MIT LIBRARIES
IllllllliHI
iiii..
3 9080 02618 3266
Pji:
R.liiiii':'
2005
Digitized by the Internet Archive
in
2011 with funding from
Boston Library Consortium IVIember Libraries
http://www.archive.org/details/europeanunemploy00blan2
DEWEY
HB31
.M415
.
.
Massachusetts Institute of Technology
Department of Economics
Working Paper Series
EUROPEAN UNEMPLOYMENT:
THE EVOLUTION OF FACTS AND IDEAS
Olivier
Blanchard
Working Paper 05-24
October 10, 2005
Revised: November 5, 2005
Room
E52-251
50 Memorial Drive
Cambridge,
This
MA 021 42
paper can be downloaded without charge from the
Network Paper Collection at
Social Science Research
http://ssrn.conn/abstract=825885
MASSACHUSETTS INSTITU^
OF TECHNOLOGv
JUN
2 G 2G0S
LIBRAE
European Unemployment: The Evolution of
Facts and Ideas
Olivier Blanchard
November
5,
*
2005
Abstract
In the 1970s,
in
European unemployment started
the 1980s, to reach a plateau
the average
unemployment
in
the 1990s.
increasing.
It is still
It
increased further
high today, although
rate hides a high degree of heterogeneity across
countries.
The
of shocks.
As unemployment remained
focus of researchers and policy makers was initially on the role
high, the focus has progressively shifted
to institutions. This paper reviews the interaction of facts
gives a tentative assessment of
what we know and what we
and
still
theories,
and
do not know.
MIT and NBER. Prepared for the Economic Policy Panel, London, October 2005. I thank
Giuseppe Bertola, Ricardo Caballero, Emmanuel Farhi, homas Philippon, Steve Nickell, Julio
Rotemberg, Thomas Sargent, and Robert Solow for comments. I thank the Russell Sage Foundation and the National Science Foundation for financial support.
Introduction
From the end
was
of
World War
II to
the end of the 19G0s, European unemployment
ver}' low. In the 1970s, it started increasing. It
1980s, to reach a high plateau in the 1990s.
the average European
unemployment
It
continued to increase in the
still
is
high today, although
rate hides a high degree of heterogeneity
across countries.
This has been a tough learning experience, both
makers.
When
ment was
just born,
and
Milton Friedman (1968)
level
economists and
for
for policy
the 1970s started, the concept of a natural rate of unemploy-
from operational. The following quote from
still far
"The natural rate
revealing:
is
of
unemployment
is
the
which would be ground out by the Walrasian system of general equilib-
rium equations, provided that there
characteristics of the labor
imbedded
is
in
them the
actual structural
and commodity markets, including market imper-
demands and
fections, stochastic variability in
information about job vacancies and labor
supplies, the cost of gathering
availabilities,
the costs of mobility,
and so on."
One might have hoped
in the evolution of
that, with thirty years of data, with clear difl'erences
unemployment
now have an operational theory
Many
come and
theories have
rates
and
policies across countries,
unemployment.
of
— partly — gone.
I
we would
do not think that we do.
Each has added a
layer to our
knowledge, but our knowlege remains very incomplete. To use a well worn
nnila,
we have learned a
The purpose
of this paper
ment and the theory
me
but we
lot,
is
fronts,
begin with two caveats.
my
my own
defense,
for-
lot to learn.
both on
tlie
unemploy-
and give an assessment of where we are today. Let
I
ha,ve not tried to
research— one of the
I
have a
to review the developments,
the editors unwisely encouraged
on
still
would argue that
me
to
do
so, I
be encyclopedic.^ And, because
have certainly focused too
much
results being a Stiglitz-like bibliography. For
it
is
broadly representative of the twists and
turns of our theories over the last thirty years.
I
review the basic
facts,
across time and across countries, in Section
1.
As
1.
A more cncyclopodic and very good survey, but, now 10 years old, wa,s given by Charles
Bean (1994). A more recent one was given by Stephen Nickell (1997). The standard reference
on unemployment in general, and European unemployment in particular, remains the book
by Richard Layard. Stephen Nickell, and Richard Jackman published in 1991 (2nd edition,
200.5). Another important contribution is the 1994 book by Ned Phelps.
unemployment increased
from
oil
price increases to the
topic of Section
2.
was on the
in the 1970s, the initial focus
slowdown
in
role of shocks,
productivity growth. This
As the shocks receded but unemployment remained
is
the
high, the
focus shifted in the 1980s to persistence mechanisms, from the role of capital
accumulation, to the role of insiders
in bargaining.
This
is
the topic of Section
3.
In the early 1990s, the focus shifted yet again, this time towards the role of labor
market institutions, from employment protection
This
the topic of Section
is
4.
to
unemployment insurance.
Since then, research has tried to sort out the
respective role of shocks, institutions, and interactions.
The main
directions of
The
state of play,
exploration and the open questions are the topic of Section
and whether we know enough
up
to usefully guide policy
5.
and reforms, are taken
in Section 6.
1
Basic Facts
Figure
(the 15
1
gives the evolution of the
unemployment
member
European Union)
steady increase
a rough plateau
countries of the
in
unemployment from 2%
in
rate for the
EU15
since 1960. It
the 1960s to
8%
whole
shows the
in the 1980s,
and
— with cyclical declines at the end of the 1980s and 1990s — since
then.
Figure
I960
as a
1.
EU15 unempioyment
1970
1980
rate,
1990
since 1960
2000
Source:
OECD
How much
reflects
database.
of the increase reflects an increase in the natural rate,
an increase of the actual rate over the natural rate? The answer to
that question
— around
2%
inflation rate to be an indication that
rate, this suggests that, today, the
the natural
using the
since 1970 reflects, for the
Figure
2.
1960
EU15
actual
most
tell
question
is
to
EU15
actual
1970
how
If
stable
roughly at the natural
rate
close to
is
unemployment
rate
in the natural rate.
and constructed
natural rate
1990
1980
been
2000
natural
database and text
the natural rate has increased over time?
obviously
be useful:
is
we take a
unemployment
an increase
part,
inflation has
index. If
unemployment
actual
Can one
CPI
follows that the increase in the actual
rate.-^ It
OECD
EU15
relatively straightforward: Since 2000,
is
indeed roughly constant
Source:
and how much
we
much
harder. Despite
are willing to
assume
its limits, I
that,
The answer
find the following exercise
when unemployment
natural rate, inflation will tend to increase, and
to that
is
below the
when unemployment
is
above
the natural rate, inflation will tend to decrease, we can construct a series for
the natural rate using the actual rate and the change in inflation.
2.
One may
The
results of
question however whether this relation holds at very low rates of inflation;
return to the issue in the
la-st
section of the paper.
I
such a construction are presented in Figure
rate increased in the 1970s
since then.
They suggest
2.^
that the natural
and early 1980s, and has remained roughly stable
,,
,
Heterogeneity Across Countries
Turning from the EU15 average to individual countries, Figure 3 gives the
unemployment
rates in each of the
the evolution of
unemployment
EU15
countries as of
rates in each
May
2005
EU15 country
(for reference,
shown
is
in the
appendix.)
F gure
Greece
"
EU15
3.
""
unerr ployment rates, 2005
**"*"'
""
"^"
Spain
France
^^
Germany
""'"""" *"""""*"'''*''''''^"'""""Biiiiin^H
Finland
Portugal
^^^^^'~'" ""'''*^^^^"^~"'
"""'"""'™™"' ^^^^^^^
^^^
"""""*""iiiiiii"'**** i.ii— 1—
Sweden
IIIIIW^IIHI
Belgium
Italy
II
^^^^""**'
IIIIIWIMIIII"""""""""''*
Netherlands
^^""'^"*"**
"'*^*"
Denmark
Luxembourg
United Kingdom
IIIWMIIIIIIIII
^—
in
^.
f
IIIIIIIIIIIIIWIIII Willi IIIIIII
!
*™''^™iiiiiii^iiiii
Austria
^^^
Ireland
2
C
6
4
10
8
percent
Source: Eurostat
The
figure
tries.
While
3.
The
n{ —
l)
shows the large heterogeneity of unemployment rates across counthis heterogeneity
scries for the natural rate
— a(u — u') where w
unemployment
is
has always been present,
is
it is
more marked
toda)',
constructed as follows: Start from the rcJation
it
=
the rate of inflation, u and u' are the actual and natural rates of
respectively. Rewrite the relation as u*
=
u
+
(l/a)(A7r).
The
scries for
ix'
in
Figure 2 is constructed using this relation, using a = .5 (a value consistent with econometric
estimates for Europe) and a three-year moving average of the change in CPI inflation for Att.
(A different value
for a
would change the amplitude of the movements
not the ordering of the two rates in a given year.)
in u" relative to u,
but
to the point
leading.
where talking about "European unemployment"
Unemployment
is
knv
in
many
Kingdom, the Netherlands, Denmark,
inent rates lower than the United States.
ployment
unemployment
reflects high
Germany, France,
But
have imemploy-
in the four large continental countries,
unemployment
rate
dif-
down from more than 20%
is
is
instead up from
its
low
and shows sharp regional differences between the West
and France's unemployment
Italy's
the early 1980s.
all
high average European unem-
Germany's unemployment rate
pre-reunifi-cation level,
and the East.
of mid-2()05, the United
and Austria
And
indeed mis-
and Spain. Even among these four countries, the
Italy,
ferences are striking. Spain's
in the early 1990s.'
As
countries:
Ireland,
is
Italy's rate
rates have
been high since
shows sharp regional differences between the
north and the south. France's does not.
Unemployment,
As
FIovv^s,
and Duration
a matter of arithmetic, a high
unemployment
rate
may be
the result of
high flows in and out of unemployment, or/and a high average duration of
unemployment. Figure
which gives the evolution of the unemployment rate
4,
and unemployment duration
employment by duration
unemployment
rate has
exists
year.^
France
of long term
year) has increased from
which data on the composition un-
20%
in the
large increase in duration. The figure sug-
was already higher than that of United States
more than doubled
The proportion
(for
from 1968 on) shows that the increase
come with a
gests that duration, which
late 1960s, has
in
since then,
and now stands
unemployed (unemployed
in the late 1960s, to
in the
at well over a
for
more than a
more than 40% today. From
the point of view of the unemployed, being unemployed in Europe has always
been a different experience from being unemployed
mean duration
has remained around 3 months
—
,
in the
United States
— where
and has become increasingly
so over time.
In 1994, the official number for the unemployment rate reached 24%. The definition of
unemployment and the number.s have since been revised, and the current time series have
unemployment peaking at 18.4% in 1994.
5.
The increase in duration would be even larger, were it not for the increasing role of
4.
temporary contracts since the early 19S0s, contracts which are
ensuing unemployment spells.
typically associated with shorter
Unemployment
Figure 4.
and duration
rate
France
in
E
°^
1970
1980
Source: Enquetes Emploi,
2000
1990
2010
•
Unemployment
rate
»
Unemployment
duration (months)
INSEE,
adjustment by author
series longues; with
for 1968 to 1974.
Unemployment Rates
Another
across Workers
dinicnsioii of uii(;uij)loyiiiciit
is
how
it
affects different groups, skilled
versus unskilled workers, young versus older workers,
men
versus
often-mentioned characteristic of European unemployment
employment
plots the
rate
unemployment
for
all
rate for the 15-24 age group against the overall
unem-
EU15 country
Figure
in
and Greece, indeed have very high youth unemployment
rates, in excess of
for
each
this reflects a uniquely
countries, high
unemployment
some groups, the young and the
experience of Europe
responding numbers
is
European countries
time
for the year 2004.
European pathology
is
associated with higher
To
unskilled in particular.
in this respect, I also plot in
United States
circle.
is less
for
Put simply, the points
for
and Greece indeed have much more
line
would
predict,
clear however.
unemployment
see
whether the
Figure
each year since 1960
Germany much
5
the cor-
—each year
the cross section of
are not far off the regression line one
series. Italy
than the regression
unusual
for the
represented by a small
US
the un-
which
is
countries, such as
25%. Whether
In
is
shown
5,
workers. This
Some
ployment rate
Italy
among young
is
women. One
how high
would obtain from
youtli
less
unemployment
(because of
its
ap-
prenticeship programs); on average, the experience of the
EU15 does
not appear
that unusual^
Figure
c
EU15
'
—
..•--
'
percent
US
rate,
countries and across time for the
ilTA
" "*
ss-
Youth unemployment
5.
across
*GRC
'^ "" — "
'
•
<D
CL
B
A ESP
c/f
2O
C.
ro
<D
*FIN
o
CM
^FRA
_
*BEL
>,
sf
CM
•
•
1
i^u^
.
-
*PRT
"
*SWE
s
^1,
••
«
ra
A LUX
-^BR*!!
*
•
*DEU
ment
10 T
>.
o
KDNK
•
•
••
•
*"
•
'mnL
o.
E
c
AAUT
o
zjm
-
,
Unemployment
Source:
OECD
..,„
...
.
10
8
6
—
rate,
12
percent
database
Unemployment
versus Other Labor Market Indicators
Yet another question
is
whether the unemployment rate
is
in fact a
may
in-
employment
dicator of the general state of the labor market. Faced with poor
prospects, people
good
decide not to join the labor force, students
may
decide
to stay longer at the university; these decisions will show up as lower participa-
tion rates, not as higher
unemployment
rates.
reduce unemployment numbers; and indeed
Governments may
many governments
also
want
to
have. Measures
have ranged from training programs, real or perfunctory, to generous disability
programs (the example
early retirement
of the Netherlands being the best
known), to subsidized
programs (which have played an important
role in
France since
the mid 1980s.)
6.
Bertola et
employment
that decreases
stronger
in
al
unemployment and
and OECD countries. They conclude
the employment rates of younger (15-24) and older (55-I-) have typically been
(2005) provide a detailed description of the patterns of
rates for different age groups across time
in
more unionized
countries.
