Just a few years ago, Germany was known as the sick man of Europe (Burda 2007). Starting from an average unemployment rate below 4% in the 1970s, Germany saw its rate increase to almost 9% in the period 1995-2005. As seen in Figure 1 the unemployment rate has a strong cyclical component but also a trend component that has been rising since the 1970s until the mid-2000s.
Figure 1. Quarterly unemployment rate, Germany 1970Q1-2012Q4
Source: OECD: 1970-1990, quarterly unemployment rate for West Germany; 1991-2012, quarterly harmonised unemployment rate for Germany.
In response to the dismal labour market performance, in 2003-2005 the German government implemented a number of wide-ranging labour market reforms, the so-called Hartz reforms. The first three parts of the reform package, Hartz I-III, were mainly concerned with creating new types of employment opportunities (Hartz I), introducing additional wage subsidies (Hartz II), and restructuring the Federal Employment Agency (Hartz III). The final part, Hartz IV, was implemented in 2005 and resulted in a significant cut in the unemployment benefits for the long-term unemployed. Overall, the Hartz reforms constitute one of the most ambitious attempts in recent history of restructuring the labour market of an advanced economy.
An unpopular success
Figure 1 suggests that the Hartz reforms were quite successful. Between 2005 and 2008 the unemployment rate fell from almost 11% to 7.5%, barely increased during the Great Recession, and then continued its downward trend reaching 5.5% at the end of 2012. This view is shared by many economists in Germany and confirmed in our recent macroeconomic study of the Hartz reforms based on a calibrated search model (Krebs and Scheffel 2013). Specifically, we find that the Hartz IV reform reduced the non-cyclical unemployment rate in Germany by 1.4 percentage points. We further find that the Hartz I-III reforms decreased the non-cyclical unemployment rate in Germany by 1.5 percentage points. Thus, our analysis suggests that the entire reform package let to a permanent reduction in the German unemployment rate by almost three percentage points!
Most economists are probably not surprised to hear that a policy reform that dramatically cuts unemployment benefits and improves a very inefficient matching process will reduce the unemployment rate substantially. Perhaps more surprising is that despite their apparent success, the Hartz reforms have always been very unpopular among the German public. This unpopularity has been documented in surveys, but the best evidence comes from the upcoming National Election in Germany to be held on 22 September 2013. There is no major party that dares to run on a platform that openly endorses the Hartz reforms. Indeed, several parties are trying to win votes by promising to roll back the Hartz reforms, including the Social Democrats who initiated the reforms in 2003-2005 under the leadership of Chancellor Gerhard Schroeder.
To some observes, the recent German experience with labour market reform poses somewhat of a puzzle. On the one hand, we have an economic profession that overall declares the Hartz reforms a success. On the other hand, we have a German public that is overwhelmingly critical of the reforms. It seems that this is yet another case of economists not understanding real people or real people not understanding economics.
Winners and losers
In our study we propose an explanation of the unpopularity of the Hartz reforms that assumes full rationality of economic actors. Our explanation starts from the simple and well-known idea that labour market reforms create winners and losers, and that the unemployment rate is not a sufficient statistic for identifying how gains and losses are distributed across the population. Specifically, we conduct a welfare analysis assuming moderately risk-averse households and find that the median employed household gained from the Hartz IV reform taking into account that the Hartz reforms reduced the tax burden (social-security taxes). However, Hartz IV resulted in a significant cut in the unemployment benefits for the long-term unemployed, and this group therefore experienced a welfare loss. Most interestingly, the short-term unemployed also lost even though Hartz IV did not reduce their unemployment benefits, but the prospect of less insurance in the case of becoming long-term unemployed is enough to make them resist the reform. Finally, low-skilled workers with precarious jobs are in a similar position as the short-term unemployed and many of them are likely to lose as well. Indeed, our study suggests that the Hartz IV reform reduced real wages, a prediction that is consistent with the data – see the evolution of the real wage in Germany depicted in figure 2.
Figure 2. Real wage and real GDP per Capita (1992 = 100), Germany 1992-2011
Source: Statistisches Bundesamt: annual real wage index (series: Reallohnindex) and annualreal gdp per capita (series: Bruttoinlandsprodukt) normalized to 1992.
The welfare losses experienced by unemployed workers and workers with precarious jobs provide a simple explanation of the unpopularity of the Hartz reforms, and in particular Hartz IV. Clearly, those who expect to lose from the reform will oppose the reform, and our quantitative welfare analysis shows that this segment of the German population is large, though not the majority. However, even if the reform losers are not the majority, it can well be that unions and politicians will support their concerns by promising to roll back the reforms. For example, if the reform losers are a clearly defined group with individually large and sometimes dramatic losses, but the reform winners are a diffuse group with only small and indirect gains for each individual, then a rational politician interested in winning elections will run on a platform that calls for reversing the reforms or not implementing them in the first place.
Lessons for other countries
We believe that two important lessons can be drawn from the German experience. First, even relatively well-implemented and “successful” reforms will meet resistance when people are risk-averse and gains and losses are distributed unevenly. This is a simple yet important insight to be kept in mind when proposing labour-market reforms in some EU countries to reduce currently high levels of unemployment. Second, there is no reason to believe that there is a magic reform that works for all countries equally well. For example, in our study we find that a Hartz-type reform of the unemployment insurance system in France or Spain would have relatively modest effects on the French or Spanish unemployment rate. The reason for this finding is simple. The benefits paid to the long-term unemployed are already low in France and very low in Spain, and reducing these unemployment benefits to even lower levels is not likely to have large incentive effects. In contrast, the German unemployment-insurance system was very generous to the long-term unemployed before the Hartz IV reform, and in this case efficiency gains from implementing the reform were quite large.
Burda, Michael (2007) “German recovery: it’s the supply side”, VoxEU.org.23 July.
Krebs, T and M Scheffel (2013) “Macroeconomic Evaluation of Labour Market Reform in Germany,” forthcoming, IMF Economic Review.