Dualism in Spain: El Contrato nico El Contrato nico Samuel - - PowerPoint PPT Presentation

Dualism in Spain: El Contrato Único El Contrato Único

Samuel Bentolila

CEMFI

Workshop on “Beyond Dual Labor Markets. The Time of Legislation”

fondazione RODOLFO DEBENEDETTI

Università Bocconi, Milan, March 17th 2010

Roadmap Roadmap

• 1. Spain’s idiosyncratic response to the Great

Recession 2 Labour market problems and potential

• 2. Labour market problems and potential

solutions

• 3. The single contract proposal

• 1. Spain is different

Similar GDP bust across developed countries, but largest unemployment rise in Spain g p y p

GDP (growth rate, %) 2007 2008 2009 2010 (f)

(g )

( )

Spain 3.7 1.2

• 3.6 -0.3

Euro area 2.6 1.2

• 3.9 1.4

US 2.0 1.1 -2.5 3.0

Spain (growth rate, %)

Employment 3.1 -0.5

• 6.8 -2.5

Labour Force 2.8 3.5 0.8 -1.2

Unemployment rate

Spain 8.3 11.3 18.0 19.3 Euro area 7.5 9.5 10.0 10.2 US 4.9 5.8 9.7 10.0

The wild ride...

Harmonized unemployment rate

20 20 15 15 10 10 5 5 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 France Euro Area Spain Netherlands

…and its composition

2007 2009 2007 2009 Unemployment rate 8.3 18.0

Natives 6 4 16 0 Natives 6.4 16.0 Immigrants 12.2 28.4 Male 6.4 17.7 Female 10.9 18.4 Youth (< 25 y.o.) 18.2 37.9 Long-term unemployed (% U) 23.7 28.3

Job destruction flows (*) 1 469 Job destruction flows (*) 1,469

Construction 854 Manufacturing 470 Manufacturing 470 Temporary 1,324

(*) Absolute changes in 2007-2009 in thousands.

• 2. Urgent need of labor market reforms

g

• In recessions with different origins [oil shocks (1975,

g ( 1979), EMS collapse (1990s), financial shock and bursting of housing bubble (2007)] the Spanish economy always reacts by massively destroying employment

• Reducing the weight of the residential construction

sector and restoring credit flows to firms and consumers will not be enough to solve a structural bl hi h l l problem: high structural unemployment rate

(vs. other bubble economies IRL: 12.8 %, UK: 7.8%, DK: 4 3%) 4.3%)

Labor market adjustment via quantities... j q

(Percentage change)

8 4 8 4 4 2007 2008 2009 8

• 4
• 8

Labor cost per worker Consumer price index Gross Domestic Product Employment p y

... overwhelmingly borne by temporary employees g y y p y p y

(Change in emploment in thousands)

800 400

• 400
• 800
• 1200

Permanent employees Temporary employees

...instead of adjusting wages (UK) or working hours (GE) j g g ( ) g ( )

8

United Kingdom

8

Germany

4

4

• 4
• 4
• 8

GDP Employment

• 8

GDP Employment

GDP Employment CPI Hourly wage

GDP Employment CPI Hourly wage

Little response of real wages (2009) p g ( )

• Average increase in wage bargains: 2.6%

Average CPI increase:

• 0 3%

Average CPI increase:

• 0.3%

Real wage increase: 2.9% Employment change:

• 6.8%
• Construction:

R l i 3 9% Real wage increase: 3.9% Employment change:

• 23.0%
• A story: Interaction of employment protection legislation

and collective bargaining institutions g g – Insiders insulated from unemployment by two-tier labor market (Bentolila-Dolado, 1994) – Resistance to labor market reforms (Saint-Paul, 1996)

Diagnosis: Spanish labor market institutions are far from working properly far from working properly ... however there is great resistance to labor market reform: Excess of vague proposals (“structural reforms” ( structural reforms , “new industry specialization”) Deficit of specific proposals