There
is
thus httle question that a complete examination of the labor mar-
ket requires looking not only at
and employment
rates J
But
it
unemployment
is still
rates,
but also at participation
the case that, in the case of European
unemployment, focusing on the unemployment rate
is
not misleading. This
is
because, in general, depressed labor markets have shown not only higher un-
employment, but
and employment
also lower participation
example, the participation rate of men, aged 55 to 59,
85%
in
the late 19GOs to below
70%
retirement, precisely at the time
Figure
6.
in
in
rates.
To take one
France, went from
the mid 1980s, reflecting subsidized early
when unemployment was
Unemployment and
increasing.
participation rate in Spain
CO
1960
'0
1980
•
Unemployment
o
2000
1990
rate
A Participation rate
Figures 6 and 7 provide more time series and cross section evidence on this
point. Figure 6 shows the evolution of the
rates for Spain since 1960.
It
unemployment and participation
shows how the large increase
how
associated with a decrease in participation, and
has been associated in turn with a large increase
the
in
more recent decrease
in participation.
the employment rate against the unemployment rate across
men
7.
for the year 2003,
The
participation rate
is
unemployment was
Figure 7 plots
EU15
covmtries for
and shows a strong negative correlation between the
defined as the ratio of
to the population of working age.
to the population of working age.
The employment
llie
rate
sum
is
of the
employed and unemployed
defined as the ratio of
employment
two.
Figure
7.
Unemployment and Employment
2003,
NLD
•
EU15
rates of
men
countries
.DNK.(3BR
AUT .IRL
1
i
i
"BEL
10
Unemployment
rate,
percent
m-
Having
laid the basic facts,
The
2
Initial
I
now
look at the history more closely.
Rise in Unemployment.
The Role
of Shocks
Along a balanced growth path, the wage consistent with stable employment
must grow
if
at the rate of Harrod-neutral technological progress.'"* In addition,
the prices of the other factors of production increase, the wage must de-
crease so as to maintain zero net profit for firms. Call this
wage the "warranted
wage." Call the wage set in bargaining the "bargained wage".
If,
for given labor
market conditions, the bargained wage grows faster than the warranted wage,
equilibrium employment will decline, and the natural rate of unemployment will
8.
The reason
ipation rates of
for focusing
women
on men
in
the figure
is that,
cross-country variations
reflect in part cultural differences.
A
in
the partic-
low participation rate of
women
an indication of a liadly functioning labor market.
9.
Much of our intuition and most of our models are based on the assumption that technological progress is Harrod-neutral and that there is a. balanced growth path. What happens
is
not, per se,
if
not
is
largely unexplored, but
may
well be relevant.
10
This proposition
in the 1970s.
the key to understanding what happened to
is
European countries were
which impHed a slowdown
hit
by a
unemployment
series of adverse shocks,
shocks
the rate of growth of the warranted wage:
in
Just Hke the rest of the world, European countries were hit by two major
price increases, the
first
one triggered by the Arab
embargo
oil
oil
of 1973-1974,
the second by the Iranian revolution in 1979 and the Iran-Iraq war of 1980.
Figure 8 gives the price of
shows
4 times
oil,
in dollars
by the early 1980s, the
that,
its level
and
8.
Nominal and
2005
terms of
its
1965
1970
1975
less visible initially
1980
real price of
down
crude
1985
1990
.
1995
2000
2005
,
but eventually more important, both
considerably.
By
which had been high
in the
in
1950s and 1960s,
the late 1970s, the rate of Harrod-neutral tech-
nological progress (constructed using the Solow residual,
labor share), which had run at more than
down
oil
impact on growth and on unemployment, was also at work. Total
factor productivity (tfp) growth,
slowed
stood at nearly
dollars
Source. U.S. Department of Energy.
Another shock,
in dollars,
oil,
at the start of the 1970s.
Figure
1960
in real (U.S.) terms, since 1960. It
real price of
5%
to 2%, In other words, the annual rate of
in
and dividing
it
by the
the 1950s and 1960s, was
growth
of
warranted wages had
decreased by 3 percentage points, a dramatic decline. Figure 9 gives five-year
11
averages of estimates for the
Italy,
and the UK)
— the EU5
similar across countries
Figure
EU
countries (Germany, France, Spain,
for short. It
shows that the dechne was largely
five
major
10
Rate of Harrod-neutral technological progress,
9.
EU5
Centered five-year averages, 1968 on
00 -
q
• ITA
• ESP
• FRA
—
• GBR
• FRA
.D^E^
r
1
^
s
m
——
.
.
—
• GBR
fj
tm
• ESP
• ff*A
• DEU
"to
o
-
• FRA
• ITA
1970
1975
1980
• ITA
• sas
• ESP
• DEU
• ESP
Ji^
• DEU
• FRA
• DEU
1965
• GBR
1985
1990
•
1995
ITA
2000
Source: Blanchard Wolfers (2000) database.
These two shocks would have required slowdowns
wages to avoid an increase
labor unrest in
Italy,
unemployment. In
countries
growth of actual
both came
after a period of
fact,
— May
1968 in France, Spring 1969 in
the end of dictatorships in Portugal and Spain in 1974 and 1975
workers had asked
of lower
in
in
many European
in the rate of
for increases in
—
in
which
wages. Not surprisingly, the joint outcome
growth of warranted wages and higher wage demands was an increase
unemployment. By the end
increased to 5%, up from
2%
of the 1970s,
unemployment
for the
at the start of the decade; Spain's
EU15 had
unemployment
rate exceeded 10%, Prance's and Italy's exceeded 6%.
The development
this story,
10.
of a conceptual frame,
and the econometric
fleshing out of
were largely the work of Michael Bruno and Jeffrey Sachs, who put
For a tentative explanation, see Peter Temin (2002).
12
it
together in a series of articles and then in a book in 1985." Their book can
be seen as a
natural rate.
first
attempt to put together a working theory of movements
They argued
tliat
in
the
the rise in unemployment could be explained of
shocks interacting with two types of
rigidities, real
and nominal (Box
1
gives
the basic algebra):
"Real wage rigidities" captured the speed at which real wages adjusted
•
to changes in
warranted real wages, the speed at which,
employment, workers would
wages
in
for given un-
example accept a slowdown
for
response to a productivity slowdown.
ment, the higher and the longer lasting the
The
in actual
slower the adjust-
efTects of adverse
shocks on
unemployment.
"Nominal wage
•
rigidities"
captured the speed to which nominal wages
adjusted to changes in prices.
The
slower the adjustment, the larger
the decrease in the real wage in response to an unanticipated increase
in prices.
And by
implication, the slower the adjustment, the
monetary authorities could use
inflation to reduce real
fore limit the increase in actual
unemployment
more the
wages and there-
response to an adverse
in
supply shock.
Differences in real
ilar
A
and nominal
could explain why, despite largely sim-
rigidities
shocks, different coiintries experienced different increases in unemployment.
smaller increase in
unemployment could be due
to smaller real rigidities, re-
sulting in a smaller increase in the natural rate; or
nominal
flation,
a
rigidities,
an unemployment rate below the natural
more aggressive use
unemployment
Where
it
could be due to larger
allowing policy makers to achieve, through the use of in-
of
monetary
rate; or
it
could be due to
policy, leading to higher inflation
and an
rate below the natural rate.
did these differences in real and nominal rigidities themselves
come from?
Differences in real rigidities were naturally traced to differences in the structure
of collective bargaining.
Sweden, with an unemployment rate of 2.2% at the
end of the decade, was seen as a poster child
centralized bargaining and strong unions.
for the case for corporatism,
An important
that of Lars Cahnfors and John DriffiU (1988),
11.
who
i.e.
contribution here was
argued, both theoretically
Other researchers share the credit, among them Robert Gordon (for example Gordon
ModigHani and Tommaso Padoa-Schioppa (1977), Branson and Rotemberg
1975), Franco
(1980).
,
,
13
and empirically,
that, in the face of adverse supply shocks, countries with either
very centralized bargaining or very decentralized bargaining would fare better
than those with intermediate bargaining structures. With centralized bargaining
in particular, the parties at the bargaining table could see the
implement the wage adjustment required
need
and
for
employment.
to maintain
Differences in nominal rigidities were also traced to collective bargaining, albeit
to different aspects of
The
it.
High indexp^tion
countries, played a central role.
monetary policy
many European
degree of indexation, present in
to limit the increase in
prevented the use of
in effect
unemployment, or required a very high
rate of inflation.
must be seen
Overall, this initial strand of research
as a
major achievement.
Macroeconomists had entered the 1970s without a model of the natural
and had not anticipated
stagflation.
working model of the natural
the increase in
rate,
By
rate,
the end of the decade, there was a
And
and stagflation was well understood.
unemployment was explained by adverse shocks
intera.cting
with
country-specific collective bargaining structures.
Continuing Unemployment. Sources of Persistence
3
Unemployment continued
EU15
The
in 1980, to
8%
to increase
at the
further increase in the
end
throughout the 1980s, from
of the decade, with a
half of the 1980s
first
peak of 9.5%
was
still
5%
for
the
in 198G.
easy to explain.
Partly accomodating monetary policy in response to the adverse shocks of the
1970s had led to a large increase in inflation: In 1980, EU1.5 inflation was 12.5%.
Throughout Europe, governments and
through
By
tight
1986, the
monetary
EU15
inflation rate
through a large increase
central
policy, starting
in the
banks decided to reduce
inflation
UK
in 1979.
with Mrs Thatcher
in the
was down to 3%. This was achieved however
unemployment
rate
—
reflecting
an increase
in
the actual rate of unemployment over the natural rate.
For the rest of the decade however, inflation was roughly stable, an indication
that the actual
unemployment
—around 8-9%)
rate
explain:
for the
rate
was now
close to the natural
unemployment
EU15. This high natural rate was more
As can be seen from Figure
8 earlier,
difficult to
by the mid-1980s, the sharp
14
increases in oil prices
growth was
But
it
wage
still
had been
much
very
present,
appeared increasingly
setters
largely reversed.
and now
unlikely,
decline in productivity
well understood
more than
would not have adjusted
The
to the
and documented.
ten years after the decline, that
new
reality of lower productivity
growth.
This led researchers to focus on persistence mechanisms, on why the
initial
ad-
verse shocks might have very long lasting effects on unemployment. Research
focused mainly on two mechanisms, capital accumulation, and the role of insiders in collective bargaining.
Capital accuniidation was already one of the themes of Bruno and Sachs.
It
was
the focus of a major project later on, directed by Charles Bean and Jacques
Dreze (1991). The basic
down
in productivity
logic
was straightforward:
growth or an increase
in
If,
in
response to a slow-
the price of non labor inputs,
bargained wages did not adjust fast enough, employment decreased.
ment decreased,
so did the profit rate.
And
If
as long as the profit rate
employ-
was below
the user cost, capital decreased over time, leading to a further decrease in em-
ployment.
The dynamics
and deep increase
in
of capital accumulation could therefore lead to a long
unemployment.^^
This had interesting and highly relevant imphcations
for
monetary
policy;
In that context, expansionary monetary policy potentially played two roles.
was, as before, to decrease real wages and limit the decrease in employ-
first
ment
for
rate,
and
a given capital stock.
b,v
The second was
to decrease the real interest
implication the user cost; by doing so
capital accumulation,
Both channels had
and
clearly
been at work
—
forecasts of inflation at the time
in the
banks
in
limited the decrease in
employment over
time.
second half of the 1970s. In-
were negative, and
— using
so were ex-ante real interest rates in
European countries (Blanchard and Summers
By symmetry,
it
so the further decrease in
flation steadily increased; ex-po.st real interest rates
most
1984).
a monetary contraction, such as that engineered by most central
the early 1980s, also had two
effects.
The
flrst
one was to increase
real wages, and thus decrease employment given the capital stock.
was
The
to increase the real interest rate,
The second
and thus decrease capital accinnulation,
12. The focus here is on the effect on the increase in the natural rate. The decrea.se in investment demand could well lead to an even larger increase in the actual unemployment rate.
15
and by implication, further decrease employment. Again, both channels were
clearly at
work
in the first half of the 1980s. Inflation
real interest rates
was sharply
were much higher than they had been
In short, the delayed reaction of monetary policy,
first
and
lower,
earlier.
accomodating and
later
contractionary, could explain wliy the effects of the initial shocks were in
fect delayed.
Under
this interpretation,
ef-
with a more neutral monetary policy,
the increase in unemployment would have been higher
initially,
but shorter in
duration.
An
interesting twist to the theory
in their
study of Germany.
was suggested by Hellwig and Neuman (1987)
bargained wages were set by looking at labor pro-
If
ductivity growth rather than at the underlying rate of technological progress,
then a vicious cycle could easily emerge. Suppose workers asked
for too
high
wages. Firms would respond by reducing employment, thereby increasing the
capital-labor ratio, and thus increasing labor productivity
— relative to the un-
derlying rate of technological progress. This might trigger further wage de-
mands, further decreases
and
in
employment, further increases
so on, leading to a potentially very large increase in
The second
in labor productivity,
unemployment.
focused on collective bargaining.
line of research
It
was based on
the idea that wage bargaining typically takes place between employed workers
(or,
more
specifically, their
union representatives) and firms, and that the un-
employed are not represented
at the bargaining table.
This "insider-outsider"
theory was developed by Lindbeck and Snower (as summarized for example in
their 1989 book),
and applied
to
unemployment,
then by Olivier Blanchard and Lawrence
first
Summers
by Robert Gregory (1986),
(1986) and by Nils Gottfries
and Henrik Horn (1987).
