12
Let us consider an example from Chakravarty et al. (2019), a study that explores the effect of a
vocational training program in Nepal on subsequent labor market outcomes among program
beneficiaries.
• The study specifically explores the issue of the labor market returns to vocational training
in the context of a low-income country.
• In the text, the authors contextualize the study findings by discussing previous studies
examining the effect of vocational training from other countries and income
classifications:
o “Most of the vocational training literature on similar programs in high- or middle-
income countries finds low or insignificant effects (Card et al., 2010; Kluve,
2010; Dar and Tzannatos, 1999).”
• Then, the authors proceed to compare the study’s findings and research question relative
to other very closely related studies on the same research question:
o “Our paper provides evidence that large program impacts of vocational training
programs – particularly in the low-income context – are possible, despite most of
the existing evidence from middle-income countries suggesting otherwise.
Based
on the experience of middle-income countries, only Maitra and Mani (2017), Reis
(2015), and Alzua (2016) find positive impacts on the probability of any
employment and any earnings in India, Brazil, and Argentina, respectively. On
the other hand, Honorati (2015), Card et al. (2010), Attanasio et al. (2011, 2015),
Ibarraran (2015), Hirshleifer et al. (2015), Acevedo et al. (2017), Diaz and Rosas
(2016), and Galasso et al. (2004) find either mixed, muted, or no impacts at all
from vocational programs on various labor market outcomes. That large effects
may be particularly pronounced in low-income contexts is confirmed by two
studies conducted in Liberia and Uganda: Adoho et al. (2014) randomly assign a
similar intervention like the one studied here and detect an increase of 47 percent
in non-farm employment and 80 percent in earnings among young Liberian
women
. Similarly, Bandiera et al. (2017) find positive impacts on income-
generating activities of 48 percent (which were almost entirely driven by self-
employment), but no positive impacts on wage-employment in Uganda.”
• The authors also point out how their study and its findings relate to other closely related
studies on one important dimension: program impact differences by gender of the
beneficiary:
o “Our second contribution to the existing literature relates to the pattern of
different returns to vocational training between women and men. Although
Blattman and Ralston (2015) point to a stylized fact that proposes vocational
training has higher returns for women, McKenzie (2017) reviews recent
vocational training programs in low-income and middle-income countries and
argues that previous studies, which formally test for equality by gender, can either
not reject similar impacts for men and women, or have found significantly higher
impacts for men. In stark contrast, our study does formally test for equality by
gender, and it unambiguously shows robust evidence that vocational training in
our context yields higher returns for women.”
o “We highlight that our results are likely driven by the socio-cultural norms in
Nepal, which shape gender roles that identify women with more restrictive