Household Shocks and Education Investments in Madagascar
| Published date | 01 December 2016 |
| Author | Thomas F. Walker,Peter J. Glick,David E. Sahn |
| Date | 01 December 2016 |
| DOI | http://doi.org/10.1111/obes.12129 |
792
©2016 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd.
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 78, 6 (2016) 0305–9049
doi: 10.1111/obes.12129
Household Shocks and Education Investments in
Madagascar*
Peter J. Glick†, David E. Sahn‡ and Thomas F. Walker§
†RAND Corporation, 1200 South Hayes Street, Arlington, VA 22202, USA (e-mail:
pglick@rand.org)
‡Cornell University, Institute for the Study of Labor (IZA) and Centre d’Etudes et de
Recherches sur le D´eveloppement International (CERDI), B16 MVR Hall, Ithaca, NY
14853, USA (e-mail: David.Sahn@cornell.edu)
§The World Bank, 1818 H Street NW, Washington, DC 20433, USA (e-mail:
twalker@worldbank.org)
Abstract
This paper investigates the impact of exogenous shocks to household income, assets and
labour supply on children’s school attendance in Madagascar. The analysis uses a unique
data set with 10 years of recall data on school attendance and household shocks.We find that
the probability of a child dropping out of school increases significantly when the household
experiences an illness, death or asset shock. We propose a test to distinguish whether the
impact of shocks on school attendance can be attributed to credit constraints, labour market
rigidities, or a combination of the two. The results suggest that credit constraints, rather
than labour market rigidities, explain the inability of households in Madagascar to keep
their children in school during times of economic stress.
I. Introduction
Madagascar has made considerable progress in raising its net primary school enrolment
rate from 67% in 2001–2002 to 98% in 2004–2005 (World Bank, 2008). This rapid in-
crease reflects the decision to eliminate school fees for public primary education in 2002
following a prolonged political crisis. The provision of free school supplies (a book bag, a
textbook, and a kit containing pencils and other items) to all primary school children is also
considered to have playedan impor tant role. Nevertheless, as of 2011, primary completion
rates remained around 70%, and dropout rates around 60% (World Bank, 2015). Delayed
enrolment also remains a concern, with nearly a third of children enrolling in school after
their sixth birthday and over 8% after age nine.
JEL Classification numbers: I25; J220; D13; E24.
*We would like to acknowledge the useful comments of George Jakubson and two anonymous reviewers. This
document is an output from a project funded by the UK Department for International Development (DFID) and the
Institute for the Study of Labor (IZA) for the benefit of developing countries.The views expressed are not necessarily
those of DFID or IZA.
Household shocks and education investments in Madagascar 793
From a policy perspective, it is important to understand the factors preventing children
in Madagascar from progressing further in their education. This is especially crucial given
that Madagascar’s poverty headcount at $1.25/day PPP was 88% in 2010 (World Bank,
2015). Education is important in determining success in the labour market, both in terms
of finding a job in the small but expanding formal labour market and in terms of wages
received. In rural areas, where 80% of workersreside, increases in education are associated
with higher hourly earnings, even among agricultural workers (Hoftijzer and Paci, 2008;
Glick, 1999). It is in this context that weexplore whether unanticipated health and economic
shocks cause children to drop out of school or delay enrolment, thereby compromising
their future welfare and employment prospects. Understanding the impact of shocks on
education decisions – and the channels of this impact – could help in designing safety nets
and other policies to insulate investments in education from such shocks.
A few recent studies attempt to measure the impact of unanticipated shocks on education
investment; the literature is discussed in detail in section II. Our paper builds on this
literature by employing a discrete hazard model to test the response of school attendance
in Madagascar to various types of idiosyncratic household shocks. This model captures the
tradeoff in allocation of children’s time between labour, leisure and schooling, whichin the
absence of complete markets will be affected by shocks to household resources (Becker,
1965; Jacoby and Skoufias, 1997).2A novelty of our study is the unusually detailed data
set used, which comes from a nationwide survey of schooling and educational attainment
in Madagascar. Ten years of recall data were collected on various idiosyncratic shocks,
mostly events expected to cause economic stress. These include illness and death of the
household head and spouse; unemployment spells of parents; unanticipated loss of crops,
land, livestock and other assets; and lower-than-expected business income. We find strong
evidence that such shocks result in higher rates of dropout and lowerrates of enrolment. We
also examine the impacts on enrolment and dropout of access to secondary schools, local
primary school quality and a government-sponsored health programme called Secaline,a
(primary) school-based nutrition programme.
Another innovation of the paper is a test of whether and how credit and labour market
rigidities moderate the effects of shocks on schooling decisions. Previous studies have
illustrated how, in the presence of incomplete markets, parents may be compelled to with-
draw their children from school in response to shocks because they cannot borrow enough
to both sustain household consumption and continue investing in their children’seducation.
We propose that such a response might also reflect labour market constraints: the family
may be unable to obtain (or mayincur nontrivial costs in obtaining) replacement labour for
family enterprises if the household’s labour supply is diminished as the result of health or
mortality shocks.3Based on the premise that there is gender differentiation in household
tasks, we develop a partial test between the two effects. Our results support the conclusion
that credit constraints, rather than labour market rigidities, explain the transmission of
economic shocks to schooling decisions.
2‘Complete markets’ here includes not only access to formal savings and credit instruments, but also informal
arrangements that would fully insure the household against unanticipated shocks to income and assets (see, e.g.
Townsend,1994; Udr y, 1994).
3This need not mean hiring paid workers; in some situations, households may be ableto draw on extended family
or friends for temporary assistance.
©2016 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
To continue reading
Request your trial