Is gender equality conducive to economic growth of developing countries?

Abstract

Gender equality is generally advocated on the basis of fairness and social justice. Its intrinsic value is widely recognized among academic and policy circles. However apart from its intrinsic value, it may also be considered important due to its significant implications for economic and development outcomes. In existing literature, despite a lot of discussion on its intrinsic value, relatively less focus has been paid to its functional aspect. This study is an attempt to empirically investigate the effect of gender equality on economic growth. By using panel data for developing countries of the world, we have found that gender equality has positive and significant effect on economic growth of developing countries.

Keywords:

• gender equality
• economic growth
• developing countries

Public Interest Statement

Gender equality is advocated on the basis of fairness, justice and social equity. Apart from its significance as an important end, gender equality can also work as a mean to achieve different development goals. This paper has investigated the effects of gender equality on economic growth of developing countries. The paper argues that promoting gender equality can be beneficial for the economic performance of developing countries.

1. Introduction

Gender equality is an important and one of the widely discussed issue all over the world. It can be perceived as a situation where there is no discrimination on the basis of gender and both men and women have equal treatment in all spheres of life. Promoting gender equality is one of the sustainable development goals (SDG) set by the United Nations and treated as an important public policy agenda by different countries of the world. Nonetheless, women are still behind men in different aspects of life particularly in developing countries of the world (UNDP, 2020). Gender equality is advocated on the basis of different types of arguments. At first place, it is considered important due to its intrinsic value by arguing that any discrimination on the basis of gender is against the basic principles of equity, social justice and fairness. Hence, promoting gender equality is important and should be viewed as an end in itself.

The second set of arguments draws upon the instrumental role of gender equality for the well-being and human development of society. Under this set of arguments, gender equality can help to achieve a number of development goals. The positive effects of gender equality on human development of a society are asserted by arguing that curbing gender discrimination would lead to women’s better educational attainments, better health outcomes, better employment opportunities, better earnings and greater involvement in decision making. With better education, better income and greater control over resources, women are more likely to spend a larger proportion of their income on education and health of their children. Female’s education may also bring down fertility rate and result will be better quality children who are well nourished, healthier and well educated. Thus, gender equality may prove beneficial for next generation’s educational attainments and health outcomes. In existing literature, both micro-level and macro-level studies provide support for this notion (Bobonis, 2009; Hoddinott & Haddad, 1995; Imai et al., 2014; Muchomba, 2017; Santoso et al., 2019).

The third set of arguments is related with the functional implications of gender equality for macroeconomic performance of a country in terms of its economic growth. It is argued that economic outcomes can be significantly affected by existing gender relations in the society through various direct and indirect channels (Hill & King, 1995; Klasen, 2000, 2002, 2006; Klasen & Lamanna, 2009; Klasen & Minasyan, 2017; Knowles et al., 2002; Mishra et al., 2020; Seguino, 2000a, 2000b; van Staveren, 2011; Walters, 1995). Hence, in addition to different factors such as endowment of natural and human resources, physical capital, population growth, technological advancement and institutional arrangements, the role of gender relations can be an important source to explain cross country growth differentials. However, two opposing point of views regarding the relationship of gender equality and economic growth have been reported by existing research studies. The first point of view suggests that gender equality helps to enhance economic growth (Brummet, 2008; E. King et al., 2007; Klasen & Lamanna, 2009; Mitra et al., 2015; Perrin, 2021). The second point of view suggest that despite being intrinsically important, gender equality does not necessarily promote economic growth rather it may be detrimental for economic growth (Blecker & Seguino, 2002; Busse & Spielmann, 2006; Seguino, 2000a, 2000b). The inconclusive and sometimes even contradictory empirical findings regarding the relationship of gender equality and economic growth largely seems to be due to two reasons. First, different proxies such as gender wage gap, gender inequality in education and gender differences of labour force participation have been used to capture the phenomenon of gender inequality by different research studies. Secondly, the sample used by different studies is different as the data of different set of countries have been utilized in these studies. For instance, gender inequality as measured by gender gap of labour force participation has been found detrimental for economic growth and low female labour participation has been identified as one of the main cause of low economic growth of the countries of North Africa as compared with East Asia; a region with comparatively higher female labour force participation (Esteve-Volart, 2004; Klasen & Lamanna, 2009). Similarly, the negative effects of gender inequality in education and earnings on economic growth have been postulated by different studies (Baliamoune-Lutz, 2006; Benavot, 1989; Brummet, 2008; Galor & Weil, 1996; Hill & King, 1995; King et al., 2007; Klasen, 2000; Klasen & Lamanna, 2009; Knowles et al., 2002; Lagerlof, 1999). Such effects have been supported (Benavot, 1989; Hill & King, 1995; Klasen, 2002; Knowles et al., 2002) by arguing that gender biases in education and earnings can affect the fertility decisions of households as well as educational attainments and health status of children which in turn can affect economic growth. Though the contradictory findings have also been reported where women’s education have been found negatively associated with economic growth (Barro & Lee, 1994; Barro & Sala-I-Martin, 1995; Perotti, 1996). The reported effects of gender wage differentials on economic growth are also of mixed nature in literature. Such differentials were found to be having positive effects on the economic growth of economies of South Asian region where lower wages for women as compared to men helped producers to produce cost-effective goods in export oriented industry by utilizing lower paid female workers. This helped countries of the region to increase their competitiveness by lowering per unit production cost and it boosted their exports. Increase in exports provided incentives for producers to invest more as it enhanced their profitability which further led to increase economic growth of these countries (Blecker & Seguino, 2002; Seguino, 2000a, 2000b). However opposing point of view is also present according to which gender wage gap can slow down the pace of economic growth through different channels (Baldwin & Johnson, 1992; Galor & Weil, 1996; Mitra-Kahn & Mitra-Kahn, 2008; Schober & Winter-Ebmer, 2011).

