Transcript The disappearance of European Unemployment

What Happened to European
Mass Unemployment?
Tito Boeri
Bocconi University
26/06/2008
1994 OECD Jobs Study
• “The labour market has become particularly
worrying in Europe… (…) in comparatively
inflexible Europe, on the other hand, both
relative employment and unemployment rates
deteriorated”.
• “The high incidence of long-term unemployment
in most EC countries is associated with low
inflow rates into unemployment. The opposite
relationship – low incidence of long-term
unemployment and high inflows into
unemployment – holds for North America”.
Since then
as a % of the labour force
12 6
10
4,5
83
6
1,5
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Unemployment EU 15
Long term Unemployment EU 15
Outline
• A European dream… turning into
a European nightmare
• Why? A closer look at transitions
across labour market states
• Can we “cheat” the employment
labour productivity tradeoff?
A European dream
“The Community shall have as its task (…)
to promote throughout the Community (…)
a high degree of convergence of economic
performance, a high level of employment
and of social protection, the raising of the
standard of living and quality of life, and
economic and social cohesion and
solidarity among Member States”.
Rome Treaty, March 25, 1957
1985 - 1995
FI
.5
1
SE
FR
GR
DE
IT
0
ES
LU
DK
PT
BE
UK
IE
-.5
Log (Unemployment rate 1995 / Unemployment rate 1985)
Convergence: Did the countries with
high unemployment experience the
strongest decline in unemployment?
NL
1
1.5
2
Log (Unemployment rate 1985)
2.5
3
Convergence
.5
AT
DE
0
PT
GR
BE
FR
SE
-.5
NL
UK
DK
IT
FI
ES
-1
Log (Unemployment rate 2006 / Unemployment rate 1996)
1996 - 2006
IE
1
1.5
2
Log (Unemployment rate 1996)
2.5
3
.5
1
No convergence
1985 - 1995
IT
GR
FR
ES
0
DE
LU
DK
PT
BE
UK
IE
-.5
NL
1.5
2
2.5
3
Convergence
1996 - 2006
.5
Log (Unemployment rate 1985)
AT
DE
0
PT
GR
BE
-.5
FR
SE
NL
UK
DK
IT
FI
-1
1
Log (Unemployment rate 2006 / Unemployment rate 1996)
Log (Unemployment rate 1995 / Unemployment rate 1985)
FI
SE
1
1.5
2
IE
ES
2.5
3
Log (Unemployment rate 1996)
Convergence in unemployment rates also
within EU countries since the mid 1990s
Squared root of Sum of Squares
(Nuts II regions)
2,5
2,0
1,5
1,0
0,5
0
Total
Between
Within
Δ Empl 2006-96
It is not inactivity. It is employment
0,3
IRE
-ΔU<ΔE
0,25
ES
0,2
0,15
FI
NL
0,1
PT
SE FR
GR
IT
BE
UK
0,05
AT
DK
Δ Unempl 2006-96
DE
0
-0,3
-0,2
-0,1
0
0,1
0,2
0,3
-0,05
-0,1
-ΔU>ΔE
Δ Empl = - Δ Unempl
-0,15
Unemployment and Employment expressed as a fraction of the working age population
Whatever measure of
labour slack we use
Measure of labour slack (in millions)
16
14
12
10
8
6
4
2
0
1994
1995
1996
1997
Unemployed workers
EU Countries: DK, BE, FR, IRE, IT, GR, ES, PT.
1998
Discouraged workers
1999
2000
Lisbon is no longer a mirage
Employment to population rates and the distance from
the Lisbon Employment Target
1995
80%
2006
70%
60%
50%
40%
30%
20%
10%
0%
DK
NL
Source: Eurostat
SE
UK
AUT
FI
IE
PT
DE
EU15 ES
FR
LU
GR
BE
IT
The dream came true?
• Since 1995, the UE15 has 4 millions less
people unemployed.
• Decline seen in 11 countries out of 15.
• Associated with 21 millions more jobs.
• Reducing cross-country and within country
unemployment differentials.
