This was a Project we took in our Econometrics course with Prof. Hemanshu Kumar in the Spring 2018 Semester. Using Regression with STATA, we attempted to establish a causal relationship between gender equality and economic growth.

1. ECONOMETRICS RESEARCH
PROJECT
Aashay Verma and Ashna Kedia
Professor Hemanshu Kumar
Course: Econometrics 206-1
Date of Submission: 10th May, 2018
Word Count: 1200

2. Introduction
“...investing in women is not only the right thing to do. It is the smart thing to do. I am deeply
convinced that, in women, the world has at its disposal, the most significant and yet largely
untapped potential for development and peace.”
Ban Ki Moon, UN Secretary General 2007-2016
One of the United Nations' Millennium Development Goals is to promote gender equality
and empower women, and many advances have been made in this direction over the past few
decades. However, gender inequality is still pervasive. The lack of an equal opportunity leads
to an inefficient use of human capital in the economy and hinders growth. There are a
growing number of studies exploring the relationship between gender equality and economic
growth, the results of which have had policy implications for countries around the world.
This paper uses cross-country data across 109 countries to ask whether gender equality does
have a significant effect on gender growth. The results suggest that this is true for the
indicators that measure equality in terms of health, education and wage whereas it may not be
as significant for indicators that capture equal representation and participation in the
workforce.
Literature Review
Over the past two decades, there have been many studies examining the effect of gender
equality on economic growth through cross-country regression analyses. These papers take
account of issues faced in econometric analyses at this highly aggregated level with varying
degrees of methodological precision. GDP per capita is a reflection of changes in the
economic well-being of a population and has been used by several empirical papers including
this paper as the measure of economic growth (Kabeer and Natalie, 2013, p.8). Studies such

3. as Klasen and Lamanna’s argue that gender gaps in education negatively impact the economy
and improvements in female health, income and education are associated with better child
development outcomes and hence higher economic growth (2009, p.108). The broadest
indicator of women’s labour market activity is the female labour force participation rate.
According to Morrison et al., an increase in this rate leads to an increase in overall
employment which leads to higher output and growth (2007, p.10). It has been suggested that
women are less prone to corruption and therefore increasing access of women in decision-
making bodies leads to efficient allocation of resources which may lead to higher growth
(King and Mason, 2001, p.95). All these papers agree that gender equality does have an
impact on economic growth.
Data
Variable
Name
Description
lnpercapgdp Per Capita GDP (2015): Dependent Variable; Transformed logarithmically
gdistd Gender Development Index (2005): Consists of equality in terms of health,
income and education; Standardised
labpartfem Labour Force Participation Rate
femparl Percentage of women in the country’s Parliament
open Aggregate of Import and Export as a share of GDP

4. The research project has been carried out on a country level basis with a sample size of 109
observations with fully published secondary data. The present study has sourced data from
the World Bank and the UN Human Development Reports.
Estimation Strategy
The main estimation equation modelled the question using:
𝐥𝐧𝐩𝐞𝐫𝐜𝐚𝐩𝐠𝐝𝐩 = 𝜷 𝟎 + 𝜷 𝟏 𝒈𝒅𝒊𝒔𝒕𝒅 + 𝜷 𝟐 𝒍𝒂𝒃𝒑𝒂𝒓𝒕𝒇𝒆𝒎 + 𝜷 𝟑 𝒇𝒆𝒎𝒑𝒂𝒓𝒍 + 𝒖
After adding openness as a control variable, the estimation equation became:
𝐥𝐧𝐩𝐞𝐫𝐜𝐚𝐩𝐠𝐝𝐩 = 𝜷 𝟎 + 𝜷 𝟏 𝒈𝒅𝒊𝒔𝒕𝒅 + 𝜷 𝟐 𝒍𝒂𝒃𝒑𝒂𝒓𝒕𝒇𝒆𝒎 + 𝜷 𝟑 𝒇𝒆𝒎𝒑𝒂𝒓𝒍 + 𝜷 𝟒 𝒐𝒑𝒆𝒏 + 𝒖
Initially, the paper ran a regression equation without the control variable for the openness of a
country. After running the second regression equation with openness, it was found that while
the results for gdistd and labpartfem were significant at all levels of significance (p-value =
0), those for femparl were significant only at the 10% level, and finally, the results for open
were statistically insignificant.
An F-Test was then run to test for the joint significance of femparl and open. This test gave a
p-value of 0.18, which meant that the effect of open was more significant when it was tested
jointly with femparl than when it is tested for independently.
Results
Table B reports the results from the main logistic regression which seeks to explain the effect
of gender equality on economic growth. The project’s primary null hypothesis was that GDI
has no impact on Per Capita GDP. As seen in the table, holding participation in the labour
force and the parliament, and the openness of an economy fixed, an increase in the standard