Existing studies have used different indicators such as gender wage gap, gender employment gap and gender education gap for studying the implications of gender inequality for economic growth. But because of the multidimensional nature of the phenomenon of gender inequality, none of these indicators can truly depict the nature and extent of gender inequality in a society. The existing gender relations in a society, on one hand, are linked with societal norms, values, culture, customs and traditions and on the other hand, these relations may also be affected by public policy being exercised (Pearse & Connell, 2016). Keeping in view the multiplicity of the phenomenon of gender inequality, a holistic measure covering different dimensions related with gender inequality is needed for studying its economic and development implications. This study aims to serve this objective by using an index of gender equality which has been developed by Indices of Social Development (ISD) database at International Institute of Social Studies, The Hague, Netherlands. Gender equality index (GEI) which has been constructed by taking into account 21 actionable as well as perception based indicators is one among the different Indices of Social Development (ISD) developed by ISS. Although some other composite gender indices such as Gender Inequality Index by UNDP, Social Institutions and Gender Index by OECD, Women’s Economic Opportunity Index by Economist Intelligence Unit and Global Gender Gap Index by World Economic Forum are also available yet the use of variety of qualitative and quantitative variables in the construction of GEI and its coverage for a large set of countries and for a reasonably large span of time¹ makes it a suitable choice for our study (Stotsky, 2016; van Staveren, 2013; van Staveren et al., 2014). We have used GEI to test our proposition that gender inequality has worked as an obstacle for the economic growth of developing countries of the world which are characterized by higher gender inequality as compared with developed countries and by promoting gender equality, developing countries can boost their economic growth. Thus, we want to empirically test the proposition: Does gender equality promote economic growth in developing countries?

2. Literature and theory

Different research studies have attempted to theorize gender as an important variable in the determination of economic growth and different direct and indirect channels have been suggested through which women’s status in the society may have important implication for economic growth. Intuitively, gender gap in education and literacy seems to be a detrimental factor for economic growth for a variety of reasons. If women are denied equal access to educational opportunities, then there may be a decrease in the overall stock of human capital in society. It can also result in inefficient allocation of resources if under the influence of social norms and cultural values a household’s investment decision for children’s education is biased in favour of less talented boys against more talented girls. Low human capital and misallocation of resources can slow down economic growth. Fertility decisions of households may also be affected by female’s education. Educated women are more likely to have fewer children which may result to increase in female’s labour force participation and hence can stimulate economic growth through the expansion of economic activities. A large body of literature provides empirical and theoretical support for the notion that gender bias in education can deteriorate economic growth directly as well as indirectly through its effects on fertility and next generation’s education (Baliamoune-Lutz, 2006; Benavot, 1989; Galor & Weil, 1996; Hill & King, 1995; King & Mason, 2001; King et al., 2007; Klasen, 2000; Klasen & Lamanna, 2009; Knowles et al., 2002; Lagerlof, 1999; Morrison, 2007; Schultz, 2002; Thomas, 1997).

However, opposing view has also been presented by some of empirical studies. Barro and Lee (1994), Barro and Sala-I-Martin (1995) and Perotti (1996) for example, report a negative sign of the coefficient of female education in their regression of economic growth. The findings reported by these studies have been questioned by subsequent studies (Brummet, 2008; Dollar & Gatti, 1999; Klasen & Lamanna, 2009; Knowles et al., 2002) on methodological grounds. Dollar and Gatti (1999), for instance, argues that results presented by R. J. Barro and Lee (1994) may be problematic due to the existence of multicollinearity arisen from strong correlation between male and female schooling in their data. They further point out that negative effect of female schooling on economic growth may disappear when a dummy variable for Latin America and East Asia is included in growth regression. They suggest that these puzzling findings may be due to combination of low economic growth and high female education in Latin America and high economic growth and low female schooling in East Asia. But this low economic growth in Latin America may be associated with some other factors instead of high female education. Similarly high economic growth in East Asia cannot be termed as an effect of low female schooling. Klasen and Lamanna (2009) points out that surprising effects of gender inequality for economic growth may be due to the reason that endogeneity of female and male education has not been considered by these studies. It is also argued that female’s education can affect economic growth by generating positive externalities for economic growth (King et al., 2007). By utilizing data of 72 developing and developed countries over the period 1965–1984, Brummet (2008) found that gender differences in mean years of schooling had a negative and significant effect on growth of annual GD per capita. Primary school enrolment was also found to be statistically significant to explain the differences of GDP per capita growth whereas secondary school enrollment was statistically insignificant. Karoui and Feki (2018) argue that gender inequality has retarding effect on economic growth of African countries whereas increase in school enrollment rate of girls can help to boost up economic growth.

