Alesina and Giavazzi have a point

by Henry on October 8, 2007

Alberto Alesina and Francesco Giavazzi at “VoxEU”:http://www.voxeu.org/index.php?q=node/596.

Our point is that the goals that are traditionally held dear by the European left – like protection of the economically weakest and aversion to excessive inequality and un-earned rewards to insiders – should lead the left to adopt pro-market policies. What has often been the norm in Europe from the 60s until recently – heavy market regulation, protection of the status quo, an enormous public sector which rewards not the very poor but the most-connected and requires highly distortionary taxation, universities which often produce mediocrity in the name of egalitarianism (while the very rich get a good education anyway, somehow) –all end up decreasing efficiency and justice at the same time. … A good example can be found in the labour market. In Italy, Spain, and France, the labour market is split. The young are hired with temporary contracts which offer no social security and no prospects. When the contract expires, the employer opts not to renew it, so as not to run the risk of having to convert temporary hires into permanent employees who would de facto immediately acquire the right never to be fired. Reforms that eliminate this duality by making the entire labour market flexible with an appropriate scheme of unemployment compensation would not only reduce unemployment but, most importantly, would favour the really poor and the young entry-level workers.

This is true – but there is an important qualification. Some degree of market liberalization _would_ almost certainly help lower inequality and help the weakest in countries such as Italy. Anyone who’s spent time there, knows how difficult it is for young people to get jobs and acquire independence from their parents (people in their late 20’s and early 30’s still have sex in cars because they are living at home and can’t otherwise get away from their parents), and how much people’s life success depends on connections to the powerful. But the insider-outsider argument for egalitarian liberalization is weaker in countries a little further north such as France, much weaker in countries like Germany, and not convincing at all in the Scandinavian countries. When I read Alesina and Giavazzi’s book, “The Future of Europe: Reform or Decline?” a couple of years ago, I thought it was interesting, but flawed. It seemed to me that behind the broad rubric, ‘Europe,’ lurked a rather more particular jeremiad against ‘Italy and its problems.’ This essay suggests that I was right.

More generally, when you read any semi- or quasi-popular book about the European economic model, it’s almost always a useful mental exercise to think about which country or small set of countries the author had in mind when he/she was writing the book. European countries vary quite substantially in how they organize their economies – there really _isn’t_ a European model as such. Lefties who write about Europe with an idealized Sweden in mind are as guilty of this as rightwingers fulminating about the faults of the French state. Books about the European Union, and genuinely comparative books about the differences within Europe are a different matter, of course.

{ 31 comments }

1

stostosto 10.08.07 at 8:36 pm

But the insider-outsider argument for egalitarian liberalization is weaker in countries a little further north such as France, much weaker in countries like Germany, and not convincing at all in the Scandinavian countries.

And the Scandinavian countries, too, display a wide variety in their labour market particulars.

Btw, to the extent that they reallyh are successful in Alesina and Giavazzi terms, it has a lot to do with some of their favoured labour market policies — egalitarian liberalization, as you call them — already being in effect. The level of unemployment compensation may be rather high (if mostly for the low wage segment), but there is a high and very real institutional pressure on them to actively seek employment. Some of it in the form of blatantly idiotic obligatory job training programmes…

Also job protection is way lower than in France, Italy, or Germany. Thus, Denmark’s much-touted “flexicurity” arrangement makes it very easy to fire people indeed.

2

dsquared 10.08.07 at 8:41 pm

all end up decreasing efficiency and justice at the same time.

blackboard economics in action!

Of course, in an economics textbook French labour policies give it lower productivity and higher inequality than if it were to adopt the UK/US/Australia model, but in actual fact, they don’t.

3

Maria 10.08.07 at 8:45 pm

The characterisation of “an enormous public sector which rewards not the very poor but the most-connected” struck me as rather parochial and specific to Italy, and maybe some other southern European countries. Needing ‘pull’ to get a public sector job isn’t the norm in northern Europe.

For example, in France, there is a highly formalised, open and merit-based process for divvying up the best public sector jobs amongst ENA graduates. In the UK, the ‘fast stream’ scheme recruits quite effectively from top flight academic programmes. In Ireland, you don’t get in the door without passing standardised exams open to anyone.

4

dsquared 10.08.07 at 8:48 pm

The young are hired with temporary contracts which offer no social security

what does this even mean in the context of a welfare state?

5

stostosto 10.08.07 at 8:51 pm

dsquared:

But isn’t the high French labour productivity to some extent the flip side of their labour market model? High wages pricing low productivity workers out of employment and at the same time giving employers an incentive to invest in labour saving technologies?

