires

The number of layoffs related to unemployment and new hires also shows that the corona crisis is already hitting the labour market much harder than in the 2009 financial crisis (see Figure 3). In April this year alone, more than 100,000 more people registered as unemployed than in the same month last year. The reaction to the corona crisis is therefore significantly stronger than the monthly changes throughout the course of 2009. However, it was possible to stop this development quite early.

The corona crisis affects not only people who become unemployed, but also people who are already unemployed and looking for work. In April this year, new hires from unemployment decreased significantly compared to the previous year (-78,000). This decline was much stronger than was the case in all the months of 2009.

So far, unemployment is concentrated heavily within the hospitality industry

In this context, it is also helpful to consider the entries into unemployment by sector (see Figure 4). Here, the developments in the hospitality industry are particularly striking. This is because the share of additional job losses registered in March and April this year amounts to over two percent of those employed in the sector. This share is significantly higher than in the first half of 2009 and currently far higher than in any other industry.

Even amongst “other economic service providers”, which also includes temporary work, the proportion so far has been significantly below one percent. In this sector, the far greatest increase in unemployment relative to employment was observed in the 2009 recession (over 7%). Overall, layoffs are likely to increase significantly in the current recession.

Thus far, all sectors have relied more on short-time work than layoffs

To date, the proportion of people in a sector for which short-time work has been reported, and the proportion of people in a sector that have already been laid off and thus become unemployed have been considered separately. However, in order to be able to evaluate the further development of the labour market for each industry, it is of interest to compare both parameters – because one can thus deduce how much a sector (at least up to now) is relying on short-time work to avoid layoffs.

This relationship can be seen in Figure 5. For both recessions, it shows the ratio of people for whom short-time work was registered, relative to the number of people entering unemployment in the respective sector (changes in the previous year).

If the bar representing a sector shows a value greater than one, the number of people for whom short-time work was reported exceeds the number of people who became unemployed in the same period (more precisely: the changes compared to the corresponding period of the previous year). The longer the bar, the more a sector is relying on short-time working instead of layoffs. For example, a value of 10 would mean that the number of people on short-time work exceeds the number of people who had become unemployed tenfold.

Already in the financial crisis of 2009 (shown with turquoise coloured bars) it became apparent that the majority of economic sectors in Germany focused far more on short-time work than on layoffs. Already then, the number of short-time work registrations in the manufacturing sector exceeded that of unemployment reports by a factor of 17 – by far the highest value in all industries.

In the current corona-recession (shown with blue bars for the months of March and April 2020), the factor for the manufacturing sector is in fact 296. In the second quarter there have been hardly any unemployment reports in relation to short-time work registrations. This is probably also due to the fact that the corona crisis was still relatively new and typically relatively few people are laid off at the beginning of a recession. However, it is clear that in the current crisis the manufacturing sector is once again focusing more heavily on short-time work relative to layoffs than all other sectors. Likewise, the other sectors have also been relying much more on short-time work than on layoffs than they did in the financial crisis.

In Figure 5, the axes are set such that the bars for both crises in the manufacturing industries are exactly the same length – despite the strong level of difference. In this way, it can be seen for each industry whether the current ratio of short-time work and unemployment reports has grown faster or slower than in the manufacturing sector compared to the financial crisis. The former is the case when the blue bar is longer than the turquoise. The latter applies if the blue bar is shorter than the turquoise.

In many industries, the bars are of a similar length, so the ratio of short-time work announcements to unemployment reports has changed in a similar way to that in the manufacturing sector. There are, however, clear differences in five sectors: agriculture and forestry, the hospitality industry,  financial and insurance service providers, “other economic service providers” and amongst “other services; art, entertainment and recreation; private household services”.

For agriculture and forestry, financial and insurance service providers and “other economic service providers,” this is due to the fact that there have been relatively few additional unemployment registrations compared to 2009 (see also Figure 4). By contrast, in the hospitality industry and in the areas of “other services; art, entertainment and recreation; private household services” there were an extremely high number of registrations for short-time work. In these industries, short-time working is therefore an alternative to terminating employment relationships, which had hitherto been used frequently by companies in these sectors.

In the construction industry, however, registrations for short-time work have so far increased less in relation to the unemployment reports than in manufacturing.


Short-time work is being used on a massive scale in the corona crisis. Never before have so many companies reported a need for short-time work to the Federal Employment Agency. And more applications for short-time work have been submitted for significantly more employees than during the financial crisis.

