The ECB released an extensive report on the Euro area labour market recently. It’s worth a read and contains a detailed statistical analysis of the labour developments since the start of the Euro crisis. One interesting part of that analysis is a calculation of Okun’s law while decomposing GDP into its components and computing the impact of each one on total unemployment change. The exercise is quite elegant and explains a large part of the change in unemployment (R² = 0.57). Coefficients for each component are computed, as well as the elasticity by utilizing the weights of each component on aggregate GDP:
What is very impressive is the fact that the computed elasticities cast a significant blow on the ‘internal devaluation’ policy viability since the (private) consumption elasticity is almost an order of magnitude larger than the export and import elasticity (which are equal and up to a point cancel each other since a part of any exported product is imported). That’s probably attributed to the fact that domestic, consumption oriented sectors such as services are much more labour-intensive than mostly capital-intensive export sectors. As a result, export-led recovery, at least in terms of employment is not a viable solution for the Euro imbalances since the austerity measures will lead to sustained long-term unemployment which will not be absorbed easily by the tradable sector and create serious social problems as well as lower the potential growth rate.
One can use the computed coefficients and the Greek GDP components weights (for the 2000 – 2012Q2 period) to examine what the calculated change in unemployment would be. I ‘ve done the exercise for the 2008Q2 – 2012Q2 period in the following table:
The computed change is almost 2/3 of the actual change so the regression is actually able to explain a very large part of the increase in the Greek labour unemployment. Private consumption and investment account for 5.9% and 5.2% of the change while the imports squeeze only lowers the unemployment rate by 1.8% and exports (especially compared to 2008) do not have any meaningful effect. Even if Greece were to observe a large export-led recovery, the very low elasticity of no more than 0.02 (which is equal to the import elasticity) would not produce any significant result on the unemployment rate, especially since at least a part of exports must be imported. The ultimate employment recovery must be domestically demand led to have any measurable change.
One can also use the above elasticity numbers to calculate the projected unemployment increase from the 2013 austerity measures, using the Greek budget projections for GDP components change:
The government projection seems reasonable. Since the regression can only explain 2/3 of the change, the actual increase could be up to 3%. What is optimistic is the projection of only -3.7% change in investment since the corresponding change in 2012 was -15%. A more realistic change of -10% leads to an unemployment increase of 2.5%.