Children and Coupled Joblessness in Europe: Labour Supply, - PowerPoint PPT Presentation

Children and Coupled Joblessness in Europe: Labour Supply, Fertility, and Comparative Differences Juho Härkönen European University Institute Dep. of Political and Social Sciences E-mail: Juho.Harkonen@IUE.it May 2006

Background Employment is polarized across couples: the fraction of dually employed couples is on the increase, while in many couples both partners are without work Employment polarization is more severe in some countries (especially Ireland and the UK) than in others Jobless households have high poverty risks and household joblessness is an important risk factor behind child poverty Despite the increasing interest in jobless couples, no studies have so far focused explicitly on coupled joblessness and children

This paper The starting point is couples and the differences in dual joblessness between couples with and without children: Do children decrease the risk of dual joblessness (see table) or is it a matter of selectivity? Dual joblessness in Europe 12 10 8 Childless couples 6 With kids 4 2 0 d k d a m y e e y l n K a r n c l n n i c i a r u a g a U a n e a a t p m s i t u g m a e I l l S e n u t n l r r r i e r F r A G o e F e I B P D G The approach used here provides a direct way of looking at the risks of dual joblessness of couples, and an indirect way of examines childrens’ risks of living in jobless households

Research questions 1. Does having a child affect a couple’s risk of dual joblessness? 2. Do couples with higher/lower risks of dual joblessness have more/less children? 3. Are there comparative differences? Can they be explained?

Theoretical background Childbearing and labour supply Labour supply theory predicts that having a child decreases the labour supply of the mother, but may increase that of the father The effect on the labour supply of the mother can be expected to be stronger → having a child increases the risk of dual joblessness Socio-economic characteristics and fertility A common prediction states that female socio-economic resources suppress fertility, whereas male resources boost fertility → a prediction of selectivity at the level of couples hard to make Easterlin: couples postpone or limit fertility when confronting or expecting economic problems → couples with higher risks of dual joblessness at time t likely to postpone or limit fertility

Institutional effects Policies, which help families combine child care and paid work are expected to decrease the effect of children on dual joblessness. The impact on selective fertility is less clear. Policies aimed at reducing the direct costs of children can increase the effect of children on dual joblessness, and encourage fertility among couples with higher risks of dual joblessness Means tested social benefits expected to increase the effect of having a child on dual joblessness Flexible labour markets may either increase or decrease the effects, open question.

Data European Community Household Panel (ECHP), waves 1-8. 12 countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Portugal, Spain, UK Total number of couples = 27,008, ranging from 1,000 in Denmark to 3,687 in Italy Dependent variable: Dual joblessness (both partners jobless, dummy) Independent variables: Number of children, Age of youngest child (0-1, 2-3, 4-5), Age of husband (linear and squared), Unemp. rate, Either partner student, Either partner been long-term jobless Macro-level indicators: 0-2 year-olds in public childcare, Maternal employment support –index (Gornick et al), Support for costs of children (Bradshaw and Finch), Percentage part-time employment, Labour market rigidity, Female economic activity, Egalitarian values, GDP Methods Effect of having a child on dual joblessness: Fixed effect logit models (control for unobserved heterogeneity) Selection into having a child on the basis of the risk factors: Comparison of the estimates of the fixed effects model and a pooled logit model → an indicator of the bias arising from the correlation between the child variable and the error term

Results: Does childbearing increase the risk of dual joblessness? In eight countries, having a child increases the risk of coupled joblessness In four countries, childbearing decreases the risk, but the estimate is significant in only one country (Greece) The results do not show any clear patterning according to welfare regime lines Some surprises, especially Greece (negative estimate), but also Austria and Belgium Effects of children on dual joblessness, fixed effects logit estimates 1 0,8 0,6 0,4 n.s. 0,2 p<0.05 0 d k d m e y e l n a y K a r n c n c l i n i a U a r u a g n a e a a t p m s i t u g a m I l l e e n u t S r -0,2 n l r r r i e F r o A G e F e I B P D G -0,4 -0,6

