Impact of the Rise in Immigrant Unemployment on Public Finances^* by Pablo Vazquez^**, Mario Alloza^*** Raquel Vegas ^*** and Stefano Bertozzi^***** DOCUMENTO DE TRABAJO 2008-15

Serie Inmigración CÁTEDRA Fedea – Banco Popular

March 2009

* This study has been funded by the Bureau of European Policy Advisers of the European Commission.

** Pablo Vazquez is the Executive Director of the Foundation of Studies of Applied Economics in Madrid.

*** Mario Alloza and Raquel Vegas are a research fellow at the Foundation of Studies of Applied Economics in Madrid.

**** Stefano Bertozzi is a member of the Bureau of European Policy Advisers at the European Commission. The ideas contained in this article are entirely those of the author and do not necessarily reflect the views of the European Commission.

Los Documentos de Trabajo se distribuyen gratuitamente a las Universidades e Instituciones de Investigación que lo solicitan. No obstante están disponibles en texto completo a través de Internet: http://www.fedea.es. These Working Paper are distributed free of charge to University Department and other Research Centres. They are also available through Internet: http://www.fedea.es.

ISSN:1696-750X

Jorge Juan, 46 28001 Madrid -España Tel.: +34 914 359 020 Fax: +34 915 779 575 infpub@fedea.es

Abstract

The
current
slump
is
having
a
heterogeneous
impact
on
the
EU
economies
regarding
their GDP
 and
 employment
 growth
 responses.
 The
 impact
 of
 immigrants’
 unemployment
 on public
 finances
 of
 EU
 countries
 depends
 on
 three
 factors:
 (i)
 the
 sensitiveness
 of
 the economy
 to
 the
 business
 cycle,
 (ii)
 the
 share
 that
 migrants
 represent
 over
 total
 labour force
population
and
(iii)
the
benefits
structure
of
their
unemployment
benefits
programs. Our
 results
 confirm
 that
 the
 impact
 of
 the
 rise
 in
 immigrants’
 unemployment
 on
 the unemployment
 benefit
 burden
 during
 the
 next
 few
 years
 is
 likely
 to
 be
 sizeable. Unemployment
benefit
burden
is
expected
to
peak
in
2009
after
an
increase
in
2008,
and to
slow
down
slightly
in
2010.
We
find
that
Latvia,
Estonia
and
France
are
the
ones
more likely
to
suffer
a
higher
public
finance
burden
from
the
rise
in
immigrants´
unemployment. Other
 economies
 such
 as
 Germany,
 Finland,
 Spain,
 Ireland,
 Italy
 or
 Austria
 would
 also register
a
noticeable
increase
in
their
public
burden
although
to
a
lesser
extent.

1. INTRODUCTION ....4 2. DESCRIPTIVE ANALYSIS ....6 2.1. Differential Effects of the Crisis on EU Labour Markets....6 2.2. Migration in the EU....7 2.3. Differences in the unemployment benefits systems....11 3. ECONOMETRIC ESTIMATION ....17 3.1. Data....17 3.2. Modelling strategy....17 3.3. Results....18 4. CONCLUSIONS....22 5. REFERENCES....24

1. INTRODUCTION

International
migration
has
been
drawing
increasing
attention
 from
policy
makers
in
the EU,
 reflecting
 the
 impact
 that
 this
 phenomenon
 has
 on
 the
 economy,
 foreign
 affairs
 and society
in
general.
The
net
number
of
immigrants
in
EU‐27
has
increased
threefold
during the
last
decade
leading
to
heterogeneous
patterns
across
the
member
states
which
range from
 long‐
 standing
 destination
 countries,
 to
 new
 gateway
 destination
 or
 even
 new
 net immigration
countries.

Figure
1:
Evolution
net
migration
to
the
EU

The
pattern
of
 the
recent
immigration
wave
is
somewhat
different
 to
 the
previous
ones: immigration
from
countries
outside
the
EU
appears
to
be
much
more
sizeable
than
intra‐EU
 flows.
 Moreover,
 migration
 towards
 southern
 countries
 has
 hastened
 considerably during
 recent
 years,
 making
 them
 as
 important
 as
 the
 more
 traditional
 immigration countries.
EU
destination
countries
have
relied
on
immigration
as
a
device
to
lessen,
or
at least
 to
postpone,
 their
ageing
population
problem,
 to
offset
shortages
regarding
specific skills
and
to
improve
overall
labour
market
flexibility.

However,
higher
migrant
workers flexibility
 often
 implies
 higher
 volatility
 regarding
 their
 labour
 status:
 immigrants employment
and
 unemployment
 rates
exhibit
 higher
fluctuations
 than
 natives’
 response to
changes
in
the
business
cycle.

In
the
current
context
of
economic
downturn
driven
by the
 global
 financial
 crisis,
 this
 fact
 plays
 not
 only
 against
 migrants
 –
 who
 experience
 a surge
in
their
unemployment
hazard
rate
and
a
worsening
of
their
standard
of
living
in case
of
becoming
unemployed
‐
but
also
against
the
host
economies
whose
public
budgets are
hit
as
a
result
of
the
subsequent
increase
in
the
unemployment
benefits
burden.

Quantifying
 the
 increase
 in
 unemployment
 benefits
 paid
 to
 immigrants
 in
 the
 context
 of the
current
slump
is
the
goal
of
this
document.

