To the Italian reader, the chapter by Joachim Möller poses a question which is not easy to answer but nevertheless cannot be avoided: Why did the Hartz reform of the labour market succeed in reducing the unemployment rate while the reforms of the Italian labour market (which started at the end of the 1990s but, as it seems, have not yet been completed) have had the only effect of increasing its volatility? What is surprising is not only the drastic reduction in the total German unemployment (in 2005 the rate was 11.3 per cent, in 2012, only seven years later, it more than halved at 5.5 per cent), but the simultaneous fall of long-term unemployment when the Hartz labor market reform was about to be completed (2003-2005). In the same years (2005-2012) the Italian unemployment rate increased from 7 to 11 per cent. To be sure, as Möller makes it clear, the ‘flexibilisation’ of the German labour market was not without costs for workers. On the one hand, the Hartz reform acted on long-term unemployment: it entailed a weakening of the welfare status of the long-term unemployed in such a way as to increase their willingness to accept job offers even at lower wages. On the other, it also acted on short-term employment through the liberalisation of temporary contracts and the deregulation of work agencies. For instance, the OECD index for the strictness of employment protection legislation for temporary workers decreased from 2 to 1.25. Bearing these costs produced at least two positive outcomes. As mentioned above, one of them is the strong reduction of long-term unemployment. This is an important achievement as it meant the reversal of the systemic unemployment which afflicted Germany, and Europe, since the oil shocks of the 1970s. The other positive effect was wage moderation which raised German competitiveness boosting investment and exports. On the other hand, the Hartz reform also produced negative, perhaps undesired, consequences. Labour market flexibility, as often happens, implied an increase in the share of temporary workers, particularly among the young workers. Another side-effect of the reform was a worsening of wage inequality, i.e. a strong rise in wage dispersion. All in all, this is Möller’s conclusion, that the positive effects more than offset the negative ones. And, above all, even if weakened, the German model, one of the pillars of the German society, was not substantially undermined. These very

conclusions pose again the question raised at the beginning: Why did the German reforms of the labour market have such an outcome while in Italy they failed? It should be underlined that the labour market reforms carried out in both countries were very similar: liberalisation concerned only the legislation for temporary contracts while leaving untouched the strictness of regulation for regular contracts and collective dismissals. Even if difficult, this is a question worth trying to answer. I suggest three elements of reflection as ingredients of a possible answer. The detailed and convincing reconstruction of the recent evolution of the German economy after the Hartz reform offered by Möller can hardly be disputed. Nor can the assessment he gives of the positive effects of the Hartz reform. For instance, in the years following the reform until 2008 the German economy saw an exceptional expansion of production and employment. However, labour market flexibility does not seem to work symmetrically in his historical account. For when it comes to the reaction to the 2009 Great Recession, a different explanation of the German performance is given. Certainly wage moderation induced by the Hartz reform helped the competitiveness of exportoriented firms, thus giving support to employment. Nevertheless, the explanation of employment stability during the recession is an ‘unprecedented level of within-firm flexibility’. In a sense, one could say that at the basis of the virtuous performance of the labour market there was the German model, especially in the form of social partnership and internal flexibility, rather than the Hartz reform, which favoured instead external flexibility. This is the first difference with respect to the Italian labour market reforms based almost exclusively upon external flexibility. Moreover, in conjunction with the decision of keeping the workforce within firms, the data also shows a marked acceleration in capital accumulation in Germany. Labour hoarding per se, especially when financed through government subsidies (by means of Kurzarbeitergeld or Cassa Integrazione Guadagni) is subject to ambiguous interpretation since by definition it cannot be a longterm decision. It may be seen as a first step toward a future massive layoff. Or, if the crisis is perceived as temporary, it will be accompanied or followed by a restructuring of the production process. Investment dynamics can be used as litmus paper to distinguish between the two situations. In the years immediately after the 2009 recession, when Germany and Italy were affected by an analogous fall in gross domestic product (more than 5 per cent) and investment (more than 11 per cent), gross fixed capital formation rapidly recovered in Germany in 2010 and 2011 (at some 6 per cent); in Italy it first stagnated and then declined. Although there is still no data to break down aggregate investment into its components in those years, what occurred just after the much more modest 2003 recession leads to the hypothesis that Germany privileged information and communications technology (ICT) investment at the expense of non-ICT investment, while the reverse happened in Italy. The evolution of total factor productivity in the two economies in the same years provides support for this supposition.