The
basic idea was straightforward. Suppose that unions set the
to the firms'
the
demand
for labor.
employment prospects
wage so
And
of the currently employed.
that, in expected value,
wage subject
suppose that unions cared only about
Then, they might
set the
employment remained the same. Because
of
unexpected shocks, employment would be sometimes smaller and sometimes
larger than expected. In other words,
and
for
a given labor
force, so
employment would
follow a
random
walk,
would unemployment. There would no longer
be a natural rate of unemployment to which the economy would return; unem-
ployment would exhibit "hysteresis" not returning
,
to
any particular value, but
16
being determined instead by the wiiole
liistory of
shocks to the economy.
This extreme form of the theory was provocative, and rightly criticized as being
too strong:
Empirically,
in
it
implied that movements in the labor force would not be reflected
employment; but a strongly established
fact
is
that, even in
economies with
— due
high unemployment, exogenous movements in the labor force
phy
or repatriation, such as the return of
dence of former colonies
movements
for
in
(for
Europe from the 1970s
Theoretically, even
if
— translate
fairly quickly into
[1992])
also,
why would
and
hysteresis be relevant
at other times?
the unemployed do not participate in bargaining, there
unemployment
will afTect the
outcome. The
that, given the positive probability of finding themselves
employed workers should and
The
after the indepen-
on, but not elsewhere
are at least two reasons to think
first is
European nationals
example Hunt
employment. Empirically
demogra-
to
will care
unemployed,
about the state of the labor market:
higher the unemployment rate, the more careful they will be in setting the
wage.
The second
is
that wages are not set unilaterally by unions, but rather
by bargaining between imions and
firms.
\memployed; the higher the unemployment
These
And
firms can threaten to hire the
rate, the
criticisms suggested that the central role of
gaining implied persistence of unemployment
but typically not hysteresis. The
weak, but was not zero; even
if
effect of
in
more
relevant the threat.
employed workers
in bar-
response to adverse shocks,
unemployment on wages might be
the unemployed were not present at the bar-
gaining table, high unemployment
still
led the
economy
to return to the natural
rate, albeit slowly.
An
important extension to
this line of
argument was provided by Richard La-
yard and Stephen Nickell (1987), focusing on the effects of high unemployment
on human capital
—following
They pointed out
that, in
an argument
European
first
developed by Phelps
countries, high
unemployment
implied very high average unemployment duration (recall Figure
duration was likely to lead to
of the long
the
term unemployed
unemployment
loss of skills, loss of
in effect
4).
morale, and thus
in 1972.
typically
Such high
make many
unemployable. In that case, the higher
rate, the higher the duration, the higher the loss of skills,
the lower the pressure on wages from a given
unemployment
rate.
Separating
the unemployment rate between short-term unemployment (the ratio of those
17
unemployed
less
than a year to the labor force) and long-term unemployment
unemployed
(the ratio of those
and Nickell indeed showed
to
more than a year
for
to the labor force), Layard
what seemed
that, in Phillips curve type relations,
matter was short-term unemployment, not long-term. This provided a po-
tential explanation for
why
persistence was higher in
Europe than,
say, in the
United States (where the proportion of long term unemployed was and
low),
.
.
very
.
,
.
Overall, these developments again represented progress.
movement on employment and
capital,
real
wages and the
ing,
were important extensions of the
number
real interest rate,
The
focus on the joint
on the role of monetary policy through
on the implications of
initial
collective bargain-
framework. They also made clear a
of holes, theoretical and empirical, in our understanding of
mination. Were wages in collective bargaining set with an eye to
(more
specifically, the rate of
to the
dynamic
Looking
effects of capital
at bargaining
How
more
wage
TFP
deter-
growth
Harrod-neutral technological progress) or labor
As we saw
productivity growth?
ing?
is
earlier,
the answer makes a
lot of
accumulation on unemployment
closely,
how
did
unemployment
affect
difference
for
example.
wage bargain-
did employment protection, which clearly affects the probability that
employed workers
will find
thomsolves unemployed, affect the outcome? This
takes us to the next stage, the shift in focus towards labor market institutions.
4
Stubbornly High Unemployment. The Role of Institutions
In the 1990s, average
10.4%
for the
EU15
European unemployment remained very
high, peaking at
average in 1993, and ending at 7.6% in 2000 (a cyclical
peak; the unemployment rate stands at 8.6% in mid-2005.) But this average
reflected
•
an increasing heterogeneity of evolutions across countries:
Unemployment remained high
unemployment
rate,
about 10% (8.5%
Italy.
which had remained relatively low
1990s, steadily increased after reunification:
at
and
in France, Spain,
in
it
now
Western Germany, twice
Germany's
until the early
stands (mid 2005)
as
much
in
Eastern
Germany).
•
Unemployment decreased
Netherlands,
all
from high
to
under
5%
in
the
UK,
levels in the early 1990s.
Ireland,
and the
(Belgium, with an
18
—
unemployment
Flemish provinces
in the
the
is
rate of 8%,
unemployment
an interesting case; the unemployment rate
— those close to the Netherlands—
rate in the
Wallon provinces
— those
is
5%, while
close to France
11.0%).
Unemployment remained
•
is
gal.
And, while
it
relatively low in Austria,
went up sharply
in
Norway, and Portu-
Sweden, Denmark, and Finland, the
behavior of inflation suggests that this wtia mostly a cyclical movement
an increase
in the actual
unemployment
unemployment sharply declined
Finland
With
ers
still
these evolutions, a clear shift in focus took place, both
in the
—and
has high unemployment.
and among researchers,
ment
rate over the natural rate
thereafter; of the three countries, only
for
two reasons.
First,
major continental countries made the
among
mak-
policy
continuing high unemployearlier explanations,
based
on adverse shocks and persistence, increasingly implausible: Could shocks
the 1970s and the 1980s
And
in
still
have such strong
effects in the 1990s
in
and 2000s?
second, given the continued large commonality of shocks, the differences
unemployment
rates across countries pointed to differences in institutions as
central to any explanation of
unemployment.
The most dramatic evidence
of this shift in focus
Study."'"' Ill-adapted labor
market
the source of high unemployment.
institutions, the
And
was the 1994
OECD
OECD
".Jobs
report argued, were
the report went on to advocate reforms,
from the design of unemployment insurance and employment protection, to a
reduction of the tax wedge and the
labor market policy programs.
extremely influential.
of
The
minimum
The
wage, to better training and active
report was
— and
its
general line
still
is
notion that "labor market rigidities" are at the core
European unemployment has gained wide acceptance among policy makers.
In parallel,
on the academic research
was made
easier
side, the shift in focus
towards institutions
by the emergence of a new and richer framework to think
about unemployment, a framework based on
flows,
matching, and bargaining.
For some time already, Christopher Pissarides. building on earlier work
Diamond on
13.
To be
first
Peter
search and bargaining (1982), had explored models of the labor
market which explained unemployment
the
bj'
the labor market as a result of a
importance of institution.s was already an important theme
book by Layard, Nickell, and Jackman.
historically fair, the
edition (1991) of the
in
in
19
process of creation and destruction, large flows of workers in the labor market,
and a complex matching and bargaining process between firms and workers
example, Chris Pissarides
[1985]). His
and the development and extension
Mortensen
rightly so.
model
of the
in
of
its
much more
strengths was to allow for a
in 2000),
a series of articles with Dale
example 1994) made the framework extremely
(for
One
1990 book (with a second edition
(for
and
influential,
specific analysis of
the role of institutions, both theoretically and empirically (Box 3 gives a more
formal description):
The framework
by large flows
firms. In
France
roughl,y as
as in the
why
for
many
example, 1.5% of
are created
may
from
rates of separations
separate from a
many
this
and high
is
is
and there
will
reasons other than job destruction
4%
per
month
(Pierre
Actual unemployment
is
the
in
typically have
away from the
more
From the
this rate of
point of efficiency, there
unemployment
is
unlikely to be optimal however,
nature of bargaining. Even
to walk
is
a complex
is
an optimal
clearly positive.
and depends on the
the absence of collective bargaining, both the firm
some bargaining power. The worker can threaten
job, but walking
so the higher the
away and
unemployment
rate.
finding another job
The
tighter the labor market, the lower the
is
costly,
firm can threaten to
the worker; but doing so and replacing the worker by another
more so the
Cahuc and
always be workers looking for jobs (unemployment) and jobs
unemployment, and
and the worker
much
2004).
looking for workers (vacancies).
rate of
by
month and
about the same percentage
In such a labor market, the process of matching workers and jobs
one,
characterized
rates of hires
firm, the flows of workers are typically
higher. In France, they are of the order of
Andre Zylberberg
labor market
firms,
jobs are destroyed each
all
— interestingly,
United States. As there are
a worker
The
started from a basic fact:
— high
is
fire
also costly, the
unemployment
rate.
This
has two main implications. First, the bargained wage depends on the labor
market prospects of workers and
and strengthens
in
firms:
firms. Second, labor
High unemployment weakens workers
market institutions
also play a central role
wage determination: The more generous the unemployment insurance, the
less costly
it
is
for the
worker to look
employment protection, the more
From
for
costly
another job.
it is
The
higher the level of
for the firm to fire a worker.
a methodological viewpoint, this framework led to major progress:
20
It
allowed for a
careful analysis of the implications of
than could be done before. Take
institutions
tion.
more
The framework made
tion, to the extent that
it
for
complex labor market
example employment protec-
three broad predictions. First,
employment protec-
increased the cost of laying off workers, was likely to
decrease layoffs, and thus to reduce the flow of workers entering unemployment.
Second, by increasing the costs to firms, and more importantly, by strengthening the bargaining power of workers,
bargained wages, and
Third, given that the
in
was
it
likely to lead to
an increase
in
turn to an increase in the duration of unemployment.
unemployment
rate
is
the product of the flows into
unem-
ployment and unemployment duration, lower flows and higher duration implied
that the effect of employment protection on the unemployment rate
ambiguous. All three implications have proven to
ple Olivier
Blanchard and Pedro Portugal
fit
[2001]).
itself
was
the facts well (for exam-
Employment
protection
probably one of the main factors behind the long unemployment duration
is
in
Europe; differences in employment protection seem however largely unrelated
to differences in
It
uninnploymcut rates across countries.
mapping between the
allowed for a better
increasingly available panel-data
microeconomic evidence on firms and households, and macroeconomic models.
Take
for
example unemployment insurance. The framework points to two sep-
arate effects of insurance on unemployment.
The
first is
through
its effect
on
search intensity, and thus the matching between unemployment and vacancies.
The second
is
through the reservation wage: Higher unemployment benefits
make unemployment
less painful
and are
likely to lead to
an increase
in the
bargained wage. Both effects in turn imply an increase in equilibrium unem-
ployment duration, and thus an increase
theory,
much
in the natural rate.
Guided by search
empirical work has looked into the effects of the schedule of unem-
ployment benefits on search by the unemployed. The findings
better calibration of our
in turn allow for a
macro models. (There has been however
micro work on the other channel, namely the
effects of
little
empirical
unemployment
ance on bargained wages. This reflects a more general shortcoming, a
insur-
still
poor
empirical understanding of wage determination in environments such as Europe
where both individual and
It
collective bargaining are likely to play a role.)
gave new macro tools to interpret facts and look at the sources of unemploy-
ment. In particular,
it
gave a way to combine the evidence from the Phillips
curve with the evidence from the Beveridge curve
— the relation
between un-
21
employment and
vacancies. Conceptually, the Beveridge curve evidence tells us
about factors that
matching
affect
in
the labor market, whereas the Phillips
curve evidence teUs us also about factors that
affect bargaining.
A
shift in
the
Phillips curve not associated with a shift in the Beveridge curve points to factors related to bargaining; a joint shift points to factors related to matching.
I
initially
hoped that the
joint use of these
two tools would prove powerful
(Olivier Blanchard
and Peter Diamond 1989),
been disappointed,
at least in its application to
a recent examination, and a slightly
Olivier Blanchard 1990);
unemployment
Beveridge curve across countries; one reason
much from
may be
Europe
in
more optimistic conclusion,
Nickell et al [2002]). It has proven hard to learn
have
I
(for
see Stephen
the shifts in the
that data on vacancies are
often of poor quality.
Did the
shift in focus
towards institutions give us the key to the evolution of
European unemployment, across countries and time? The
at the data, at the
end
of the 1990s, gave a
mixed
Differences in institutions appeared able to explain
unemployment
was
first
first
systematic look
an.swer:
much
of the differences in
rates across countries either in the 1980s or in the 1990s. This
shown
in a cross-country regression
quantitative indexes for a
number
of labor
by Stephen Nickell
market institutions
in 1997.
for the
Using
mid- and
late-1980s, he found that, together, they did a good job of explaining differences
OECD
across 20
in his regression
countries.
Among
were the duration of unemployment benefits (which increased
unemployment), and the degree
decreased
Changes
of
in institutions did
not appear able however to explain the evolution
rates over time.
"employment
I
institutions, relying
on
if
the initial increase in
unemployment
first to
his earlier
in the 1960s should
unemploy-
be explained by much
less
friendly" institutions than 40 years ago. In 2000, .Justin Wolfers
took a careful look
not the
Even
to shocks rather than institutions, the difference between
ment today and unemployment
and
of coordination in collective bargaining (which
it).
unemployment
was due
the most economically significant variables
do
at the relation
between unemployment, shocks and
on a panel data approach over countries and time.
so. In
We
were
an important book pubhshed in 1994, and building
work with a number
of collaborators (Hian
Teck Hoon, George
Ned Phelps had
articulated a theory
Kanaginis, and Gylfi Zoega in particular)
22
of
it
movements
to the
in the natural rate
based on shocks and institutions, and taken
data using panel data. Our contribution was
and
in part to construct
include time series for a larger set of institutions than Phelps.
And
our
first
pass at the time series evolution of institutions, was not very encouraging:
EU5
Figure 10. Replacement rates,
a:
Average
• f.A
• FRA
EM
•
• SEU
• OCU
M=.