It
is
important
to
start
by
highlighting
that
the
EU
countries
are
rather
heterogeneous
in terms
of
 the
 factors
 that
comprise
expenditure
on
unemployment
benefits.
First,
despite the
fact
that
the
global
crisis
is
having
negative
repercussions
for
all
the
EU
economies,
its impact
 will
 differ
 across
 member
 states
 depending
 on
 their
 openness
 to
 global
 markets, country‐specific
 characteristics
 and
 domestic‐adjustments
 needs.
 These
 heterogeneous features
 could
 lead
 to
 different
 patterns
 in
 the
 evolution
 of
 unemployment
 rates
 and therefore
to
disparity
in
the
financial
burden
caused
by
the
automatic
stabilizers,
such
as unemployment
benefits.

Secondly,
the
size
of
the
unemployment
benefit
burden
caused
by
immigrants
hinges
not only
on
 the
relative
business
cycle
position
of
 the
EU
economies
but
also
on
 the
specific characteristics
 of
 their
 unemployment
 benefits
 systems
 ‐
 degree
 of
 generosity
 and qualifying
 conditions
 required
 to
 be
 covered‐
 and
 the
 socioeconomic
 characteristics
 of immigrants
currently
working
 there
and
in
 the

 future
–number
of
migrants,
years
since arrival,
skills
and
legal
status.

In
the
next
section
we
estimate
the
increase
in
the
unemployment
benefit
burden
among the
 different
 EU
 countries
 based
 on
 current
 prospects
 about
 their
 GDP
 growth6,
 under alternative
hypothesis
about
the
characteristics
of
the
migrants
that
become
unemployed as
a
result
of
the
crisis.

These
 alternative
 scenarios
 are
 based
 on
 the
 assumption
 that
 virtually
 no
 net
 migrant’s flow
 towards
 EU
 countries
 will
 take
 place
 in
 the
 next
 few
 years.
 Given
 certain
 data shortages,
 additional
 assumptions
 are
 needed
 about
 the
 average
 immigrant’s characteristics.
 Specifically,
 we
 assume
 that
 these
 average
 characteristics
 and
 their average
 unemployment
 benefits
 claims
 remain
 unaltered
 from
 2006
 to
 the
 end
 of
 the outlook
horizon
period.

6 The
Commission’s
economic
forecast
published
in
November
projects
EU
economic
growth
to
drop
sharply
to 1.4%
in
2008.
It
was
2.9%
in
2007.
In
2009
the
EU
economy
is
expected
to
grind
to
a
stand‐still
at
0.2%
before recovering
to
1.1%
in
2010.
The
equivalent
figures
for
the
euro
area
for
the
period
are
1.2%,
0.1%
and
0.9%. In
2007
it
was
2.7%.

Risks
seem
to
be
tilted
toward
the
downside.
A
sharper
than
expected
adjustment
could lead
 to
 higher
 unemployment
 rates
 and
 therefore
 higher
 migrant’s
 unemployment benefits
 burdens.
 Further,
 it
 is
 possible
 that
 the
 global
 economic
 downturn,
 rather
 than having
no
lasting
impact
on
the
pace
of
net
immigration
towards
the
EU
countries,
it
may exacerbate
 it,
 if
 other
 traditional
 destination
 countries
 outside
 the
 EU
 experience
 more acute
adjustments
in
the
forthcoming
future
that
discourage
immigration
towards
them

2. DESCRIPTIVE
ANALYSIS

2.1. Differential
Effects
of
the
Crisis
on
EU
Labour
Markets

After
a
strong
increase
in
the
EU‐27
GDP
annual
growth
during
2006
and
the
first
half
of 2007,
the
EU
economy
started
to
lose
momentum.

As
a
result
of
the
global
slump,
GDP annual
growth
slowed
down
in
2008
and
a
worse
performance
is
expected
for
2009.
Even under
this
common
adverse
scenario
there
are
significant
differences
among
GDP
growth of
the
member
states,
ranging
 from
7%
in
Slovakia,
4.4%
in
Slovenia
and
Czech
Republic to
‐1.6%
in
Ireland
and
‐1.3%
in
Estonia.

Table
1:
GDP
annual
growth
across
EU
members
and
selected
countries,
1999­2010
(Source: Eurostat
&
EC
DG­ECFIN)

Following
a
similar
pattern,
employment
across
EU‐27
countries
experienced
a
noticeable increase
in
both
2006
(1.6%)
and
2007
(1.8%),
followed
by
a
strong
deceleration
in
the next
two
years
(see
European
Commission
Autumn
Forecast
2008).

Unemployment
rates
prospects
also
record
worsening
expectations
of
the
labour
market, with
 a
 noticeable
 rise
 both
 in
 2009
 and
 2010.
 This
 increase
 is
 expected
 to
 place
 the unemployment
rate
over
its
natural
trend
(as
measured
as
the
non‐acceleration
wage
rate of
unemployment;
source:
AMECO‐EC)
during
these
years.

Figure
2:
GDP
and
Employment
Growth
in
EU­27 Across
 EU‐27
 members,
 according
 to
 the
 EC
 Forecasts,
 the
 Spanish
 labour
 market
 is expected
 to
 be
 the
 most
 affected
 by
 the
 business
 cycle
 (Table
 2).
 In
 fact,
 Spanish unemployment
rate
could
increase
its
unemployment
rate
by
2.5
points
in
2008,
reaching 13.8%
 in
 2009,
 and
 15.5%
 in
 2010.
 Although
 to
 a
 lesser
 extent,
 other
 countries
 could follow
the
same
mood,
such
as
Ireland,
Greece
or
Eastern
Europe
countries
(e.
g.
Latvia, Lithuania
or
Estonia).

