Foreign Aid and Institutional Quality towards Reducing Gender-Based Violence

Abstract

Studies have shown a possible link between women achieving some degree of economic independence through empowerment and a decline in gender-based violence (GBV). However, inadequate funding has been a major challenge in developing economies, while foreign aid has been seen as an alternative funding source. Foreign aid for promoting women’s rights and gender equality is improving in African countries. Yet, according to official statistics, the frequency of GBV in Africa is among the highest in the world. Given this, this paper examines the impact of women’s aid and institutional quality on factors that can reduce GBV using both cross-sectional autoregressive lags to investigate short- and long-run implications and the Panel Vector Correction Model to examine the shock effect of aid on other variables. Also, the mediating role of institutional quality and women’s aid are considered. The metrics used for the factors that reduce GBV are female children out of school, the secondary school gender parity index, the genital mutilation prevalence rate, and the metrics for institutional quality are the rules of law and government effectiveness. The implications from the findings show that enhancing the legal structure is very important in Africa, given the low coefficient values to reduce discrimination against the right to education; more should be done to increase the enrolment of female students through the maximisation of women’s aid in order to achieve the short- and long-term objectives of reducing gender violence. The findings also show that the rule of law significantly aids in the reduction of female genital mutilation in the short and long run; government effectiveness is insignificant. Also, they show that women’s aid also reduces female genital mutilation, albeit at a low rate. The results of this study call for strong enhanced government support and funding to end the practice of female genital mutilation and discrimination against female education in the short and long run.

1. Introduction

The United Nations (1993) defines gender-based violence (GBV) against women as any act of violence that causes or is likely to cause any damage or suffering to women; this includes threats, coercion, or the unfair deprivation of liberty, whether in public or private life. GBV comprises emotional abuse and lack of economic participation and physical assault (Bannister and Moyi 2019). To further stress, GBV is also defined as an abuse of social, economic, and political rights based on standard international guidelines, such as the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW). The subject of women’s empowerment has existed since the 1970s. Women comprise about half of the adult population in emerging nations, yet it is noted that they lack vocational skills and adequate education (Asaleye and Strydom 2022); this makes women unable to uphold the control of resources and makes them vulnerable to GBV. The Sustainable Development Goals (SDGs) include women’s empowerment as a priority and a key indicator of societal transformation (Awoa Awoa et al. 2022). Violence against women at initial stages aims to control and subjugate women to the rules and norms enforced by the aggressor based on societal, cultural, and structural factors rather than to cause harm (Acosta 2020). As a result, this violence persists throughout time and results in physical, mental, and sexual assaults.

Men and women have been affected by GBV, but women are substantially more susceptible since violence perpetuates and reflects pre-existing gender disparities in economic, political, and social participation (World Bank 2019). While there has been a decline in GBV in developed economies due to the closing of the gender gap in economic participation, discrimination against women and girls is still pervasive in African nations (Olojede et al. 2020). Due to the character and cultures of the African nations, the region has not succeeded in its battle against GBV. However, there are strong campaigns and efforts against GBV. Intimate partner violence against women affects 30 per cent of women globally (WHO 2021), with the worst cases occurring in Southeast Asia and sub-Saharan Africa (Orpin et al. 2020; WHO 2021). The prevalence of GBV against women also appears to be linked to women’s low economic independence and educational attainment (Obiagu 2023).

UN Women (2022) recognised six contributing points for the GBV increase in Africa, including a lack of government support; a lack of resources; a lack of knowledge; a shortage of education; the acceptance of cultural norms and beliefs; and, lastly, gender discrimination. Likewise, according to UN Women (2022), gender equality will always be an unfulfilled dream until the international community shows more dedication. More so, the report by UN Women stated that it would take an additional 286 years to overhaul legislative structures to foster, implement, and evaluate gender equality in society at the present pace of change. Approximately 12.5 per cent of women and girls between the ages of 15 and 49 experienced physical or sexual abuse from a romantic relationship in the previous year, and development must proceed 17 times more than its current situation in both 2022 and the last ten years to stop child marriage by 2030. Another form of GBV is female genital mutilation, which is considered violence against women and girls (Tordrup et al. 2022; Williams-Breault 2018). Over 200 million girls and women have been estimated to be victims of female genital mutilation globally (World Health Organisation 2023). Numerous human rights, such as the Universal Declaration of Human Rights, the Convention on the Rights of the Child, and the Convention on the Elimination of All Forms of Discrimination Against Women, stressed that female genital mutilation violates human rights. The six contributing factors that cause GBV, as stated by UN Women (2022), can be addressed by government effectiveness, the rule of law, and adequate financing. The importance of quality institutions and adequate financing for women’s empowerment has been emphasised in the empirical literature (Asaleye and Strydom 2022; Edgell 2017; Hidrobo et al. 2016; Kabeer 1999; Sen 1990). Likewise, studies have stressed that women’s empowerment may reduce the rate of GBV in societies (Kabeer 1999; Sen 1990; Hidrobo et al. 2016).

