Why are nominal wages still rising in every European economy?
Tyler Cowen links to a Eurostat study that show nominal wages to be rising in every single European economy, even those deep in Depression:
Among the Member States for which data are available for the second quarter of 2012, the highest annual increases in hourly labour costs for the whole economy were registered in Romania (+7.1%), Finland (+4.9%), Bulgaria and Latvia (both +4.8%), and the lowest in Ireland (+0.4%), Spain and the Netherlands (both +0.5%).
This is obviously very different from the 1930s, and I’d be interested in knowing why. Perhaps some of my European readers have some insights into the Eurozone labor markets.
PS. Tyler looked at the same data and reached the opposite conclusion. I’m not sure why, but he seems to have focused more on unit labor costs, whereas I focused on hourly nominal wage costs. Still, given the big slowdown in NGDP growth (and outright declines in many countries, it appears wage stickines sis a big problems, particulrly in Spain.
Update: Mark Sadowski pointed out that this post was somewhat sloppy. There is no 21012:2 data for Greece, where wages probably are declining. And a few other countries saw declines during previous years. Mea culpa. Still, given the big slowdown in NGDP growth (and outright declines in many countries), it appears wage stickiness is a big problem, particularly in Spain.
HT: Saturos
Tags:
18. September 2012 at 11:33
“Tyler looked at the same data and reached the opposite conclusion” is often my response to his posts on adjustment efforts in Europe.
18. September 2012 at 11:40
Cuz 25% fewer people are working??
18. September 2012 at 11:47
Price stability tends to reduce price flexibility. If you’re pretty sure that inflation is going to go up by 2% a year rather than 20% one year and -20% the next, you’re much more likely to make a fixed-increase contract, and much less likely to accept a nominal pay cut.
Increases in the size of government as a share of GDP make a big difference as well.
I’d note, too, that NGDP fell a lot further in the Depression, and nominal wages lagged NGDP significantly in the 1930s, to the extent that there were substantial real wage increases for those who still had work.
18. September 2012 at 11:51
Maybe people with lower salaries are more likely to lose their jobs? Thus average salaries go up even though the salary of an individual does not improve.
18. September 2012 at 12:48
“This is obviously very different from the 1930s, and I’d be interested in knowing why. Perhaps some of my European readers have some insights into the Eurozone labor markets.”
This is very sloppy Scott.
As you very well know US nominal wages fell during the contraction and rose during the expansion in the 1930s. Average Hourly Earnings (Twenty-Five Manufacturing Industries for United States) increased by about 60% from 1933 to 1937, by which point they were 20% higher than they were in 1929:
http://research.stlouisfed.org/fred2/series/M08142USM055NNBR?cid=33069
Also, nominal hourly wages may be rising right now in almost all EU countries (note there is no data for Greece) but even your own link shows periods of falling nominal hourly wages for Ireland, Greece and Portugal. And if you go to Eurostat you’ll find that Estonia, Latvia, Lithuania, Poland, the Czech Republic, Slovakia, Hungary, Slovenia, Romania, Cyprus, Malta, Austria, Germany, the Netherlands, Ireland, Finland, Italy, Portugal, Spain, Greece and the UK all suffered nominal hourly wage declines at some point or another. The only exceptions appear to be Belgium, France, Luxembourg, Denmark, Sweden and Bulgaria.
The hardest hit country is Lithuania, which saw its nominal hourly labour cost index fall 13.4% from 2008Q3 through 2010Q2. As of 2012Q2 it is still about 7% below its previous peak nearly four years ago.
This is clearly another one of Tyler Cowen’s exercises in pointing out that the rubble is bouncing and mistaking that for genuine progress.
18. September 2012 at 13:27
You are focused on whether nominal wages are sticky, I was considering whether internal devaluation on the real wage proceeds very much at all. Some people say it won’t. It seems it does. Doesn’t mean this is the preferred way to go, obviously it isn’t.
18. September 2012 at 13:29
By the way, a number of the commentors are jumping on the normative issue — mood affiliation perhaps — even though my post explicitly disavowed that. It’s keeping them from learning the positive lesson here, a common syndrome.
18. September 2012 at 14:10
Yes, Scott, stickiness is a BIG problem in Spain. Probably the most stickiness labor market in te W. But that is a reason for a high NAIRU, but not for the 25% of unemployment now, as you know.
18. September 2012 at 14:11
Tyler, I certainly agree on the mood affiliation problem.
I’m not a big fan of using “real wages,” as I think W/NGDP is the more important variable from a “sticky wages model” point of view. But I will admit that my first reaction was too hasty; the actual data is more ambiguous than I first thought.
18. September 2012 at 14:30
In many European countries wages are indexed to inflation, and labor unions are strong.
In the Netherlands (and possibly other countries) there is a system were people with temporary contracts can be easily let go – these are usually young people with low salaries – whereas people with permanent contracts are very difficult (and costly) to lay off.
In a slump the typical company will cancel all temporary contracts.
Obviously, the average wage will increase if you lay off the cheap workers and keep the expensive ones.
My €0.02
18. September 2012 at 16:30
Maybe labor costs include mandatory severance pay? This tends to be high in Southern Europe, and given the large increase in unemployment it could be an important factor. The numbers for Portugal seem hard to understand otherwise.