By
contrast,
the
current
downturn
seems
to
have
a
negligible
effect
on
the
labour
market of
 other
 countries.
 For
 example,
 Bulgaria,
 Cyprus
 or
 Slovakia
 are
 expected
 to
 improve employment
conditions
during
both
2008
and
2009.

2.2.Migration
in
the
EU

During
1999‐2007,
net
migration
has
increased
substantially
(almost
a
95%),
accelerating especially
 in
 Southern
 Europe
 countries
 (Spain,
 Italy,
 etc.)
 The
 consequences
 of
 this phenomenon
 have
 been
 recently
 analysed
 by
 the
 EC
 (European
 Commission
 2008), concluding
 that
 immigration
 could
 foster
 economic
 growth
 and
 increase
 labour‐
 market flexibility.
The
latter
would
be
improved
not
only
by
having
an
increase
in
labour
input
but also
through
the
characteristics
of
the
incoming
labour
supply.
This
is
so
since
immigrant workers
exhibit
a
higher
degree
of
dynamism
compare
to
native
EU
workers.
In
addition, incoming
 migration
 flows
 have
 helped
 to
 ease
 shortages
 of
 low‐skill
 jobs
 in
 several industries.
Overall,
migrant
employment
seems
to
complement
rather
than
substitute
EU workers.

According
 to
 he
 EC’s
 report,
 the
 potential
 contribution
 of
 immigrants
 is
 not
 yet
 fully realized
since
there
persist
problems
regarding
to
their
integration
in
the
labour
market (higher
unemployment
rates
and
tend
to
be
more
exposed
to
jobs
of
lower
quality)
Hence, migrants
 find
 more
 difficult
 to
 make
 effective
 use
 of
 their
 human
 capital,
 resulting
 in underutilisation
 of
 their
 skills
 and
 mismatches
 between
 their
 actual
 job
 and
 their qualifications.

Table
2:
Unemployment
Rate
across
EU
countries
(Source:
Eurostat
and
EC
DG­ECFIN).

The
concentration
of
immigrants
varies
strongly
across
the
EU.
Figure
3
shows
the
relative size
of
foreign‐born
working‐age
population
in
the
host
country
labour
market.
There
are some
 countries
 in
 which
 foreign
 active
 population
 plays
 an
 important
 role,
 like Luxembourg,
 Austria,
 Sweden,
 Cyprus
 and
 Spain
 (with
 a
 share
 of
 foreign
 working‐age population
above
10%).

Another
divergent
feature
of
immigration
across
EU
members
is
the
period
elapsed
since immigrants´
arrival.
Hence,
there
are
countries
with
an
old
tradition
on
migration
hosting, such
 as
 Luxembourg,
 France,
 Estonia
 or
 Latvia,
 while
 there
 are
 others
 where
 this phenomenon
has
started
only
recently
(e.g.
Spain
or
Cyprus).
Immigrants
who
have
been residing
 for
 a
 long
 in
 the
 same
 country
 are
 more
 likely
 to
 be
 entitled
 to
 receive unemployment
 benefits,
 which
 made
 the
 host
 country
 Social
 Security
 Funds
 more vulnerable
to
a
shock
that
lowers
immigrant
employment.
Moreover,
the
longer
they
have stayed
in
the
host
country,
the
higher
their
average
contribution
to
Social
Security
systems will
be,
increasing
therefore
their
expected
unemployment
benefits.
Conversely,
the
lower they
 have
 stayed
 in
 the
 host
 country,
 the
 lesser
 is
 the
 social
 and
 labour
 attachment
 of immigrants.
Consequently,
these
immigrants
are
more
likely
to
return
to
their
countries
of origin,
shortening
the
potential
unemployment
benefit
burden
in
the
host
country.

Figure
3:
Foreign­born
as
a
share
of
working­age
population
by
years
of
residence,
2007.

Traditional
destination
countries
are
currently
the
ones
exhibiting
a
higher
share
of
older foreign
 workers
 (above
 45)
 in
 the
 total
 foreign
 active
 population
 (Table
 3).
 On
 the opposite
 side,
 countries
 that
 are
 experiencing
 the
 migration
 phenomenon
 in
 the
 recent years
 (Mediterranean
 countries,
 among
 others)
 show
 a
 lower
 proportion
 of
 older
 active foreigners.
This
is
an
important
fact
when
considering
that
old
workers
are
more
prone
to suffer
from
unemployment.
Assuming
that
these
workers
have
been
employed
for
a
long time,
an
unemployment
shock
affecting
this
group
would
cause
a
large
impact
on
public finances
since
these
workers
have
access
to
higher
unemployment
benefits.

Table
3:
Proportion
of
foreign
active
population
older
than
45
in
total
foreign
actives,
2007 (Source:
authors´
calculations
from
Eurostat
data).

An
 important
 issue
 worth
 noticing
 is
 that
 the
 immigration
 inflow
 up
 to
 2007
 took
 place while
 the
 EU
 economy
 was
 in
 a
 cyclical
 upturn,
 with
 growing
 working‐age
 population. However,
with
the
arrival
of
the
crisis,
migration
trends
remain
uncertain.