The economic implications of gender bias in employment have also been documented by researchers and a mix kind of empirical evidence has been produced. Indicators such as female share of the total labour force, the ratio of female to male labour force participation rate, and the share of the female working age population in formal sector employment have been used to measure gender employment gap. A positive and significant effect of the female share in formal employment on economic growth has been reported by Klasen (2000). The coefficient of share of female in the total labour was positive but statistically insignificant.

By using a panel data over the period of 1961–1991 for Indian states, Esteve-Volart (2004) have found that female to male ratio in total workforce and women’s share in managerial positions have positive and significant effect on the economic growth of Indian states. The states which observed the highest growth rates were those states which had the highest gender equality in employment. Equality of economic opportunities and equality of economic and political outcomes have been found positively and significantly related with economic growth in a panel-data analysis of the countries of the world (Mitra et al., 2015). By utilizing cross country and panel growth regressions, Klasen and Lamanna (2009) have reported the negative effects of gender inequality in education and employment on economic growth. They argued that gender inequality in terms of education and employment is one of the major causes of slow economic growth in different regions of the world. According to them, slow economic growth in Middle East and North Africa as compared with East Asia may be due to the differences of female labour force participation rate and female’s education among these regions. Contrary to this, a negative effect of female share in labour force has been reported by Baliamoune-Lutz and McGillivray (2007) in their sample of Arab and Sub-Saharan African countries.

Mix kind of empirical results of the effects of gender wage gap on economic growth have been reported by different empirical studies. Seguino (2000a), Seguino (2000b) and Blecker and Seguino (2002) have reported a positive and significant effect of gender wage gap on economic growth of semi-industrialized export-oriented economies. The underlying mechanism put forward by these studies suggest that lower wages for women as compared to men in manufacturing sector reduces the per unit cost of production. The lower wages for women particularly in export-oriented industry increase the profitability of producers and competitiveness of a country in its external sector. The increased profitability of producers stimulates investment and increased competitiveness of the country leads to export expansion. Export earnings also provide resources to purchase modern and sophisticated technology which can be used in the production. This helps to stimulate economic growth through increase in exports (Erturk & Cagatay, 1995). The same argument has also been supported by Busse and Spielmann (2006). Thus, a point of view emerges that despite being intrinsically important, gender equality does not necessarily contribute to economic development (Duflo, 2012). Contrary to Duflo’s assertion who considers gender-affirmative policies as distortionary in their nature, Kabeer (2020) argues that empowering women through such policies can actually work as a catalyst for overall economic development. Empirical findings of Schober and Winter-Ebmer (2011) and Mitra-Kahn and Mitra-Kahn (2008) assert that gender wage gap may have adverse effects on economic growth. Some earlier studies (Baldwin & Johnson, 1992; Galor & Weil, 1996) have also described the channels through which gender wage gap can have a negative effect on economic growth. These channels include reduction in female labour force participation which may be the result of women’s hesitation to participate in economic activities due to their lower wages as compared to men. And an increase in fertility which may be a result of lower opportunity cost of having a child due to lower female wages. Increase in fertility rate would increase dependency ratio and lower the saving rate in an economy which may lead to reduced economic growth. Thus, the growth effects of gender discrimination in labour market can vary across countries depending upon the context. Gender discrimination where women have lesser opportunities of securing an employment than men can reduce female labour force participation. As a result of which, overall economic activities would be reduced and economic growth would remain low. Gender wage gap can also have the same negative effects on economic growth if such gap works as discouragement for female labour force participation. On the other hand, if producers can exploit gender wage differentials to increase their profitability and exports of their country by producing goods with lower average cost then such discrimination would have positive effects on economic growth. Nonetheless, positive growth effects of gender discrimination can not be used to justify such discrimination due to well recognized intrinsic value of gender equality.

Implications of economic growth and development for gender equality and women empowerment have also been discussed in different studies. This has led to the emergence of dominant point of view that overall economic development would lead to promote gender equality. Hence countries and societies with more economic development are likely to have better situation of gender equality. Duflo (2012), for example, argues that economic development would reduce gender discrimination particularly by reducing gender biases in intrahousehold distribution of food, health care and education. She asserts that gender discrimination at household level can possibly be an outcome of the resource constraints. Limited resources are to be distributed among household members and women can possibly be the first victim of cut in food, health and education expenditures. Economic development would ease resource constraint and households would no longer be forced to exercise discrimination against women. Kabeer (2016, 2020), however, argues that gender equality would not necessarily be an inevitable outcome of economic development because gender inequality is a complex phenomenon and its roots can be traced in the history, culture and social norms. She advocates for gender-affirmative policies to tackle the issue. It is pertinent to note that the positive effects of economic growth on gender equality are far less consistent in literature whereas the positive effects of gender equality on economic growth are more robust in different empirical studies. It implies that promoting gender equality in different dimensions of human lives such as education, health, employment and political participation can lead to better economic performance of the countries and can be considered as a win-win situation (Kabeer, 2016; Kabeer & Natali, 2013).