6

Bob B 10.08.07 at 9:04 pm

For years, to my recollection, the OECD has been delicately making the point in its regular surveys of the French economy that the combination of the relatively high statutory minimum wage and high employment taxes on employers to fund social security benefits have the predictable effect of discouraging employers from offering permanent jobs to young people.

“A certain degree of duality in the labour market, where very well protected workers (‘insiders’) are together with the unemployed and workers in insecure jobs (the ‘outsiders’) tends to foster exclusion. Present targeted policies relieve poverty quite effectively but need to be more employment-oriented. Better results from scarce resources would result from smaller increases in minimum wages. Increases in, combined with better targeting of in-work benefits would help to reduce poverty further.”
http://www.oecd.org/dataoecd/61/43/38840164.pdf

For comparative standardised unemployment rates for young workers in EU countries for the age range 15-14 years, try:
http://epp.eurostat.ec.europa.eu/portal/page?_pageid=1996,39140985&_dad=portal&_schema=PORTAL&screen=detailref&language=en&product=EUROIND_LM&root=EUROIND_LM/euro_lm/lm_un/lm011rt

By EU standards, the unemployment rate for young people is higher in France than in almost all other EU countries. From the perspective of impeccably orthodox, mainstream economics that is not surprising. France has some outstanding, internationally renown economists but the political classes in France tend to ignor them.

7

Bob B 10.08.07 at 9:21 pm

There seem to be problems accessing the Eurostat website – which is not an unusual experience. However, try this report from EU Business dated 3 April 2006:

“France’s youth unemployment rate fell slightly in the past year, according to new government figures, but it remains one of the highest in the European Union, with ethnic minorities especially hard hit.

“Unemployment among under 25-year-old jobseekers stood at 22.2 percent in February, having fallen 0.9 percentage points over 12 months, according to the latest figures released by the employment ministry.

“Overall unemployment, seasonally adjusted on the International Labour Organisation (ILO) definition, was stable at 9.6 percent.”
http://www.eubusiness.com/Employment/060403154452.c0egg72s/

Otherwise, try googling on: “Eurostat unemployment”.

8

dsquared 10.08.07 at 10:28 pm

High wages pricing low productivity workers out of employment and at the same time giving employers an incentive to invest in labour saving technologies?

well on your first, the “batting average effect” just can’t be big enough to explain more than a fraction of the observed gap, and on the second, I would take some convincing that incentivising investment in industrial capital is an adverse, unintended effect of national policy. If French employers have a surplus to invest in machine technology then that surplus has come from somewhere, and if the French economy is specialising in high value-added activities then that’s probably a good thing.

9

Bob B 10.08.07 at 11:12 pm

But none of that helps to explain why the official standardised unemployment rate in France is persistently high compared with Britain or Germany nor why successive French governments have been unable to curb the budget deficit as required by the Eurozone’s Stability and Growth Pact:

“The French government on Wednesday unveiled a budget for 2008 that does little towards balancing public finances only days after the prime minister warned the country was ‘bankrupt’.

“The first budget under Nicolas Sarkozy’s presidency puts on hold France’s efforts to trim its public deficit and debt in spite of the protests of its eurozone partners, who want Paris to stick to promises made by the previous government.”
http://www.ft.com/cms/s/0/7e5bb248-6c27-11dc-a0cf-0000779fd2ac.html

10

Andrew Levine 10.08.07 at 11:57 pm

>the official standardised unemployment rate

What standardised rate would that be? Standardised to what?

11

Bob B 10.09.07 at 12:35 am

“What standardised rate would that be? Standardised to what?”

EU country governments used to define “unemployment” in different ways, usually relating to eligibility for state unemployment benefits although the eligibility criteria for benefits differed between countries. This meant that inter-country comparisons of unemployment rates were mostly meaningless without applying complicated reconciliation methods.

The “standardised unemployment rates” relate to standardised polling surveys in which samples of respondents are asked an agreed set of questions at regular intervals about personal employment status and working intentions if they are not currently working. The polling methods and survey questions were agreed through the International Labour Organization (based in Geneva) so the standardised unemployment rates are sometimes referred to as ILO Unemployment Rates and that is the practice applied in EU countries with the results published via Eurostat, the official EU organization for data collection and publication. The US also collects and publishes unemployment data on the ILO basis so comparisons can be made between the EU and the US.

This is about the best we can do for comparative unemployment figures for EU countries in the absence of a harmonised social security system for the EU – when some EU countries, notably Britain, are averse to harmonising social security systems.