• DEU
• DEU
• FRA
!'..
since 1960
• DEU
mofiu
• DEU
im
• ESP
t^
'"
• £5P
.tt-J
• OBR
• eSC
1
1
':"'
..:
•
ITA
..-A
• IT*
S60
1970
• IIA
b:
EfiJt
2000
1S90
1980
•
m
...
Maximum
• ESP
•ESP
• FBA
.FPA
• E5P
.,s=
• irA
• 5a>
• DEU
•'••
• aeu
• 03"
.,:*
• ITA
•'"
960
1970
19S0
2000
1990
Source: Blanchard and Wolfers (2000)
we gave
Figures 10 and 11 reproduces two of the time series
replacement rates and
for
employment protection
country, for each five-year period since 1960.
Figure 10 were constructed from an
OECD
in that paper, for
respectively, for each
The replacement
data
set,
rates
EU5
shown
which measiured the
in
ratio
23
-
of pre-tax social insurance and social assistance benefits to the pre-tax wage, for
various categories of unemployed workers, depending on income, family status,
and duration of unemployment. Figure 10a
summary measure
replacement rates, the
gives an unweighted average of these
striking are the different evolutions of the five countries,
a
common
trend. Figure 10b provides a different
maximum
showing the
OECD. What
is
and the absence
of
often used by the
replacement rate over
all
and more relevant angle by
categories for each country and
each subperiod. Again, no clear trend emerges. Clearly, some of the
replacement rates increased
maximum
the early 1980s, but they have declined since
in
then.
Figure 11.
-^ -
eESP
•
Employment
• BSP
• BSP
protection index,
• esp
ITA
• ITA
• ESP
• ITA
• DEU
• DEU
• ESP
EU5
since 1960
• ITA
• ESP
• DEU
• DEU
X
• DEU
• FRA
,,
„
c
c
g
• FRA
^
E
o
—
—
o
rFRA-
• EBR
• DEU
• FRA
• FRA
• ITA
• DEU
Q-
Ev-
• FRA
-
HI
• DEU
• FRA
• GBR
• GBR
• GBR
o
• GBR
• GBR
• GBR
• GBR
• GBR
-
1960
1980
1970
2000
1990
Source: Blanchard and Wolfers (2000)
The indexes
of
employment protection shown
by combining two sources, the
in Figure 11
series constructed
by Ed Lazear (1990)
period before 1985, and the indexes constructed by
and the 1990s. Again, what
is
striking
is
were constructed
tire
OECD
for
for the
the 1980s
the absence of a clear trend, and the
heterogeneity of evolutions across countries.
A
more systematic construction
ular Michele Belot
sions. In panel
of time varying
and Jan van Ours
measures by others
(in partic-
[2001]) suggested roughly similar conclu-
data regressions of unemployment rates on institutions across 20
24
and allowing
countries since 1960,
country and time dummies, none of the
for
labor market institutions appeared significant.
Figure 12. Tax wedge, 2000 versus 1960, by country
ABEL
IFIWSWE
*AUT
'ESP
I'feSft'*
m
A PfiCSA
So
20
Tax wedge
10
Source:
OECD
(courtesy of
1
40
30
960
Luca Nunziata)
In this context, one variable deserves particular mention because
back
in discussions.
The
"tax wedge,"
workers and the cost of labor
for
gone up
in
i.e.
the difference
for firms, divided
often comes
by the wage, has
steadily
it
is
blamed by firms and policy makers
often
as
is
given in
Box
4):
On
(a
theoretical grounds, taxes or social
contributions that treat income equally whatever
unemployment
it
one
major sources of unemployment. Most economists are more skeptical
formal discussion
affect
it
between take-home pay
most European countries since the 1960s. In many countries,
stands at above 30%, and
of the
50
its
source (labor income or
benefits) should not affect the cost of labor to firms,
unemployment. The same should be true
and thus not
for taxes or social contributions
which come with corresponding benefits, such as retirement contributions, so
long as they are not redistributive.'
in
the tax wedge
14.
to a
Major
fits
effects of the ta-x
minimum wage
'
On
empirical grounds, while the increase
the general increase in unemployment,
wedge are
fioor. In this case,
likely to
for
does poorly
be present only for wages which are at or close
additional contributions by firms cannot be shifted to
workers, and thus lead to an increase in cost. For this reason,
decreased the tax wedge
it
many European
countries have
low wages since the late 1980s, somotiirics by substantial amounts.
25
explaining differences in unemployment across countries. This
in
Figure 12, which plots the tax wedge (defined as the
sum
EU15 country and
for
in
shows
all
little
countries in 2000 than
relation to
1960 and
the United States.
All the points are above the 45 degree line, indicating that the tax
higher in
in
of payroll taxes paid
by employers and employees and income taxes paid by employees)
2000 for each
shown
is
was
it
unemployment
in 1960.
rates.
But the ranking
wedge
is
of countries
Three of the four countries with
the highest tax wedge, Finland, Sweden, and Austria, are also countries with a
low natural
To take
rate.
stock;
We
ended the 1990s with a much better framework
unemployment. But we
also
ended with many questions. Even
to study
the earlier
if
shocks were no longer the main source of unemployment, they clearly were
responsible for the initial increase.
unemployment
less
at the
employment
constructing?
friendly? If so,
One can
Institutions
institutions were primarily responsible for
end of the century,
is it
why was
it
because they had become steadily
not reflected in the series we were
see the research since then as exploring different answers
to these questions. This
5
If
is
the topic of the next section.
and Shocks. Current Directions of Research
Giving a clear description of current research
is
always harder than giving one of
past research; research appears to go in
many
eventually pan out.
main directions
I
see roughly three
directions, only
some
of
at this point.
which
The
will
first is
an exploration of the role of other shocks, other institutions, other interactions.
The second
third
of
is
is
a
more
careful exploration
an attempt to look not only
unemployment, employment,
at
and measurement
unemployment, but
capital,
wages and user
of institutions.
The
at the joint behavior
costs. I take
them
in
turn.
5.1
Other Shocks, Other Institutions, Other Interactions?
Another
line of research
institutions
and shocks,
has extended the
initial
panel data examination of
to look at other shocks, other institutions, other inter-
actions:
26
There are potentially many more relevant institutions than those
•
cluded in the
initial regressions
many
Researchers have examined the effects of
others, from measures of
product-market regulation, to measures of home ownership
suggested by
Andrew Oswald
train crossings in FVance that said;
well.
''A
There used
train
not implausible that, in the same way that
may
oil
hid the decline in productivity growth, the
demand away from
—due
slowdown
(I
for
example
at
low skilled workers, or at increased
economy, through
bai'riers,
return to this particular theme below.)
There are potentially many interactions between shocks and
•
is
in productivity
deregulation of domestic goods markets, the decrease in trade
and globalization.
It
price increases initially
to higher competition in the world
turbulence
be a sign at
to
hide another".
growth also hid other shocks. Researchers have looked
shifts in labor
— a variable
(1997).
There are potentially many shocks as
•
in-
by Nickell and Blanchard and Wolfers.
Recall that the initial focus of research by
interaction between adverse supply shocks
institutions.
Bruno and Sachs was on the
and the structure
of collective
bargaining. Recall that the focus of the research on persistence was on
the strength of insiders; this strength clearly depends on institutions such
as
employment protection and unemployment
There are
•
also potentially
theme explored
The
for
lective bargaining, a
layoffs
ing
—may be
The
may depend
for
example on the structure of
effects of
employment protection
— which may increase
prisingly high labor turnover
A
partial
—which
layoffs,
a hypothesis explored
numbers
in
to explain the sur-
Europe.
and a few more, have been explored through panel data
summary
of the results
is
reduce
by collective bargaining focused on reduc-
by Giuseppe Bertola and Richard Rogerson (1997)
All these
col-
theme explored by Francesco Daveri and Guido
yiartly offset
wage dispersion
interactions between institutions, a
example by David Coe and Dennis Snower (1997).
effects of taxation
Tabcllini (2000).
many
benefits.
given in
Dean Baker
regressions.
et al (2002).
Some
correlations are intriguing; the conclusion by Stephen Nickell et al (2005) that
time series for institutions do a better (but
of the evolutions of
Blanchard Wolfers
the
number
still
mediocre) job of fitting some
unemployment across time than
series
is
intially
perhaps the most interesting.
of potential shocks, institutions,
It is
and interactions
suggested by the
clear
is
however that
sufficiently large
that the ability of such panel data regressions to
matters
is
limited.
Such regressions allow us
correlations; they are unlikely to
institutions
is
tell
tell
us what exact combination
to check for simple
and
partial
us about which combination of shocks
and
responsible for unemployment (for a similar view, see Freeman
2005).
Of
all
the hypotheses listed above, at least one deserves a longer treatment. ^^
It
the idea that higher competition in the goods market, lower trade barriers and
is
higher integration of goods markets across countries, higher globahzation and
outsourcing, are
all
leading to a more turbulent environment, an environment
with more job destruction and job creation.
more turbulent,
and lead
existing labor market institutions
to substantially higher
was rarely binding before
and most of us believe
Unemployment
off,
benefits,
become
dysfunctional
that, indeed, there
but the data just do not show
is
is
becomes binding and
which were not very costly
costly, requiring higher contribu-
to higher costs of firms.
than there was thirty years ago. There
it,
may become
unemployment. Employment protection, which
so long as few workers were laid
and leading again
the environment becomes
as firms rarely laid off workers,
increases the cost of firms.
tions
When
The
general story
is
appealing,
more economic turbulence today
one catch however.
We may
all
believe
it...
This puzzle showed up early on, when European unemploj'nient was just rising
in the late 1970s.
didate.
But
it
Increased turbulence already seemed to be a plausible can-
turned out that the measures of reallocation we could construct
— typically measures based on the standard deviation of rates of change of
employment, either across sectors or across regions — showed no trend increase.
then
The evidence
(1986),
for a
who
as of the early 1980s
is
well
summarized
in
Johnson and Layard
construct a table of standard deviations by industry or by region
number
of countries: Half of the standard deviations are higher in 1979
than they were
in 1960, half are lower. In all cases, the
changes are small.
I
could not locate an update of this table for the 1980s and 1990s, but the series
I
have seen
for a
few countries yield the same conclusion: There
is
no apparent
increase.
15.
Another hypothesis worthy
of a longer treatment
is
the presence and the role of skilled-
mention that, while
one ob,serves that, in countries with high unemployment, unemployment
high across the skill spectrum (although obviously higher for low-skilled workers).
biased technological progress.
relevant,
I
shall not
do
it
here, except to
it is
is
surely
typically
28
One may reasonably argue
crease in reallocation
is
that these measures are too raw. Perhaps, the in-
taking place mostly within industries or regions, rather
than across industries or across regions. In that respect, measures of job flows
based on plant-level data, along the
are clearly preferable.
far
enough
in time.
The
But
lines of the
practical issue
is
work by Davis
et al (1996)
that they typically do not go back
to the extent that they do, they also
show
little
sign
of increased turbulence. Figure 13 gives the evidence for France, based on two
studies, one by
Nocke (1994)
(1999) for 1990 to 1996.^''
tion (the
sum
of
all
for
The
1985 to 1990, and the other by Duhautois
lower line shows the evolution of job destruc-
employment changes
at plants with decreasing
employment,
divided by total employment); the upper line shows the evolution of job reallocation, defined
clear:
is
At
cis
the
sum
least starting
of job destruction
and job creation. The conclusion
from when data becomes
no evidence of an increase
available,
is
namely 1985, there
in turbulence.-"'
Figure 13. Job destruction and job reallocation
France 1985-1996
0)
n]
o
1985
1990
1995
The
juxtapo.sation of the two scries, constructed using shghtly different data and methodimply that there may be an artificial break in the series between 1990 and 1991.
17. One might argue that, in the case of France, turbulence has increased, but its effects on
flows has been offset by increasing employment protection. This suggests looking at data for
the United States, where employment protection is low and has not increased. The evidence
there is that, if anything, there has been a decrease in flows relative to total employment over
the last thirty years (Steve Davis ot al [2005], Figure 4 for the private sector since 1990, and
Figure 5 for manufacturing since 1947).
16.
ology,
29
—
Source: See text.
Is
the argument therefore settled? No, for two reasons, one empirical, the other
theoretical:
The
empirical reason
is
that other
measures of turbulence send a
measure of
different
sales volatility constructed
less
appropriate
flows. For
example, the
by Diego Comin and Thomas Philippon
set,
measure of variabiUty over time since the
evidence on
show a steady
increase
late 1960s. Reconciling the
job flows and increasing sales variability remains to be done.^*
flat
Until then, the discrepancy should
The
message from job
Compustat data
(2005), based on the firms in the
in this
— admittedly conceptually
theoretical reason
is
make
more
us
careful
about conclusions.
that one can construct models in which turbulence
is
not necessarily reflected in higher job flows. Such models have been explored
by Ljunr|vist and Sargent
in
a series of contributions
In their formalization, increased turbulence
is
specificity of skills associated with particular jobs.
involontary job change
example, 1998, 2005).
(for
reflected in
The
an increase
implication
is
in the
that an
associated with a larger drop in the wage distrib-
is
worker than was the case
ution facing a
laid-ofi'
unemployment
benefits are linked to past wages, the
in the past. In this case,
if
unemployed may have
high reservation wages, and remain unemployed for a long time. Furthermore,
if
skills
deteriorate through unemployment,
become trapped
into
some
unemployment. Differences
of the
unemployed may even
in the generosity of the
unem-
ployment insurance system, Sargent and Ljunqvist argue, may therefore explain
why Europe
is
doing so
increase in turbulence
much worse than
— an example of the interaction between institutions and
shocks. While the theory
skills for laid-off
5.2
A
workers
is
is
appealing, direct evidence of a larger decrease in
however so
very limited.
institutions are typically multidimensional.
quantitative indexes
is
not easy:
employment insurance systems,
18.
far
Closer Look at Institutions
Labor market
for
the United States in facing the same
The two
series differ in
coverage
in
How
if
the
many
Reducing them
to
does one compare for example two unfirst
wa)'s
has more generous unemployment
— plant versus firm
level data,
manufacturing
the long series for job flows, large firms for Compustat, employment versus sales.