Results: Do couples with higher/lower risks of dual joblessness have a higher/lower probability of having a child? Positive estimate: Couples with higher risks of dual joblessness have a higher propensity of having a child. Negative estimate: the opposite In five countries, couples with higher risks of dual joblessness postpone or forgo childbirth Only in Ireland and couples with higher risks of dual joblessness also (significantly) more likely to have a child Selection: Do couples with higher risks of dual joblessness have more/less children? 0,4 0,2 0 y d k d a m e e y l n a K r c n n n i c l i a r u a g a U n a e a a t m s i m t u p g a e I l l e n u t S n l r r r e r r i A F G o n.s. e F e I B G P D -0,2 p<0.05 -0,4 -0,6 -0,8

The two measures have a strong negative correlation at the country level This suggests that in countries with high labour market penalties of childbearing, couples are more likely to postpone or forgo births, when labour market prospects are grimmer .5 IRL S election DK GR FIN 0 S F P UK D -.5 I A BE -1 -.5 0 .5 1 Effect (FE) r=-0.848 (p=0.001)

These country-level measures had a moderate and/or significant correlation with the FE estimate Fig 2.3 Fig 2.4 1 1 A A BE BE UK Ch ild e ffe ct (FE ) UK Ch ild e ffe ct (FE ) .5 .5 D D I I P P S F S F 0 0 IRL IRL FIN FIN DK DK GR GR -.5 -.5 0 5 10 15 20 10 20 30 40 Support for costs of kids Share of part-time employment (%) r=0.407 (p=0.1893) r=0.653 (p=0.021) Fig 2.7 1 A BE UK Ch ild e ffe ct (FE ) .5 D I P F S 0 IRL FIN DK GR -.5 2.2 2.4 2.6 2.8 3 Egalitarianism r=-0.456 (p=0.136)

The share of part-time employment seems to have an especially important (positive) impact on the effect of childbearing on dual joblessness. Suggests that part-time work not interesting for dually jobless couples with children (low entry wages) Egalitarian social values seem to reduce, and generous family policies seem to increase the effect of kids (although not significant) Northern and Southern regimes have lower risks, thus suggesting different paths to same outcomes. However, as seen above, no clear country clusterings Ordinary least squares (OLS) regressions with the point estimate of the child effect as the dependent variable. Model A Model B Model C Model D Model E Model F Model G Part-time 0.027* - - - 0.024† 0.042* 0.031 employment (0.010) (0.012) (0.013) (0.020) Egalitarianism - -0.652 - - -0.561 - -0.232 (0.403) (0.329) (0.558) Costs of kids - - 0.028 - 0.005 - - (0.020) (0.019) EPL - - - - - 0.188 - (0.117) Nordic - - - -0.769* - - -0.294 (Ref. Cont) (0.284) (0.497) Southern - - - -0.487† - - -0.055 (0.232) (0.350) Liberal - - - -0.203 - - -0.388 (0.284) (0.319) Constant -0.492 1.851 -0.102 0.507 0.978 -1.281 0.145 R 2 0.427 0.208 0.166 0.525 0.581 0.555 0.668 Sources : Eurostat (2003) European Community Household Panel, waves 1-8; see above. † p<0.10; * p<0:05; ** p<0.01; N=12

Conclusions Childbearing is an important “trigger event” inducing downward labour market mobility for couples in many European countries, possibly leading to important drops in (material) well-being In many countries, couples experiencing or anticipating employment difficulties restrict fertility These two effects vary, however, across countries so that in countries where the child penalty is the highest, couples seem more risk-averse. Although the country differences do not cluster clearly according to borders drawn by regime theory, the Nordic and the Southern countries have smaller child penalties The child penalty is higher in countries, in which part-time employment is more widespread. Part- time employment, while celebrated as a way of combining paid work and family duties, may be incompatible with the needs of jobless couples with children