A Proposal to Restart the Spanish Labour Market Labour Market

21 April 2009 (Signed by 100 academic economists) (Signed by 100 academic economists)

crisis09 es www.crisis09.es

Proposals for labor market reform based on: 1. Rigorous theoretical analysis 2 Large international empirical evidence 2. Large international empirical evidence 3. Improving efficiency without lowering social protection Four proposals

• A. Employment protection (severance pay)

B. Passive labor market policies (unemployment benefits) p ( p y )

• C. Active labor market policies (placement, training)
• D. Collective bargaining

Today I will focus on the first proposal

• 3. The single contract proposal

Employment Protection Legislation (EPL)

(I d 0 100 181 i S D i B i 2009 W ld B k)

Indicator Spain OECD

(Index 0-100, 181 countries. Source: Doing Business 2009, World Bank)

p Hiring restrictions 78 26 k d Working time rigidity 60 42 Firing restrictions 30 26 EPL rigidity 58 31 Severance pay (weekly

wages, 10-year tenure)

56 26

Labor market effects of firing costs Labor market effects of firing costs

• Employment and unemployment (Bentolila-Bertola, 1990; Bertola,

1994 B i i D l 2006) 1994; Bassanini-Duval, 2006):

– Ambiguous effect on aggregate unemployment rate (except for detrimental effects on investment ) (except for detrimental effects on investment ) – ↑Employment rate of prime age workers and ↓ female and youth employment rates and youth employment rates

• ↓ Employment volatility and mobility of workers
• ↑ Turnover rate of temporary workers → ↑ Unemployment

↑ Turnover rate of temporary workers → ↑ Unemployment (dismissal costs gap) (Blanchard-Landier, 2002; Cahuc-Postel Vinay,

2002, Bentolila et al., 2009)

• ↓ Productivity growth (Hopenhayn-Rogerson, 1993; Mortensen-

Pissarides, 1994; Bassanini-Nunziata-Venn, 2008; Dolado-Stucchi, 2009)

Dual labor market: Leader in temporary work

40 40

Temporary employment rate (percentage of employees)

30 35 30 35 25 25 15 20 15 20 5 10 5 10 5 5 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Italy France EU (15 countries) Portugal Spain

Huge labor market turnover

• Transitions: 2008:IV (LFS): 4.45 m. workers changed labor status

ΔE (net) = UE+IE-EU-EI = 1.363-1.861 = -0.5 m. EMP. → UNEMP.: 1.026.000 (Temp.: 81%)

• UNEMP. → EMP.:

708.000 (Temp.: 87%) EMP NON AC 835 000 EMP. → NON-AC.: 835.000 NON-AC.→ EMP.: 655.000

(SP: 3.9% PWA/month approx. , vs. US: 5.4%, FR: 1.8% PWA/month)

• No. of contracts:

(2007): 18.6 m. → 0.80 contracts/employee (2008): 16.6 m. → 0.76 contracts/employee ( ) / p y (2009) : 14.1 m. → 0.76 contracts/employee

• Low conversion rate (temp.-perm.): 5.1%; < 8% since 1994

Low conversion rate (temp. perm.): 5.1%; 8% since 1994

(despite generous job subsidies: García Pérez-Rebollo, 2009)

• Very short-term temp contracts : 2 7 m < 7 days and 5 m <30 days (2008)
• Very short-term temp. contracts : 2.7 m.< 7 days and 5 m. <30 days (2008)

Other effects of temporary contracts

• Uneven incidence on the labour force (2008 average = 28%):

30% Females vs. 26% Males 47% Youth (< 29 y.o.) 52% Immigrants … but also on prime-age workers (30-50 y.o.): 25% p g ( y )

• Reduce productivity: ↓ training and effort ↑ absenteeism and over-

education + generates technology adoption based on low-skill labor g gy p

(Dolado-Stucchi, 2008, Dolado-Jansen-Jimeno, 2009; Bentolila-Dolado-Jimeno, 2009)

• Other effects (Dolado-García Serrano-Jimeno, 2002):
• ↓ Youth unemployment rate (+)
• ↓ Long-term unemployment (+)

g p y ( )

• ↑ Wage inequality (−)

Why so much turnover?

Dismissal cost gap: 45 (perm.) vs. 8/0 (temp.) days p.y.s.

