Gender Development in India Dimensions and Strategies
Gender Development in India Dimensions and Strategies
Gender Development in India Dimensions and Strategies
N. Santosh Ranganath
Faculty Member
Dept. of Commerce and Management Studies
Dr. B.R. Ambedkar University
Srikakulam, Andhra Pradesh – 532410
"Ensuring gender parity has become a great challenge for the modern world.
Gender equality forms part of development strategy in many countries of the world
including India. When both men and women have equal access to services and resources,
enjoy equal rights, and get equal opportunities to develop skills and capabilities then the
development of the country is fast and balanced.
The debate over the effect of gender equality on growth centers on two key issues:
Is the relationship between gender equality and development positive? Does gender
equality affect growth, or vice versa, or both - or do common factors determine both
simultaneously? In order to answer the first question, recent studies have attempted to
reconcile findings of a positive relationship with the contrasting results of Barro and Lee
(1994) by examining alternative econometric specifications of the growth equation. One
method has been to identify "outlier" countries that could be influencing the Barro and
Lee findings. Stokey (1994) suggests that these apparently puzzling findings might derive
from the East Asian "tigers", which have had very fast growth but comparatively lower
female education. Lorgelly and Owen (1999) use statistical techniques to identify outlier
countries. Excluding them from the growth Regression is sufficient to weaken the
significance of the partial correlation between female education and growth, though it
remains negative. Dollar and Gatti (1999) find that, by omitting regional dummy
variables from their model, they are able to reproduce the estimates of Barro and Lee.
This is because the Latin American economies grew more slowly than predicted
by the other variables in the growth equation at the same time that the region had high
female secondary attainment. Hence, if the regional dummy variables are excluded from
the growth regression, Latin America's poor growth gets attributed to the female
education variable, resulting in a negative co- efficient on female education. Moreover,
because of the high multicollinearity between male and female education in the Barro-
Lee specification, there is very little independent variation to allow a disentangling of
their separate effects on growth. Excluding the male education variable switches the sign
of the coefficient on female education from negative and significant to positive and
insignificant (Lorgelly and Owen 1999). Knowles, Lorgelly, and Owen (2000) use a
different specification of the growth regression.
They average data over the entire estimation period of 1960-90 for the
explanatory variables to estimate a long-run relationship. They find that the effect of
female education on output per worker is significantly positive. But when they re-
estimate their model with base period values of human capital stocks as Barro and Lee
did, they reproduce the contrasting Barro-Lee results of a negative significant relationship
with respect to female education, and a positive one with respect to male education.
The second issue is concerned with the direction of causality. If the positive
relationship between gender equality and growth is due to a two-way relationship or
underlying common factor determining both simultaneously then ordinary least squares
regressions of income growth on measures of gender equality would be biased. This is
particularly true when different variables pertain to the same period. Recent studies have
attempted to unravel the cause and effect relationship between gender equality and
economic growth-but have run into problems of measurement and statistical inference.
The most common technique is estimation using instrumental variables.