The German Federal Labour Office publishes monthly data on employment performance such as the number of officially unemployed (ALO), the number of those employed subject to social contributions (SVP), and the number of people participating in ‘Kurzarbeit’ schemes (KUG - the German state supported STWA).
Data from 2005 through to August 2008 shows that in ‘normal’ periods the correlation between KUG and ALO is positive (0.57) and that between KUG and SVP negative (-0.54). These results offer no surprise. However, there appears to be a time lag between the take-up of KUG and the corresponding changes in the general labour market situation. This is because the correlations get much stronger if ALO resp. SVP were plotted against earlier KUG take-up instead of KUG of the same period (see Chart 2).
These findings seem to confirm less optimistic views of a rather complementary relationship between unemployment (lay-offs) and STWA scheme take-up, reviving Calavrezo‘s notion of high unemployment following on from STWA (Calavrezo et al, 2009, for France): STWA is being taken up immediately during a downturn while the labour market still needs time to adjust. In Germany, the ‘period of full reaction’ seems to be quite long - around two years - as the correlation between ALO resp. SVP and KUG gets weaker again after lagging KUG by more than two years.
These findings are suggestive of less optimistic labour market prospects in the short term, although it is still much too early to assess the impact of the crisis on employment.
3.1.5. An econometric analysisThe hypothesis that STW schemes reduce the variability of employment has been tested by Arpaia et al. (2010) using an equation in which the dependent variable is the annualised change in employment in the industry sector, and the explanatory variables are one lag of the dependent variable itself, the change in the industry’s value added, a dummy signalling the 2008-2009 recession, a variable that combines this dummy with another dummy signalling countries with STW schemes, constant and country-fixed effects which allow for country-specific factors other than the incidence of STW. Table 3 presents the empirical results for the 27 EU Member States from the first quarter of 1990 to the fourth quarter of 2009.
The coefficient of annualised changes in value added indicates that a 1.00 percentage point change in production leads to a 0.11 percentage point change in employment growth. The coefficient on dummy crisis indicates that, during the 2008-2009 recession, a further fall (by 0.5 percentage points) in employment growth has been registered as compared with the average annual fall registered over the entire period. However, this additional fall has been counterbalanced in countries with STW schemes. Indeed, the coefficient of the multiplicative dummy (Dummy crisis x Dummy STWA) is significant and positive.
Although these estimates indicate that STW schemes have had an impact on the variability of employment during the economic crisis, further analysis is needed in order to determine whether the employment initially saved by STWA will persist after the crisis.