30
benefits, but also
more
compare two systems
conditionality of benefits on search effort?
of
employment
when the
protection,
first
How
does one
includes higher
protection for some workers, and lesser protection for others?
In the process of looking at the effects of institutions,
vinced that existing measures fully capture what
work with
to explore, in on-going
Da.niel
is
Cohen and
have become
I
less
con-
going on. This has led
Cyril
Nouveau
me
(2005), the
evolution and the determinants of labor market institutions in France since the
What
1950s. ^^
emerges
a
is
more complex picture than that given by quantita-
tive measures.
What we
find
major changes
is
that the increase in unemployment in the mid-1970s led to
ance was
Under the
in institutions.
therefore the increase in
made
initial
assumption that the shock, and
unemployment, was temporary, unemployment
insur-
more generous, and employment protection was
substantially
sharply increased. Early retirement programs were put in place to "make room"
for
young workers. As high unemployment turned out
pressures on the
unemployment system, and the
to persist,
both financial
some
realization that
of the
earlier
measures probably contributed to unemployment, have led most of the
initial
changes to be reversed. But the reversal has not taken the form of a
return to earlier institutions.
The
decrease in employment protection has come
form of the introduction of two types of labor contracts, traditional and
in the
highly protected permanent contracts, and new, less protected, temporary contracts.
what
Whether such a reform
is
certain
is
that
it
actually decreases
unemployment
is
ambiguous;
has created a dual labor market, with protected and
marginal workers.
So, while existing time series for labor
market institutions
in
France show
change since the 1960s, a closer look at history suggests that, at
institutions indeed
1980s.
became
less
employment-friendly
While things turned around starting
reforms have had perverse
effects, either
in
in the early
the 1970s and early
1980s, man\' of the
because of poor design, unanticipated
consecjuences, or political constraints.^'^ Institutions today are less
19.
For an exercise in the .same spirit for
20.
In addition,
fallacy,
little
least for France,
Germany
since 1990,
.see
employment
Conny Wunsch
(2005).
France has suffered from a number of policies based on the lump-of-labor
from the generous early retirement programs put in place in the 1980s aimed at
freeing jobs for
new workers
—
— to the 35-hour laws at the end of the 1990s — aimed at creating
jobs by decreasing hours per worker. Both of these programs have turned out to have large
budgetary
costs.
31
friendly than tliey were in the early 1970s.
do not know whether the concln.sions reached from similar studies of other
I
European conntries
will
One
why
of the reasons
be similar.
I
suspect that the message
in
in institutions,
which has been partly and poorly undone
Employment,
All the theories
more
general:
the shocks of the 1970s and 1980s have led to high
some European
unemployment
5.3
is
countries today
Wages and
Capital,
we have discussed have
is
that they triggered a change
in these countries.
Interest Rates
unem-
testable implications not only for
ployment, but also for capital accumulation, wages,
profits,
and
interest rates.
For example, an increase in bargained wages, for given labor market conditions,
should lead not only to an increase in unemployment, but also to a decrease in
the labor/capital ratio, a decroa.se in
th(^ profit rate,
and, for a given user cost, a
decrease in capital accumulation over time. Yet, few of these theories have been
more than data on unemployment and through the estimation
tested using
unemployment
This led me,
in the late 1990s, to
of looking jointly at capital,
perform a conceptually simple exercise, that
employment, wages,
profits,
and user
use this information to try to identify shifts in either "labor
lation giving warranted wages as a function of
level of
side,
(the re-
capital,
and the
employment,
I
assumed that firms chose
capital
rate,
with
then constructed shifts
labor
the period 1970-1995.
demand and
My
organizing the empirical evidence
tually
On
the labor
convex
to
the supply side,
level of technology, the un-
labor supply for 14
OECD
I
coun-
papers were primarily an exercise aimed at
in
a simple but interpretable way.
A
concep-
more ambitious attempt was made by Ricardo Caballero and Mohamad
Hammour
21.
On
other factors showing up as shifts in the relation.
all
in
~^
and labor subject
both investment and factor proportions.
assumed that bargained wages depended on the
employment
tries for
and
technology) or "labor supply" (the relation giving bargained wages as
costs of adjusting
I
costs,
demand"
a function of labor market conditions) (Blanchard 1997, 1998).
demand
of
ecjuations.
(1998),
who
constructed a structural model starting more explicitly
Semantics are not settled here. The relation giving warranted wages is often called the
it gives the prices set by firms given wages and other variables. The
"price setting relation" as
relation giving bargained
the wages set
in
wages
is
often called the ''wage setting" relation, because
it
gives
bargaining, given the price level, actual or cxficctcd, and other variables.
32
from bargaining and institutions such as employment protection, and allowing
for
endogenous technological progress. Their model was not estimated, but
Hammour
brated, and Caballero and
used
it
employment, productivity and factor prices
Both
of
On
the one hand,
Bruno and Sachs and the
shifts"
—
that
visible,
of the
I
had
dynamics suggested by the early work
work on the
role of capital accumulation,
early 1970s were characterized
interest rates
cali-
capital,
France.
in
increases in bargained wages given
is,
and
effect of profit rates
later
The
clearly present in the data.
supply
many
on
and the evidence more complex than
exercises proved interesting,
expected.
to look at the evolution
were
by "adverse labor
unemployment. The
on capital accumulation were also clearly
with low interest rates delaying the slowdown in capital accumulation
to the 1980s.
These labor supply
shifts
were largely reversed starting in the mid
1980s. Countries, such as the Netherlands
and Ireland, which had seen a major
decrease in unemployment, also showed a large decrease in wages in efficiency
units
—wages divided by the index of Harrod neutral progress.
The
reversal of adverse labor supply shifts should have led to a decrease in
unemployment over
time. But, the data suggested, something else
starting in the early 1980s. At a given
stock,
employment was
result
was a decrease
lower:
wage
(in efficiency units)
There was an adverse
at
work
and given capital
shift in labor
European
the labor share in most
in
was
demand. The
countries, starting
in the early 1980s. Figure 14a gives the behavior of the labor share in France,
one of the countries where the decline was the most dramatic,
sector,
for the business
from 1965 to 2001. The labor share, which had gone up by
down by 12% percentage
points from 1970 to 1981, then went
to the early 2000s;
has remained roughly at that
it
and 14c show the proximate causes
5 percentage
points from 1980
level since. ^^
Figures 14b
of the evolution of the labor share. Figure
14b shows the evolution of the wage
(in efficiency units),
and
figure 14c
shows
the evolution of the ratio of employment (in efhciency units) to capital, since
1965.
22.
in
There arc many
the figure
is
i.s.sucs
adjusted
also payroll taxes
and other
reconstructed by the
measurement a-ssociated with the labor .share. The .scries used
employment. Labor income includes not only the wage but
of
for self
OECD
since 2000,
decrease; the basic evolution
due to composition
effects,
by firms. Some of the data have been
and the current series for France shows a smaller
social contributions paid
is
still
the same.
the result of a
France, this composition effect
is
Some
of the evolution of the labor share
shift to sectors
with lower labor shares. Again,
small. (Alain de Serres et
al,
is
for
[2002].)
33
wage
Figure 14. Share, employment and real
France
a:
b:
.-,.,...;
q
Real wage per efficiency
unit
"
p
-
-
r\
r^
p
V
p
^
5
c:
o
Labor share
-
-
-
w
Ratio of
V
W
\ y^-^
/
employment
In
"
-
/
efficiency units to capiital
- --
—
rotn
;
1
s
IS'
UJ
^^-^"
^
(^J
I
^^^-^
y
1
960
1980
1970
2000
1990
Source: Olivier Bianchard (2000), updated
In the second half of the 1970s, the wage (in efficiency units) went up, and the
ratio of
result
employment
(in efficiency units)
was an increase
went down over time
in the labor share,
and
this
is
expect in response to an adverse labor supply shock
gained wage
for
in response; the
exactly what
— an increase
we would
in the bar-
given labor market conditions. Since then however, the wage
has come down; since 1990,
it
has remained roughly at
its
1970
level.
The
ra-
34
,
tio of
employment
a given
wage
is
to capital has not recovered however:
what mechanically explains the lower labor
algebra of the labor share
Why
is
employment lower
is
Lower employment
given in
Box
at a given
share.
at
(The basic
5.)
wage? In
my
1997 and 1998 papers,
I
considered various candidates and converged on a decrease in "labor hoarding"
due perhaps to higher competition and tougher corporate governance, as the
more
likely one.
Under
this explanation however, the decrease in excess labor
should have led to an increase in
profit,
an increase
and an eventual recovery of employment; so
employment has not taken
in capital
accumulation,
and
far the increase in capital
place, at least not in France or in
Germany, where
a similar evolution of the labor share has taken place.
An
alternative interpretation
They argued
was given by Caballero and
that the decline in the labor share below
instead an increase in the marginal
crease in the marginal
wage
for
wage
Hammour
its initial level
relative to the average
(1998).
reflected
wage (an
in-
a given average wage will lead to a decrease in
employment, and thus a decrease
in the labor share).
Caballero and Hammour's
conclusion was therefore that the low labor share reflected the firms' desire to
decrease labor beyond what the average cost of labor would suggest. And, they
argued, this reluctance of firms to hire labor could be traced to a worsening
of labor
market
institutions. Their explanation leads to a
view of the future:
A low
much
less optimistic
labor share does not lead to higher incentives to invest,
nor to an increase in employment.
I
see the labor share puzzle as largely unsolved.
has been
much
smaller in the
UK
The decrease
in the labor
and the United States than
share
in continental
Europe, pointing indeed to factors specific to continental Europe: Institutions
are a natural starting point. At the
same
time, within continental Europe, the
decrease in the labor share has taken place both in countries that have reduced
unemployment
(the Netherlands for example), and in countries that
high unemployment (Prance for example).
I
still
have
also see the puzzle as a potentially
major piece of the story of European unemployment, and one on which more
work should be done.
35
Do We Know Enough
6
At the end
to policy
of this tour, one
may
makers about how
to Give Advice?
know enougli
ask whether we
to reduce
unemployment.
the proper degree of humility. In this last section,
I
I
to give advice
believe
we do
summarize what
I
— with
think we
know and we do not know.
A
6.1
General Story Line
Going back over the
increase in
common
There
is
last thirty years, there
unemployment
shocks, from
oil
in
price increases to the
much question
not
is
little
question that the
slowdown
initial
to adverse and largely
Europe was primarily due
in
productivity growth.
that different institutions led to different initial
outcomes. Whether collective bargaining led to a decrease
the growth of
in
bargained wages, whether inflation could be used to reduce real wage growth,
all
played a central role
There
is
not
much
in
determining the
unemployment.
question that the increase in unemployment led, in most
countries, to changes in institutions as
increase in
size of the increase in
most governments
unemployment through employment
protection,
tried to limit the
and
to reduce the
pain of unemployment through more generous unemployment insurance.
There
is
not
much
question that, since the early 1980s, because of financial
pressure and intellectual arguments, most governments have partly reversed
the initial change in institutions. But this reversal has been partial, and some-
times perverse.
unemployment
The
different paths chosen
rates across
may
well explain the differences in
European countries today.
Despite the twists and turns of research, the sediments from the successive theories are
nearly
all
relevant.
bargaining emphasized by
insider effects
The
role of shocks
initial theories,
and the interaction with
collective
the role of capital accumulation and
emphasized by the theories focusing on persistence, the
specific institutions clarified
by flow-bargaining models,
all
role of
explain important
aspects of the evolution of European unemployment.
6.2
It is
know
Which
Institutions?
one thing to say that labor market institutions matter, and another to
exactly which ones and how.
36
Humility
is
needed here, and there
is
no better reminder than the comparison
between Portugal and Spain. Both experienced revolutions and wage explosions
in the 1970s (the
Portuguese labor share reached 100%
in the
mid
1970s...);
both
have, at least on the surface, rather similar institutions, including high employ-
ment
protection. Yet, Spanish
in the
unemployment has been very
high, exceeding
20%
mid-1990s, whereas Portuguese unemployment has remained low, with
a high of 8.6%
the mid-1980s, and a decrease thereafter.
in
Many
researchers,
including myself, have tried to trace the differences to differences in shocks or
institutions (for a recent attempt, see
much more than
that our explanations are
And
Olympia Bover
the history of the last thirty years
sad endings,
first
is
et al 2000).
am
ex-post rationalizations.
not sure
^'^
a series of love affairs with sometimes
with Germany and German-like institutions
ment started increasing
I
there in the 1990s...
— then
— until unemploy-
with the United Kingdom
and the Thatcher-BIair reforms, then with Ireland and the Netherlands and the
role of national agreements,
Denmark, and
and now with the Scandinavian countries, especially
concept of "flexisecurity"
its
Nevertheless, even
if
.•^'^
one cannot pretend to have much confidence about the
optimal overall architecture, much has been learned about the effects of the
various pieces, especially from the large
number
of empirical micro-studies
and
natural or designed experiments. As a review of the relevant research would require another survey,
much more about
let
me just mention
a few directions of research.
the incentive aspects of unemployment insurance on search
and unemployment duration, be
intensity
employment
We know
benefits, or the
it
the length and time shape of un-
form of conditionality or training programs
example the surveys by Peter Fredriksson and
(see
Bertil Ilolmlund (2003)
on
unemployment insurance, and by John Martin and David Grubb (2001) on
ac-
for
tive labor
market
policies).