3. Methodology and data

In order to study the effects of gender equality on economic growth, data of developing countries of the world has been used. Our definition of developing countries is based upon the World Bank’s classification of countries whereby low income and lower-middle income countries have been treated as developing countries. The study has included all those low income and lower-middle income countries in the analysis for whom the data is available. The list of the countries used in the analysis is available in appendix (Table A1). The data used in the study spans over 25 years (1988–2012) with five-year intervals. We have used the growth rate of real GDP per capita (Gr) as dependent variable. Independent variables include initial income measured by one period lagged (five years lagged) value of log of GDP per capita (Y), investment (Inv) measured by gross capital formation as a percentage share of GDP, the variable of schooling (Sch) measured as mean years of schooling for population aged 15 years and above, government consumption (GC) taken as percentage share of GDP, trade openness (TO) measured by value of sum of imports and exports as percentage share of GDP, financial development measured by credit to private sector as percentage share of GDP and gender equality (GE) measured by gender equality index (GEI). The variables used are five-year averages except initial income, mean years of schooling and GEI. In case of GEI, the underlying variables used in the construction of the index are already averages of five years whereas data for mean years of schooling is available with five-year intervals. Data spread over 1988–2012 has been utilized for calculating five-year averages of our variables at five points of time i.e. 1990, 1995, 2000, 2005 and 2010. In the construction of averages for the year 1990, data for the period of 1988, 1989, 1990, 1991 and 1992 has been used. Same method has been used for the construction of averages for each point of time where each corresponding year lies in the middle of five-year period. Data is from World Development Indicators (WDI) except the data of mean years of schooling and GEI. Mean years of schooling is from R. J. Barro and Lee (2013) and GEI has been taken from Indices of Social Development database (ISD) constructed and maintained by International Institute of Social Studies (ISS), The Hague. GEI has been constructed by taking into account 21 quantitative as well as qualitative indicators (see Table A2 in appendix for list of indicators) related with gender relations in the society and by using matching percentile methodology (Frank et al., 2011; Lambsdorff, 2007). Its value ranges from 0 to 1 and a higher value implies higher gender equality. Detailed discussion on the methodology used to construct these indices is available in Foa and Tanner (2012). The rationale of the choice of GEI over other available indices and proxies for measuring gender equality lies in its unique methodology and its wide coverage for a large set of countries over a large period of time (Stotsky, 2016; van Staveren, 2013; van Staveren et al., 2014).

Panel data has been used by the study which has certain advantages over cross-sectional or time series data as it has both the cross-sectional and time specific dimension. Data is with some missing observations which makes our data set an unbalanced panel data. One way to estimate panel data is to use the following general form of panel least square as given in equation (1)(1) $Y_{\mathrm{i}\mathrm{t}}={\alpha }_{\hspace{0.17em}}+{\beta }_1{X}_{1\mathrm{i}\mathrm{t}}+{\beta }_2{X}_{2\mathrm{i}\mathrm{t}}+\dots \dots ..+{\beta }_k{\mathrm{X}}_{\mathrm{k}\mathrm{i}\mathrm{t}}+{\mu }_{\mathrm{i}\mathrm{t}}$(1)

(1) $Y_{\mathrm{i}\mathrm{t}}={\alpha }_{\hspace{0.17em}}+{\beta }_1{X}_{1\mathrm{i}\mathrm{t}}+{\beta }_2{X}_{2\mathrm{i}\mathrm{t}}+\dots \dots ..+{\beta }_k{\mathrm{X}}_{\mathrm{k}\mathrm{i}\mathrm{t}}+{\mu }_{\mathrm{i}\mathrm{t}}$(1)

However, the estimates of Ordinary Least Square can be inconsistent because of possibility of cross-sectional heterogeneity (Tolsma et al., 2009). Alternative solution in such situation is to use Fixed Effects (FE) or Random Effects (RE) Model. In FE Model, cross-sectional specific, time specific or both of these two effects can be used. Cross-sectional FE are applied to control for time-invariant characteristics of cross-sectional units. Thus, while applying cross-sectional FE, intercept is assumed to vary and slope is treated as constant across cross-sectional units. Whereas time-specific FE are applied to control for those effects which remains same across the cross-sectional units but vary over the period of time. Random effects assume that every cross-section is different in its value of error term (Porter & Gujrati, 2009). The Hausman test (Hausman, 1978) is applied to decide about the choice and appropriateness of FE or RE Model. The Hausman test suggests the use of FE Model in the case of this study. The general form of cross-sectional FE Model has been provided in equation (2)(2) $Y_{\mathrm{i}\mathrm{t}}={\alpha }_{\mathrm{i}}+{\beta }_1{X}_{1\mathrm{i}\mathrm{t}}+{\beta }_2{X}_{2\mathrm{i}\mathrm{t}}+\dots \dots ..+{\beta }_k{\mathrm{X}}_{\mathrm{k}\mathrm{i}\mathrm{t}}+{\mu }_{\mathrm{i}\mathrm{t}}$(2) .