GBV was a major global issue before the COVID-19 outbreak and became more severe afterwards. The ability for many women to escape violent circumstances has been further hampered by rising economic insecurity. The likelihood of child marriage and female genital mutilation have also increased because of the economic and social instability brought on by COVID. Given the problem of financial constraints in Africa, foreign aid is seen as an alternative source of revenue. Strong evidence suggests that foreign aid is effective when implemented properly (Elbert 2013), and one of the factors responsible for this is the role of institutions. Institutions include government bodies, regulatory agencies, legal frameworks, and other structures that shape the functioning of a society or a specific sector (Acemoglu et al. 2005; Asaleye et al. 2023), which are measured by institutional quality (Kaufmann et al. 2010). A country’s improved institutional quality implies well-designed, transparent, and accountable means that are capable of achieving long-term growth and development without corruption or inefficiency (Acemoglu et al. 2005). The effectiveness of the nation’s institutional quality is important for social and economic growth and achieving any Millennium Development Goals (MDGs) (Asadullah et al. 2014). The Foreign Assistance Act was created by the US Congress in 1961 to manage long-term economic and humanitarian aid to developing nations. Reducing poverty, strengthening democracy, establishing market economies, fostering security, dealing with crises, and enhancing the quality of life are among the objectives of USAID in emerging economies. Increasing funding for female empowerment is one of the biggest factors driving positive change against discrimination against women (UN Women 2022). Also, strengthening institutions in emerging economies may help law enforcement tackle gender-based violence more effectively. In addition, most empirical studies have focused on the impact of COVID-19, health-related issues, and political empowerment on GBV (Hicks et al. 2016; Hornset and de Soysa 2022; Bechange et al. 2021; Johnson et al. 2020; Maruta et al. 2020; Muluneh et al. 2020; Tsapalas et al. 2021). Studies on women’s empowerment, institutional quality and foreign aid are still growing. Given this, this study examines the impact of foreign aid and institutional quality on women’s empowerment to reduce GBV in Africa. The specific objectives are as follows:

• To investigate the short- and long-run implications of foreign aid and institutional quality on women’s empowerment and GBV;
• To analyse the mediating role of institutional quality and foreign aid on women’s empowerment and GBV;
• To examine the shock effect of foreign aid on women’s empowerment and GBV.

This study proposes an interactive effect of institutional quality and foreign aid on women’s empowerment and GBV. For example, improved institutional quality regarding government effectiveness and the rule of law will make the environment conducive for policy implementations. Also, external funding in the form of foreign aid diverted to promote women’s empowerment and reduce GBV should be invested accordingly to meet target objectives. Because of this, the effects of institutional quality and foreign aid on women’s empowerment and GBV may be more beneficial when combined than when applied alone.

This article begins with an introduction, followed by a Literature Review presentation in the second part. The third segment discusses the Theoretical Framework and Model Specification, while the fourth part presents and discusses the Empirical Results. Lastly, the fifth part summarizes the Conclusion and Policy Recommendations.

2. Literature Review

Numerous studies have examined the dynamic relationship between gender-based violence (GBV) and health-related issues. By contrast, few studies have focused on how women’s empowerment can reduce GBV, together with the role of the institutions and foreign aid. Although a few studies concerning GBV and women’s empowerment reported indecisive outcomes, this may be due to the choice of estimation, region, and type of empowerment. The literature review in this section is carried out in three sub-sections: first, to establish the connection between women’s empowerment and GBV; second, to provide an account of the previous studies regarding empowerment, foreign aid, institutional quality, and GBV; and third, to show gaps identified in the empirical literature.

2.1. The Connection between Women’s Empowerment and Gender-Based Violence

In Africa, the male occupies a more important place in most cultures. As a result, gender-based violence (GBV) has been indirectly allowed and is seen as acceptable behaviour in most societies, in order to preserve male dominance over women. Theoretically, resource theory shows the likelihood of reducing GBV via women’s empowerment. According to Kabeer (1999) and Sen (1990), it is presumable that women rely on males for financial assistance and put up with some degree of violence in exchange in places where the marriage or home is viewed as a setting where resources are negotiated. By this, the violence risk towards women may decrease as their financial dependency diminishes if their income rises through empowerment. Another argument in support of this is given by Hidrobo et al. (2016), who stressed that domestic violence often takes place because of women’s low self-esteem and frustration on the side of the men to meet daily needs and provide for the family; this could also be addressed by women’s empowerment. Increasing women’s involvement in assistance to meet family needs would relieve men of financial stress, which could therefore reduce GBV. Contrary to these arguments, men may believe that their social, political, or family positions are in danger when a woman is financially empowered, which may result in an attempt to regain control by abusing their partners, using violence to obstruct women’s participation in the market, confiscating their money, or imposing control over how money is managed (Hidrobo et al. 2016).

The Sustainable Livelihoods Model also shows the connection between women’s empowerment and gender-based violence; this model stresses economic empowerment through human and physical capital as an essential factor in improving livelihoods (United Nations Development Programme 2015). Regarding GBV, women gaining financial autonomy through empowerment may reduce GBV (Kabeer 2005). Conversely, financial dependence may worsen vulnerability to GBV. Social and human capital, integral to the model, underline the importance of education and community support in the form of foreign aid (Duflo 2012; Asaleye et al. 2023). In addition to this, educated and socially connected women often possess enhanced decision-making abilities and are more likely to resist GBV. This model also stresses the environmental dimension, which can be significant with regard to a safe and sustainable physical environment for women to thrive without fear.

It is worth mentioning that GBV against women extends beyond just physically hurting the victim. A woman’s self-esteem is psychologically damaged and diminished by the fear of more abuse, making it difficult for her to protect herself or stop the abuser. Additionally, it amounts to a breach of her human rights, outlined in the 1948 Universal Declaration of Human Rights. Women’s empowerment can increase women’s self-esteem and promote human rights (Asaleye and Strydom 2022). Empirical studies have shown a connection between women’s empowerment and GBV (Bolis and Hughes 2015; Cherry and Hategekimana 2013; Gupta et al. 2013; Magar 2003). Magar (2003) shows that empowerment helps married women handle male oppression in their homes and, in the long run, reduces GBV. In a similar study by Gupta et al. (2013), it was reported that women empowered through group savings programs in rural Côte d’Ivoire were less affected by GBV.