We know
more about the
contributions on low wages (for example
(2001) on the French experience).
ment protection, and the
23.
now
.\long the
fallen
same
lines,
below 10%
—
i.s
effects
effects of decreasing social
Bruno Crepon and Rosenn Desplatz
We know more
about the
effects of
employ-
on the labor market of introducing temporary
the rapid decrease of the unemployment rate
in
Spain
—which has
also hard to trace back to either shocks or dramatic changes in insti-
—
is the coincidence of very low productivity growth
zero measured
growth in Spain over the last 15 years and decreasing unemployment.
24. For descriptions of events and reforms see Stephen Nickell and Jan van Ours(2000) for the
Netherlands and Ireland, Patrick Honohan and Brendan Walsh for Ireland (2002), David Card
and Richard Freeman (2004) for the UK, and Niels Westcrgaard-Nielscn (2001) for Denmark.
tutions. Yet another puzzle
tfp
—
contracts at the margin while keeping
employment protection the same
workers (Olivier Blanchard and Augustin handler (2002)
From both
consensus
It
macro evidence and
the
body
this
for
most
for France).
of micro-economic work, a large
— right or wrong — has emerged:
holds that
modern economies need
to constantly reallocate resources, includ-
new products, from bad
ing labor, from old to
to
good
firms.
At the same time,
workers value security and insura.nce against major adverse professional events,
job loss
in particular.
While there
a trade-off between efficiency and insurance, the experience of
is
the successful European countries suggests
important
in essence
is
need not be very steep.
it
This means providing unemployment insurance, generous
tional
on the willingness of the unemployed to train
available.
This means employment protection, but
to firms to
make them
What
is
to protect workers, not jobs.
in the
internalize the social costs of
in level,
but condi-
and accept jobs
for
if
form of financial costs
unemployment, including
unemployment insurance, rather than through a complex administrative and
judicial process.
This means dealing with the need to decrease the cost of low skilled labor
through lower
social contributions paid
need to make work attractive to low
tax rather than a
minimum
by firms
skill
at the low
wage end, and the
workers through a negative income
wage.
for
example
a description, see Conny
Wunsch
This consensus underlies most recent reforms or reform proposals,
Germany
in the recent
Hartz reforms
2005), or the
"Camdessus Report" on reforms
in
These measures are probably
would they be enough
all
(for
in
France (2004).
desirable. If they
were to be implemented,
to ehminate the European problem?
I
see at least two
reasons to worry.
Collective Bargaining and Trust
6.3
The
first
worry
is
that these reforms deal only with a subset of the institutions
that govern the labor market.
An
unemployment was the importance
some
early
theme
of the research
of collective bargaining.
And
on European
it is
a fact that
of the successful countries, the Scandinavian countries in particular, have
38
very different structures of collective bargaining from, say, France or
much more
of
an emphasis on national,
trilateral, discussions
between unions, business representatives, and the
Italy,
with
and negotiations
state.
This raises two questions. First whether countries such as France or Italy need to
modify the structure of
also
did, the results
know
Sweden
as in
we explored the hypothesis that
perhaps tracable to differences
firms, explain
We
would be the same
or
Denmark.
the answer to either of the two questions. In work with
(2003),
firms,
Second whether, even
collective bargaining.
some
differences in trust
in
of the difference in
I
if
they
we do not
think
Thomas Philippon
between unions and
economic models between unions and
unemployment
rates across countries.
found that various measures of trust, from strike intensity in the 1960s to
survey measures of trust between firms and workers, could explain a substantial
fraction of differences in
unemployment
across
European
countries. ^^
these findings reflect causality from lack of trust to unemployment,
start.
The question
is
Even
it is
just a
UK
whether trust can be created. The example of the
where the unions have not only become weaker but have
also
if
changed attitudes,
suggests that trust cannot be taken as an immutable country characteristic.^^
Low Inflation, the Natural and
Unemployment
6.4
Since 2000, European
inflation.
the Actual Rate of
unemployment has been associated with roughly constant
This would suggest that the current high unemployment rate
reflects
a high natural unemployment rate, rather than a large deviation of the actual
unemployment
rate above the natural rate. This
on
justifies the focus
constant inflation
is
inflation
indeed the assumption which
is
by the European Central Bank; Maintaining
then equivalent to maintaining unemployment close to
its
natural rate; this natural rate can only be reduced by labor market reforms,
and
this
is
not the responsabihty of the central bank.
One may however
question this assumption. Inflation in the
ning under 2%, and close to
25.
A
0%
in countries
EU15
is
now
run-
such as Germany. At these low
Cahuc and Algau [2005] takes another step in that direction, and
and empirically, that the efficiency cost of social insurance depends on
recent paper by
argues, theoretically
civic attitudes.
26.
This section can be seen as a variation on the old theme of whether there are different
national models, adapted to different countries, an "Anglo-Saxon" model, a "Scandinavian"
model, and so on.
•
'
•
„,
,7
:
39
inflation rates,
it
is
not implausible that nominal
rigidities
matter more, that
workers for example are reluctant to accept nominal wage cuts
— a hypothesis
explored, for example, by Akerlof et al (1996). In such an environment,
be that an unemployment rate above the natural rate may lead
than declining
of
demand might
inflation.
Put another way,
inflation.
it
may be
that, in fact,
The experience
different,
argument
unemployment
is
unemployment below the natural
for
an expansion
where unemployment has steadily decreased
of Spain,
for a
in inflation,
can
more expansionary monetary
that institutional reforms encounter less opposition
is
are growing and
rate
low rather
in this light.
Another, conceptually
policy,
may
decrease unemployment without leading to steadily higher
without major labor market reforms and without an increase
be read
to
it
itself.
This argument
is
rate
may
actually help decrease the natural
an old one (Blanchard
such a "two-handed" approach in Europe) but
issue however
is
whether,
in fact,
when economies
decreasing. In other words, a decrease in
et al (1985)
is still
growth and the decrease
already argued
relevant today.
in
One
unemployment do
not alleviate the political need for reform, and thus delay rather than encourage
reforms.
The experience
often delayed reforms,
of the late 1990s in
is
Europe, where a cyclical expansion
not reassuring in that respect. Developing this last
point would take us to the political
economy
of labor
market reform, and
this
should be the topic of another survey.
40
Box
Real and Nominal Rigidities
1.
The purpose of
ments
these and the following boxes
in the text. I
one wants
some of
the argu-
have made the choice of presenting simple, related, hut ad-
hoc models. References
if
to formalize
is
to derive
micro-founded models
to explicitly
optimal policy
— are given
when
This box shows the role of real and nominal rigidities
— which axe needed
available.
shaping the
in
effects of
adverse shocks to warranted wages on unemployment.
Conisider firms with constant returns to labor (we leave aside capital for the
time being):
so,
using logs:
J/
„
;
where y
tfp or
log output,
is
"'_.
and a
log employment,
is
is
log productivity (either
w—
'
,
lu is
.
the log nominal wage,
=
p
and p
a
'
=
,
the price level (constant terms are
is
a( — 1)
'
.
Assume
out throughout for notational simplicity).
we can focus on
firms, the
given by:
a
so
+n
a
goods market or a constant markup, the wage paid by
"warranted real wage",
left
is
=
labor productivity; the two are the same here). Assuming either compe-
tition in the
where
n
.
..
+
further that a follows:
e
the effects of a negative realization of
e,
a decrease in pro-
ductivity.
Assume
the bargained wage
is
given by:
w - p = Ea — Pu
The bargained wage depends on expected productivity Ea and
ing function of the
unemployment rate
u.
Assume
adjusts over time to actual productivity according
Ea = XEa{-l) +
I taJce the
{I
is
a decreas-
that expected productivity
to:
- X)a
speed of adjustment of expected to actual productivity,
(1
—
A) as
given here. In more explicit models, the A hke parameter has been derived
41
from learning
in
a Bayesian environment in
firms
u'liicA
and workers have
to
assess whether sliocks are temporary or permanent, or from staggering of wage
decisions (a la Taylor (1980) or Calvo (1983)), or both.
Combining the equations
u
An
unexpected decrease
Combining the equation
warranted and
for the
in
=
tlie
bargained wage gives:
— Ea)
—-^{o.
productivity leads to an increase in unemployment.
for expected productivity with the equation
above gives
the behavior of the natural rate of unemployment:
u
= Au(-l)--e
A permanent decrease in productivity increases equilibrium
alently, the ''natural rate" of
and
ft
unemployment)
capture the two dimensions of real
for
some
rigidities.
unemployment
time, but not forever. A
The higher
A, i.e
the .slower
the adjustment of expectations, the longer lasting the effects of the shock.
lower
Now
ft,
(equiv-
The
the larger the effect of the adverse shock on unemployment.
introduce nominal
rigidities.
To do
replace the equation for the bar-
so,
gained wage by:
ru
The nominal wage
is
set
= Ep + Ea -
'
ftu
on the basis both of the expected price level and
expected productivity. Combining the equations for the warranted and the bargained wage gives:
u
=
-^[(<^
-
E"')
+
(P
- Ep)]
(Ignoring the unexpected productivity term and moving terms around gives the
conventional expectational Phillips curve, p
= Ep — ftu). The
of this equation
nominal
is
that, in the presence of
can, to the extent that
effects of
it
central implication
rigidities,
can increase p — Ep, potentially
monetary policy
offset the
adverse
adverse productivity shocks. Put another way, expansionary monetary
policy can offset the increase in the natural rate by maintaining the actual rate
below the natural
rate.
The
precise form of
monetary policy required
depends on the formation of price expectations and the
not essential to the argument
made
rest of the
to
do so
model;
it is
here.
42
a
In short, this box has
adverse supply shocks
shown how,
will
depend on
real
micro- founded treatment of these issues
Box
The
2.
Erst persistence
its
is
and nominal
(An
rigidities.
explicitly
given in Blanchard and Gali (2005)).
Persistence Mechanisms
mechanism focuses on
plications for the warranted wage.
and
response of unemployment to
in genera.!, the
capital accumulation
The second focuses on
and
its
im-
collective bargaining
implications for the bargained wage.
Capital accumulation, and the two effects of monetary policy
Assume
that, instead of the
assumption of constant returns
in the previous box, the production function
labor,
and constant returns
is
Cobb-Douglas
is
we made
in capital
and
to scale:
y
where k
to labor
—
a{a
-\-
n)
-\-
[1
— a)k
the log of the capital stock, and a
the index of Harrod neutral
is
technology ("technology for short). Assuming perfect competition in the goods
market or a constant markup, the wage paid by firms, the "warranted
is
real
wage"
given by (up to a constant term, that I ignore):
lu
— p = (a ~
For a given capital stock, the higher
product of labor, the lower
The
is
l){n
is
—
k
-i-
a)
-\-
employment, the lower
is
the marginal
the warranted real wage.
profit rate associated with
a given real wage
is
given by the factor price
frontier relation (up to a constant term, again ignored)
1
where n
is
-Q
[w — p
—
a)
the log of the proBt rate.
43
Let r be the log of the user cost of capital. Ifn
time. If
TT
is
greater than
w—
the bargained wage
is
p
a
~
is
given by:
7-
-{
Box
— Ep = Ea —
Let n the log of the labor force, so u
wage
-a
I
=
given, as in
xu
warranted real wage
k decreases over
r,
k grows over time. In the long run, the profit rate
r,
mu.st be equal to the user cost, so the warranted real
Assume
than
is less
n -
n.
1,
by:
'
f3u
'
Then, using the equation for the
in the
short run and the equation for the bargained wage
—
—
gives:
p
+
{a
l){n
+
k
a)
—
{a
~
l)u
= Ep + Ea —
[3u
Or, reorganizing:
u=~-1
-a+
-\{a
- Ea) +
{p
- Ep) + {a - l){n - k +
In the short run,
unemployment depends, as
and p — Ep. But
it
stock, the lower the
in the
demand
for labor, the higher the
unemployment
ative
a — Ea. For
will initially
expectations of productivity adjust to the
in the previous box. But,
is
proBt,
and so
may
In the context of this model,
affect
p
—
and as
new
lower
before,
it
is
is
level.
at work.
it
effect
is
what we saw
So long as employ-
capital accumulation. Thus,
is
unemployment
worth returning to the role of monetary
expansionary monetary policy, to the extent
also reduces the real interest rate
can also reduce the
— Ep.
take a long time.
by having actual unemployment remain below the natural
in the
This
Ep. can reduce real wages and Umit the increase
extent that
a neg-
initially, to
go up, and then come down as
now, another mechanism
returns to normal, but this
policy. Finst,
— Ea,
rate.
the time being, ignore nominal rigidities and the term p
Other things equal, unemployment
lower, so
.
depends on the capital stock. The lower the capital
also
consider a permanent decrease in productivity, leading,
is
'
previous box, on a
Now
ment
a)]
/5
in
rate.
it
can
unemployment,
Second, to the
and therefore the user
cost,
it
on capital accumulation, and thus reduce the increase
natural rate over time.
What happened
in
the second half of the 1970s can
44
be interpreted
Had monetary
in tiiis ligiit.
would have been higher, and the decrease
If.
policy been tighter,
unemployment
accumulation larger.
in capital
however, monetary policy turns contractionary, then both effects work
money
reverse. Tight
And
natural rate.
and
tion,
to
leads to an increase in the
in
unemployment rate over the
high real interest rates lead to a decrease in capital accumula-
an increase in the natural
This can be seen as what happened
rate.
during the disinflationary episodes of the early 1980s.