(2) $Y_{\mathrm{i}\mathrm{t}}={\alpha }_{\mathrm{i}}+{\beta }_1{X}_{1\mathrm{i}\mathrm{t}}+{\beta }_2{X}_{2\mathrm{i}\mathrm{t}}+\dots \dots ..+{\beta }_k{\mathrm{X}}_{\mathrm{k}\mathrm{i}\mathrm{t}}+{\mu }_{\mathrm{i}\mathrm{t}}$(2)

The specific form of econometric model used in the study is as given in equation (3)(3) $G{r}_{\mathrm{i}\mathrm{t}}={\alpha }_{\mathrm{i}}+{\beta }_1{Y}_{\mathrm{i}\mathrm{t}-1}+{\beta }_{2}In{v}_{\mathrm{i}\mathrm{t}}+{\beta }_3\mathrm{S}\mathrm{c}{\mathrm{h}}_{\mathrm{i}\mathrm{t}}+{\beta }_4\mathrm{G}{\mathrm{C}}_{\mathrm{i}\mathrm{t}}+{\beta }_5\mathrm{T}{\mathrm{O}}_{\mathrm{i}\mathrm{t}}+{\beta }_6\mathrm{F}{\mathrm{D}}_{\mathrm{i}\mathrm{t}}+{\beta }_7\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{t}}+{\mathrm{e}}_{\mathrm{i}\mathrm{t}}$(3) . Cross-sectional FE have been used to control for time-invariant characteristics of the countries included in the empirical analysis. FE Model with robust standard errors has been applied to cope with the problem of heteroskedasticity.

(3) $G{r}_{\mathrm{i}\mathrm{t}}={\alpha }_{\mathrm{i}}+{\beta }_1{Y}_{\mathrm{i}\mathrm{t}-1}+{\beta }_{2}In{v}_{\mathrm{i}\mathrm{t}}+{\beta }_3\mathrm{S}\mathrm{c}{\mathrm{h}}_{\mathrm{i}\mathrm{t}}+{\beta }_4\mathrm{G}{\mathrm{C}}_{\mathrm{i}\mathrm{t}}+{\beta }_5\mathrm{T}{\mathrm{O}}_{\mathrm{i}\mathrm{t}}+{\beta }_6\mathrm{F}{\mathrm{D}}_{\mathrm{i}\mathrm{t}}+{\beta }_7\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{t}}+{\mathrm{e}}_{\mathrm{i}\mathrm{t}}$(3)

In addition to FE Model, Two Stage Least Square (2SLS) approach and Generalized Method of Moments (GMM) approach have also been used. While doing so, the variable of gender equality (GE) has been treated as endogenous variable. First lag of gender equality (GE_t-1) and institutional quality (IQ_t-1) whereas fourth lag of percentage of population with complete secondary school education ($S{S}_{\mathrm{t}-4})$ have been used as instrumental variables. The state of gender relations in any society is expected to be affected by the effectiveness of formal institution and cultural as well as social norms (Inglehart et al., 2017; Jayachandran, 2021; Milazzo & Goldstein, 2019; Pearse & Connell, 2016). Education is an important tool to bring changes in cultural and social values of any society (Chatard & Selimbegovic, 2007; Kabeer & Natali, 2013). That’s why, in addition to institutional quality, we have also used the variable of percentage of population with complete secondary schooling as instrumental variable (Henry et al., 2013). An index of government stability, measured and reported by International Country Risk Guide (ICRG) has been used as a measure of institutional quality. This index, measured on a scale of 12 points, assesses the ability of governments to stay in office and to carry out their declared programmes successfully. The data of percentage of population with complete secondary schooling (SS) is from R. J. Barro and Lee (2013). While the efficacy of formal institutions is expected to affect gender equality in a society with relatively a lesser or no time lag, the changes in cultural and social norms which can affect gender relations in the society are expected to take place during relatively longer period of time. This justifies the use of first lag of institutional quality and fourth lag of percentage of population with complete secondary schooling as instrumental variables. Under identification test, Hansen statistic and Endogeneity test have also been conducted for the validation of the instruments used in the analysis.

Table 1 provides the descriptive statistics of the variables used in our study.

Table 1. Descriptive statistics

Variable	Mean	Std. Dev.	Min	Max
Gr	2.245	2.617	−7.997	13.509
Yt-1	6.54	.658	5.111	7.964
Inv	21.533	7.005	3.009	46.815
Sch	4.907	2.402	.8	11.282
GC	12.309	3.904	3.686	22.924
TO	72.286	33.333	16.383	222.258
FD	24.022	16.497	2.11	99.51
GE	.645	.098	.212	.935

Pairwise correlation of variables of the study is provided in Table 2.