Similar to the study of Gupta et al. (2013), Cherry and Hategekimana (2013) documented that Rwandan cooperative initiatives minimise female GBV. More so, Bolis and Hughes (2015) state that empowering women is important. Still, its implication on domestic abuse may have conflicting effects on the likelihood; therefore, it depends on the situation and circumstances. Race, class, and sexual orientation are just a few examples of identity variables that affect GBV and are embedded in societal power structures, which women’s empowerment can also influence. The outcome of the findings of Bolis and Hughes (2015) state that the empowerment process of women to reduce GBV is able to have a significant effect, this can be translated that improving the financial situation of women would have a negative effect on GBV. Bolis and Hughes (2015) concluded that women’s economic empowerment could reduce violence against women if it takes the appropriate channels and undergoes a growth process.

Nevertheless, despite the strong connection that is present between women’s empowerment and the reduction in GBV, the type of empowerment initiative matters, and this is usually most constrained by financial availability. In this situation, women’s empowerment may not yield desirable outcomes, as observed in the study of Bannister and Moyi (2019). Bannister and Moyi (2019) used quantitative data from questionnaires given to married women who participated in neighbourhood economic empowerment groups and community stakeholders in order to investigate incidences of GBV and its relationship to women’s empowerment in Makhokho, Western Kenya. According to Bannister and Moyi (2019), there is no evidence to indicate a link between GBV and empowerment, and the relationship between GBV and other factors, including age, education, and the number and age of children, are insignificant. According to Bannister and Moyi (2019), there is a significant connection between the frequent drinking habits of a woman’s husband and GBV, and the authors concluded that efforts to lower GBV should aim to reduce the alcohol intake of the husband.

Consequently, men and women are unequal in power, level of education, and economic participation. An individual, company, or community may change and attain progressive achievements because of empowerment. Empowerment encourages engagement in various productive activities and provides more power, income, and self-esteem (Magar 2003). The fundamental components of empowerment include knowledge, abilities, and political awareness. This includes psychological attitudes such as confidence and self-efficacy. According to Kabeer (1999), women’s empowerment is the process through which they gain the ability to make wise decisions about their lives. In addition, feminist theorists argue that, for women to be fully empowered, gender roles require a transformational process in order for women’s traditionally inferior position no longer exist, especially in the family (Kieffer 1984; Zimmerman 2000). This suggests that empowerment through developmental growth processes has a greater likelihood of reducing GBV.

From the preceding information derived from a synthesis of female economic empowerment from the literature and sociocultural ideas in Africa, women’s economic empowerment can reduce GBV. Women’s household bargaining power and ability to end abusive relationships are increased by it. Also, it helps women acquire abilities that enable them to navigate gender power dynamics in the home, and family poverty declines. Culturally, in the African context, this may help communities change their perspectives and power abuse between the sexes as well as lessen the acceptability of female tolerance for domestic violence.

2.2. Empowerment, Foreign Aid, Institutional Quality, and Gender-Based Violence

The idea of empowerment relates to how men and women hold power in society and how they might access and manage resources (Kamal 2011). There is a strong connection between investment in female human capital, resources (in the form of foreign aid), and GBV. According to Kabeer (1999), women’s empowerment is essential due to deprivation, financial constraints, and lack of quality institutions, which have limited females’ decision-making. The ability of women and girls to equitably and securely participate in economic activities is impacted by bias against women in social, cultural, and economic settings. Financial resources often constrain efforts to promote women’s empowerment in the economy. GBV in these situations can be referred to as the inability of females to establish power because of disparities that occasionally forcibly dissuade or prevent women from standing up for their rights and contributing to or benefitting the economy (Camey et al. 2020).

Foreign aid is seen as an alternative source of resources to finance and promote women’s empowerment. According to Grown et al. (2016), although many donor organisations’ policy statements acknowledge the complexity of gender equality and women’s empowerment, the systematic implementation of gender mainstreaming across developmental sectors has proven challenging. In this scenario, institution quality is important to ensure foreign aid achieves its economic purpose. Institution quality is a metric used to evaluate a nation’s institutions and system of government. Likewise, as mentioned earlier, promoting empowerment at early stages could significantly affect long-term outcomes in reducing GBV. The study by de Lange et al. (2012) shows that awareness and empowerment in order to reduce gender inequality at the primary school level reduce GBV in the long run.

Regarding foreign aid, the study by Hicks et al. (2016) found that the reallocation of assistance flows in favour of initiatives linked to education and health has been associated with the empowerment of women in national legislatures. Maruta et al. (2020) investigated the impact of institutional quality and sectoral foreign aid on growth in developing economies. Maruta et al. (2020) reported that South American countries benefit more from education aid, Asian countries benefit more from health aid, and African countries benefit more from agricultural aid. The result is region specific and dependent on the existing degree of institutional quality. The study by Olojede et al. (2020) emphasises that women are frequently violence victims because of perpetuating gender disparities, which has hindered women’s career progress. The success of female empowerment has been recognised in the empirical literature, which calls for more investment in this area. For example, a study by Mohamed (2022) stated that, in nations where poverty and gender inequality are persistent, female human capital, particularly in education and health, is crucial for economic success.

The study by Edgell (2017) stated that foreign aid initiatives are important for developing nations to implement laws that uphold gender equality by international standards. The scholar claims two causal factors link foreign aid and women’s empowerment. Firstly, governments may employ quotas based on gender as an effort to elevate their position in the global hierarchy, possibly as a means of securing foreign aid. Secondly, effective foreign aid initiatives created to support women’s empowerment may lead to nations adopting gender quotas. Using data on foreign assistance from non-OECD, the author investigated these two causal processes. Edgell’s (2017) findings indicate that initiatives to empower women may affect the acceptance of quotas. However, developing nations relying on foreign aid are more prone to adopt gender quotas as political ploys than as the consequence of sustained liberalisation measures to reduce inequality.