Proposal: Single Contract with increasing severance pay

(Example: SC 12-36, max. 24 months)

50 35 40 45

• f service

25 30

s per year o

10 15 20

ys of wages

5 10

Day

1 2 3 4 5 6 7 8 9 10

Years of service

Pros of single contract Pros of single contract

• By default, all new contracts become open ended

( f l ) (except for temporary replacement) l f d l b

• Marginal cost of conversions: gradualism vs. big gap

(“the ramp that runs through the wall”)

• Average dismissal cost could remain the same or

increase job durations increase increase, job durations increase

• Rationale for increasing severance pay: Compensation
• Rationale for increasing severance pay: Compensation

for loss of specific human capital investment, psychic cost of dismissals (Blanchard-Tirole, 2003) ( )

Some simulation results Some simulation results

Longitudinal Sample of Working Lives (MCVL 2007) g p g ( ) (Gárcía-Pérez, 2010) S b d f d l d

• Scenario based on projections of dismissal and re-

employment rates (33-severance pay contract)

• Minimal increase in average firing cost (+0.5% in

year 1, +1.5% in year 10) y , y )

• Significant increases in job duration

(graph →)

Increase in job duration 10 years after introduction of SC (male and female) SC ( a e a d e a e)

60,00% 50,00% 40,00%

Mujeres Hombres

20 00% 30,00%

Hombres

10,00% 20,00% 0,00% 18 23 28 33 38 43 48 53 58

Using France as a benchmark Using France as a benchmark

Bentolila-Cahuc-Dolado-Le Barbanchon (2010) ( )

• Search and matching model w/ perm. and temp. contracts

calibrated before (2006-07) and during the crisis (2008-09)

• Similar labor market institutions (EPL, Unemp. benefits,

Collective bargaining) but in France unemployment rate has Collective bargaining) but in France unemployment rate has

• nly risen by 2 p.p.
• Exception: Red-tape dismissal cost gap France < Spain

(1.33 quarters v. 2 quarters) (Temp. rate around 15%) ( q q ) ( p )

• If Spain had adopted French EPL, its unemployment rate

would have risen by 4-5 percentage points less

Is the single contract politically viable? Is the single contract politically viable?

• Temporary contracts were introduced in 1984 as a two-

i (i id id ) d i f fl ibili i tier (insider-outsider) device to foster flexibility in a rigid labor market (inherited from Franco)

• Further reforms in: 1994 (restrict temp. contracts), 1997

(introduce new perm contract with lower dismissal (introduce new perm contract with lower dismissal costs, 33, but restricted), 2002 (avoid judges), 2006 (widen coverage of new contract)… 2010? ( g )

• Median (union) voter argument

(graph →) ( ) g (g p ) (close to 50% if perms. with lower firing costs support the reform)

Median voter and resistance to EPL reform Median voter and resistance to EPL reform

Indicators of political viability of labor reform (share of employees plus unemployed)

0 60 0 60 0,60 0,60 0,50 0,50 0,40 0,40 1987-II 1989-II 1991-II 1993-II 1995-II 1997-II 1999-II 2001-II 2003-II 2005-II 2007-II 2009-II Temp+Unemp Temp+Perm33+Unemp

Is the single contract legally viable? g g y

Bentolila and Jansen (2010): Concerns about viability

• Equality (2 contracts with different severance pay)

– Spanish Constitution allows it for employment promotion p p y p reasons – It already happens: 45-day and 33-day contracts

J di i l t ti

( O C 158 Ch A 30)

• Judicial protection (ILO Conv. 158, EU Charter Art. 30):

– Still protected against: violation of fundamental rights (e.g. discrimination), disciplinary dismissal (45 days) discrimination), disciplinary dismissal (45 days) – Law already allows employers to avoid going to court since 2002: disciplinary dismissal + recognition of unfairness f l l ( %) l f – Perversion of law: Disciplinary (33%), Non-renewal of temporary (54%) → Only 13% of dismissals now protected – So: Single contract would protect all workers g p

Thank you Thank you for your attention! for your attention!