Education, Skills, and Labor Market Outcomes by Gender: Evidence from Pakistan

1
Education, Skills, and Labor Market Outcomes by
GenderEvidence from Pakistan

• By
• Geeta Kingdon
• Måns Söderbom

2
Introduction

• Pakistan is an international outlier in gender
  gaps in education
• Far from narrowing, gender gap in primary
  enrolment rose by 30 points 1985-95, superseding
  even Afghanistan where gap rose by 18 points over
  same period 1.
• While somewhat narrowed, the gender gap remains
  very high
• Raises the Q why girls educational outcomes are
  so inferior to boys
• One plausible explanation among others is if
  girls face much lower economic incentives for
  acquiring education than boys
• We test this potential explanation using PIHS
  1998-99 2001-02
• 1 Computed from Conly, Shanti (2004) Educating
  Girls Gender Gaps and Gains, Population Action
  International, http//www.populationaction.org/res
  ources/publications/educating_girls/ggap_graph02.h
  tm.

3
Part of wider work

• Investigating education-earnings relation in
  Pakistan / Ghana
• For this presentation, focusing on Q Can educ be
  a path to gender equality in the labour market?
• E can have direct effect on earnings also
  indirect effect via occup
• Goals are to examine
• Whether E/skills improve women and mens
  occupational attainment
• Whether E/skills raise women and mens earnings
  (within occupation)
• Compare women and mens returns to E/skills
• Separately by age group
• Pakistani labour market - sharply segregated by
  gender

4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7

• Next we ask
• How occup changes with age (transitions)
• How occup changes with education level
• Find occup choices of women much narrower,
  suggesting
• sharp gender division of labour
• conservative attitudes to womens econ work
• This limits the extent to which educ can be a
  path to gender equality in economic outcomes

8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
Conclusions

• RORE/skills greater for women in all occupations
  and in both age groups.
• Could reflect
• scarcity of educated women
• existence of jobs which require (or are reserved
  for) edu women
• Gender pattern of returns is welcome and provides
  strong economic incentives to acquire schooling.
• Positive story counterbalanced with overall
  returns to employment being far lower for women

16

• Even though RORE may be high for women (slope 3
  times as steep), have much lower earnings (lower
  intercept)
• Aslam (2007) a large part of gender wage gap not
  explained by gender diff in productivity
  endowments, due to pot. discrimination (though
  more work is needed to get better measures of
  experience, quality of S)
• Education of women reduces that wage gap, i.e.
  theres less discrimination among educated
• Thus, total labour market returns are higher for
  men even though returns to edu are higher for
  women
• This gives parents an investment motive for
  investing more in sons
• Even if return higher for girls, the part of it
  which accrues to parents is lower
• To reduce its gender gaps in educ, it needs not
  only supply-side measures eg improving school
  supply for girls
• Also needs to address the demand side
• ease credit constraints (e.g. attendance-contingen
  t transfers to girls)
• labor market policies to reduce gender-diff
  treatment by employers
• Address conservatism of attitudes to womens educ

17
(No Transcript)
18
(No Transcript)
19

• RORE invariably greater for the older group
• Plausible explanation is filtering down of
  occupations
• Successive cohorts of workers at a given educ
  level enter less and less skilled jobs (Knight,
  Sabot and Hovey, 1992)
• When our old group got their jobs, primary
  completers in scarce supply and primary may have
  been sufficient to obtain white-collar job. Those
  who obtained such jobs remain in them today. But
  rapid expansion of supply of educ, primary
  completers among young today fortunate to even
  get a low paying blue-collar job.
• So, over time, there is a compression of wages by
  education level.
• Thus, ROR for a given level of educ lower for
  younger workers because they perform less skilled
  tasks than older persons with that level of educ

20
(No Transcript)
21
(No Transcript)