Table 2. Correlation matrix

Variables	Gr	Y	Inv	Sch	GC	TO	FD	GE
Gr	1.00
Yt-1	−0.14	1.00
Inv	0.44	0.14	1.00
Sch	0.25	0.43	0.25	1.00
GC	−0.21	0.12	0.18	0.06	1.00
TO	0.03	0.27	0.37	0.39	0.33	1.00
FD	0.14	0.44	0.37	0.44	0.13	0.30	1.00
GE	0.18	0.20	0.28	0.51	0.09	0.39	0.32	1.00

4. Results and Discussion

The results of our FE, 2SLS and GMM estimations have been reported in Table 3.

Table 3. Gender Equality and Economic Growth (Dependent Variable: Growth Rate of Real GDP Per Capita (Gr))

Variables	FE Estimations	2SLS Estimations		GMM Estimation
	Gr	Gr	GE	Gr	GE
Yt-1	−8.824***	−0.718***	−0.012	−0.712***	−0.012
	(−6.59)	(−2.67)	(−1.18)	(−2.74)	(−1.18)
	[1.338]	[0.269]	[0.011]	[0.260]	[0.011]
Inv	0.087*	0.201***	0.000	0.200***	0.000
	(1.82)	(5.33)	(0.04)	(6.21)	(0.04)
	[0.048]	[0.038]	[0.001]	[0.032]	[0.001]
Sch	1.605***	0.064	0.008	0.059	0.008
	(5.72)	(0.43)	(1.66)	(0.40)	(1.66)
	[0.281]	[0.149]	[0.005]	[0.148]	[0.005]
GC	−0.025	−0.056	−0.003**	−0.053	−0.003**
	(−0.38)	(−1.27)	(−2.28)	(−1.19)	(−2.28)
	[0.065]	[0.044]	[0.001]	[0.044]	[0.001]
TO	−0.001	−0.026**	0.001**	−0.025**	0.001**
	(−0.04)	(−2.10)	(2.56)	(−2.07)	(2.56)
	[0.012]	[0.012]	[0.000]	[0.012]	[0.000]
FD	0.049***	−0.009	0.001	−0.009	0.001
	(3.61)	(−0.82)	(1.59)	(−0.92)	(1.59)
	[0.014]	[0.011]	[0.000]	[0.010]	[0.000]
GE	4.032**	13.101**		13.635**
	(2.03)	(2.21)		(2.48)
	[1.985]	[5.922]		[5.505]
$G{E}_{\mathrm{t}-1}$			0.342***		0.342***
			(4.56)		(4.56)
			[0.075]		[0.075]
$I{Q}_{\mathrm{t}-1}$			0.011***		0.011***
			(4.16)		(4.16)
			[0.003]		[0.003]
$S{S}_{\mathrm{t}-4}$			0.002**		0.002**
			(2.48)		(2.48)
			[0.001]		[0.001]
Constant	45.788***	−3.397	0.340***	−3.801	0.340***
	(5.71)	(−0.96)	(3.81)	(−1.08)	(3.81)
	[8.021]	[3.556]	[0.089]	[3.533]	[0.089]
R^2	0.48	0.39	0.54	0.38	0.54
Hausman test	(139.40) 0.000
Under identification test		11.023 (0.00)		11.023 (0.00)
Hansen J statistic		2.090 (0.3537)		2.090 (0.3537)
Endogeneity test		3.435 (0.06)		3.435 (0.06)

Note: ***, ** and * denote the significance at 1%, 5% and 10% level, respectively. standard errors are in square brackets while t- statistics are provided in parentheses.

Our results presented in Table 3 reveal that initial income is negatively and significantly associated with economic growth. This confirms the existence of conditional convergence among developing countries. Investment is positively and statistically significantly related with economic growth. The variable of schooling has positive and significant relationship with economic growth in fixed effects model but the variable turns out to be insignificant in 2SLS and GMM estimations. Government consumption and trade openness are negatively related with economic growth. Financial development shows a positive and significant effect on economic growth in fixed effects estimations but the coefficient of financial development is statistically insignificant in 2SLS and GMM estimations. The variable of gender equality has positive and significant relationship with economic growth in fixed effects, 2SLS as well as GMM estimations. This confirms our proposition that promoting gender equality is conducive to economic growth of developing countries. It is also evident from our empirical results that gender relations in a society are affected by the quality of formal institutions as well as overall level of education in the society. Effective institution can be instrumental to create an inclusive environment for all segments of society and women can also benefit from such inclusive environment. Increase in the overall education in a society can help to bring social changes required for gender equality. Thus, education can have dual effects. On one hand, it can boost economic growth by increasing the stock and quality of human capital. On the other hand, it can improve the situation of gender equality which would lead to increase in economic growth.