5. deviation of the GDI leads to an increase in Per Capita GDP by 85%. The GDI is not only
practically significant but also statistically significant; with a p-value of 0, it rejects the
project’s primary null hypothesis that GDI has no impact on Per Capita GDP, at all
significance levels.
The result also reports that a 1 percentage point increase in female labour force participation
rate leads to a 3% decline in the GDP per capita. Although this does not cause a large
reduction in the GDP per capita, it is still surprising to see that the increase in the labour force
participation rate by women has a negative effect on the growth of an economy. Whilst
discrimination in the labour market leads to a reduction in average labour productivity, one
reason for the negative relationship could be that women are often involved in unpaid labour
which is included in the labour force participation rate but is not included in the GDP per
capita leading to the results being skewed (e.g.: Women involved in unpaid work on
agricultural fields).
An increase in female representation of the parliament by a percentage point leads to a 1%
increase in the GDP per capita. This is not a very high amount and therefore this increase is
not too significant practically.
One can also see that the control variable only leads to a small difference in the Beta values
of the independent variables.
The R squared for the equation is .6466 which demonstrates that 64% of the variation in the
per Capita GDP can be explained by the independent variables. Keeping aside the limitations
of this project, this result shows a large effect of GDI on per capita GDP.
Conclusion
This paper suggests that the challenge of increasing economic growth of a country is linked
to the role played by women in the society to a considerable extent. Discrimination towards

6. women in health, education and income not only harm the women concerned but also impose
a cost on the entire society. According to the results of this study, a greater degree of
confidence can be attached to certain measures of gender equality than others.
One must note that this paper consists of several limitations: omitted variable bias, limited
sample size, non-normal distribution of residuals and identification problem.
One of the primary limitations that this paper faces is due to the sample size of 105 countries
as this skews data towards developed countries that publish complete data. This data is also
not completely random as it only involves countries that have a parliament. Countries have
disparate cultures and systems and openness is only one control variable for these vastly
different nations.
Variables that impact per capita GDP such as legal rights of women could not be included in
the regression equation leading to omitted variable bias.
This paper also faced an identification problem due to the reverse causality between gender
equality and economic growth. To avoid this issue, the paper made use of lagged variables so
that the dependent variable could not affect the independent variables.
This paper used the Central Limit Theorem to assume that the residuals were distributed
normally although Graph A demonstrates that this is not the case.
Consequently, the results of this paper cannot be deemed entirely accurate.
Works Cited
Kabeer, Naila and Luisa Natalie. “Gender Equality and Economic Growth:Is there a Win-
Win?” LSE, 2013, http://www.lse.ac.uk/gender/assets/documents/research/choice-
constraints-and-the-gender-dynamics-of-lab/Gender-Equality-and-Economic-

7. Growth.pdf. Accessed 9 May 2018.
King, Elizabeth and Andrew Mason, “Engendering Development Through Gender Equality
in Rights, Resources, and Voice.” The World Bank and Oxford University Press,
2001,http://siteresources.worldbank.org/PGLP/Resources/Engendering_Development.
pdf. Accessed 10 Apr. 2018.
Klasen, Stephen and Francesca Lamanna. The Impact of Gender Inequality in Education and
Employment on Economic Growth: New Evidence for a Panel of Countries, Chicago
Policy Review, 2009, http://chicagopolicyreview.org/wp-
content/uploads/2014/09/Gender-and-Economic-Growth.pdf. Accessed 10 May 2018.
Morrison et al. “Gender Equality, Poverty and Economic Growth.” The World Bank,
2007,https://openknowledge.worldbank.org/bitstream/handle/10986/7321/wps4349.p
df?sequence=1&isAllowed=y. Accessed 10 May 2018.
Appendix
Table A: Summary Statistics

8. Table B: Regression Results
Graph A: Normality of Residuals