References:
http://www.bls.gov/opub/mlr/2000/06/art1full.pdf
http://laborsta.ilo.org/applv8/data/iloce.pdf

12

notsneaky 10.09.07 at 12:44 am

well on your first, the “batting average effect” just can’t be big enough to explain more than a fraction of the observed gap

Do you have a source for this? Honest curiosity here. Also for how big the productivity gap is.

13

DRR 10.09.07 at 12:45 am

That’s a question worth repeating. Why are the unemployment rates in France so high as compared to that of their neighbors and other European countries?

14

notsneaky 10.09.07 at 12:50 am

It’s also worth remembering that productivity and efficiency aren’t the same thing. You can have higher productivity (defined as say output per work hour) and lower efficiency (lots of unemployed people who don’t enter either the numerator nor the denominator). Which is what blackboard economics predicts. But that slip up is probably attributable to Alesina and Giavazzi since it’s pretty common.

15

Bob B 10.09.07 at 1:38 am

“Why are the unemployment rates in France so high as compared to that of their neighbors and other European countries?”

That’s a good question. For top notch, mainstream attempts to answer it try the OECD Survey of France 2007:
http://www.oecd.org/document/15/0,3343,en_2649_34569_38808143_1_1_1_1,00.html

– with the associated Policy Brief:
http://www.oecd.org/dataoecd/61/43/38840164.pdf

Also, I’ve just found this recent (20 September 2007) online essay about France’s persistently high unemployment rate by Olivier Blanchard, an internationally renown French economist who teaches at the MIT:
http://www.voxeu.org/index.php?q=node/561

16

Bob B 10.09.07 at 1:51 am

This is recent illuminating chart from the UK’s Office of National Statistics showing comparisons for productivity per worker between major economies in 2006:
http://www.statistics.gov.uk/cci/nugget.asp?id=160

17

Cranky Observer 10.09.07 at 1:58 am

> The characterisation of “an enormous public sector
> which rewards not the very poor but the
> most-connected” struck me as rather parochial and
> specific to Italy, and maybe some other southern
> European countries. Needing ‘pull’ to get a public
> sector job isn’t the norm in northern Europe.

EADS.

Cranky Observer

Yes, it is a real company that produces real (and often quite good) products. But it is a public sector entity nonetheless, and it is not managed according to pure competitive principles (or even pure business principles)

18

dsquared 10.09.07 at 6:44 am

Radek: I looked the figures up on Eurostat for a comment on the Guardian website a while ago. Basically, we’re talking about an 8% difference in the employment rate, a difference in output per employee of 13% and a difference in output per hour worked of 20%. If you do the arithmetic on the assumption that the productivity numbers are purely batting average effects, it implies amazingly (implausibly) low productivity of the marginal workers.

Also ontopic is the good old French youth unemployment fallacy. The main reason why young people in France have a very high unemployment rate is that they have a very low labour force participation rate, because they have a very high rate of participation in full-time education. (I also don’t accept the implied assumption that unemployment in France and similar European countries is involuntary – these are all countries with high replacement ratios and generous unemployment benefit terms).

19

notsneaky 10.09.07 at 7:21 am

Ah, you were talking about the differences between UK and France, not UK/US/Australia and France. You’re right then – UK does come out looking pretty bad in terms of productivity, efficiency and a whole lotta other stuff. But the comparison with US I think is different (though not as much as some people think) – there the blackboard economics starts looking somewhat better. Which suggests that it makes as much sense to talk about the “Anglo-Saxon model” as it does about a “European” one.

20

Bob B 10.09.07 at 10:02 am

“Which suggests that it makes as much sense to talk about the ‘Anglo-Saxon model’ as it does about a ‘European’ one.”

One important insight is that there isn’t a uniform European alternative to the Anglo-Saxon model. Try this influential paper on: Globalisation and the reform of the European social models, prepared by André Sapir for the think-tank Bruegel, and presented at the ECOFIN Informal Meeting in Manchester in September 2005.

It argued that there is not one European social model, but rather four – the Nordic, Anglo-Saxon, Mediterranean and the Continental:

• The Nordic model (welfare state, high level of social protection, high level of taxation, extensive intervention in the labour market, mostly in the form of job-seeking incentives)
• The Anglo-Saxon system (more limited collective provision of social protection merely to cushion the impact of events that would lead to poverty)
• The continental model (provision of social assistance through public insurance-based systems; limited role of the market in the provision of social assistance)
• The Mediterranean social welfare system (high legal employment protection; lower levels of unemployment benefits; spending concentrated on pensions)

http://www.euractiv.com/Article?tcmuri=tcm:29-146338-16&type=News

Andre Sapir’s paper on: Globalisation and the Reform of European Social Models (November 2005), is here:
http://www.bruegel.org/Public/fileDownload.php?target=/Files/media/PDF/Publications/Policy%20Briefs/PB200501_SocialModels.pdf