Elbert (2013) stressed that, for aid to be effective and successful in South Sudan, it should prioritise empowering women in line with its international strategy and innovative objectives. According to the scholar, foreign aid would be more effective and consistent if prioritised to achieve specific objectives following society’s needs. A strong argument from the study of Elbert (2013) is that, if foreign aid is not tailored to the nations’ needs, less achievement will be recorded. Likewise, the role of the institutions matters for women’s empowerment being able to reduce GBV. Evidence from the literature has shown that promoting aggregate empowerment without improving the institution’s quality may not reduce GBV. Amaral et al. (2015) demonstrate this argument in India by investigating the Mahatma Gandhi National Rural Employment Guarantee Scheme (NREGS), which aims to empower women and reduce GBV. Amaral et al. (2015) state that NREGS influences women’s negotiation power in rural areas. However, GBV did not decline with the implementation of NREGS, because other factors, such as institutional quality, were not considered. Also, NREGS empowers both males and females.

Furthermore, GBV negatively impacts women’s career growth in several ways. According to Maslow’s (1954) hierarchy of needs theory, human needs are listed in ascending priority. Women will advance in their professional growth if the basic human safety requirements (security, liberation from anxiety, liberation from risk, the prevention of suffering, and safeguarding from danger) are met with the right life interventions; this demonstrates the importance of the role of institutional quality (most especially, the role rule of law plays in reducing GBV). Nevertheless, an anomaly will be established, and the real functionality of the person’s performance attainment is influenced when GBV challenges these demands. At this moment, women’s performance and professional growth are negatively impacted; addressing this is believed to be possible with women’s empowerment. According to Zurbriggen et al. (2007), GBV frequently thwarts women’s efforts to find jobs, develop skills, and pursue education, impeding their independence chances. It could have a consequence on sustainability and financial stability. However, evidence has strongly argued that women’s empowerment reduces GBV (Cherry and Hategekimana 2013; Gupta et al. 2013; Kabeer 1999; Magar 2003).

2.3. A Summary of the Literature Review and Identification of Gaps

Studies concerning aid and women’s empowerment focused on policymakers and donors (Hicks et al. 2016; Hornset and de Soysa 2022), while others focused on health, institutional quality, and economic growth (Bechange et al. 2021; Johnson et al. 2020; Maruta et al. 2020; Muluneh et al. 2020; Tsapalas et al. 2021). Another strand focused on growth, women’s political empowerment, and resource control (Awoa Awoa et al. 2022; Mohamed 2022). Concerning studies on GBV, more attention is given to issues of COVID-19, conflict-affected settings, socioeconomic development, and psychological effects (Dlamini 2021; Hossain et al. 2021; Olojede et al. 2020; Jatmiko et al. 2020; John et al. 2020; Acosta 2020). Although the efficacy of aid has been discussed in empirical studies, very little attention has been given to how foreign aid advances gender equality, empowers women, and reduces GBV (Grown et al. 2016).

There is a strong connection between GBV and women’s empowerment, both theoretically and empirically (Bolis and Hughes 2015; Cherry and Hategekimana 2013; Gupta et al. 2013; Hidrobo et al. 2016; Kabeer 1999; Magar 2003; Sen 1990). However, a few studies still have diverse views (Bannister and Moyi 2019), which may be due to the type of women’s empowerment, the region, and technique for estimations. But, women’s empowerment at the early stages of development is presumed to have a greater effect on reducing GBV. Likewise, the connection between foreign aid and institutional quality on the impact of women’s empowerment and GBV has also been established (Camey et al. 2020; de Lange et al. 2012; Edgell 2017; Elbert 2013; Grown et al. 2016; Kabeer 1999; Kamal 2011; Maruta et al. 2020).

The choice of variables used in this study is rooted in the empirical literature; this study used three dependent variables as follows: two variables as a proxy of women’s empowerment in the early growth process (children out of primary school and the gender parity index of secondary school enrolment), because empowerment at the early stage of development is presumed to have a greater effect on reducing GBV through the transformative process (Bolis and Hughes 2015; Kabeer 1999; Zimmerman 2000; Cherry and Hategekimana 2013; Gupta et al. 2013). Likewise, the independent variables are foreign aid and institutional quality. Empirically, it has been established that foreign aid and institutional quality influence women’s empowerment and GBV (Camey et al. 2020; de Lange et al. 2012; Edgell 2017; Elbert 2013; Grown et al. 2016; Kabeer 1999; Kamal 2011; Maruta et al. 2020). Other control variables from empirical literature are used in this study due to their importance: income per capita, female employment, and female wages and salaries (Asaleye and Strydom 2022; Egert et al. 2020; Ouedraogo and Stenzel 2021; Kabeer 1999). Egert et al. (2020) show a positive relationship between income per capita, education in European countries, and the Organisation for Economic Co-operation and Development. Asaleye and Strydom (2022) show that labour outcomes in terms of increasing labour and wages increase human capital in the short and long run. Likewise, Ouedraogo and Stenzel (2021) reported that economic conditions such as income per capita influences GBV in sub-Saharan Africa. Also, Kabeer (1999) stated that women’s empowerment is essential to reduce deprivation.

Studies have documented that institutional quality and foreign aid transmitted three main effects on the economy: the shock impact, short-run effects, and long-run effects (Asaleye et al. 2023). As a result of the channels of transmission and significance of the study, this current work can be distinguished from other studies by the method utilized and its objectives which involve (i) examining the impact of foreign aid and institutional quality on women’s empowerment and GBV using cross-sectional autoregressive lags, (ii) examining the shock effect of foreign aid on the indicators of women’s empowerment and GBV using the Panel Vector Correction Model, and (iii) investigating the mediating role of institutional quality and foreign aid on women’s empowerment and GBV.

3. Theoretical Framework and Model Specification

The theoretical framework used in this study is the Sustainable Livelihoods Model (SLM), a comprehensive and multidimensional approach to understanding and addressing socio-economic issues (United Nations Development Programme 2015). The key components are livelihood assets (which covers financial and human capital to social and natural resources), capabilities (the abilities of individuals to pursue diverse livelihood strategies), and finally, outcomes (this focuses on economic gains, social well-being, and overall quality of life). The SLM in connection to women’s empowerment and gender-based violence can be explained from the perspective of restricted capabilities. For example, lacking specific assets or restricted capabilities may render individuals more susceptible to violence.