2.3.Differences
in
the
unemployment
benefits
systems.

The
impact
of
immigration
unemployment
on
public
finances
of
EU
members
depends
on: (i)
 the
 sensitiveness
 of
 the
 economy
 to
 the
 business
 cycle,
 (ii)
 
 the
 share
 that
 migrants represent
over
 total
labour
 force,
and
(iii)

 the
structure
of
 their
unemployment
benefits programs.

Table
 4
 shows
 the
 contribution
 sources
 to
 unemployment‐benefit
 programs
 and
 the contribution
 weighted
 by
 the
 share
 of
 foreigners
 in
 the
 active
 population.
 Columns
 1‐3 contain
the
contributions
made
by
workers
and
employers
to
the
unemployment
fund.
In column
5,
we
compute
a
proxy
of
the
contribution
made
by
immigrant
workers
per
each €10,000
 of
 revenue
 collected,
 through
 weighting
 the
 total
 contribution
 revenues
 by
 the share
of
foreign
working‐age
population
(e.g.
for
each
€10,000
collected
in
Austria,
€64.6 come
 from
 immigrants,
 according
 to
 their
 share
 in
 the
 active
 population).
 Since immigrants’
 contributed
 earnings
 are,
 on
 average,
 lower
 than
 nationals’,
 this
 measure
 is an
 upper
 bound
 of
 the
 Social
 Security
 total
 revenues
 dependence
 from
 immigrants´ contributions.

The
countries
where
immigrants´
imputed
contributions
are
higher
are
Cyprus,
Spain
and Ireland.
By
contrast,
Eastern
European
countries,
such
as
Romania,
Slovenia
or
Bulgaria, exhibit
 much
 lower
 imputed
 contributions,
 mainly
 due
 to
 their
 low
 share
 of
 foreign workforce.

Table
4
Contributions
to
the
UB
system,
and
share
of
foreign
employment
in
EU
countries (Source:
 authors´
 calculations
 from
 Eurostat
 and
 International
 Social
 Security
 Association data).

The
 degree
 of
 generosity
 of
 unemployment
 benefits
 schemes
 can
 be
 analyzed
 from different
 perspectives.
 Some
 authors
 (see
 Burda,
 1993
 and
 OECD,
 1994)
 describe generosity
of
unemployment
benefits
systems
by
means
of
cross‐country
comparisons
of the
benefits
entitlements
that
a
hypothetical
worker
would
claim
if
becoming
unemployed. An
 alternative
 way
 is
 to
 consider
 the
 degree
 of
 coverage
 of
 those
 workers
 registered
 as unemployed.

Both
methods
are
unsatisfactory
since
they
overlook
key
factors
such
as
tax treatments
 of
 unemployment
 benefits,
 unemployment
 benefit
 duration
 and
 eligibility conditions
for
claiming
unemployment
benefits.

To
overcome
these
shortcomings,
we
rely
on
Allard´s
(2005)
index
of
generosity
that
takes into
 account
 all
 the
 above‐mentioned
 features
 for
 21
 OECD
 countries.
 According
 to
 this index,
 Denmark,
 Germany,
 Finland,
 France
 and
 Ireland
 are
 found
 to
 have
 the
 more generous
systems.
In
view
of
the
existing
differences
in
unemployment
benefits
schemes and
immigration
rates,
one
can
argue
that
Austria,
Germany
and
Spain
seem
to
be
the
EU countries
 whose
 unemployment
 benefit
 burden
 could
 be
 more
 strongly
 affected
 (in absolute
 terms)
 by
 immigrants´
 unemployment.
 This
 is
 so
 since
 they
 have
 high immigration
 rates
 and
 their
 benefit
 schemes
 are
 classified
 among
 the
 most
 generous (Germany
is
in
the
first
position
in
Allard´s
ranking
of
EU
countries).

Table
5:
Unemployment
Benefit
Indicator
Index
(Source:
Allard,
2005)

Note: high score in the indicator suggests a high degree of generosity in the country.

Yet,
due
to
the
complexity
of
the
different
schemes,
indexes
like
Allard
´s,
which
dates
back to
2003,
omit
recent
important
reforms
affecting
the
total
unemployment
benefit
burden. Therefore,
 an
 updating
 is
 required
 whenever
 country
 specific
 legislation
 changes modifying
 eligibility
 conditions
 have
 taken
 place
 since
 2003.
 For
 this
 reason,
 Table
 6 summarizes
 the
design
and
main
characteristics
of
 the
unemployment
benefit
programs, as
of
to
date,
based
on
the
information
provided
by
Social
Security
Administration
Office
of Retirement
and
Disability
Policy.

To
 sum
 up,
 Table
 7
 shows
 cross‐country
 comparisons
 of
 the
 relevant
 characteristics
 to explain
 the
 unemployment
 benefit
 expenditure
 attributable
 to
 immigration.
 Considering this
 information
 we
 can
 classify
 European
 countries
 attending
 to
 their
 exposure
 to
 a potential
shock
in
immigrant
unemployment.

Table
6:
Social
Security
Programs
in
EU
countries
(Source:
Social
Security
Administration).

Table
 6
 (cont.):
 Social
 Security
 Programs
 in
 EU
 countries
 (Source:
 Social
 Security Administration).

