How to assess European leave policies regarding their compliance

Report
How to assess European leave
policies regarding their compliance
with an ideal leave model
By Helene Dearing
Vienna University of Economics and Business (WU)
Institute for Social Policy
Presented at the 11th seminar of the International Network on Leave Policies and Research
18 September 2014
Definitions
Parental Leave
= maternity, paternity, parental and child care leave = “leave”
well-paid leave ≠ total leave (paid and unpaid)
Gender Equality in the Division of Labour
= men and women contribute equally to employment and
family work, on average
(Fraser (1997), Crompton (1999), Gornick and Meyers (2003))
Previous Literature
Policy Typologies:
Haas (2003) & Wall (2007)
Fuzzy-Set Analysis:
Haas and Rostgaard (2011) & Ciccia and Verloo (2012)
Composite Indicators:
Gornick and Meyers (2003), Ray et al. (2010), Javornik
(2014)
Motivation
Comparison with an ideal model because of…
…trend towards
standardization
…inverted
U-shaped
effects
Methodological Approach
Stage 1:
• Defines ideal leave model
• Develops EGDL indicator that measures degree of
compliance (EGDL = „Equal Gender Division of
Labour“)
Stage 2:
• Sensitivity analysis with regard to assumptions
that cannot be based on empirical evidence
Empirical Literature I
…with regard to the effects of leave on
mothers’ employment:
1. The duration of total leave positively affects employment
participation and working hours (inverted U-shaped effect)
(Akgunduz and Plantenga (2012), Genre et al. (2010), Pettit and Hook (2005), Misra et al. (2011))
2. The duration of total leave negatively affects wages,
occupational segregation and job-related training
(Akgunduz and Plantenga (2012), Lalive and Zweimüller (2009), Puhani and Sonderhof (2011))
3. The duration of paid leave negatively affects return to work
(Grunow et al. (2011), Ronsen and Sunström (2002), Ondrich et al. (2003), Lalive and Zweimüller (2009)
Bergemann and Riphahn (2011)Pronzato (2009), Lapuerta et al. (2011))
Empirical Literature II
…with regard to the effects of leave on
fathers’ family work:
1. Introduction of a fathers’ quota positively affects their
parental leave take-up
(Duvander and Johansson (2012), Ekberg et al. (2013))
2. Provision of payments positively affects fathers’ parental
leave take-up
(Pull and Vogt (2010), Lapuerta et al. (2011))
Data
• Annual report of the Leave Network (Moss 2013)
• Focus on 27 European countries
• Variables of interest:
– Duration of total leave (paid and unpaid)
– Duration of well-paid leave
(paid at above 66% of average income)
– Share of well-paid leave reserved for fathers
Index on duration of total leave I
Index capturing the effect of total leave duration on the gender division of labour
1.0
0.9
0.8
0.7
"Moderate" duration of leave
(= 14 months)
Minimal
duration of
leave
(= 0 months)
0.6
0.5
0.4
0.3
Maximal duration of
leave (= 49 months)
0.2
0.1
0.0
Total leave
duration in
months
Index on duration of total leave II
Source: Author’s analysis based on data provided by Moss (2013)
Index
values Luxembourg
1.0
Slovenia
0.9
Index based on the duration of total leave
Netherlands
Italy Iceland
Ireland
Sweden
Greece
UK
0.8
Denmark
Austria
0.7
0.6
0.5
0.4
0.3
Switzerland
Belgium France PortugalSlovakia
Hungary Estonia
Norway
Lithuania
Czech Rep.Germany Russia Spain
Finland
Croatia
0.2
0.1
0.0
Poland
European
Countries
Index on duration of well-paid leave
Source: Author’s analysis based on data provided by Moss (2013)
Index
values
Index based on the duration of well-paid leave
1.0
Germany
Sweden
Norway
0.9
Slovenia
0.8
Poland
Denmark
Finland
0.7
Iceland
0.6
Estonia
0.5
Croatia
0.4
Spain
0.3
0.1
0.0
Greece
Portugal
Italy
France
Switzerland
0.2
Ireland
Belgium
Luxembourg
Austria
Russia
Netherlands
Czech Rep. Hungary
UK
Slovakia
Lithuania
European
countries
Index on fathers’ leave
Source: Author’s analysis based on data provided by Moss (2013)
Index based on the share of well-paid leave reserved for fathers
Index
values
1.0
0.9
0.8
0.7
Croatia
Iceland
0.6
Portugal
0.5
Finland
0.4
0.3
Germany
Norway
Sweden
France
Belgium
0.2
0.1
0.0
Spain
Poland Lithuania Denmark
Croatia Ireland Luxembourg Switzerland Czech Rep.