This framework stresses that women can achieve livelihoods through an investment that positively affects their growth process. Cherry and Hategekimana (2013) emphasise the importance of women’s empowerment at the grassroots level and its ability to reduce GBV. Based on this, the forms of women’s empowerment considered in this study are primary and secondary enrolment, and the GBV indicator used is female genital mutilation. Globally, female genital mutilation is acknowledged as an assault on women’s and girls’ human rights (World Health Organisation 2023), also regarded as a form of violence against women and girls (Tordrup et al. 2022; Williams-Breault 2018). The models with the incorporation of foreign aid, institutional quality indicators, and control variables are given as follows:

$$WE1=f(IPC,AID,EMP,WAG,IG1,IG2)$$

(1)

$$WE2=f(IPC,AID,EMP,WAG,IG1,IG2)$$

(2)

$$GBI=f(IPC,AID,EMP,WAG,IG1,IG2)$$

(3)

In Equations (1)–(3), the women’s empowerment indicators are $WE1$ and $WE2$, which are proxies for female children out of primary school and the gender parity index of secondary school enrolment. The GBV indicator is represented by $GBI$, which is a proxy for female genital mutilation prevalence. $AID$ is the women’s aid. Two indicators are considered for institutional quality: government effectiveness and the rule of law, represented by $IG1$ and $IG2$. The control variables are income per capita, female employment, and female wages and salaries, represented by $IPC$, $EMP$, and $WAG$, respectively. The explicit forms of Equations (1)–(3) are given as follows:

$$WE{1}_{i,t}=a_1 + a_{2}IP{C}_{i,t}+a_{3}AI{D}_{i,t}+a_{4}EM{P}_{i,t}+a_{5}WA{G}_{i,t}+a_{6}IG{1}_{i.t}+a_{7}IG{2}_{i.t}+u_{i,t}$$

(4)

$$WE{2}_{i,t}=b_1 + b_{2}IP{C}_{i,t}+b_{3}AI{D}_{i,t}+b_{4}EM{P}_{i,t}+b_{5}WA{G}_{i,t}+b_{6}IG{1}_{i.t}+b_{7}IG{2}_{i.t}+e_{i,t}$$

(5)

$$GB{I}_{i,t}=c_1 + c_{2}IP{C}_{i,t}+c_{3}AI{D}_{i,t}+c_{4}EM{P}_{i,t}+c_{5}WA{G}_{i,t}+c_{6}IG{1}_{i.t}+c_{7}IG{2}_{i.t}+v_{i,t}$$

(6)

In Equations (4)–(6), the respective intercepts are $a_1$, $b_1$, and $c_1$. Also, $a_2,a_3,\dots ,a_7$; $b_2,b_3,\dots ,b_7$; and $c_2,c_3,\dots ,c_7$ are respective parameters for $IPC,AID,EMP,WAG,IG1$, and $IG2$. The error terms are $u_{i,t}$, $e_{i,t}$, and $v_{i,t}$.

The interactive effect between foreign aid and institutional quality is also established in Equations (7)–(9), referred to as models 1, 2, and 3. The equations are used to achieve objectives 1 and 2. The interactive effect equations are given as follows:

$$WE{1}_{i,t}={\alpha }_1 + {\alpha }_{2}IP{C}_{i,t}+{\alpha }_{3}AI{D}_{i,t}+{\alpha }_{4}EM{P}_{i,t}+{\alpha }_{5}WA{G}_{i,t}+{\alpha }_{6}IG{1}_{i.t}+{\alpha }_{7}IG{2}_{i.t}+{\alpha }_{8}AID\ast IGQ+{\epsilon }_{i,t}$$

(7)

$$WE{2}_{i,t}={\beta }_1 + {\beta }_{2}IP{C}_{i,t}+{\beta }_{3}AI{D}_{i,t}+{\beta }_{4}EM{P}_{i,t}+{\beta }_{5}WA{G}_{i,t}+{\beta }_{6}IG{1}_{i.t}+{\beta }_{7}IG{2}_{i.t}+{\beta }_{8}AID\ast IGQ+{\mu }_{i,t}$$

(8)

$$GB{I}_{i,t}={\lambda }_1 + {\lambda }_{2}IP{C}_{i,t}+{\lambda }_{3}AI{D}_{i,t}+{\lambda }_{4}EM{P}_{i,t}+{\lambda }_{5}WA{G}_{i,t}+{\lambda }_{6}IG{1}_{i.t}+{\lambda }_{7}IG{2}_{i.t}+{\lambda }_{8}AID\ast IGQ+{\omega }_{i,t}$$

(9)

Similarly, the respective intercepts are ${\alpha }_1$, ${\beta }_1$, and ${\lambda }_1$. Also, ${\alpha }_2,{\alpha }_3,\dots ,{\alpha }_7$; ${\beta }_2,{\beta }_3,\dots ,{\beta }_7$; and ${\lambda }_2,{\lambda }_3,\dots ,{\lambda }_7$ are respective parameters for $IPC,AID,EMP,WAG,IG1$, and $IG2$. The respective interactive parameters are ${\alpha }_8$, ${\beta }_8$, and ${\lambda }_8$. The error terms are ${\epsilon }_{i,t}$, ${\mu }_{i,t}$, and ${\omega }_{i,t}$. Scholars have hypothesised and examined interactive effects among variables using CS-ARDL (Mehmood 2022; Ullah et al. 2023; Vo et al. 2022). A positive relationship is expected between all the independent variables and dependent variables. Li and Xue (2022) and Adejumo et al. (2021) documented a positive relationship between education enrolment, employment, and financial empowerment. More so, studies by Ameyaw et al. (2021), Doucet et al. (2020) and Williams-Breault (2018) reported that financial assistance and sufficient economic power would reduce female genital mutilation. Likewise, Arendse (2012) emphasises the role of law and government effectiveness in promoting basic education. Jacobson et al. (2023) show how law can be used to influence female genital mutilation.