Table
7:
Cross­country
comparisons
of
immigration
and
generosity
of
UB
systems
(Source: Authors´
calculation
from
Eurostat
and
Allard,
2005).

Legend: + above EU-27 average, - below average. Missing values in columns 2, 3 and 4 are not available from the original source. *Measured as the percentage of foreign-born immigrant with more than 7 years of residence over the total foreign-boirn population. **European mean only considers countries reported in Allard (2005)

It
 seems
 that
 Eastern
 Europe
 countries
 (Czech
 Republic
 and
 Bulgaria)
 are
 the
 least jeopardized
 to
 experience
 a
 dramatic
 increase
 in
 the
 public
 burden
 as
 a
 consequence
 of the
 current
 economic
 downturn.
 However,
 other
 Eastern
 Europe
 countries
 in
 a
 similar situation
(Estonia)
are
more
vulnerable
to
an
unemployment
shock
(due
its
higher
share of
immigrants).

Economies
 with
 a
 high
 share
 of
 foreign
 population
 and
 with
 a
 labour
 market
 more sensitive
to
the
business
cycle
face
a
riskier
situation.
This
is
the
case
of
Spain,
Greece
and Belgium,
though
in
the
first
two
immigration
has
been
a
relatively
recent
phenomenon,
a fact
 that
 moderate
 the
 negative
 outlook
 (as
 a
 consequence
 of
 less
 benefit‐entitled immigrants).
Countries
with
a
more
generous
UB
benefits
schemes
(e.g.,
Germany,
France) are
also
highly
exposed.

3. ECONOMETRIC
ESTIMATION

3.1.Data

The
 data
 is
 obtained
 from
 the
 EU‐SILC
 (Community
 Statistics
 on
 Income
 and
 Living Conditions)
elaborated
by
Eurostat.
This
survey
analyzes
the
welfare
and
living
conditions of
private
household
across
25
EU‐members
(plus
Norway
and
Iceland)
during
2004‐06. The
choice
of
this
data
source
is
driven
by
the
fact
that
EU‐SILC
pays
attention
to
both
the economic
and
social
roots
of
unemployment,
enriching
therefore
the
analysis.
However,
a drawback
 is
 that,
 due
 to
 the
 sampling
 design,
 the
 survey
 underestimates
 the
 number
 of immigrants
 receiving
 unemployment
 benefits.
 To
 overcome
 this
 drawback,
 a
 factor correction
 has
 been
 used
 to
 re‐scale
 the
 predictions.
 Relevant
 socioeconomic
 variables concerning
household
characteristics
and
personal
situation
have
been
selected.

3.2.Modelling
strategy

Unemployment
 benefit
 coverage
 can
 vary
 over
 time
 and
 across
 countries
 due
 to:
 (i) changes
in
the
composition
of
workers
that
claim
unemployment
benefits,
(ii)
changes
in the
rules
determining
eligibility
or,
(iii)
by
a
combination
of
both.
If
we
have
had
sufficient detailed
data
to
simulate
benefit
entitlement,
the
ideal
approach
would
have
been
to
build an
unemployment‐benefit
prediction
model
 that
would
 take
into
account
 the
differences in
 unemployment
 benefits
 rules
 that
 determine
 eligibility.
 However,
 EU‐SILC
 does
 not contain
 sufficient
 detailed
 information
 to
 carry
 out
 this
 simulation.
 Hence
 we
 adopt
 a different
 approach,
 attempting
 to
 decompose
 changes
 in
 coverage
 using
 standard multivariate
regression
analysis.

In
 particular,
 to
 estimate
 the
 evolution
 of
 the
 unemployment
 benefit
 burden
 among
 the different
EU
states,
we
proceed
in
three
steps.

First,
we
compute
the
probability
of
receiving
unemployment
by
means
of
a
probit
model for
each
country.
Controls
are
 the
demographic
and
socio‐economic
 factors
(gender,
age, marital
 status,
 education,
 citizenship
 and
 an
 index
 of
 the
 degree
 of
 household
 economic difficulties
in
making
ends
meet),
in
addition
to
labour‐market
characteristics
(experience, economic
 sector,
 type
 of
 contract,
 etc.)
 of
 those
 workers
 receiving
 benefits
 during
 the period
 under
 analysis,
 plus
 aggregate
 time
 variables
 (national
 GDP
 annual
 growth)
 to capture
 cyclical
 effects.
 Individuals´
 working
 histories
 are
 not
 available
 in
 the
 data
 but they
could
be
considered
to
be
a
function
of
all
the
above‐mentioned
controls.

To
 compute
 the
 probability
 of
 receiving
 unemployment
 benefits
 in
 2009‐
 2010,
 we
 use predicted
probabilities
obtained
from
the
previous
estimations,
evaluated
at
the
means
of the
observed
individual
controls
in
2004‐2006
and
the
EUROSTAT
GDP
growth
 forecasts in
each
country7.

Secondly,
 we
 need
 to
 compute
 the
 unemployment
 benefits
 that
 each
 individual
 would claim
 in
 case
 of
 becoming
 unemployed.
 
