Europe’s Unemployment Problem – Perhaps Half As Big Again

andrew wattWe are getting used to bad news about the labour market situation in Europe.

Record-breaking 12% unemployment in the euro area. More than 26 million men and women unemployed across the EU27. More than a quarter of the workforce jobless in Spain and Greece. Youth unemployment rates (which need careful interpretation) around twice as high.

You could be forgiven for thinking that it couldn’t possibly be much worse – but you would be wrong. A useful Eurostat overview (here), based on the 2012 results of the European Labour Force Survey shows, in addition, a worrying degree of and rising trend towards underemployment and hidden unemployment.

Let’s start with the standard international definitions used by Eurostat according to which the unemployed are all those without work but actively seeking it and available for work in the short term. The unemployment rate is this number divided by the sum of the unemployed and the employed. There are other categories, though, that can sensibly added to this figure. First of all there are the part-time workers who actually want, and would be able to work, longer hours. In fact more than a quarter (21.4%) of the 43 million part-timers in Europe want longer hours. That percentage is up from 18.5% in 2008.  It is also worth noting that in Germany, widely seen as having a robust labour market and, according to some, even suffering from labour shortages, there are 1.8 million workers in this category.

And that is not all.

An additional 8.2 million are available for work but at the time of the survey are not actively seeking. These are known informally as “discouraged workers” who are assumed to recommence job search once labour market prospects pick up. Then there is another, smaller category of people who, conversely, are seeking work but are not immediately available (for example because they have a job offer that starts in the near future)*. There are 2.3 million in this group, meaning that more than 11 million people are not in the unemployment statistics – they are considered to be “economically inactive” or out of the labour force – but can reasonably be counted in a broader measure of underemployment. This is up from 9.7 million in 2008. Adding these two groups together, Eurostat calculates that there is an additional unused labour supply of 4.6% in the EU27 and 4.9% in the euro area. In Italy, where unemployment is already extremely high, these two groups represent almost three million people and more than 12% of the labour force.

Using these figures we can make a rough and ready calculation of the “real” rate of underemployment in Europe. According to the European Labour Force Survey, average hours of part-timers are a tick under twenty per week, whereas full-timers work 41.5 hours a week. We don’t know exactly how the working time preferences are distributed, but a reasonable starting point would seem to be to assume that those part-timers wanting longer hours have the same average hours as all part-timers and that they want to move to the average of full-timers. In  other words we add somewhat more than half of the 9.2 million involuntarily part-time workers to the unemployed total.** We then add the people in the two sub-groups of the economically inactive both to the numerator (unemployed) and denominator (unemployed plus employed).

If I haven’t made any stupid excel mistakes*** the underemployment rate in the EU27 could be considered to be up to 16.7% and in the euro area as high as 18.0%. In other words the unemployment rate captures only about two-thirds of the extent of European underemployment.

Or the unemployment problem is perhaps half as bad again as you already thought.

* The full definition of this latter groups is as follows: Persons seeking work but not immediately available are those aged 15-74 neither employed nor unemployed who actively sought work during the last 4 weeks but are not available to work in the next 2 weeks. For completeness this category also includes three smaller groups: those who found a job to start in less than 3 months and are not available to work in the next 2 weeks; those who found a job to start in 3 months or more; those who passively sought work during the last 4 weeks and are available to work in the next 2 weeks. Passive job search is e.g. waiting the results of a job interview.

** Clearly this is a simplification. It might be that those with above average part-time hours are more likely than low-hours workers to want longer hours, in which case there would be a smaller number of “full-time equivalents”. I don’t see any strong grounds for assuming this, though. Arguably those with the shortest hours have the greatest (financial) incentive to lengthen hours. It might be that moving to the full-time average exaggerates the number of additional hours that part-timers would like to work. Determining this would require more detailed survey evidence. (I would be happy to receive hints from readers.) Given, though, that it is conceivable that some will want to work longer than 41 hours per week, the full-time average is in no way a maximum. Still, the actual-desired hours gap is probably somewhat lower than assumed here, but all in all this measure can serve as a reasonable first approximation. Not considered at all here are the working-time preferences of full-time workers.

*** To reduce the chances of an R-R moment, the calculations (I hope self-explanatory in the light of the text; the decimal point is a comma) are below. The underemployment rate is (U+FTE+ANS+SNA)/(E+U+ANS+SNA)