The CS-ARDL was used to estimate Equations (7)–(9). First, the autoregressive distributed lag (ARDL) is given as follows:

$$X_{i,t}={\displaystyle \sum _{I=0}^{K_j}{\mathsf{\Phi }}_{I,i}}{Y}_{i,t-1}+{\displaystyle \sum _{I=0}^{G_n}{\mathsf{\Psi }}_{I,i}{K}_{i,t-1}}+{\epsilon }_{i,t}$$

(10)

From Equation (10), using cross-sections, Equation (10) is modified to (11); according to Chudik and Pesaran (2015), this is vital because of cross-sectional dependence. The CS-ARDL is given as follows:

$$X_{i,t}={\displaystyle \sum _{I=0}^{R_j}{\mathsf{\Phi }}_{I,i}}{Y}_{i,t-1}+{\displaystyle \sum _{I=0}^{R_n}{\mathsf{\Psi }}_{I,i}{K}_{i,t-1}}+{\displaystyle \sum _{I=0}^{R_q}{\mathsf{\Omega }}_{i,I}^1{\overline{M}}_{i,t-1}}+{\epsilon }_{i,t}$$

(11)

In Equation (11), $\overline{M}$ is the mean value, which is given as follows: ${\overline{X}}_{i,t-1}$ and ${\overline{K}}_{i,t-1}$; to take control of the existing properties. $R_j$, $R_n$, and $R_q$ are the lags of the independent variables. $X_{i,j}$ is the dependent variable that represents the women’s empowerment indicators and GBV. The CS-ARDL comprises two impacts: the short- and long-run estimations. The long run, the average group estimation, and the long-run value are derived by

$$\widehat{G}AE=\frac{1}{N}{\displaystyle \sum _{i=1}^N{\widehat{G}}_i}$$

(12)

$$\widehat{G}CS-ARD{L}_i\frac{{\displaystyle {\sum }_{I=0}^{G_N}{\mathsf{\Psi }}_{I,1}^{K_j}}}{dx}{\widehat{\mathsf{\Phi }}}_{I,i}$$

(13)

Likewise, the second impact—that is, the short-run effect—is stated as follows:

$$\Delta X_{i,t}={\vartheta }_i\left[X_{i,t-1}-{\gamma }_i{K}_{i,t}\right]-{\displaystyle \sum _{I=1}^{R_{j-1}}{\mathsf{\Phi }}_{I,i}{\Delta }_I}{X}_{i,t-1}+{\displaystyle \sum _{I=0}^{R_{n-1}}{\mathsf{\Psi }}_{I,i}{\Delta }_I{K}_{i,t-1}}+{\displaystyle \sum _{I=0}^{R_q}{\mathsf{\Omega }}_{i,I}^1{\overline{M}}_{i,t-1}}+{\epsilon }_{i,t}$$

(14)

In Equation (14), ${\vartheta }_i=-(1-{\displaystyle \sum _{I-1}^{R_j}{\widehat{\mathsf{\Phi }}}_{I,i}})$ and ${\gamma }_i=\frac{{\displaystyle {\sum }_{i=0}^{R_n}{\mathsf{\Psi }}_{I,i}^{R_n}}}{\widehat{\vartheta }}$ Also, $\widehat{G}AG=\frac{1}{N}{\displaystyle \sum _{i=1}^N\widehat{G}}$ and the symbol $\Delta$ represents the first difference operator. The speed of adjustment is also included, which must be negative and significant (Mehmood 2022).

The third objective, the shock effect, is achieved using the panel vector correction model (PVECM). The PVECM is given as follows:

$$\Delta N_{it}={\chi }_0+{\displaystyle \sum _{i=1}^p{\chi }_1\Delta N_{i,t-i}}+{\displaystyle \sum _{i=0}^q{\chi }_2\Delta M_{i,t-1}}+{\chi }_{3}E{C}_{i,t-1}+u_{i,t}$$

(15)

The independent variables are income per capita ($IPC$), female employment ($EMP$), female wages and salaries ($WAG$), women’s foreign aid ($AID$), government effectiveness ($IG1$), and the rule of law ($IG2$), while the indicators of women’s empowerment and GBV are included separately in the empirical analysis (that is $WE1$, $WE2$, and $GBI$). Let us assume that the indicators of women’s empowerment and GBV are represented by $REG$. Following the study by (Asaleye et al. 2023), with slight modifications to achieve the aim of this current study, the variables are ordered in the VECM specification as follows:$AID,IPC,IG1,IG2,REG,WAG,EMP$. Also, in Equation (15), $EC$ is the error correction term.

The research hypotheses for this study are stated in null forms as follows:

• There is no significant relationship between foreign aid, institutional quality, and women’s empowerment or GBV in both the short and long run.
• Institutional quality and foreign aid do not mediate the relationship between each other and women’s empowerment or GBV.
• Foreign aid does not have a significant shock effect on women’s empowerment and GBV.

This study chose the cross-sectional autoregressive distributed lag (CS-ARDL) to achieve the first two objectives due to its strength and ability to investigate short- and long-term implications (Pesaran et al. 2001). Also, it accounts for heterogeneity across different variables; this method accommodates diverse characteristics among variables and facilitates robust inferences among them (Pesaran et al. 2001). The Panel Vector Error Correction Model (PVECM) was used to achieve the third objective. The PVECM captures the dynamics of time series data across multiple variables. Also, it is a vigorous tool for analysing how various variables respond to shock and restore equilibrium over time (Baltagi 2005). It is preferable to traditional vector autoregression because it considers the stationarity properties of the series.