 Given
 data
 shortages
 on
 individuals’
 complete working
histories,
we
are
not
able
to
compute
individuals’
unemployment
benefit
claims according
 the
current
 rules
in
 the
country
of
 residence.
To
overcome
 this
hindrance,
we use
employed
and
unemployed
workers´
characteristics
in
the
sample
in
order
to
impute potential
 unemployment
 benefit
 for
 those
 who
 will
 lose
 their
 jobs
 in
 the
 future.
 The matched
characteristics
are
gender,
cohort
of
birth,
civil
status,
migrant
status,
education, years
since
first
job,
country
of
residence,
household
economic
difficulties
in
making
ends meet
 and
 other
 characteristics
 related
 to
 the
 location
 of
 the
 family’s
 house
 (noise
 and crime
in
the
area,
and
degree
of
urbanization
measured
by
population
density).

Thirdly,
 we
 pose
 alternative
 scenarios
 about
 the
 characteristics
 of
 the
 immigrants becoming
 unemployed.
 Given
 that
 complete
 working
 histories
 are
 not
 available
 in
 the data,
we
cannot
identify
precisely
which
of
them
are
eligible.
Hence,
different
alternatives about
the
proportion
of
eligible
prospective
unemployed
migrants
are
considered.
Then, through
randomization,
migrants
affected
by
the
crisis
are
chosen
among
the
total
stock
of migrants
in
each
country
according
to
each
scenario.

Formally,
the
total
unemployment
benefits
burden
(TUB)
in
the
country
i
during
the
year
t is
computed
as

\[T U B = \sum_ {x = 1} ^ {n} p _ {x _ {i t}} U _ {x _ {i t}}\]

that
 is,
 TUB
 is
 the
 sum
 of
 the
 expected
 unemployment
 benefits
 of
 all
 migrants
 likely
 to become
unemployed
in
country
i
during
the
year
t.

3.3. Results

Five
alternative
scenarios
are
considered.
In
each
of
them
the
negative
effect
of
the
crisis
is increased
 by
 raising
 the
 number
 of
 workers
 susceptible
 of
 becoming
 unemployed.

7 It
relies
on
the
assumption
that
workers’
characteristics
remain
unchanged
at
their
average
values
in
2004‐ 2006
during
the
outlook
horizon
period.

Scenario
 1
 considers
 an
 unemployment
 shock
 that
 affects
 the
 33%
 of
 the
 immigrant population
younger
than
30
years,
who
represent
the
most
likely
group
to
suffer
from
the termination
of
temporary
job
contracts
under
weak
labour‐market
conditions.
Scenario
2 accounts
 for
 a
 harder
 impact
 on
 immigrant
 unemployment,
 where
 66%
 of
 the
 young immigrant
 population
 loses
 their
 jobs.
 In
 Scenario
 3,
 not
 only
 66%
 of
 young
 immigrant workers
 get
 unemployed
 but
 also
 66%
 of
 those
 aged
 above
 45.
 Scenario
 4
 entails
 an increase
 up
 to
 100%
 in
 the
 unemployment
 rate
 for
 unskilled
 immigrant
 workers
 in Scenario
 3.
 Finally,
 in
 Scenario
 5,
 we
 set
 an
 upper
 bound
 to
 the
 estimated
 effects
 by considering
an
unemployment
shock
that
affects
all
the
immigrant
workers.

As
 Table
 8
 shows,
 immigrants’
 unemployment
 impact
 on
 TUB
 is
 rather
 limited
 under Scenario
 1.
 Due
 to
 heterogeneity
 in
 immigrants’
 characteristics,
 these
 figures
 vary considerable
 between
 countries
 with
 Luxembourg
 (where
 the
 foreign‐nationals
 over population
reach
39%
in
2006)
suffering
the
largest
impact
by
far,
i.e.,
an
increase
of
22% in
 TUB
 during
 2006‐2008.8
 Allowing
 for
 a
 larger
 tougher
 job‐destruction
 shock
 (66%
 of young
immigrant
population)
in
Scenario
2
implies
that
TUB
could
double
in
most
of
the countries.

Widening
the
migrant
population
affected
by
the
shock
to
the
66%
of
migrants
older
than 45
under
Scenario
3
gives
rise
to
wider
differences
across
countries.
In
this
way,
countries such
as
Latvia,
Estonia
or
France
record
a
large
increase
(about
200%)
in
TUB
due
to
their high
share
of
workers
above
45
years
of
age
in
foreign
employment
(recall
Table
3),
while the
effect
is
milder
in
countries
with
less
older
foreign
workers,
like
Spain
(8%).

Scenario
4
extends
the
shock
to
low
skilled
immigrants
(without
a
university
degree).
The effect
under
this
scenario
is
more
homogeneous,
though
United
Kingdom
and
Italy
seem
to be
the
most
sensitive
to
the
shock.

As
mentioned
above,
Scenario
5
is
not
meant
to
be
realistic
but
it
is
just
used
to
set
up
an upper
 bound
 on
 the
 sensitivity
 of
 a
 country’
 s
 TUB
 to
 immigrant
 unemployment.
 EU members
such
as
Luxembourg,
Latvia,
Estonia
and
France
are
the
most
prone
to
suffer
a larger
burden.
Other
economies
as
Austria,
Finland,
Germany,
Ireland
Spain,
or
Italy
would register
an
increase
of
around
20‐25%
in
TUB
under
this
scenario.

TUB
 is
 expected
 to
 peak
in
 2009,
after
 increasing
in
 2008,
 and
 to
slow
 down
slightly
in 2010.
 Results
 should
 to
 be
 read
 with
 caution
 because
 are
 mainly
 driven
 by
 current prospects
about
GDP
growth
dynamics
during
the
following
two
years,
which
have
turned to
be
exceptionally
uncertain.