• Supposedly more “social cohesion”, but…
…Europeans are unhappy
Fully satisfied as a % of
respondents
Satisfaction with work or main activity in EU10
18
17
16
15
14
13
12
1994
1995
1996
1997
1998
years
Source: ECHP
1999
2000
2001
Declining job satisfaction notably in the countries
with the strongest unemployment decline
Job satisfaction in EU15
(% of employees expressing satisfaction with their working
conditions)
Denmark
Finland
Ireland
UK
Austria
Belgium
Luxembourg
Germany
2000
1995
Netherlands
Sweden
EU15
France
Portugal
Italy
Spain
Greece
50%
60%
Source: www.eurofund.europa.eu
70%
80%
90%
100%
Reducing unemployment is not a
popular business
• Under Berlusconi II (June 2001- May
2006) 1,354,320 jobs were created. Yet
support for the Govt fell by 43%.
• Under Prodi II (May 2006 - December
2007) 432,512 jobs were created. Yet
consensus fell by 51%.
• Aznar had to go in spite of 4,982,050 jobs
created and halving the Spanish
unemployment rate.
Employment
Unemployment
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1996
1997
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
Employment as a % of population in working age
60%
12,5%
58%
11%
56%
9,5%
54%
8%
52%
6,5%
50%
5%
Unemployment as a % of labour force
Italy
Employment
Unemployment
75%
25%
65%
20%
55%
15%
45%
10%
35%
5%
Unemployment as a % of labour force
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
Employment as a % of population in working age
Spain
Outline
• A European dream….. turning
into a European nightmare
• Why?
– a simple explanation
– looking at flows
• Can we cheat the employment
labour productivity tradeoff?
The simple explanation
• Lower unemployment could simply be
related to demographics.
• Insofar as unemployment rates are higher
for the young people than for the other age
groups, the ageing of Europeans may
involve a reduction of unemployment.
• Is this the reason why we no longer see
mass unemployment in Europe?
No. It isn’t
1995-2004 Variation of Unemployment as a % of the Working Age Population
-7
-6
-5
-4
-3
Within group variation in U rate
-2
-1
0
Pure demographic effect
1
Also more migrants, but they have higher
unemployment rates than natives
1995-2004 Variation of Unemployment as a % of the Working Age Population
SE
ES
PT
NL
LU
IE
GR
FR
FI
DK
BE
AT
-7
-6
-5
-4
Migration composition effect
-3
-2
-1
0
1
Variation in U rate of natives and migrants
Inflows/Outflows as a % of the working age population
-0,6
Inflows EU15
1991
1989
1987
Outflows EU15
-0,8
U rate EU15
0,8
0,6
0,4
0,2
11
0
-0,2
8
5
Unemployment as a % of the labour force
2006
2003
2001
1999
1997
1995
1993
-0,4
1986
Unemployment declined with larger
unemployment inflows
15
Let us look at transition matrices
Employment t
Employment t-1
Unemployment t-1
Inactivity
t-1
93,5
30,0
3,7
Unemployment t
Inactivity
3,0
3,5
47,3
22,6
1,8
94,6
Yearly Average
2001 – 2004
Out of 100 hundred unemployed, 30 find a job
…and 23 leavewithin
the labour
altogether
a yearmarket
….
another 47 remain unemployed….
t
What do these transitions imply in terms
of the long-run unemployment rate?
12%
10%
8%
6%
4%
2%
0%
1991-1995
1991-1995
Current Unemployment rate
2001-2004
2001-2004
Unemployment rate in the long run
1985 – 1988
Employment t
Employment t-1
Unemployment t-1
Inactivity t-1
93,1
29,0
2,8
Unemployment t
3,2
56,5
1,7
Mobility Index = (S- tr(M))/(S-1)
Inactivity t
3,7
14,1
95,5
where S denotes the number of
labour market states and tr(M)
the trace of the transition matrix
28%
In Europe more mobility across labour market
states than 15-20 years ago
2001 – 2004
Employment t
Mobility Index
Employment t-1
Unemployment t-1
32%
Inactivity t-1
93,5
30,0
3,7
Unemployment
3,0
47,3
1,8
t
Inactivity t
3,5
22,6
94,6
AT
0
PT
GR
BE
FR
-.5
NL
IT
FI
-1
log(U rate 2006 / U rate 1996)
.5
Increase in mobility in the countries with the
strongest unemployment declines
ES
IE
-.1
0
.1
.2
.3
.4
log(Average Mobility Index 1996-2004/Average Mobility Index 1985-1995)
Why did mobility increase?
• Multivariate analysis (across countries and over
time) of the determinants of mobility indexes.