Before the estimations, a preliminary analysis was carried out on the series. First, descriptive statistics and correlation tests were performed, followed by cross-sectional dependence (CD), which is vital before the stationarity test. CD often occurs because of closely related policies and programmes that the countries may apply in accordance with the region (Asaleye and Strydom 2022). The result can be spurious if the test is not considered (Westerlund 2007). Concerning this, the Pesaran (2015) CD test was used to determine cross-sectional dependence. Also, after the unit root test, this study tested for the slope of heterogeneity using Swamy’s (1970) approach.

Furthermore, the Westerlund and Edgerton (2008) approach was preferred to the Kao and Pedroni technique, because the latter ignored CD in its test, and heterogeneity was not included. The information about the series is given in

Table 1. Information about the data.

Symbol	Variable	Description	Source
Dependent Variables
$WE1$	Female children out of primary school	The percentage of primary school-aged children not enrolled in primary or secondary school.	UNESCO Statistics
$WE2$	Secondary school gender parity index	The ratio of girls to boys enrolled in public and private institutions at the secondary level.	UNESCO Statistics
$GBI$	Female genital mutilation prevalence rate	Women aged 15 to 49 who have partially or entirely removed the female external genitalia.	UNICEF Data
Independent Variables
$AID$	Women’s aid	Financial Assistance by Women’s rights organisation and movements and government institutions.	Organisation for Economic Co-operation and Development, Credit Reporting System (CRS)
$IG1$	Government Effectiveness	Reflects opinions on the credibility of the government’s adherence to policies and the quality of policy development, execution, and effectiveness.	Worldwide Governance Indicator (WGI)
$IG2$	Rule of Law	Reflects the beliefs of how much people trust and abide by social norms, including the effectiveness of the enforcement of the courts and the possibility of crime and violence.	Worldwide Governance Indicator (WGI)
Control Variables
$IPC$	GDP per capita growth	Percentage of the country’s growth rate in constant local currency.	World Bank data.
$EMP$	Female employment in the industry	Female working age involved in any activity for pay or profit.	International Labour Organisation
$WAG$	Female wages and salaries	Female workers for paid employment occupations.	International Labour Organisation

Source: Authors’ computation.

. The estimation period was from 2002 to 2021, involving twenty-seven African countries. The rationale for the estimation period and countries used was due to data availability. The information about the countries is presented in

Table A2. Countries used in this study.

List of Countries
Benin, Burkina Faso, Cameroon, Central African Republic, Chad, Cote d’Ivoire, Eritrea, Ethiopia, Ghana, Guinea, Guinea-Bissau, Kenya, Liberia, Mali, Mauritania, Niger, Nigeria, Senegal, Sierra Leone, South Africa, South Sudan, Sudan, Tanzania, Togo, Uganda, Zambia, Zimbabwe.

in the Appendix A section.

4. Empirical Results and Discussion

4.1. Preliminary Analyses

The results of preliminary analyses are presented in the Appendix A section. The data used in this study are in logarithmic form; this aids in interpretation using the concept of elasticity and reduces the risk of heteroskedasticity and autocorrelation. The results of the descriptive statistics and correlation analysis are presented in

Table A1. Descriptive statistics and correlation results.

Descriptive Statistics Results
	WE1	WE2	GBI	EMP	IPC	WAG	AID	IG1	IG2
Mean	1.894	0.798	1.226	0.752	1.989	0.946	−0.191	−0.979	−0.875
Median	1.390	0.834	1.502	0.801	2.259	1.009	−0.150	−0.950	−0.900
Maximum	11.650	1.097	1.973	1.511	28.676	1.563	1.200	−0.200	−0.190
Minimum	−0.007	0.323	−0.523	−0.268	−22.312	−0.149	−3.889	−1.640	−1.650
Std. Dev.	2.041	0.175	0.757	0.483	4.587	0.394	0.777	0.343	0.371
Obs.	349	349	349	349	349	349	349	349	349
Correlation Results
	WE1	WE2	GBI	EMP	IPC	WAG	AID	IG1	IG2
WE1	1
WE2	0.158	1
GBI	0.387	0.163	1
EMP	−0.217	0.346	−0.115	1
IPC	−0.006	−0.125	−0.010	−0.059	1
WAG	0.013	0.680	−0.100	0.599	−0.151	1
AID	−0.051	0.545	−0.166	0.172	−0.019	0.459	1
IG1	−0.282	0.342	−0.363	0.380	0.153	0.222	0.375	1
IG2	−0.315	0.304	−0.276	0.332	0.032	0.136	0.466	0.767	1

Source: authors’ computation.

; evidence shows that income per capita has the highest average value (1.989), and government effectiveness has the lowest average value (−0.979). Also, the correlation analysis shows that the series used in this study are not highly correlated.

Table A3. Cross-sectional dependence and slope of heterogeneity results.

Cross-Sectional Dependence Test (Pesaran CD)
Variables	Values	Probability
WE1	4.672 ***	0.000
WE2	18.462 ***	0.000
GBI	−0.038	0.969
EMP	−2.605 ***	0.009
IPC	5.969 ***	0.000
WAG	15.48 ***	0.000
AID	7.409 ***	0.000
IG1	2.030 ***	0.004
IG2	5.232 ***	0.000
Results of slope heterogeneity
WE1 Specification: Model 1
Variables	Values	Probability
Delta tilde	14.109 **	0.010
Delta tilde Adjusted	30.114 ***	0.000
WE2 Specification: Model 2
Variables	Values	Probability
Delta tilde	12.41558 ***	0.0042
Delta tilde Adjusted	27.03123 ***	0.0000
GBI Specification: Model 3
Variables	Values	Probability
Delta tilde	14.406 ***	0.0041
Delta tilde Adjusted	24.902 ***	0.0000

*** and ** show significance @ 0.01 and 0.05, respectively; Source: authors’ computation.

shows the cross-sectional dependence (CD) and slope of heterogeneity results. The outcome of the results shows the presence of CD and confirms the heterogeneous slope at a five per cent significance level in all three specifications, that is, when the female children out of primary school variable, the secondary gender parity index variable, and female mutilation prevalence rate variable are used as dependent variables. The unit root test presented in

Table A4. Unit Root Test.

