Teenage mothers and education policy

1
Teenage mothers and education policy

• by
• Sandra Black,
• Department of Economics, UCLA
• NBER and IZA-Bonn
• Paul J. Devereux,
• Department of Economics, UCD
• IZA-Bonn
• Kjell G. Salvanes,
• Department of Economics, Norwegian School of
  Economics
• Center for Education Research-CEP and IZA-Bonn

2
Research questions and motivation

• Teenage mothers are associated with negative
  outcomes such as low educational, earnings,
  welfare dependence etc
• What determines teenage motherhood is not well
  understood
• For instnance we know that low-educated women are
  associated with the likelyhood of a teenage birht
• But will increased education in our case
  compulsory schooling encourage delayed
  childbearing?
• Why do we care?
• Will government policy have an effect on teenage
  childbearing

3
Research questions and motivation

• Our approach
• Measure the causal effect of changes in
  compulsory schooling laws on teenage childbearing
  for Norway and the US
• Institutional differences
• Understand mechanisms through which this
  relationship works

4
Outline

• Introduction
• Institutional differences
• Relevant Literature
• Norwegian and US School Reforms
• Identification Strategy
• Data
• Results
• Specification/Robustness Checks
• Mechanisms/interpretations
• Conclusion

5
Institutional differences

• Similarities and differences between the US and
  Norway
• Very high GDP per capita
• High education level
• The US system is relatively unsupportive for
  teenage mothers
• The system in Norway very generous
• Single parents get support to take care of the
  child until the age of 10
• The government enforce payment from fathers
• The government pay all education expences for
  the mother
• The government provides subsidized housing
• Sinlge mothers get double child allowance

6
Previous literature

• Negative adult outcomes of teenage mothers
• Lower education, less work experience, welfare
  dependence, lower birth weights, higher rates of
  infant mortality, and higher rates of
  participation in crime (Ellwood, 1988 Jencks,
  1989 Hoffman et al., 1993 Kiernan, 1997).
• Negative outcomes of children (Francesconi, 2004,
  Hunt, 2003)
• The effect of education policy in reducing teen
  fertility (McCrary and Royer, 2003).
• All births in California and Texas from 1989-2001
• A selected sample for those who had children

7
The Norwegian School Reform

• Mandatory School Reform passed by Parliament in
  1959
• 3 Goals
• To increase the minimum level of education in
  society by extending compulsory schooling from 7
  years to 9 years.
• To smooth the transition to higher education
  (standardized curriculum)
• To enhance equality of opportunities both along
  socio-economic dimensions and geographical
  dimensions
• All municipalities must implement reform by 1973
• Cohorts affected 1947 - 1959

8
Table 2 Impact of the reformDistribution of
Education Two Years Before and After the Reform
9
The Norwegian School Reform

• What caused different municipalities to adopt the
  reform?
• Government wanted adoption to be representative
  of the national geography.
• Some evidence (Lie 1973, 1974) that proximity to
  adopting municipality can somewhat explain
  adoption pattern.
• Little evidence suggesting municipality
  characteristics determined timing.

10
Figure 1The Number of Municipalities
Implementing the Education Reform, by Year
11
Figure 2Reform implementation in Poor vs Rich
MunicipalitiesBased on Average Family Income.
12
Figure 3Reform Implementation in High vs. Low
Education MunicipalitiesBased on Average Years
Fathers of Education in the Municipality
13
Changes in US compulsory school laws

• The period 1924-1974
• Five possible restrictions on educational
  attendance
• 1. maximum age by which a child must be enrolled
• 2. minimum age at which a child may drop out
• 3. minimum years of schooling before dropping out
  
• 4. minimum age for a work permit
• 5. minimum schooling required for a work permit.

14

• Following Acemoglu and Angris (2001) we assign
  compulsory school attendance laws to women on the
  basis of state of birth and the year the
  individual was 14 years
• We do several check testing the relationship
  between compulsory schooling laws and early
  fertility

15
Identification Strategy and Specification

• The US model
• Where
• Cohort is full set of year of birth indicators
• State is a full set of state indicators
• White is a indicator for wheter the woman is
  white
• compulsory is a vector of dummies with a min
  dropout age for less than 16 as the omitted
  category
• Norway model
• Where
• compulsory is 1 if the individual was affected by
  the education reform (minimum dropout age of 16)
  and 0 otherwise (min dropout age of 14).

16
Norway Data

• Merged Administrative Registers and Census Data
  from Statistics Norway
• Covers entire population of Norwegians aged 16-74
• Municipality of individual's mother in 1960
• Education Information
• Educational attainment reported by Educational
  Establishment
• Augmented with information from 1970 Census
• Timing of Reform by Municipality
• Have Indicators for 545 out of 728
  municipalities.
• Women born between 1947 and 1958
• Omit mothers with births before 15

17
US data

• IPUMS extracts from the decennial Census
  1940-1980
• 1 of 1940-1960, 2 of 1970 state samples,5 of
  1980 state samples
• Only possible to identify children living in the
  household
• Restrict our Census samples to women aged between
  20 and 30 and then calculate age at first birth

18
Table 1 Descriptive StatisticsUnited States
19
Table 1 Descriptive StatisticsNORWAY
20
Table 2 Effect of Compulsory Schooling Laws on
the Probability of First Birth by A Certain Age
21
Table 3 Effect of Compulsory Schooling Laws on
the Probability of Birth Urban/Rural
Distinction
22
Robustness checks

• Include state/municipality specific trends
• Since samples for the US varies across cohorts,
  we weight the cohorts equally
• Testing effects of future school laws
• Alternative measures for the US on compulsory
  schooling laws

23
Different mechanisms

• The incarceration effect
• A new compulsory schooling law may change the
  optimal fertility age schooling is more costly
  if you are a young mother
• Human capital effect
• Increased school changes your preferences
  regarding fertility age
• If the incarceration effect is the only effect
  of compulsory schooling laws, there should be no
  effect on behavior at ages above which the
  schooling laws bind.

24
Table 9 Effect of Compulsory Schooling Laws on
the Probability of Birth Conditional on Not
Already Having a Child
25
Table 9 Effect of Compulsory Schooling Laws on
the Probability of Birth Conditional on Not
Already Having a Child
26
Concluding remarks

• We find that minimum schooling laws have a
  significant negative effect on the probability of
  having a child as a teenager both in the US and
  in Norway
• The results are quite strong and very similar
• In the US the compulsory schooling laws reduced
  the probability of a birth at 18 and 19 by 4.7
  percent
• In Norway the effect is 3.5 percent
• Our results suggest that the mechanisms both
  include an incarceration effect and human
  capital effect