On the reasons for the UK’s relatively low productivity performance compared with other major European economies, this is one of the main conclusions in a research project at the LSE :

“The persistent productivity gap between the UK and the two big continental European economies can mainly be ‘explained’ by the fact that they have more capital invested per worker and their workers are more skilled.”
http://www.lse.ac.uk/collections/pressAndInformationOffice/newsAndEvents/archives/2004/UKs_ProductivityGap.htm

21

Z 10.09.07 at 12:26 pm

For the sake of scientific exactitude, I remark that the division of eureopean economies in four areas (Nordic, Anglo-saxon, Continental, Mediterranean) is not due to André Sapir but to a series of works by Boyer and most notably Bruno Amable. Highly recommended, if I may say so.

22

Tom T. 10.09.07 at 12:28 pm

Re: 18. How is cause and effect determined? What if the French youth are staying in school in higher rates because there aren’t jobs available when they leave?

23

Bob B 10.09.07 at 12:31 pm

“the division of eureopean economies in four areas (Nordic, Anglo-saxon, Continental, Mediterranean) is not due to André Sapir but to a series of works by Boyer and most notably Bruno Amable.”

Please post the links.

I cited the Sapir paper because it is readily accessible and was circulated to ECOFIN ministers in 2005 during the UK’s presidency of the EU.

24

Jonathan 10.09.07 at 1:30 pm

More on the limits of labour market flexibility in explaining European unemployment with this detailed review of the econometric evidence:

http://www.cepr.net/index.php?option=com_content&task=view&id=66&Itemid=8

It doesn’t suit Alesina & Giavazzi (and others) to notice but the EU is a single market, largely operating with one currency. Much is made of limited remaining barriers, but essentially the trend for two decades or more has been for major market liberalisation. It hasn’t delivered noticeably better performance, indeed actual gains have typically been below predicted. And so the mantra continues that more liberalisation must be needed, it must all be down to remaining barriers to economic activity. Alternative explanations for European economic performance are ruled out by assumption at the start.

25

Alex 10.09.07 at 1:53 pm

I think this piece of mine stands up rather well on the subject of exaggerating France’s problems for internal political purposes, as does this onethis one of Emmanuel’s on the media.

You have to wonder whether, even if 22 is true, if they go into higher value-added jobs in the end it’s not a blessing in disguise.

26

stostosto 10.09.07 at 3:00 pm

Re ##20, 21 & 22:

I have sometimes seen Gøsta Esping-Andersen credited with a seminal attempt at categorising welfare states, in 1990.

27

Z 10.09.07 at 3:32 pm

Bob b, Bruno Amable has published a daunting number of works, most of them are available at his web site, which in turn is the first google hit on his name.

Stostosto, sure people categorised economic systems before Amable. However, that particular classification (Nordic…) is due to him (or so I believe, I could be wrong of course).

28

Bob B 10.09.07 at 3:54 pm

Thanks for the links and references although I’m more interested here in the analysis than in scholastic priorities and it’s important to get away from the mistaken notion that there’s a uniform model for capitalism on mainland Europe that is not Anglo-Saxon.

To put that another way, some EU economies with high tax burdens are notably more successful than others. We need to take account of that and attempt to ascertain why.

29

dsquared 10.09.07 at 4:12 pm

#22: possibly, although obviously being at university is a) more fun and b) more useful than simply hanging around on the dole (or for that matter, being in prison, which matter surely ought to be brought up if we’re comparing US and European unemployment rates).

30

stostosto 10.10.07 at 12:52 pm

Re #24

It’s not simply a question of more or less liberalisation. It’s how you liberalise, how you regulate, how you tax, and how you protect. This is the crucial point in considering different welfare models. The Scandinavian countries, for instance, have notoriously high taxes and welfare levels. At the same time they have relatively well-functioning labour markets that on some dimensions are quite free-market based.

In particular, it’s easy to fire people (particularly in Denmark), and there are strict requirements on unemployed that they seek and take available jobs.

Regarding the EU as a single market: Well, yes, but by no means a single labour market. There are all sorts of local particularities and oddities.

31

stostosto 10.10.07 at 12:59 pm

dsquared #18:

(I also don’t accept the implied assumption that unemployment in France and similar European countries is involuntary – these are all countries with high replacement ratios and generous unemployment benefit terms).

So, the straightforward recommendation would be to cut back on benefits, right? This of course is the standard right wing line on unemployment…

Comments on this entry are closed.