The eurozone’s labour market Cinderella story

At face value, eurozone unemployment almost looks like a Cinderella story. With barely any increase in unemployment, it is currently the belle of the global labour market ball, at least compared to many other developed economies. When the clock strikes midnight, however, and short-time work schemes come to an end, the fairy tale is unlikely to continue. We expect a second wave of job losses towards the end of the year and going into 2021 although the peak in unemployment should still remain below the highs seen in the aftermath of the financial crisis.

This crisis is not like the others. Even economists who've written books about how recessions are all broadly similar admit to this. One of the ways through which this plays out is the unemployment rate. Even though the economy has seen an unprecedented shock, unemployment has only inched up so far, from 7.1% in February to 7.7% in June. This is in no way reflective of the historic decline in economic output experienced in the first and second quarters. By contrast, in the US, the unemployment rate increased from 3.5% to 14.7% in April, after which it declined but still remained in double digits, at 10.2% in July. One of the driving factors behind the difference is the short-time work schemes that eurozone economies have implemented. This has suppressed the amount of people being laid off on a large scale.

Short-time work schemes were introduced in Germany more than 100 years ago and gained enormous popularity during the financial crisis in 2008/9. Back then, they helped the German economy to exit the crisis quickly. Other eurozone economies also implemented similar schemes and have been motivated to do so during the current crisis. The Support to mitigate Unemployment Risks in an Emergency (SURE) programme, a newly-introduced policy tool at the European level, is the best illustration of how popular such schemes are. However, this policy instrument works best when it is applied in sectors which quickly return to full strength so that employers do not lose time and money laying off and re-hiring employees. The Dutch short-time work scheme is a good illustration of this as it works with multiple rounds. The first round which started at the beginning of the crisis was used by just under 140,000 businesses, but the second version of the short-time work scheme which started recently has drawn significantly less demand than the first, at just 36,000 businesses so far, as better economic prospects reduced the need to apply again. While short-term work schemes are an effective tool to smooth unemployment for cyclical purposes, the tool will be less effective in sectors which are undergoing structural change, posing a risk for unemployment in the recovery phase.

Another factor contributing to the artificially low unemployment rate is the decline in people looking for work. People who have not been actively looking for jobs are not officially counted as unemployed. This is especially the case in Italy and Spain, where the increase in the number of unemployed has been accompanied by an unusual drop in the active population. Adding the “dropouts” in Italy, Spain and Portugal to the number of unemployed would raise the eurozone unemployment rate by about 1.5%. Other eurozone countries have seen the opposite effect though, and as such, the overall active population decline in the eurozone has had a negligible impact on the unemployment rate. Still, it does add to the risk of unemployment trending higher in the months to come even though the effect will likely fade as economies reopen. A lot of downside risk to the labour market outlook therefore remains.

All large eurozone economies have put short-time work schemes in place for companies to bridge the economic fallout caused by the lockdowns. In Germany, around 15% of workers are on short-time schemes, compared to just above 30% in Italy. This means that government is supporting employment on a massive scale at the moment. To simply add those numbers to the amount of unemployed would overstate the current “actual” unemployment situation, so it still does not tell us much about the state of the eurozone labour market in this unusual crisis.

To get a better sense of what the unemployment situation would look like if governments had not massively subsidised work in the first months of this crisis, we create a hypothetical unemployment rate using Okun’s law. This simple historical relationship between GDP and unemployment also helps us sketch a path of the quarters ahead under our baseline GDP growth scenario, which is helpful in estimating a possible peak in unemployment under current circumstances. As chart 1 shows, unemployment would have gone up much quicker in the hypothetical scenario, the second quarter unemployment rate would be 9.1% instead of the 7.7% we experience now.

For the months ahead, the hypothetical unemployment rate would increase to around 11% based on our base case for GDP growth which does not include a second lockdown. This is still lower than the peak of 12.1% reached in 2013 during the euro crisis. A possible explanation for why the current crisis, even without government measures, would not have a more negative impact on the labour market could be the short-lived nature of the initial shock. A strong economic rebound followed by a levelling off is required to prevent unemployment from accelerating further. In the aftermath of the financial crisis, austerity measures, plus a second wave of the crisis (aka the euro crisis) led to a protracted increase in unemployment in many countries.

Even without short-time work schemes, eurozone unemployment would peak below its 2013 record, and thanks to the schemes, it is logical to assume that the peak will be lower as they flatten the cyclical curve. However, the big unknown in this prediction is the structural element of the current crisis. Besides the regular cyclical element, there is a significant risk of companies laying off workers to adapt to a new post-Covid reality. This could lead to a more long-lasting increase in unemployment, and will determine how many employees currently subject to short-time work schemes actually return to full employment.

So far, we have already seen businesses starting to restructure, with big names in sectors that are likely to experience a more structural impact from the virus standing out. The European Restructuring Monitor - a database that tracks restructuring events – indicates that incidents of businesses laying off workers for internal restructuring purposes has indeed shot up since May. It is not the case though that the job losses indicated by the dataset are dominated by the most obvious sectors impacted more long-term by the coronavirus. Judging from recent company announcements, some 25% of the jobs currently under short-term work schemes, in the worst hit sectors, could eventually be lost.

The manufacturing and financial services sectors have seen a large number of job losses from internal restructuring, which hints at the fact that in some industries, businesses are using the corona crisis to push through already necessary reforms. Other sectors high up on the list are more specifically affected by the crisis, like hotels, restaurants and retail. This is and will be an important factor contributing to the unemployment rate creeping up, despite significant government support. It is definitely not a given that companies will be able to adapt to a post-corona world in a way that improves productivity, but structural change could well push up unemployment in the short- to medium term.

So, from a cyclical perspective it is unlikely that unemployment reaches 11%, but structural factors are an unknown for the coming quarters and could mean higher unemployment. If 10% of all employees currently subject to short-term work schemes do not return to full employment, this would add about two to three percentage points to eurozone unemployment.

At what point the peak in unemployment will be reached is, in our view, very dependent on the end of short-time work schemes, returns to the labour force from discouraged workers and restructuring to adapt to a post-coronavirus world. New waves of unemployment are not easy to time, but the official end of short-time work schemes could signal potential starting points. Most eurozone employment schemes are set to come to an end between the end of the third quarter 2020 and the first quarter of 2021. For Germany, they are now set to end in the first quarter of next year. This is also true for certain sectors in Italy, while for most, they are set to end in December. However, it is possible that governments once again extend the length of their respective schemes, especially as the EU's SURE programme allows for short-time work funding to do this. France has extended until early 2022 and Italy has already requested €28 billion from SURE for current funding, but could request more to extend. Whether all companies would actually make use of another extension is, however, a different story as the cyclical need for the programmes will be a lot smaller, as the rebound in economic activity has begun. This makes the peak in unemployment especially hard to time, but it seems logical that the run-up in unemployment will not end before mid-2021 based on the timing of short-time work schemes. In a country like Germany, the forthcoming elections in autumn 2021 should also play an important role in the decision on whether to extend short-time work schemes again.

All of this means that the turn of the year could mark the start of a second wave of adverse labour market developments. This will be the labour market’s Cinderella midnight moment; the point at which the labour market starts to slow down the recovery. Another reason for the European Central Bank to stay fully in the quantitative easing game and a clear reminder that the V-shaped rebound is unlikely to continue.