Mario Draghi is so wrong about the European labour market…

As far as I know the European Central bank has a monetary policy mandate, not an ‘economic policy’  mandate. In his monthly press conference Mario Draghi, head of the bank, however stated, as usual:

“In terms of economic policies, product market reforms to increase competitiveness will facilitate the creation of new businesses, support the tradable goods sector and foster job creation, while high unemployment rates require decisive structural reforms to reduce rigidities in labour markets and to increase labour demand.”

Easy fire, easy hire? But as far as I’m concerned European labour markets are not rigid at all – not structurally. Absolutely massive numbers of people are changing jobs each year, new jobs are created, old ones disappear. Look here for the Irish ‘job churn’ statistics.  And look here for recent Eurostat data on the rise in the activity ratio (the employed and unemployed as a % fo their age group). Except for young people, activity ratios are rising everywhere. Though I do have to admit that the amount of people changing jobs declined sharply, after 2008…,,, at least in the Netherlands. The crisis is the real rigidity. 

Remarkably, this rigidity is not noticed by the ECB. The Bank does however look at the ‘Beveridge curve’, the relation between vacancies and unemployment. If vacancies as well as unemployment increase at the same time, this is supposed to be a sign of structural rigidities which require ‘structural reforms’. What do these researchers find, according to a new ECB study? This:

This paper studies unemployment and vacancy developments in the euro area at the aggregate and country level over the Great Recession. The recent crisis has had a heterogeneous impact on euro area labour markets, leading to significant employment losses, especially in some sectors. The extent to which the rise in unemployment and particularly long-term unemployment reflects growing mismatch across euro area labour markets is one of the biggest questions facing euro area labour market policy makers. This paper attempts to shed light on this question by analysing developments in euro area Beveridge curves over the past 20 years, at both the aggregate level and on a disaggregated basis for all euro area countries. Using a simple model of Beveridge curve developments, we test for statistical significance of observed developments and  notice a significant shift in the euro area Beveridge curve since the onset of the crisis, but considerable heterogeneity at the country level. At the extremes, country level didifferences include a significant outward shift in the Beveridge curve for Spain and France, an inward shift for Germany, while. some euro area countries reveal no significant changes in the responsiveness of unemployment to vacancy  developments over the course of the crisis. We include an examination of factors underlying the observed developments across the countries.

At the moment, however, vacancy data for many European countries are not good enough for any kind of econometric analysis. As the Spanish labour market was, before 2008, clearly the most dynamic of the entire European Union I decided to take a look at the vacancy data myself, especially as I had done this before which had led me to the decision not to use them for any kind of graph on this blog, as I did not trust them enough (gut feeling). My gut feeling turned out to be right, at least for Spain. The outward shift of the Spanish Beveridge curve mentioned in the quote above is clearly a figment of the data as the older Spanish series is lousy, insensitive and way too low (the series is probably not based upon actual measurements but upon administrative data). Remember: about 50% of pre 2008 total Eurozone job took place in Spain, mainly caused by buoyant demand in the construction sector! There is no way that the Spanish vacancy rate in this period was lower than in 2009.

Update: I included the piece below on 07/09/2013

To circumvent this problem the authors do not use vacancy data but also :

“data … taken from the European Commission’s regular Confidence Surveys – specifically the aggregated responses from the question relating to employers’ perceptions of labour shortages as limits to business.”

What is their conclusion?

“From the above, it seems that workforce characteristics and the sectoral dimension explain much of the pattern of Beveridge curve movements for the countries in our analysis. Institutional variables – employment protection, the use of temporary contacts; trade union density and  trade union coverage of collective bargaining arrangements and replacement ratios- which characterise national labour markets are a further set of variables of particular interest to policy makers.”

To state the first sentence otherwise: the typical Romanian construction worker in Spain, there are about 1 million Romanians over there, does not qualify readily for precise nursing work in the medical sector. About the second sentence the authors state

“Institutional variables often do not work well in econometric analyses”

This is shown clearly by their graph 7 (I did not manage to copy-save-insert it in any way) which shows the influence of EPL (Employment Protection Legislation) on the increase of unemployment since 2008. The relationship is extremely noisy and (you can always draw a line…) basically flat. However – if you would leave out Greece and Spain from the graph (there is some reason for this: both countries have extreme unemployment of 26-28% which is about 10% higher than even in Portugal) the relationship becomes a lot stronger – but negative. Understatement: this is not consistent with the Draghi remarks.

Aside: the increase of unemployment in especially Spain is also due to a rise in average hours, a fast increase of average productivity (a weight effect: construction had a low average productivity and an increase in the activity rate) 

Another observation on the EU labour market: job growth in Ireland was high – mainly caused by a massive increase in agricultural employment and amongst 45+… Amazing.