Insider Effects, Hysteresis, and Persistence
Assume
real
real
and so the warranted wage are the same
that technology,
wage paid by firms —equivalently the
wage— given
w —p =
—
{a
\){n
—
To focus on the dynamics from
wage
setting.
k
+
a)
+
=
a
(cv
—
Suppose the nominal wage
where
m
is
is
lets
is
l)(n
collective bargaining,
Think of wages as being
chooses the wage, and then
of the union
employment and the
relation between
source of persistence, and assume the capital stock
to
The
by:
is
Turn
as above.
set
—
fc)
+ aa
we turn
off the other
is fixed.
by a monopoly union that
the firm decide about employment.
chosen so that, in expected value, the membership
employed:
;''
w
\
En =
log membership. Suppose that
.•
7n
membership
is
•-'
given by:
m = n(-l) + 6i(rl-n(-l))
If 9
=
0,
then membership
is
just equal to
cares only about the employed. If
I
>
on employment of the unemployed, but
9
employment
>
less
0,
last period:
the union puts
The union
some weight
than on the employment of those
already employed.
Assume
that the union chooses the nominal wage, based on the warranted
relation above,
level,
wage
and based on expectations of both technology and the price
so
w = Ep +
{a
- l){n{-i) + 9{n - n{-l) -
k))
+ aEa
45
.
Combining the equations
ganizing, using u
—
il
—
u={l-
for the
warranted and the bargained wage, and reor-
n{ — l), gives:
e)u{-l) -
—^[{p
—
a
1
Unemployment adjusts
nology. Tlie lower 9
Ep)
over time to unexpected
— t^e
+
a{a - Ea)]
movements
in prices
lower the weight of unemployment
in
the higher the persistence.
The
assumed
Under that assumption, the process
6 equal to zero.
ployment rate has a unit
root:
particular value, and where
price level
and technology.
initial
It
Even
if
—
for the
unem-
rate does not return to any
depends on the history of surprises
to }:)oth the
the assumption that
is
equal to zero
is
too
the union does not care about the unemployed, they care what
could happen to their
members
if
there were adverse shocks
became unemployed. The higher the unemployment
will
bargaining
exhibits "hysteresis"
As many pointed out however,
strong.
tech-
Blanchard-Summers (1986) formulation
The unemployment
it is
and
and some members
rate, the
more
careful they
be in their wage demands. Also, unions rarely set the wage unilaterally; to
the extent that there
The higher
the
is
bargaining, firms can threaten to hire the unemployed.
unemployment
rate, the stronger the threat. All these factors
imply a positive value of 0, and thus persistence rather than
Box
3.
Flows, Matching, and Bargaining.
hysteresis.
The Beveridge Curve
and the PhiUips Curve.
Tlie purpose of this box
for understanding
most
full
easily
is
to present the basic implications of the flow
unemployment.
comparable
I
have presented
it
in a
approach
way which makes
to the theories presented in the previous boxes.
it
For a
treatment, see the Ijook by Pissarides (2000).
Gross Flows
The
is
starting point of the theory
is
that relatively stable aggregate
the result of large gross flows of job creation
employment
and job destruction. Let
:i:
and y
46
be respectively the logs of the Hows of jobs created and jobs destroyed. Assume
that X
and y are given
by:
X
=
=
y
lu
and p are the
creation
logs of the
-p) +
-Ox{iv
2x
ey{w -P) - Zy
nominal wage and the price
Job
level respectively.
decreasing in the real wage, job destruction increasing in the real
is
wage. Zx and Zy are the factors that affect creation and destruction given the
real
wage
(for
focused on
example, productivity,
earlier).
=
This
y.
in turn
,
,
stable, so creation
must be equal
This implies that the ''warranted" real wage
W-p =
-.:.
,,.,,,..
,
we
,,,
.
;,
employment must be
In steady state,
struction: X
the cost of capital, which
oil prices,
=
Z
to de-
satisfies:
J-^{Zx + Zy)
determines steady state gross flows x and
y.
Matching, Unemployment, and Vacancies
At any point
time, there are workers looking for jobs (the unemployed),
in
and jobs looking
by a "matching
vacancies.
is
function'',
Assume
'>:,'.,
h
for workers (vacancies).
".-v^-
„>'-'.
i^
respectively.
U
The higher
of vacancies, the
is
characterized
which relates hires to the stock of unemployed and
this function
the log of hires,
The matching process
,^
and
the
of the form:
is
ctU
V
+
- a)V + Zm
arc the logs of
number
more matches,
[l
''
"'-
'-
'-
unemployment and vacancies
of unemployed, or the higher the
the
more
hires.
Zm captures
all
number
the factors
that affect the efficiency of the matching process. For example, a decrease in
the search intensity of the unemployed, or an increased
skills
of the unemployed and the
skills
mismatch between the
desired by firms, both lead to a decrease
in 2^,.
The
relation between
unemployment and vacancies
Beveridge curve. Shifts
in
for
a.
given h
is
called the
the Beveridge curve correspond to changes in
z,n.
47
The
U and V
relation between h,
can also be written
{h-U) = {l-a){V-U) +
U
h—
is
Zrr,
the log of the ratio of hires to unemployment, thus the log of the prob-
per period of finding a job when unemphypd,
ability
as:
or,
minus the log of average unemployment duration. The
unemployment duration
that
is
put yet another way,
relation therefore says
a decreasing function of the ratio of vacancies
unemployment.
to
Bargaining, and the Determination of the Bargained
we think of wages
If
and
firms, the
h — U.
as being the result of bargains between individual workers
outcome
the easier
it is
will
for a
depend on the state of the labor market. T'he higher
worker to find a job
if
unemployed, and so the stronger
the worker will be in bargaining. Symmetrically, the higher h
for a firm
is
Wage
to 611 a vacancy,
and so the stronger the firm
— V,
will
be
the easier
it is
in bargaining.
This suggests a wage relation of the form:
w-Ep
=
0[{h
-U)-
where the nominal wage depends, as
{h
before,
-
V)]
+
zb
on the expected price
difference between the labor market prospects of the worker, h
firm,
h — V.
z^
stands for
all
it
if
— which strengthen the worker
or employment protection — which makes
costly for the firm to replace a worker by another.
Note that the wage equation can be rewritten
w-Ep=
Or
U, and of the
unemployed
bargaining and increase the wage,
more
the
the other factors that affect bargaining; for example,
the level of unemployment benefits
in
—
level,
—a
this suggests the correct labor
the bargained
wage
is
Zb
from the matching function:
w - Ep = -^[{h Note that
as:
-p{U -V) +
equivalently, using the relation derived
1
,
U) -
z„,]
+
Zb
market variable
in the
equation for
not the unemployment rate (the variable traditionally
used in such equations), but either the ratio of unemployment to vacancies, or
48
the probability of exiting
unemployment
unemployment
(or its inverse, average
duration).
Combining the
ranted wage
first
of the two wage equations with the equation for the war-
earlier gives a Phillips curve relation:
p- Ep= -0{U -V)-z + Zh
The Natural Unemployment Rate
Combining
Ep =
this equation
and the equation
for the
p, gives a characterization of equilibrium
h^U
=
-^[(1
Ecpiilibrium duration (recall that
- a)z +
U—
h
/ir„
warranted wage, and assuming
unemployment duration:
-
(1
-
a)zk\
the log of duration) depends on the
is
factors that shift the warranted wage, those that shift the
and those that
affect
,
bargaining (some factors
may
be
matching function,
common
to the different
z's.)
The equihbrium
flow h
must be equal
job creation, so h
to
determined by the warranted wage. This
in turn
=
x where x
is
determines the natural rate
of unemployment as the product of equilibrium duration and the equilibrium
flow.
Note how,
in principle,
we can
learn about the sources of the shifts in the natural
rate by looking both at the Beveridge curve relation
in Zjn
— and at
Box
4.
the Phillips curve — which
tells
— which
tells
us about shifts in z
and
Zf,.
The Tax Wedge and Unemployment
Suppose that there are two types of
taxes.
(Assume both are paid by workers
rather than by the firm, but, except in the presence of a binding
it
us about shifts
minimum
wage,
does not matter whether these are paid by the worker or by the firm.)
49
A
•
tax levied on
the
same
income
all
here), be
consumption; the two are taken
(or
unemployment
it
benefits or labor income,
T]
to
.
he
(For
example, an income tax, or a VAT.)
A
•
tax levied on labor income only,
value) equal to Xt2. A
maybe
with benefits (in present discounted
r-y,
equal to zero,
with any benefit to the worker. It
may
be
the
if
may
not associated
ta.x is
he close or equal to one
goes towards financing a retirement account.
if it
Consider the choice between unemployment and employment. If unemployed,
the worker receives b
—
ti
—
>
f{u), /(.)
0, /'(.)
benefits,
and J(u) captures the private
assumed
to
receives
ti
,
—
T2
+
\t2. His surplus
Vy,,
•
= w-b+
from working
-
f(u)
[l
-
unemployment
b is
is
it
does not,
from employing the worker
where y
lu,
receives nothing.
it
therefore:
is
Thus the
is
the produc-
firm's surplus
is:
=
Vf
total surplus
for the firm
X)t2
employs the worker, the hrm receives y —
tivity of the worker. If
The
where
0,
be increasing in the unemployment rate. If employed, the worker
w—
If the firm
>
cost of being unemployed, which
from the match
y
~w
given by:
is
V = y-b + f{u)-(l-^\)T2
If
we assujue that workers receive a
wage (and
tfie
sliare
cost of labor to the firm)
w = py + {l-
(i)(b
-
is
of the surplus, so
[3
V„,
=
f5V, the
given by:
f(u))
+
72(1
-
A)(l
-
/3)
So
dCost/dTy
=
dCost/d.T2
=
(1
-
A)(l
-
i3)
This makes clear that the frequent practice of simply adding
contributions together, plus the
VAT
and/or the income tax
wedge" does not make sense (with one exception:
all
to get a
If the alternative to
not unemployment, but a job in the underground economy, then the
becomes again
relevant). For each tax,
we need
to
the social
know whether
it is
"tax
work
VAT
is
rate
levied on
50
labor income and other benefits, or just on labor income. For each tax, we also
know
the relevant value of X. In the case of social security for example,
need
to
some
redistribution
Box
5.
Assume
is
typically at work, so A
depends on the wage
level.
The Labor Share
that the production function has constant returns to scale, and Harrod
neutral technological progress, so:
y
where y,k,n, and a are the
=
fifi.ari)
levels of output, capital,
employment, and technol-
ogy respectively.
Let
fi.
= an and
(I shall
the
drop
wage
lu
= w/a
denote cuiployrncnt and the wage
"in efficiency units" in
as given, labor
demand
1=
The
ratio of
wage.
employment
The share
of labor
is
follows).
Then, assuming firms take
3(^0)
g'i.)<o
given in turn by:
'
wn
_ wn
~ y
y
consider an exogenous increase in the wage w. Suppose that in the short
run, there
The
what
given by:
to the capital stock is a decreasing function of the
.
Now
is
in cfhcicncy units
is
no scope
for substitution
between labor and
ratio of labor to capital remains the same,
to output. Thus,
What happens
it is less
w goes
up,
=
0.
ratio of labor
h/y does not change, and the labor share goes
up.
over time depends on the long run elasticity of substitution. If
than one, the labor share decreases from
higher than before the increase in the wage;
returns to
capital, so g'
and so does the
its initial value; if it is
its initial
if it is
peak, but remains
equal to one, the share
greater than one, the share returns to a lower
value than before the increa.se in the wage.
51
These dynamics can
went
first
up,
clearly explain why, after the increase in w, the labor share
and then down over
labor share decreased below
time.
But they do not
its initial level.
A
easily explain
permanent increase
a long-run elaMicity of substitution greater than one would explain
we saw
in the text, id
mid- 1980s
decreased back to or below
its original
why
in ib,
it;
the
and
but as
value from the
on.
Another element
and a lower
is
share.
therefore needed to explain the combination of a lower
One
Extend the equation
potential explanation
for the
demand
J
Then, a decrease
in z,
due
for
is
wage
a decrease in labor hoarding.
for labor to be:
g[>{)
=g{i'v,z)
example
to the reduction of x-inefBciciicy, will
decrease labor demand, and decrease the labor share. This
is
the explanation
suggested in Blanchard (1998).
Suppose instead that firms are not wage
call it
Wm,
differs
takers,
and that the marginal wage,
from the (measured) average wage. The demand for labor
is
therefore given by:
-j:
= g(wm)
In this case, an increase in the marginal wage for a given average wage, will
lead to a decrease in employment, and thus to a decrease in the labor share.
One can
in the
think of a
number
of institutions which might lead to such an increase
marginal wage: for example, additional regulations, additional workers'
rights as the size of the firm increases. This
by Caballero and
Hammour
is
the line of explanation suggested
(1998).
52
References
[1]
George Akerlof, William Dickens, and George Perry. The macroeconomics
of low inflation. Brookings Papers on
[2]
Dean Baker, Andrew Glyn, David
institutions
[3]
Charles Bean.
critical
assessment of the cross-country
working paper 2002-17, 2002.
European unemployment; a survey. Journal of Economic
Literature, 32:573-619,
[4]
Activity, 1:1-76. 1996.
Howell, and John Schmitt. Labor market
and unemployment: A
CEPA
evidence.
Economic
June 1994.
Charles Bean and Jacques
Europe's Unemployment Problem.
Dre7,e.
MIT
Press. Cambridge, 1991.
[5]
Michele Belot and Jan van Ours. Unemployment and labor market institu-
an empirical analysis. August 2001.
tions;
WP
2001-50,
CentER. Tilburg
University.
[G]
Giuseppe Bertola, Francine Blau, and Lawrence Kahn. Labor market
stitutions
and demographic employment patterns. January 2005. mimeo,
and Cornell University.
Uiiivcrsita dc Torino,
[7]
Giuseppe Bertola and Richard Rogerson. Institutions and labor reallocation.
[8]
in-
European Economic Review, 41(6):1147-1161, Jime
Olivier Blanchard.