The positive effects of gender equality on economic growth can work through various channels. Gender equality in education can ensure the access to education facilities for all individuals irrespective of their gender. It would increase the overall stock of human capital in the society. Keeping in view the importance of human capital in the process of economic growth, an increase in the stock of human capital would be helpful to boost up economic growth. Gender inequality embedded in social norms and values may affect the investment decision of households in children’s education. If, influenced by social norms, boys are given preferential treatment over girls then this decision may be biased in favour of male children. This leads towards an investment gap in male and female education, which can retard economic growth. Contrary to this, gender equality in education would also lead towards allocative efficiency. If parents’ decision of investment in their children’s education is not influenced by any gender bias, then they would invest in the education of more able and talented children irrespective of the gender of the child. This would lead towards an optimal allocation of household’s resources. Improved allocative efficiency would be helpful to boost up economic growth. Gender equality in education can also contribute to economic growth because the positive effects of female education on economic outcomes may be more pronounced as compared with male education. Female education is important not only because it can add to overall stock of human capital in the society but also because of its positive externalities. Besides direct effects of female education, it can also have indirect effects on economic growth by reducing fertility and by improving next generation’s education and health. There is substantial empirical evidence in existing literature to support the above-described channels through which gender equality in education can prove beneficial for economic growth.

Women’s increased participation in political process and their involvement in decision making at household level as well as national level can also be a crucial factor for the process of economic development. Increase in their bargaining power will be helpful to design gender-affirmative policies and gender-inclusive environment. Gender equality can also be helpful to increase economic growth by reducing gender discrimination in the labour market. It would reduce gender wage gap and gender employment gap. Increased employment opportunities and higher wages for women may also affect economic growth through human capital creation in the society. There is ample empirical evidence that women are more likely to spend a large proportion of their income on children’s education and health. With increase in women’s income and with their greater control over resources, more will be spent on children’s health and education and the result would be increase in economic growth through increase in human capital formation. On the other hand, gender inequality as measured by gender specific differences of labour force participation rate and gender wage gap can affect economic growth. Reduction in female labour force participation rate which may arise due to limited employment opportunities for women can have impeding effects on economic growth. Economic growth may also be adversely affected due to gender wage gap through reduction in female labour force participation. If women are offered wages lower than men then they may hesitate to contribute in labour market. Fertility decision is also closely linked with women’s wages and their participation in labour force. Opportunity cost of having children increases with an increase in women’s wages. This can lead to enhance economic growth by slowing down population growth. Thus, apart from its intrinsic value, gender equality may be termed as important because of its functional relevance for economic outcomes.

5. Conclusion

Gender inequality is an important feature of most of the societies of developing world where women are behind men in almost all spheres of life. The roots of gender inequality can be traced in the cultural and social norms of these societies. The patriarchal norms have restricted women’s participation in different walks of life. They have lesser access to education and health facilities, lesser control over resources and lower participation in economic activities as compared to men. This has worked as detrimental factor for economic development of these countries. Gender inequality has severely jeopardized the prospects of economic growth of the developing countries by pushing them into a vicious cycle of low economic opportunities for women and low economic growth. The study provides persuasive evidence that gender equality is important from the perspective of economic growth. The evidence points towards the untapped potential of investing in gender equality for the economic growth and development of the developing world. The issue of gender justice, protecting women’s rights to equitable treatment as men should take precedence not only because it is an end in itself but also because it can be crucial to achieve the target of economic growth and development. The objective of raising the tempo of economic development requires fundamental social and public policy changes in the direction of gender equality. Inequality of gender in the form of unequal access to labour market, earning opportunities, food, health and education as a byproduct of differential treatment of women in developing world has to be eliminated. Ensuring the availability of better opportunities of education, health and employment for women, increase in their political participation and bargaining power at household level can lead towards better socioeconomic status of women. This would be conducive for economic growth. Hence, an inclusive environment for women would create a virtuous cycle of gender equality and better economic growth.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Zahid Pervaiz

Zahid Pervaiz is Associate Professor at the Department of Economics, National College of Business Administration and Economics (NCBA&E), Lahore, Pakistan. His research interests include income inequality, inequality of opportunities, gender inequality, social exclusion, economics of health and education, social cohesion and various socioeconomic issues pertaining to economic development.

Shahla Akram

Shahla Akram is affiliated as a researcher with the Department of Economics, National College of Business Administration and Economics (NCBA&E), Lahore, Pakistan. Her research interests include economic development, income inequality, inequality of opportunities and social cohesion.

Sajjad Ahmad Jan

Sajjad Ahmad Jan is working as Assistant Professor at the Department of Economics, University of Peshawar, Pakistan. His area of research includes social exclusion, socioeconomic marginalization, vulnerability to poverty, formal and informal social security mechanisms, regional inequality and political economy of conflicts.

Amatul R. Chaudhary

Amatul R. Chaudhary is Professor of Economics at the Department of Economics, National College of Business Administration and Economics (NCBA&E), Lahore, Pakistan. Her current research focuses on women empowerment, gender inequality, dynamics of population and manpower, economic growth and human development.

