Transcript Economic and cultural determinants of unmarried cohabitation in

ECONOMIC AND CULTURAL
DETERMINANTS OF UNMARRIED
COHABITATION IN BRAZIL
Maira Covre-Sussai
Koen Matthijs
0. Outline:
1. Research questions, data and method.
2. Contextualizing the paper: Brazilian socio-economic and
cultural diversity.
3. Main results.
4. Conclusions and Limitations.
1. Research questions:
“What are the economic correlates of cohabiting rather
than getting married in Brazil?”
“How does the cultural context affect couples’ probability
of being married on the one hand or cohabiting on the
other?”
1. Research questions:
ECONOMIC ARGUMENT
?
SOCIO-CULTURAL
ARGUMENT
CULTURAL
DIFFERENCES
CHILDREN (MSC)
WOMEN INCOME
WOMEN EDUCATION
COHABITATION
SOCIALIZATION
PROCESS
RELIGION
INDIVIDUALIZATION
1. Data & Method:
- Couples level dataset from the household survey of the
Brazilian 2000 census.
- Post-marital cohabitation excluded.
- Final sample composed of 183,123 couples.
- Multilevel procedures.
2. Context: some figures (2009)
Area: Total 8,514,877 km2
Population
2009 estimate: 192,272,890
Different ethnic composition:
Source: IBGE (2009).
1. Context: ethnicity
North (N) and Northeast (NE): majority of the population
composed by indigenous and African descendents.
Central-west (CW): most equilibrated
division of ethnicities with 42% of
whites, 50% of mixed races people
and 6.5% of African descendents.
Southeast (SE) and South (S): mainly
composed by the descendents of the large
European immigration of the 19th and 20th
centuries: Italians and Germans.
1. Context: Brazilian legislation (2002)
Egalitarian power inside the family;
Matched parents’ rights in case of divorce and children’s
custody;
Cohabitation can be considered a type of marriage by the
law.
Brazilian Civil Code (2002).
1. Context: (%) Couples per type of union
(1960-2000)
70
60
50
(%)
40
30
20
10
0
1960
1970
Civil and Religious Marriage
1980
Civil Marriage
1991
Religious Marriage
2000
Cohabitation
Source: IBGE: 1970 and 2000 censuses, own calculations.
1. Context: Marriage rate in Brazil (1980-2007)
10.00
8.00
7.00
6.00
5.00
4.00
3.00
2.00
1.00
Brazil
Central-west
Northeast
North
Soultheast
6
20
0
4
20
0
2
20
0
0
20
0
8
19
9
6
19
9
4
19
9
2
19
9
0
19
9
8
19
8
6
19
8
4
19
8
2
19
8
0
0.00
19
8
Marriages (/1000 inhabitants)
9.00
Soulth
Source: IBGE: Estatísticas do Registro Civil (Civil Register Statistics), own calculations.
1. Context: Previous research
Brazilian women are more likely than their U.S. equivalents
to cohabit with male partners (Light and Ureta, 2004).
Ethnographic evidence show that Brazilian cohabitants
generally refer to themselves as married, and use the words
husband and wife to refer to their partners (Rao and
Greene, 1996).
Despite of the similarities between married and cohabiting
couples, we can also find some evidence that it is too soon
to
affirm
that
marriage
and
cohabitation
are
indistinguishable in Brazil….
1. Context: Previous research
While the average duration of a marriage in Brazil is 10
years (IBGE, 2007), half of cohabitations last no more than
6 years (Rangel, 2006).
Declines in fertility rates were sharper for couples in formal
unions than for those in consensual ones and the total
fertility among cohabiting couples is higher than among
officially married ones, even controlling for age, education
and duration of union (Lazo, 1999).
3. Results
Cohabit= +û0j
= β0 + β1Childrenij + β2Wincomeij + β3Weducationij + β4Classij + β5Mcohortij +
β6Urbratej + β7HDIj + β8Povertyj + β8Whitesj
where û0j  N(0, 2e )
is the states-level differential.
3. Results – Cultural (States) Effect
• Null model: Average cohabitation probability is 30%.
• The coverage interval for Brazilian states can range from 14% to 51%
by considering states effect.
3. Results – Children Effect
0.60
0.53
0.50
0.46
0.44
0.40
0.30
0.30
0.24
0.23
0.10
0.09
0.20
0.10
0.13
0.00
None
Up to 2
Higher Probability
Median Probability
3 or more
Lower Probability
3. Results – Female Education Effect
0.60
0.58
0.50
0.47
0.40
0.35
0.31
0.30
0.26
0.26
0.20
0.17
0.16
0.13
0.12
0.10
0.06
0.05
0.00
Illiterate
Basic
Higher Probability
Secondary
Median Probability
Lower Probability
College
3. Results – Social Class Effect
0.60
0.50
0.51
0.41
0.40
0.35
0.30
0.31
0.28
0.20
0.20
0.17
0.10
0.12
0.09
0.07
0.14
0.06
0.00
Poor
Working
Higher Probability
Middle
Median Probability
Lower Probability
Upper
3. Results – Cohort Effect
1.00
0.90
0.90
0.80
0.73
0.70
0.60
0.56
0.54
0.50
0.48
0.41
0.40
0.31
0.30
0.20
0.10
0.00
0.76
0.26
0.23
0.18
0.07
0.03
1920
0.09
0.04
1930
0.31
0.20
0.14
0.14
0.06
1940
Higher Probability
0.08
1950
Median Probability
1960
1970
Lower Probability
1980
3. Results – Religious Effect
0.80
0.70
0.69
0.60
0.56
0.50
0.48
0.49
0.44
0.40
0.31
0.30
0.25
0.25
0.26
0.11
0.11
0.23
0.20
0.10
0.11
0.14
0.04
0.00
Evangelicals
Catholics
Higher Probability
Other
Median Probability
Different Religion
Lower Probability
No Religion
3. Results – Children*Social Class Effect
Cohabitation: odds ratio
6
5
4
3
2
1
0
Up to two
Three or more
Social Class
Poor
Working
Middle
3. Results
STATE-LEVEL PREDICTOR VARIABLES EFFECT:
ETHNICITY: Whites are less prone to cohabit than non whites.
Significant part of the between-state variance can be explained by
ethnical differences.
POVERTY, URBANIZATION LEVEL, HDI: None of these effects are
significant neither add validity on the models.
4. Conclusions
The results found for the economics hypotheses were not totally
supportive:
Women with lower economic position as well as couples from the
lower classes tend to cohabit rather than get married.
Children represent a valuable marital-specific capital for the upper
classes, but their impact on the decision to get married for the lower
classes is smaller.
This result is in line with previous qualitative research which states
that it is in the Brazilian middle class that individualistic values are
nurtured (Machado, 2001).
4. Conclusions
The outcomes for the cultural argument are more consistent:
Cohabitation is more common among the younger cohorts which can
be related to the individualization of society and the
detraditionalization of family life.
Religion however was shown to (still) be a powerful mechanism of
behavioral restriction.
Couples with the same religion orientation tend to cohabit less,
mainly when Evangelicals.
Cultural differences play an important role: Significant proportion of
the probability to cohabit in Brazil is explained at state level, even
considering the level of poverty, urbanization, HDI and the ethnicity
composition.
4. Limitations
• Census data: Covers the whole country, but has limited information.
• Particularities of each family are missed.
• Cross-sectional design does not allow us to verify change in
couples’ life.
Thank you!
Maira Covre-Sussai
([email protected])
Koen Matthijs
([email protected] )