Abstract:
Internationally, static poverty measures have been used as indicators to evaluate the well-being of
individuals and have served as tools for policy formulation. There are essentially two issues related to
this, firstly static poverty measures tend to ignore the idiosyncratic and covariate shocks that
households, especially in the developing world, face that lead to them moving in and out of poverty
thus making these measures unreliable indictors of well-being. These segments can be identified as
those that are vulnerable to poverty and can only be identified if dynamic measures of poverty are
adopted such as vulnerability related measures. Secondly, there is a lack of representation of young
people in poverty assessments in general owing to the transitional phase these individuals are in which
requires assessment through dynamic poverty measures. In light of these issues, using the Young Lives
dataset for countries Ethiopia, India and Vietnam, this study estimates the vulnerability levels using
the Vulnerability as Expected Poverty (VEP) approach and assesses the impact of idiosyncratic and
covariate shocks on vulnerability of households that have young individuals (ages 15 to 16) in them.
To account for hierarchal data structures in the Young Lives dataset, this study has incorporated
Multilevel Modeling with Maximum Likelihood technique used for estimation. This study concludes
that for the case of Ethiopia and India, idiosyncratic vulnerability has the most impact on the
vulnerability of the Young Lives households which is largely driven from existing low consumption
prospects. For the case of Vietnam this study concludes that it is covariate shocks that have the most
impact on the vulnerability levels of the Young Lives households which also stems from existing low
consumption prospects.