Notes

1. Gender Equality Index has been constructed by International Institute of Social Studies, The Hague by utilizing matching percentile methodology (Frank et al., 2011; Lambsdorff, 2007). In the construction of the index, 21 variables taking from different data sources have been used. A detailed discussion on the data sources and methodology of index is available at www.indsocdev.org. Data spanning over 25 years (1988–2012) has been utilized for this purpose. The index is available at five point of times, that is, 1990, 1995, 2000, 2005 and 2010. At each point of time, the index has been constructed by taking the five years averages of underlying variables used in the construction of index in a way that each corresponding year lies in the middle of five years. For example, index for the year 1990 has been developed with the help of data of underlying variables for the year of 1988, 1989, 1990, 1991 and 1992. Same procedure has been adopted for the construction of index at other points of time. The value of index ranges between 0 and 1 where a value closer to 1 is an indication of higher gender equality.

Appendix

Table A1. List of countries

Serial No.	Country	FE estimations	2SLS estimations	GMM estimations
1	Armenia	✓	✓	✓
2	Bangladesh	✓	✓	✓
3	Benin	✓
4	Bolivia	✓	✓	✓
5	Burundi	✓
6	Cambodia	✓
7	Cameroon	✓	✓	✓
8	Congo, Rep.	✓	✓	✓
9	Côte d’Ivoire	✓	✓	✓
10	Egypt, Arab Rep.	✓	✓	✓
11	El Salvador	✓	✓	✓
12	Ghana	✓	✓	✓
13	Guatemala	✓	✓	✓
14	Guinea	✓
15	Guyana	✓	✓	✓
16	Haiti	✓
17	Honduras	✓	✓	✓
18	India	✓	✓	✓
19	Indonesia	✓	✓	✓
20	Kenya	✓	✓	✓
21	Kyrgyz Republic	✓
22	Lao PDR	✓
23	Lesotho	✓
24	Liberia	✓	✓	✓
25	Malawi	✓	✓	✓
26	Mali	✓	✓	✓
27	Mauritania	✓	✓	✓
28	Moldova	✓	✓	✓
29	Mongolia	✓	✓	✓
30	Morocco	✓	✓	✓
31	Mozambique	✓
32	Nepal	✓	✓	✓
33	Nicaragua	✓	✓	✓
34	Niger	✓	✓	✓
35	Pakistan	✓	✓	✓
36	Papua New Guinea	✓	✓	✓
37	Paraguay	✓	✓	✓
38	Philippines	✓
39	Rwanda	✓	✓	✓
40	Senegal	✓	✓	✓
41	Sierra Leone	✓	✓	✓
42	Sri Lanka	✓	✓	✓
43	Sudan	✓
44	Swaziland	✓	✓	✓
45	Syrian Arab Republic	✓
46	Tajikistan	✓	✓	✓
47	Tanzania	✓	✓	✓
48	Togo	✓	✓	✓
49	Uganda	✓	✓	✓
50	Ukraine	✓	✓	✓
51	Vietnam	✓	✓	✓
52	Yemen, Rep.	✓	✓	✓
53	Zambia	✓	✓	✓
54	Zimbabwe	✓	✓	✓

Table A2. Indicators used in the construction of gender equality index

Indicator	Source
Percentage agreeing that a married man has a right to beat his wife and children	Afrobarometer
Percentage of respondents who tend to agree or strongly agree that “women have always been subject to traditional laws and customs, and should remain so”.	Afrobarometer
Percentage of respondents who tend to agree or strongly agree that “women should have the same chance of being elected to political office as men”.	Afrobarometer
Rating on level of women’s economic rights	CIRI
Rating on level of women’s social rights	CIRI
Ratio of average female to male wages, across all available labour categories	International Labour Organisation
Percentage of women who agree that women have the same chance as men to get a good job in their country	Latinobarometer
Percentage of women who agree that women have the chance to earn the same salary as men in their country	Latinobarometer
Percentage of women who agree that women have the same chance as men to get a good education in their country	Latinobarometer
Proportion of employers and managers who agree or strongly agree that when jobs are scarce, men have more right to a job than women	World Values Surveys
Proportion of those of voting age who agree or strongly agree that on the whole, men make better political leaders than women do	World Values Surveys
Proportion of parents who agree or strongly agree that a university education is more important for a boy than a girl	World Values Surveys
Proportion of employers and managers who agree or strongly agree that on the whole, men make better business executives than women do	World Values Surveys
Percentage of labour force that is female	World Development Indicators
Ratio of females among legislators, senior officials and managers	International Labour Organisation
Ratio of females in professional jobs	International Labour Organisation
Ratio between female and male primary school enrolment	World Development Indicators
Ratio between female and male secondary school enrolment	World Development Indicators
Ratio between female and male tertiary educational enrolment	World Development Indicators
Ratio between adult female and male literacy rates	World Development Indicators
Ratio between adult female and adult male mortality rates	World Development Indicators

Source: Indices of Social Development (ISD) maintained by International Institute of Social Studies, The Hague.