8 Of
 course,
 given
 that
 a
 large
 share
 of
 the
 foreign
 population
 in
 Luxembourg
 are
 permanent
 employees
 of international
institutions
located
there,
the
results
for
this
country
are
not
likely
to
be
meaningful.

Table
8:
Estimated
increase
in
unemployment
benefits
burden
from
2006
(Source:
authors calculations).

4. CONCLUSIONS

The
current,
severe
economic
downturn
is
having
an
uneven
impact
across
EU
economies, in
particular
with
regard
to
their
GDP
and
employment
growth
responses.
Unemployment benefit
burden
is
expected
 to
peak
in
2009
after
an
increase
in
2008,
and
 to
slow
down slightly
in
2010.
The
study
identifies
three
main
factors
revealing,
with
a
certain
degree
of accuracy,
whether
a
given
EU
country
is
set
 to
 face
a
low
or
a
high
 financial
burden
as
a result
of
the
increase
in
migrant
unemployment:
(1)
the
sensitivity
of
the
economy
to
the business
 cycle;
 (2)
 the
 share
 of
 migrants
 in
 the
 country's
 total
 labour
 force;
 and
 (3)
 the nature
 and
 structure
 of
 unemployment
 benefits.
 Moreover,
 the
 age
 structure
 of
 working migrants
 and
 their
 employment
 history
 can
 have
 an
 impact
 both
 on
 the
 level
 of unemployment
benefits
to
be
paid
by
the
host
country
and
on
the
propensity
of
migrants to
return
to
their
respective
countries
of
origin.

It
can
also
be
added
that
the
welfare
system
of
those
countries
that
have
a
high
percentage of
 foreign
 labour
 and
 that
 depend
 heavily
 on
 social
 security
 contributions
 to
 fund unemployment
 benefit
 programmes
 are
 the
 ones
 that
 will
 be
 more
 vulnerable
 to
 sharp increases
 in
 unemployment.
 On
 the
 one
 hand,
 countries
 where
 payroll
 taxes
 finance unemployment
benefit
systems
will
be
more
exposed
to
a
reduction
in
tax
collection
as
a result
of
a
drop
in
the
number
of
affiliations
to
Social
Security
programmes.
On
the
other hand,
a
high
immigration
rate
would
place
a
serious
strain
on
the
public
finances
of
host countries,
 as
 migrant
 workers
 are
 usually
 more
 likely
 than
 native
 workers
 to
 lose
 their jobs
 during
 economic
 downturns.
 It
 follows
 that
 countries
 that
 have
 high
 immigration rates
and
generous
unemployment
benefit
schemes
will
be
the
ones
that
are
hit
the
most by
a
sharp
rise
in
migrant
unemployment,

In
 conclusion,
 it
 can
 be
 said
 that
 EU
 Member
 States
 that
 have
 had
 a
 large
 number
 of migrants
working
on
 their
 territory
 for
a
long
 time,
who
are
 thus
likely
 to
be
entitled
 to claim
 unemployment
 benefits,
 are
 set
 to
 face
 a
 gradually
 growing
 financial
 burden
 as
 a result
 of
 the
 increase
 in
 migrant
 unemployment.
 For
 EU
 countries
 where
 migration
 is
 a relatively
 new
 phenomenon,
 the
 financial
 burden
 will
 tend
 to
 be
 lower
 as
 a
 limited number
 of
 migrant
 workers
 will
 be
 eligible
 for
 unemployment
 benefits.
 This
 holds
 true unless
Europe's
economic
recession
deepens.
Our
projection
is
that
the
impact
of
the
rise in
 migrant
 unemployment
 on
 the
 unemployment
 benefit
 burden
 is
 likely
 to
 be
 sizeable over
the
next
few
years.
If
66%
of
young
migrant
workers
become
unemployed,
along
with 66%
 of
 migrants
 over
 the
 age
 of
 45,
 which
 is
 probable
 the
 more
 realistic
 scenario,
 the estimated
 average
 rise
 would
 amount
 to
 37%
 in
 2008
 and
 40%
 in
 2009.
 We
 have
 also prepared
a
more
pessimistic
scenario,
which
takes
account
of
a
more
profound
economic slump
 and
 more
 layoffs
 of
 migrant
 workers.
 Under
 this
 projection,
 the
 unemployment benefit
burden
will
almost
double,
rising,
on
average,
to
70%
in
2008
and
80%
in
2009.

5. REFERENCES

1. Allard,
G,
(2005)
 “Measuring
 the
changing
generosity
of
unemployment
benefits:
beyond existing
indicators”,
WP05‐18
Instituto
de
Empresa .
2. Burda,
 M,
 (1993)
 “Unemployment,
 Labor
 markets
 and
 Structural
 Change
 in
 Eastern Europe”,
Economic
Policy,
vol
16,
pp101‐137.

