The flaws inherent in aggregated data are a recurring theme in this section, and today we have a great case study: European unemployment rates. By the headline numbers, one would think the EU economy – like the US – is back to firing on all cylinders. For example:
- EU unemployment was 6.7% in November 2018 (latest data available).
- The prior cycle lows were in February – April 2008, at 6.8%.
- The seemingly obvious conclusion: the EU labor market is healthy, or at least as strong as it was at the peak of the prior cycle.
The trouble here is that unemployment rates in large individual member states do not uniformly mirror this improvement. In fact, joblessness across several important EU countries still looks quite troublesome:
- France: the November 2018 data shows unemployment at 8.9%, well above the 6.7% EU rate. Current levels are also above prior cycle lows of 7.2% in February 2008.
- Italy: unemployment currently sits at 10.5%, a far cry from the EU’s 6.7%. The prior cycle low was 5.8% in April 2007 – roughly half current rates.
- Spain: unemployment remains stubbornly high even today, at 14.7% based on the November 2018 data. The lows from May 2007 were 8.0%.
So why is the aggregate data as strong as it is, both as a snapshot measure and when compared to the last economic cycle? The easy answer is that German unemployment is just 3.3%, and that reading is far better than the 2007-2008 prior cycle trough of 7.0%. Then there are the below-average unemployment rates of the Low Countries (3.5% – 5.5%) as well as other mid-sized economies (Austria: 4.7%, Poland: 3.8%, Hungary: 3.8%).
To put some perspective around this disparity in country-level EU unemployment rates, let’s look at jobless rates across the largest 5 US states (December 2018 data):
- California: 4.2%
- Texas: 3.7%
- Florida: 3.3%
- New York: 3.9%
- Pennsylvania: 4.2%
- National unemployment: 3.9%
This data shows that unemployment in large US states collectively hover around the national number, unlike the wide dispersion we see in the EU country level readings. Some are a few ticks above (CA and PA), some a few below (FL and TX), and one (NY) is right in line. In no case (as with France, Italy and Spain) do you have wide variances (whole percentage points or more) from the aggregate unemployment rates.
All this raises several important questions:
- How does the European Central Bank consider normalizing both interest rates and its balance sheet when systematically important economies like France, Italy and Spain remain at essentially recessionary levels of unemployment?
- What are the political ramifications of high unemployment across these large countries? It seems clear that France, Italy and Spain need to do more to juice their economies, but we saw how that went when Italy proposed a more aggressive fiscal spending plan last year. And one wonders what the French government was thinking when it launched new fuel taxes…
- What happens to the EU in the next recession, whenever it is? The data clearly shows that France, Italy and Spain have yet to recover from the Great Recession, and even another year or two of slow global growth is unlikely to return these economies to full potential.
Summary: the “European Union” still has a long way to go before the second word in that name really applies. Unemployment rates are just one example, but an important one. They highlight not just economic forces, but also explain political factors as well. The aggregate data tells a happy story; the country-level information does not.