Estonia Slovenia
Austria Greece Italy Russia Netherlands Slovakia Hungary
European
countries
Baseline EGDL indicator
Baseline EGDL indicator =
(index of total leave
+ index of well-paid leave
+ index of fathers’ leave)
/3
EGDL indicator values range between 0 and 1 for each country
….indicating the compliance with the ideal leave model
Baseline EGDL indicator
1.00
Index accounting for share reserved for the father
0.80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0.60
0.40
0.20
0.00
very low
low
medium
high
Baseline EGDL indicator
1.00
Index accounting for share reserved for the father
0.80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0.60
0.40
0.20
0.00
very low
low
medium
high
Baseline EGDL indicator
1,00
Index accounting for share reserved for the father
0,80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0,60
0,40
0,20
0,00
very low
low
medium
high
Baseline EGDL indicator
1.00
Index accounting for share reserved for the father
0.80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0.60
0.40
0.20
0.00
very low
low
medium
high
Baseline EGDL indicator
1.00
Index accounting for share reserved for the father
0.80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0.60
0.40
0.20
0.00
very low
low
medium
high
Baseline EGDL indicator
1.00
Index accounting for share reserved for the father
0.80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0.60
0.40
0.20
0.00
very low
low
medium
high
Methodological Approach
Stage 1:
• Define ideal leave model: 14 months of well-paid
leave, half reserved for fathers
• Develop EGDL indicator that measures degree of
compliance (EGDL = „Equal Gender Division of
Labour“)
Stage 2:
• Sensitivity analysis with regard to different
assumptions
Sensitivity Analysis I
….regarding the assumption of an ideal duration of
14 months of leave
Assuming 12 months as ideal:
• Changes little in group composition
• Denmark moves up to high EGDL scoring
countries
Assuming 16 months as ideal:
• Changes little in group composition
Sensitivity Analysis II
Mother-centred EGDL indicator =
( 2 (index of total leave * index of well-paid leave)
+ index of fathers’ leave)
/3
…assumes that it is most important
to combine total leave with payments
Sensitivity Analysis II
Index capturing the effects of a combination of total and well-paid leave
1
0.9
0.8
Values of
combined
index
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
DUR = 5
DUR =…
DUR =…
DUR =…
DUR =…
Duration of
total leave
DUR =…
DUR =…
DUR =…
DUR =…
Duration of well-paid leave
The mother-centred EGDL indicator
1.00
Index accounting for the share reserved for the father
0.80
Index accounting for the combination of the duration
of total and well-paid leave
0.60
0.40
0.20
0.00
very low
low
medium high
The mother-centred EGDL indicator
1.00
Index accounting for the share reserved for the father
0.80
Index accounting for the combination of the duration
of total and well-paid leave
0.60
0.40
0.20
0.00
very low
low
medium high
The mother-centred EGDL indicator
1.00
Mother-centred EGDL indicator, broken down into single indices
Index accounting for the share reserved for the father
0.80
Index accounting for the combination of the duration
of total and well-paid leave
0.60
0.40
0.20
0.00
very low
low
medium high
The baseline EGDL indicator
Baseline EGDL indicator, broken down into single indices
1.00
Index accounting for share reserved for the father
0.80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0.60
0.40
0.20
0.00
very low
low
medium
high
Sensitivity Analysis III
Father-centred EGDL indicator =
(index of total leave + index of well-paid leave)
+ 2 (index of fathers’ leave))
/4
…assumes that it is most important
to provide father’s quotas
Father-centred EGDL indicator
1.00
Index accounting for share reserved for the father
0.80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0.60
0.40
0.20
0.00
very low
low
medium
high
Father-centred EGDL indicator
1.00
Index accounting for share reserved for the father
0.80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0.60
0.40
0.20
0.00
very low
low
medium
high
Father-centred EGDL indicator
1.00
Index accounting for share reserved for the father
0.80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0.60
0.40
0.20
0.00
very low
low
medium
high
Father-centred EGDL indicator
1.00
Index accounting for share reserved for the father
0.80
Index accounting for duration of well-paid leave
Index accounting for duration of total leave
0.60
0.40
0.20
0.00
very low
low
medium
high
Robustness of EGDL results
1.00
baseline EGDL scores
0.80
"opt = 12" EGDL scores
"opt = 16" EGDL scores
mother-centred EGDL scores
0.60
0.40
0.20
0.00
father-centred EGDL scores
Robustness of EGDL results
1.00
baseline EGDL scores
0.80
"opt = 12" EGDL scores
"opt = 16" EGDL scores
mother-centred EGDL scores
0.60
0.40
0.20
0.00
father-centred EGDL scores
Robustness of EGDL results
1.00
baseline EGDL scores
0.80
"opt = 12" EGDL scores
"opt = 16" EGDL scores
mother-centred EGDL scores
0.60
0.40
0.20
0.00
father-centred EGDL scores
Robustness of EGDL results
1.00
baseline EGDL scores
0.80
"opt = 12" EGDL scores
"opt = 16" EGDL scores
mother-centred EGDL scores
0.60
0.40
0.20
0.00
father-centred EGDL scores
Robustness of EGDL results
1.00
baseline EGDL scores
0.80
"opt = 12" EGDL scores
"opt = 16" EGDL scores
mother-centred EGDL scores
0.60
0.40
0.20
0.00
father-centred EGDL scores
Robustness of EGDL results
1.00
baseline EGDL scores
0.80
"opt = 12" EGDL scores
"opt = 16" EGDL scores
mother-centred EGDL scores
0.60
0.40
0.20
0.00
father-centred EGDL scores
Conclusions
• How to assess leave policies regarding their
compliance with an ideal leave model?
• Possible as an assessment in two stages:
– Stage one: defines ideal model and develops EGDL
indicator
– Stage two: provides sensitivity analysis
• Allows us to study the robustness of EGDL
results

similar documents