18. September 2012 at 18:36
There’s a book on this subject, that I’ve ordered:
Why Wages Don’t Fall during a Recession [Paperback]
http://www.amazon.com/gp/reader/0674009436/ref=sib_dp_pt#reader-link
As I understand it his thesis is that employers prefer to use other means to cut labor costs, so as to preserve employee morale. Dissatisfied workers can be very unproductive. Employers would rather reduce the number of employees, cut benefits or use more temporary labor.
If this is correct, what you would expect to see is a decrease in payroll either in absolute terms or per dollar of output, and decrease in average number of hours worked, but not one in hourly wages. And, in the US, this is what you see.
Also the median wage started falling as a percentage of the mean wage in 2003.
It stood in nominal dollars at 26,363.55 down from 26,514.38 in 2008.
The mean wage was 39,959 in 2010 up from 39,652 in 2008.
This accords with the reports that all the recovered jobs are in what were the bottom and top thirds of the income distribution of 2007.
So far these jobs haven’t returned.
18. September 2012 at 19:07
I really enjoyed reading this comments thread. Very interesting.
18. September 2012 at 20:38
Tyler, you wrote “the claim that wages are outright sticky for long periods of time, when economic pressures dictate wage declines, isn’t holding up that well”. Scott is claiming the graphs support the claim of nominally sticky wages, which is the whole reason why devaluation is supposed to increase employment.
18. September 2012 at 20:49
In Germany the workforce had to endure some cuts or a stagnation of compensation in previous years. Since the last year went pretty well unions are demanding raises…
18. September 2012 at 22:15
I tried to make a graph like Selgin did but for Sweden. I could only find wage data for manufacturing though. Not sure if this is of any interest. But it was surprising to see a fall in 2010.
http://research.stlouisfed.org/fredgraph.png?g=aPZ
19. September 2012 at 02:14
Wage data for Greece are available until 2012-I, not 2012-II. See Eurostat. Greece has, contrary to popular mythology, a very flexible labor market, with a unusual high percentage of people which are ‘self employed’. And wages are falling fast, at a rate of 15% a year. (fifteen percent a year). The Greek government has also, contrary to popular mythology, cut its deficit as fast and as much as for instance the Baltic countries. Doesn’t help of course.
19. September 2012 at 02:24
An important phenomenon in Portugal for the public sector is automatic promotions+demographics.
Let’s say that every 5 years on the job, you get a promotion with a salary increase (some public sector jobs are like that). If your workforce the demographic structure that is common in southern Europe with very few people being hired or young and a lot around 45~55 year old, you get a cost increase without changing any salary (some years, with a salary and hiring freeze, public sector wage costs still went up 3~5%).
19. September 2012 at 05:53
Mark (Sadowski),
are you able to post links to the underlying data in Eurostat for nominal hourly wages? I cannot navigate that site 🙁
The ONS has some data on nominal hourly wages for the UK, though not much confidence in the data. It does not show any year-on-year drop in hourly wages, though late 2011 saw a very sharp squeeze to 0.4% yoy growth; even sharper than the NGDP squeeze (roughly 2% yoy), interesting in light of the relatively strong jobs figures.
http://www.ons.gov.uk/ons/rel/lms/labour-market-statistics/august-2012/table-earn08.xls
19. September 2012 at 06:20
Scott,
Dutch wages may be rising but real net purchasing power is declining there. Als the gvt takes pride in conforming to EU budget guidelines resulting in a cumulative 4% decline in household purchasing power over the next four years. However the top economists from the Treasury dept (n a recent article in ESB, the country’s main economics journal) are convinced that only that will take The Netherlands onto a path of sustainable public finance, taking into account demographics and overall low real GDP growth.
Unpleasant for the Dutch but plausible within an environment of strict constraints. The pundits see one little problem: due to population ageing (and they forget a few other factors) “zorg” (“care”) costs grow much faster tha everything else and in order to stay within the EU budget constraints something will have to happen to that.
No doubt all of this can be made better through a steady (level NDGP targeting) monetary policy approach.
19. September 2012 at 14:30
Britmouse,
That’s funny, I love Eurostat and hate the UK’s ONS. Remember the time I asked you for directions there?
Here is the quarterly “table”. If there is quarterly “themes” I haven’t yet found it:
http://epp.eurostat.ec.europa.eu/tgm/table.do?tab=table&init=1&language=en&pcode=teilm100&plugin=1
I usually go to the Browse/Search Database section of Eurostat and if I have trouble finding something I use the Search Tree option. It’s very helpful with arrows pointing the way.
19. September 2012 at 17:51
Thoughts on this Bryan Caplan post (responding to a post from Eli Dourado):
http://econlog.econlib.org/archives/2012/09/long-run_unempl.html
“Akerlof, Dickens, and Perry’s “The Macroeconomics of Low Inflation” made a very strong case for this back in 1996. The last four years have strongly vindicated their position. Their key point: Firms are heterogeneous, so even when most firms are doing fine, a minority need to cut real wages. With modest inflation, these firms can cut real wages covertly via inflation. With low inflation, firms either have to cut nominal wages and incur workers’ wrath, or cut employment. ”
“[T]he main problem isn’t that unemployed workers are “stubborn” about their nominal wages. The crucial “behavioral postulates” are rather that:
1. Hiring new workers for lower wages provokes resentment and resistance from existing employees – the classic insider-outsider mechanism.
2. At this point, the unemployed will probably say “yes” to jobs even if they they perceive their nominal wages as “unfair.” But after a brief honeymoon period, these new workers would probably have low morale. This wouldn’t just hurt their productivity; it would also bring down the productivity of their better-paid co-workers.”