Two
tools for analyzing
1997.
unemployment. Issues
temporary Economics, 2:102-127, 1990. Marc Nerlove
editor,
in
Con-
MacMillan
and International Economic Association.
[9]
Olivier Blanchard.
tivity,
[10]
The medium
Olivier Blanchard.
1998 (also
NBER
Economic Ac-
•
Revisiting European unemployment: Lbiemployment,
capital accumulation,
and factor
prices.
May
1998.
Geary Lecture, Dublin
working paper 6566).
[11] Olivier Blanchard.
tions,
run. Brookings Papers on
2:89-158, 1997.
The Economics of Unemployment. Shocks,
and Interactions.
2000.
Institu-
Lionel Robbins Lectures (http://econ-
www.mit.edu /faculty /blanchar/papers. html).
[12]
Olivier Blanchard, Daniel Cohen,
and Cyril Nouveau.
The
evolution of
labor market instituticjns in France since 1950. September 2005.
[13]
Olivier Blanchard and Peter
Diamond. The Beveridge
curve.
Brookings
Papers on Economic Activity, 1:1-74, 1989.
[14]
Olivier Blanchard, Rudiger Dornbusch, Jacques Dreze, Herbert Giersch,
Richard Layard, and Mario Monti. Employment and growth:
A
two-handed
approach. Center for European Policy Studies, June 1985.
53
[15]
Olivier Blanchard
and Jordi
Gali.
Real wage
rigidities
and the
New
Key-
nesian model. June 2005.
[16]
Olivier Blanchard
and Augustin Landier. The perverse
bor market reform: Fixed duration contracts
in
effects of partial la-
France. Economic Journal,
112:F214-244, June 2002.
[17] Olivier Blanchard
rise
[18]
and
fall
of
rate.
rents,
and the
European unemployment. 2003. mimeo, MIT.
Olivier Blanchard
ment
and Thomas Philippon. The decline of
and Pedro Portugal. What hides behind an unemploy-
Comparing Portuguese and U.S. unemployment. American Eco-
nomic Review, 91(l):187-207, March 2001.
[19]
Olivier Blanchard
interest rates.
[20]
and Lawrence Summers. Perspectives on high world
Brookings Papers on Economic Activity, 2:273-324, 1984.
Olivier Blanchard
and Lawrence Summers. Hysteresis and the European
NBER
unemployment problem.
Stanley Fischer (editor)
[21]
Olivier Blanchard
of
,
MIT
Macroeconomics Annual, 1:15-78, 1986.
Press.
and Justin Wolfers. Shocks and institutions and the
rise
European unemployment. The aggregate evidence. Economic Journal,
110(1):1~33,
[22]
real
March
Olympia Bover,
2000.
Pilar Garcia-Perea,
and Pedro Portugal.
study of the Portuguese and Spanish labour markets.
A
comparative
Economic
Policy,
October 2000.
[23]
William Branson and Julio Rotemberg.
wage
[24]
rigidity.
International adjustment with
European Economic Review, 13(3):309-337,
1980.
Michael Bruno and Jeffrey Sachs. The Economics of Worldwide StagHation.
Basil Blackwell, Oxford, 1985.
[25]
Ricardo Caballero and
ity,
factor substitution
Mohamad Hammour.
Jobless growth: Appropriabil-
and unemployment. Carnegie-Rochester Conference
Series on Public Policy, (48):51-94, 1998.
[26]
Pierre
Cahuc and Yann Algan. Civic
ket institutions:
Which
attitudes
and the design
Pierre
.
of labor
mar-
countries can implement the danish Flexisecurity
model? September 2005. mimeo Universite Paris
[27]
'
_
I,
Cahuc and Andre Zylberberg. Le chomage:
Paris.
Fatalite on Necessite?
Flammarion, Paris, 2004.
[28]
Lars Calmfors and John
Driffill.
Centralization of wage bargaining and
macroeconomic performance. Economic
Policy, 6:16-31, 1988.
54
,
a utility-maximizing framework.
[29]
Guillenno Calvo. Staggered contracts
[30]
Michel Camde.ssus. Le sm'saut. Vers une nouvelle croissance pour la france.
in
Journal of Monetary Economics, September 1983.
September 2004. Documentation Francaise,
[31]
What
David Card and Richard Freeman.
economic reform delivered? 2004.
in
Paris.
have two decades of British
Seeking a Premier League Economy"
Richard Blundell, David Card and Richard Fi'eeman
[32]
David Coe and Dennis Snower.
(eds).
Policy complementarities:
fundamental labor market reform,
The
pages 1-35, March 1997.
case for
IMF
Staff
Papers 44-1.
[33]
Diego Comin and Thomas Philippon.
The
rise in
Causes and consequences. NBER. Macro Annual,
NBER
[34]
and
MIT
firm-level volatility:
20, 2005.
forthcoming,
Press.
Bruno Crepon and Rozenn Desplatz. Une nouvelle evaluation des
allegements de charges sociales sur
8,
2001.
.,
-I
.
effets des
bas salaires. Economic et Statistique,
.
,
les
,
[35]
Francesco Daveri and Guido Tabellini. Unemployment, growth and taxa-
[30]
Steven Davis, John Haltiwanger, and Scott Schuh. Job Creation and Job
tion in industrial countries.
Destruction.
[37]
MIT
Steven Davis, R. Jiison Faberman, and John Haltiwanger.
recent downturn.
IZA
New
data sources, micro-macro
flow ap-
links,
and the
discussion paper 1639, June 2005.
Europe and the United
How
States:
they affect aggregate
OECD
labor shares and the properties of the wage equations.
papers, (326), 2002. Economics Department,
[39]
The
Alain de Serres, Stefano Scarpetta, and Christine de la Maisonneuve. Sectoral shifts in
;
Policy, pages 47-104, April 2000.
Press; Cambridge, Mass, 1996.
proach to labor markets:
[38]
Economic
Peter Diamond.
Wage
working
OECD.
determination and efficiency in search equilibrium.
Review of Economic Studies, 49:217-227, 1982.
[40]
Richard Duhautois. Evolution des flux d'emplois en france entre 1990
1996;
une etude empirique a partir du
INSEE
[41]
des 'benefices reels normaux.
G9915, September 1999.
Peter Fredriksson and Berth Holmlund. Improving incentives in unemploy-
ment insurance;
April 2003.
[42]
fichier
et
A
review of recent research.
CESIfo working paper
922,
,
,
Richard Freeman. Labour ma.rket institutions without blinders: The debate
55
NBEB. working paper,
over flexibility and labour market performance.
(11286), April 2005.
[43]
Milton R-iedman. The
view, 58:1-21,
[44]
March
Nils Gottfries
Economic
to external supply shocks.
Act.ivity, 1:183-206, 1975.
and Henrik Horn. Wage formation and unemployment per-
Economic Journal, 97:877-886, December
sistence.
[46]
American Economic Re-
policy.
Robert Gordon. Alternative responses of policy
Broolcings Papers on
[45]
monetary
role of
1968.
Robert Gregory. Wages policy and unemployment
1987.
in Australia.
Economica,
53(210 (S)):S53-74, 1986.
[47]
Martin Hellwig and Manfred Neumann. Economic policy
in
Germany: Was
there a turnaround? Economic Policy, 5:105-140, 1987.
[48]
Bertil
Holmlund. The
rise
and
fall
of
Swedish unemployment.
CESIFO,
April 2003. working paper 918.
[49]
Patrick
Honohan and Brendan Walsh. Catching up with
Irish hare.
[50]
the leaders;
The
Brookings Papers on Economic Activity, (l):l-57, 2002.
Jennifer Hunt.
The impact
of the 1962 repatriates from Algeria of 1962
on the French labor market. Industrial and Labor Relations Review, April
1992.
[51]
George Johnson and Richard Layard. The natural rate of unemployment:
Explanation and poUcy. 1986. Handbook of Labor Economics, Volume
II,
0. Ashenfelter and R. Layard eds. North Holland, 1986, 921-999.
[52]
Richard Layard and Stephen Nickell. The labour market. The Performance
of the British Economy, 1987.
(editors).
[53]
Rudiger Dornbusch and Richard Layard
Clarendon Press, Oxford.
Richard Layard, Stephen Nickell, and Richard Jackman. Unemployment;
Macroeconomic performance and the labour market.
Oxford University
Press, 1991.
[54]
Richard Layard, Stephen Nickell, and Richard Jackman.
Macroeconomic performance and the labour market.
Unemployment;
Oxford University
Press, 2005. second edition.
[55]
Edward
Lazear. Job security provisions and employment. Quarterly Jour-
nal of Economics, 105-3:699-726, 1990.
[56]
Assar Lindbeck and Dennis Snower. The Insider Outsider Theory of
ployment and Unemployment.
MIT
Em-
Press, 1989.
56
[57]
Thomas
Lars Ljungqvist and
The European unemployment
Sargent.
dilemna. Journal of Political Economy, 106(3):514-550, 1998.
[58]
economic theory;
[59]
Thomas
Lars Ljungqvist and
.lohu Martin
of
OECD
A
Sargent. Jobs and
unemployment
NYU,
turbulence laboratory. 2005. mimeo,
and David Grubb.
What works and
for
in
macro-
August.
whom: A review
OECD
countries' experience with active labor market policies.
working paper, 2001.
[GO]
Tommaso Padoa-Schioppa. La
Franco Modigliani and
in
una economia con
salari indicizzati al
100% o
politica
piii.
economica
Moneta. e Credito,
30:3-53, First Quarter 1977.
[61]
Dale Mortensen and Christopher Pissarides.
tion in the theory of
unemployment.
and job destruc-
.Job creation
Review of Economic
Studies, 61-
3(208):397-416, July 1994.
[62]
Stephen
Unemployment and labor market
Nickell.
Europe versus
rigidities:
North America. Journal of Economic Perspectives, ll(3):55-74, Summer
1997.
[63]
[64]
Stephen Nickell, Luca Nunziata, and Wolfgang Ochel. Unemployment
the
OECD
27,
January 2005.
Stephen
since the 1960s:
Nickell,
The Beveridge
What do we know? Economic
in
Journal, 115:1-
Luca Nunziata, Wolfgang Ochel, and Glenda Quintini.
curve,
unemployment and wages
in the
OECD
from the
1960s to the 1990s. July 2002. Working paper, LSE.
[65]
Stephen Nickell and Jan van Ours. The Netherlands and the United King-
[66]
Volker Nocke. Gross job creation and gross job destruction:
dom: a European unemployment miracle. Economic
Policy, 30, April 2000.
study with French data. 1994. Bonn University.
[67]
OECD. The OECD
[68]
Andrew Oswald. The missing
Jobs Study.
OECD,
An
empirical
.
1994.
piece of the
unemployment puzzle?
1997.
Inaugural Lecture, Department of Economics, Warwick.
[69]
[70]
Edmund Phelps.
New York, 1972.
Edmund
Phelps.
unemployment,
MA,
[71]
Inflation theory
and unemployment
Structural Slumps.
interest,
and
assets.
policy.
W.W.
Norton;
The modern equilibrium theory of
Harvard University Press; Cambridge
1994.
Christopher Pissarides. Short run equilibrium dynamics of unemployment,
57
American Economic Review, 1985:676-690,
vacancies, and real wages.
1985.
[72]
Christopher Pissarides. Equilibrium Unemployment Theory. Basil Blackwell;
Oxford, 1990.
[73]
Christopher Pissarides. Equilibrium Unemployment Theory. Second edi-
[74]
John Taylor.
tion,
MIT
Political
Press, 2000.
Aggregate dynamics and staggered contracts.
Economy, 88(l):l-24,
The golden age
[75]
Peter Temin.
[76]
Niels Westergaard-Nielsen.
Journal of
1980.
of
European growth reconsidered. European
Review of Economic History, 6:3-22, 2002.
Danish labour market
policy: Is
it
worth
it?
Center for Labour market and Social Reserch, working paper 01-10, No^
vember 2001.
[77]
Conny Wunsch. Labour market
ments and reforms since
G alien, March
policy in
unification.
Germany:
linstitutions, instru-
Discussion Paper, Universitat St
2005.
58
Fig A1.
Unemployment
rates,
by country
1970 1975 1980 1995 1990 1995 2000 2005
1970 1975 19B0 1985 1990 1995 2000 2005
AUSTRIA
BELGIUM
1970 1975 1980 1985 1990 1995 2000 2005
1970 1975 1980 1985 1990 1995 2000 2005
GERMANY
DENMARK
1970 1975 1980 1985 1990 1995 2000 2005
1970 1975 1980 1985 1990 1995 2000 2005
SPAIN
FINLAND
1970 1975 19B0 1985 1990 1995 2000 2005
1970 1975 1980 1985 1990 1995 2000 2005
FRANCE
UNITED KINGDOM
59
Fig A1.
Unemployment
rates,
continued
1970 1975 1980 1985 1990 1995 2000 2005
1970 1975 1980 1985 1990 1995 2000 2005
GREECE
IRELAND
Sli
1970 1975 1980 1985 1990 1995 2000 2005
ITALY
1970 1975 1980 1985 1990 1995 2000 2005
1970 1975 1980 1985 1990 1995 2000 2005
1970 1975 1980 1985 1990 1995 2000 2005
NETHERLANDS
PORTUGAL
LUXEMBOURG
mi
1970 1975 1980 19S5 1990 1995 2000 2005
1970 19?5 1980 1985 1990 1995 2000 2005
SWEDEN
UNITED STATES
60
Figure
8.
Nominal and
2005
real price of
crude
oi
dollars
o
00
o
CD
ro
I
75
E
E
o
c
o
^
o
CNJ
1960
1965
1970
1975
1980
1985
1990
1995
t
I
2000
2005
61
•GO'O^
C55
mm
m
im
1
ii
wm
ililf
i'iiiliuKiiW'
'ititi^SiSiP«ii
.1