Friday, December 16, 2011

Eurozone Productivity Comparison


The above graph shows productivity for selected Eurozone countries since 2000.  The y-axis units are euros/hr worked and the data has been adjusted for inflation and seasonality.  A few observations:
  • As you would expect, the PIIGS countries are generally less economically productive than their northern European brethren.
  • Greek productivity data are not available for some reason...
  • Germany is not the most productive country - France, Holland, and Ireland are all higher.  However, it would not surprise me to find that this is an average of West German productivity that is comparable to those countries, and East German productivity that has been recovering from a much lower level.
  • The Irish are now very productive and this has risen sharply since the 2008 financial crisis.
  • Productivity generally increases over time (as one would expect with further technological advance).
  • Italy is a notable exception to the trend of improving productivity - it was higher a decade ago than now.  Hence the concern at the high debt level of the country - output is not growing enough to cover the interest on the debt.
I think the most critical point though is that the economic productivity of a society is a slowly changing variable.  Efforts to increase competitiveness are unlikely to have large impacts on productivity in the timescale of the current financial crisis.

Here are the same data expressed as percentage change over the same quarter in the prior year:


It's noticeable that productivity growth has been slowing in 2011, just as it did in the lead-in to the 2008 recession.

Update 12/19/11: Looks like part of the increase in Irish productivity may be due to compositional effects (basically layoffs in the great recession may have been concentrated in more cyclical parts of the economy with lower productivity - as opposed to it all being due to individual sectors becoming more productive).  See Krugman here.

4 comments:

James said...

Fitch: big euro fix “technically and politically beyond reach”
http://ftalphaville.ft.com/blog/2011/12/16/804161/fitch-big-euro-fix-technically-and-politically-beyond-reach/

Stuart Staniford said...

James: yeah - I saw that. They seem to live in a universe I undertand, whereas Merkel/Sarkozy/Draghi seem to be living in some other unrecognizable one...

ColdNorth said...

A comment on productivity, recalling that it is generally defined as the value of goods and services sold (or GDP) divided by hours worked.

Most assessments of performance based on productivity appear to focus on increasing revenues by:
1) producing more valuable goods and services (through innovation), or
2) producing more in the same time (better production methods).

However, a country's productivity can rise in other important ways. For example, the price of goods and services it sells can increase (think energy, food, construction materials, etc.).

Secondly, a country can increase its productivity by exiting the production and sale of lower value goods and services, resulting in short-to-medium-term social costs (un-employment etc.). This higher productivity may be overall a negative for the country and its people.

Productivity is an interesting signal about economic structure, but by itself can be misleading about overall economic success.

Stuart Staniford said...

Coldnorth - note that the productivity stats are adjusted for price level changes so those shouldn't factor into it (ie it should capture "real" trends not nominal ones).

Your "Secondly" point could be true in principle but it seems in practice that productivity dropped during the recession (indicating that companies cut output more than they cut employment).

Still and all: your basic point is well taken - productivity is a narrow measure of the efficiency with which a given society turns labor hours into goods and services: no more, no less. It says nothing about whether the goods and services are desirable or whether either the consumers or the workers were happy or found meaning in the result.