• Reduction in the strictness of Employment
Protection Legislation (EPL) for temporary
contracts increases mobility by 5%.
• Reduction of EPL strictness for regular contracts
increases mobility by 2%.
• Controlling for GDP growth and the share of
immigrants in the population.
Acceleration of reforms of EPL
18
Employment Protection
Legislation
16
N° of Reforms
14
12
10
8
6
4
2
0
Countries: EU 14
Regular contracts
Temporary contracts
15
67
66
14
65
13
64
63
12
62
11
61
10
60
1995
1996
1997
1998
1999
2000
2001
% of employees with temp. contracts
2002
2003
2004
2005
2006
Employment rate EU 15
Employment rate EU15 (%)
% of employees with temporary contracts
The growth of temporary
employment in Europe
A port of entry?…
% of employees with temporary contracts
1
0.9
0.8
0.7
France
Denmark
Netherlands
Italy
Spain
Portugal
0.6
0.5
0.4
0.3
0.2
0.1
0
18
19
20
21
22
23
24
25
26
Age
27
28
29
30
31
32
33
34
35
… or a dead end?
Spain
2004
2003
Permanent Contracts
Fixed Term Contracts
Unemployment
Inactivity
Permanent
Contracts
97,2
4,8
2,5
0,4
Fixed Term
Unemployment Inactivity
Contracts
1,1
1,0
0,7
82,6
9,0
3,5
20,1
67,0
10,4
2,4
3,9
93,3
Temporary contracts:
a longlasting phenomenon
35%
30%
25%
20%
SPAIN
ITALY
Selfemployment
15%
10%
5%
0%
Current
share of
FTC
Long-run
share of
FTC
Current
share of
FTC
Long-run
Share of
FTC
Social cohesion?
Dualism temporary-permanent contracts
creates longstanding asymmetries (this is
not properly social cohesion)….
Moreover
• Also workers with permanent contracts may suffer a
capital loss with EPL reforms as their welfare includes
expectations of future job losses. “Insiders” cannot be
fully insulated from reforms!
• Decline in their welfare associated with an increase in
the probability of job loss (λ) is
-(w-wr)/(r+λ)
where w is the market wage, wr is the reservation
wage and r is the interest rate
Easier to (re)enter the labour
market but often at lower wages
Entry wage in a new job as a % of the average wage for different groups
140%
120%
100%
80%
60%
40%
20%
0%
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Source: ECHP (BE, DE, DK, ES, FR, GR, IE, IT, LU, NL, PT, UK), EUSILC (AUT, BE, ES, GR, IE, IT, PT, UK ).
As wage increases are mostly related
to tenure
(unconditional wage-tenure profiles,source ECHP)
600
mean wage
500
400
300
200
100
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
tenure
Germany
UK
France
Italy
Spain
Sweden
20
Outline
• A European dream….. turning
into a European nightmare
• Why? A look at transitions across
labour market states
• Can we “cheat” the
employment/productivity tradeoff?
How to make Europeans happier
about lower unemployment?
• Providing more employment security is not
an option.
• More labour market risk makes risk-averse
individuals unhappy unless it is
accompanied by higher wages.
• We need higher wages to provide popular
support to reforms in Europe.
• But higher wages require higher labour
productivity.
Can Europe increase both
employment and productivity?
100%
90%
180
Hourly labour
productivity (US=100)
(left scale)
80%
160
70%
60%
Employment
140
(in millions)
50%
40%
Source: Groningen Growth & Development Centre, Total Economy Database
120
What do employment protection
reforms do to labour productivity?
• On the job destruction margin: they reduce
the number of low productivity jobs. This
increases productivity.
• On the job creation margin: they may
discourage investment in human capital as
shorter tenures reduce incentives to
training on-the-job. This reduces
productivity.
How to “cheat” the tradeoff then?
Increase wages-productivity at entry. More
investment in human capital.
• Before the job: better education systems via
compulsory evaluation of all schools
• On the job:
–
–
tenure track to permanent contracts (no
segregation into temporary contracts)
decentralised wage bargaining that rewards
productivity rather than simply tenure
Reallocation increases average
productivity and wages
wage
productivity
wage
productivity
Ls1
Ls
Ls
Ls1
w
w1
Reallocation
surplus
w1
Ld
w
Ld
Employment
Employment