European
Commission
(2008)
“Employment
in
Europe
2008”.
Directorate‐General
for Employment,
Social
Affairs
and
Equal
Opportunities. http://ec.europa.eu/employment_social/publications/booklets/employment/pdf/keah08001_en.pdf

European
Commission
(2008)
“Economic
Forecast:
Autumn
2008”.
Directorate‐General for
Economic
and
Financial
Affairs.

http://ec.europa.eu/economy_finance/publications/publication13290_en.pdf

International
Social
Security
Association
(2008)
“Social
Security
Programs
Throughout
the World:
Europe,
2008”.
Social
Security
Administration:
Office
of
Retirement
and
Disability Policy
and
Office
of
Research,
Evaluation
and
Statistics. http://www.ssa.gov/policy/docs/progdesc/ssptw/2008-2009/europe/ssptw08euro.pdf

OECD
(1994)
“The
OECD
Job
Study,
1994”.
http://www.oecd.org/dataoecd/42/51/1941679.pdf

1. 2009-15: “Impact of the Rise in immigrant unemployment on public finances”, Pablo Vazquez, Mario Alloza, Raquel Vegas y Stefano Bertozzi.
2. 2009-14: “Responding to Financial Pressures. The Effect of Managed Care on Hospitals´ Provision of Charity Care”, Núria Mas.
3. 2009-13: “Domestic Transport Cost Reductions and Firms’ Export Behaviour”, Pedro Albarran, Raquel Carrasco y Adelheid Holl.
4. 2009-12: “Compatibility with Firm Dominance”, María Fernanda Viecens.
5. 2009-11: “Pricing Strategies in Two-Sided Platforms: The Role of Sellers’ Competition”, María Fernanda Viecens.
6. 2009-10: “Scheduled Service Versus Personal Transportation: the Role of Distance”, Volodymyr Bilotkach, Xavier Fageda y Ricardo Flores-Fillol.
7. 2009-09: “Social Preferences and Strategic Uncertainty: An Experiment on Markets and Contracts”, Antonio Cabrales, Rafaele Miniaci, Marco Piovesan y Giovanni Ponti.
8. 2009-08: “Hidden Information, Bargaining Power and Efficiency: An Experiment”, Antonio Cabrales, Gary Charness y Marie Claire Villeval.
9. 2009-07: “Democracy and the curse of natural resources”, Antonio Cabrales y Esther Hauk.
10. 2009-06: “Social Interactions and Spillovers: Incentives,Segregation and Topology”, Antonio Cabrales, Antoni Calvó-Armengol e Yves Zenou.
11. 2009-05: “Chance Constrained Programming with one Discrete Random Variable in Each Constraint”,Emilio Cerdá Tena y Julio Moreno Lorente.
12. 2009-04: “Economic Value of Weather Forecasting Systems Information: A Risk Aversion Approach”, Emilio Cerdá Tena y Sonia Quiroga Gómez.
13. 2009-03: “Population Ageing, Inequality and the Political Economy of Public Education”, Francisco Martínez-Mora.
14. 2009-02: “Real Wages over the Business Cycle: OECD Evidence from the Time and Frequency Domains”, Julian Messina, Chiara Strozzi y Jarkko Turunen.
15. 2009-01: “The Determinants Of Misreporting Weight And Height: The Role Of Social Norms”, Joan Gil y Toni Mora.
16. 2008-42: “Social Security Incentives, Exit from the Workforce and Entry of the Young”, Michele Boldrin, Pilar García-Gómez y Sergi Jiménez-Martín.
17. 2008-41: “The Evolution and Main Determinants of Productivity in Brazilian Electricity Distribution 1998- 2005: an Empirical Analysis”, Francisco Javier Ramos-Real, Beatriz Tovar, Mariana Iootty, Edmar Fagundes de Almeida y Helder Queiroz Pinto Jr..
18. 2008-40: “Immigration and Housing Prices in Spain”, Simón Sosvilla.
19. 2008-39: “Modeling the Immigration Shock”, Ana Montes y Michele Boldrin.
20. 2008-38: “Immigration and the Demand for Health in Spain”, Sergi Jiménez, Natalia Jorgensen y José María Labeaga.
21. 2008-37: “Immigration and Students' Achievement in Spain”, Natalia Zinovyeva, Florentino Felgueroso y Pablo Vázquez.
22. 2008-36: “Immigration and Social Security in Spain”, Clara Isabel González, J. Ignacio Conde-Ruiz y Michele Boldrin.
23. 2008-35: “Complements or Substitutes? Immigrant and Native Task Specialization in Spain”, Catalina Amuedo-Dorantes y Sara de la Rica.
24. 2008-34: “Immigration and Crime in Spain, 1999-2006”, Cesar Alonso, Nuno Garoupa, Marcelo Perera y Pablo Vázquez.
25. 2008-33: “A Social Network Approach to Spanish Immigration: An Analysis of Immigration into Spain 1998- 2006”, Rickard Sandell.
26. 2008-32: “The Consequences on Job Satisfaction of Job-Worker Educational and Skill Mismatches in the Spanish Labour Market: a Panel Analysis”, Lourdes Badillo Amador, Ángel López Nicolás y Luis E. Vila.
27. 2008-31: “Students’assessment of higher education in Spain”, César Alonso-Borrego, Antonio Romero-Medina.
28. 2008-30: “Body image and food disorders: Evidence from a sample of European women”, Joan Costa-Font y Mireia Jofre-Bonet.
29. 2008-29: “Aggregation and Dissemination of Information in Experimental Asset Markets in the Presence of a Manipulator”, HelenaVeiga y Marc Vorsatz.
30. 2008-28: “The Measurement of Consensus: An Axiomatic Analysis”, Jorge Alcalde-Unzu y Marc Vorsatz.