Doing Gender In Context: Household Bargaining

1
Doing Gender In Context Household Bargaining
Risk of Divorce in Germany and the US

• Dr. Lynn Prince Cooke
• University of Kent - Canterbury
• Special PSID 40th Anniversary Session
• ASA Annual Meeting, Boston, 3 August 2008

2
Female EmploymentA Balancing Act

• PROS
• Increases gender equality, reducing womens
  vulnerability
• Reduces child poverty risk
• Increases current tax base
• Spurs job growth commodification of domestic
  production

• CONS ?
• Relationship effects (Becker 1981, 1985 Parsons
  1942) v Oppenheimer (1988, 1997)
• Declining fertility, threat to future tax base ?
• Need to find alternate carers

3
Equality in Context

• Welfare states vary in their reinforcement of
  gender role specialization
• Strong male breadwinner states (Lewis 1992)
  reinforce specialization through family wages,
  dependent tax credits, minimal childcare
  provision, extended maternity leave, high
  marginal tax rates, etc.
• Alternatively, policies can promote equal access
  to employment allowing women to establish
  autonomous households (Orloff 1993)

4
Doing Gender

• Womens increasing employment was expected to
  lead to a revolution in gender roles within the
  home, reducing womens specialization in these
  tasks.
• Contrary to this logic of the pocketbook,
  however, women retain primary responsibility for
  domestic tasks
• In fact, as wives earnings exceed their
  husbands, womens share of domestic tasks
  increases (Brines 1994 Bittman et al 2003)

5
Context How Gender is Done

• In U.S., both wives and husbands decrease their
  domestic hours as wives earnings increase
  (Brines 1994)
• In Australia, wives increase their domestic hours
  as their earnings increase, but husbands hours
  dont change (Bittman et al. 2003)

6
Competing hypotheses

• H1 If specialization is optimal, husbands
  greater share of domestic tasks should increase
  risk of divorce.
• H2 If gender equity is optimal, husbands
  greater share of domestic tasks should decrease
  the risk of divorce associated with womens
  employment.

7
What about doing gender?

• H3 If the compensatory domestic behavior
  represented by doing gender neutralizes gender
  deviance when women are employed, it should
  decrease the risk of divorce
• H4 If doing gender represents a display of
  mens unfair relative gender power, it should
  increase the risk of divorce

8
Analysis

• Compare
• effects of womens relative earnings and mens
  share of housework on the risk of divorce
• among couples marrying for the first time between
  1985 -1995
• in a weak and strong male breadwinner state

9
Data Method

• The GSOEP used for West Germany, where the male
  breadwinner model has been reinforced in policy
  (N 522 couples)
• The PSID used for the U.S., where policy is
  silent on the private sphere (N 368 Black 1,112
  Caucasian couples)
• Discrete time EHA of risk of divorce with robust
  standard errors

10
Benefits of PSID

• Ability to distinguish cause from effect as can
  lag independent variables to predict outcome
• Through CNEF, Comparability with GSOEP (and now
  also BHPS and HILDA) for doing cross-national
  comparisons

11
Descriptives- Mirroring Context

• West Germany
• 58 of wives are out of the labor force
• Wives spend 29 hours/week in housework
• Husbands perform 29 of housework
• 13 of wives earn more than husbands
• 7 of couples do gender

• United States
• 9 of wives are out of the labor force
• Wives spend 17 hours/week housework
• Husbands perform 33 of housework
• 24 of wives earn more than husbands
• 8 of couples do gender

12
Model 1 Main Effects on Log-odds Divorce

• Germany
  U.S.
• Male breadwinner 0.62 0.83
• Wifes earnings 0.01 0.02
• Wifes housework 0.01 - 0.03
• Husbands hw 0.01
  - 0.04
• Husbands hw-sq ns 0.0003
• Home ownership - 0.25
  - 1.23
• of children - 0.57
  0.11
• Log-likelihood -499 -378

13
Predicted Earnings-Housework Equity Effects on
Log-Odds of Divorce
14
Model 2Doing Gender Effects

• Germany U.S.
• 1 2
  1 2
• Male breadwinner 0.62 0.60 0.83 0.99
• Wifes earnings 0.01 0.01 0.02
  0.02
• Wifes housework 0.01 0.01 - 0.03
  - 0.03
• Husbands hw 0.01 0.01
  - 0.04 - 0.04
• Husbands hw-sq ns ns 0.0003 nm
• Home ownership - 0.25 -0.23
  - 1.23 - 1.23
• of children - 0.57
  -0.57 0.11 0.09
• Log-likelihood -499 -499 -378 -369

Doing Gender -0.01
-2.25
15
Conclusions

• Individual effects on couples vary in context
• Policy reinforcement of traditional gender roles
  undermines couple market and domestic flexibility
  
• Rules of exchange and equity apply in the liberal
  regime, with U.S. couples sharing breadwinning
  and housework the most stable
• Even in liberal market, however, a benefit when
  neutralizing gender deviance

16
Next Steps

• Expand comparisons (GERT)
• Countries Australia, Belgium, Finland, France,
  E/W Germany, Italy, Netherlands, Norway, Sweden,
  US
• Transitions womens into and out of both de
  facto and de jure relationships from age of 16