Those sectors that are severely affected by the pandemic containment measures are using short-time work very intensively, despite having previously had relatively little experience with the instrument. In 2009, the Federal Employment Agency paid out almost five billion euros in short-time allowance. The final value for 2020 is likely to be many times higher.

In order to use this money effectively, two current studies by Almut Balleer and co-authors as well as by Enzo Weber and co-authors recommend clear rules for the implementation of short-time work. These rules should be designed in such a way that they are transparent and incentivise the use short-time work only if companies would otherwise have laid off the employees concerned.

The reported unemployment figures for April this year (for people who have left employment) are already significantly higher than those from 2009. The bottom line, however, is that all sectors have predominantly opted for short-time work instead of layoffs. Even in the manufacturing sector, the emphasis in the current crisis has shifted even further towards short-time work. Short-time work has increased even more in sectors that are directly affected by lock-down restrictions, such as in the hospitality industry.

The number of layoffs in Germany has increased only moderately compared to countries such as the United States or the United Kingdom. Until now, the employment reaction to the economic slump remained moderate, following the low elasticity estimated in a pre-crisis study by Sabine Klinger and Enzo Weber.

The decline in new hires from unemployment in April of this year was also higher than in 2009. This could also increase unemployment considerably. A direct, temporary subsidy for new hires, analogous to short-time work for existing jobs, could therefore further bolster the labour market. Such an approach was proposed by Christian Merkl and Enzo Weber in VoxEU and is already put into practice, e.g. in the UK and in France.

Data and methods

Unemployment registrations have been taken into account until 14 April 2020. Companies that want to take advantage of short-time work must report this to an employment agency in advance. These registrations are recorded in the statistics of the Federal Employment Agency as soon as they are available electronically. On the one hand, it can be assumed that the currently available data are still under-recorded due to the large number of applications and that the actual number of notifications is higher. On the other hand, with the registrations it is not clear when the short-time work begins and whether it will actually take place – and for how many people in the company. In the financial crisis, only about one in three reporting companies actually used short-time work subsequently. There are also duplications in the data. Final data on actually implemented short-time work will only be available with a delay of several months (statistics from the Federal Employment Agency, 2020). The data on short-time job registrations and unemployment reports in March and April 2020 and in 2009 are compared by sector.


Adams-Prassl, Abi; Boneva, Teodora; Golin, Marta; Rauh, Christopher (2020): Inequality in the Impact of the Coronavirus Shock: Evidence from Real Time Surveys. IZA Discussion Paper No. 13183.

Balleer, Almut; Gehrke, Britta; Hochmuth, Brigitte; Merkl, Christian (2020): Guidelines for cost-effective use of SURE: Rule-based short-time work with workers’ consent and aligned replacement rates. In: VoxEU, 01.05.2020.

Balleer, Almut; Gehrke, Britta; Lechthaler, Wolfgang; Merkl, Christian (2016): Does Short-Time Work Save Jobs? A Business Cycle Analysis. In: European Economic Review, Vol. 84, 99–122.

Gehrke, Britta; Hochmuth, Brigitte (2019): Counteracting unemployment in crises: Non-linear effects of short-time work policy. In: Scandinavian Journal of Economics, 23.08.2019.

Klinger, Sabine; Weber, Enzo (2020): GDP-Employment Decoupling in Germany. In: Structural change and economic dynamics, Vol. 52, 82-98.

Merkl, Christian; Weber, Enzo (2020): Rescuing the labour market in times of COVID-19: Don’t forget new hires!. In: VoxEU, 07.04.2020.

Statistik der Bundesagentur für Arbeit (2020): Kurzinfo Statistik zur Kurzarbeit, 30.03. 2020.

Weber, Enzo (2015): The labour market in Germany: reforms, recession and robustness. In: De Economist, No. 163, S. 461-472.

Weber, Enzo; Bauer, Anja; Fuchs, Johann; Hummel, Markus; Hutter, Christian; Wanger, Susanne; Zika, Gerd (2020): Der Arbeitsmarkt in der schwersten Rezession der Nachkriegsgeschichte. In: IAB-Forum, 24.04.2020.

Gehrke, Britta; Weber, Enzo (2020): Short-time work, layoffs, and new hires in Germany: how the corona crisis differs from the financial crisis of 2009, In: IAB-Forum 16th of December 2020,, Retrieved: 25th of January 2021