CIPS Unit Root Test Results
	I (0)	I (1)
Variables	Value	Value
WE1	−2.221 *	−3.578 *
WE2	−2.146	−3.785 *
GBI	−0.3661	−3.731 *
EMP	−2.913 *	−3.475 *
IPC	−3.503 *	−4.738 *
WAG	−2.648 *	−2.915 *
AID	−3.209 *	−5.101 *
IG1	−1.878	−3.295 *
IG2	−3.251 *	−4.888 *

* Shows rejection of non-stationarity @ one per cent significance level; Source: authors’ computation.

confirms the mixture of the integration of order one and stationarity at the level. At the same time, the presence of cointegration was established among the variables using the Westerlund cointegration approach, presented in

Table A5. Westerlund cointegration results.

Results of Westerlund Cointegration Tests
WE1 Specification: Model 1
Statistics	Value	Z value	Probability
Gt	−3.528 ***	−3.041	0.000
Ga	−6.795	1.483	0.282
Pt	−4.739 ***	−0.182	0.001
Pa	−2.528	0.134	0.734
WE2 Specification: Model 2
Statistics	Value	Z value	Probability
Gt	−3.491 ***	−2.187	0.008
Ga	−6.982 **	−1.405	0.015
Pt	−4.608 *	−0.485	0.091
Pa	−2.073	0.943	0.491
GBI Specification: Model 3
Statistics	Value	Z value	Probability
Gt	−3.893 **	−2.045	0.023
Ga	−2.773	0.981	0.198
Pt	−4.318 ***	−2.174	0.000
Pa	−2.276 ***	0.461	0.000

***, ** and * indicate significant @ 0.01, 0.05, and 0.10, respectively; Source: authors’ computation.

.

4.2. CS-ARDL Test for Long- and Short-Run Estimation Results

Table 2. CS-ARDL test for long- and short-run estimations for Model 1.

WE1 Specification: Model 1
Short-run Elasticities Results
Variable	Coefficient	Std. Error	Z-Statistic	Probability
D(IPC)	−0.098795 ***	0.034636	−2.852411	0.0048
D(AID)	0.880785	1.229826	0.716187	0.4748
D(EMP)	−0.098219 ***	0.034604	−2.838357	0.0050
D(WAG)	1.530980 ***	0.414191	3.696312	0.0003
D(IG1)	−1.454715	1.737148	−0.837416	0.4035
D(IG2)	−3.534636 *	1.994961	−1.771782	0.0781
D(IG1*AID)	−0.633729	1.547538	−0.409508	0.6827
D(IG2*AID)	−15.10886 ***	4.939371	−3.058863	0.0026
ECT (−1)	0.851854 ***	0.092508	9.208398	0.0000
Long-run Elasticities Results
Variable	Coefficient	Std. Error	Z-Statistic	Probability
IPC	0.757451	0.518323	1.461350	0.1456
AID	−1.263156 *	0.668895	−1.888422	0.0605
EMP	−1.991732 ***	0.577508	−3.448837	0.0007
WAG	−15.28933 ***	4.947405	3.090374	0.0023
IG1	0.612300	0.793657	0.771492	0.4414
IG2	−2.025594 ***	0.740215	−2.736495	0.0068
IG1*AID	1.553143	0.991625	1.566260	0.1189
IG2*AID	−3.240095 ***	0.982992	−3.296155	0.0012
Constant	1.413796 **	0.716508	1.973175	0.0499

***, ** and * shows significance @ 0.01, 0.05, and 0.10, respectively; Source: Authors’ Computation.

shows the CS-ARDL long- and short-run estimation result when female children out of primary school (WE1) is used as the dependent variable. Income per capita (IPC), female employment (EMP), and the rule of law (IG2) play a key role in reducing the number of female children out of primary school in the short run. A one per cent increase in IPC, EMP, and IG2 causes a reduction of 0.10 per cent, 0.10 per cent, and 3.53 per cent in WE1. The outcome of employment and enrolment is aligned with the studies of Adejumo et al. (2021) and Li and Xue (2022). The effect of the rule of law in reducing female school dropout conforms with the study of Arendse (2012). Some of the factors that may be responsible for this include the free education that is practised in most African countries at this stage of schooling and the right to basic education, which is protected by international laws such as the Universal Declaration of Human Rights; UNESCO Convention against Discrimination in Education; International Covenant on Social, Economic and Cultural Rights; and the Convention on the Rights of the Child.

The interaction between IG2 and AID (IG2*AID) shows that a one per cent increase will reduce WE1 by 15.1 per cent in the short run. Unfortunately, women’s aid (AID), government effectiveness (IG1), and the interaction of IG1 and AID (IG1*AID) are not significant. And female wages and salaries (WAG) positively affect WE1; a one per cent increase in WAG will promote WE1 with 1.53 per cent in the short run. In the long run, AID, EMP, WAG, IG2, and IG2*AID will reduce WE1 with 1.26 per cent, 1.99 per cent, 15.3 per cent, 2.03 per cent, and 3.24 per cent, respectively. The ECT indicates the error correction model, which shows the model’s stability and will adjust to the equilibrium with a speed of 0.85 per cent annually. Likewise, IPC, IG1, and IG1*AID are not significant in the long run.

Table 3. CS-ARDL test for long- and short-run estimations for Model 2.

WE2 Specification: Model 2
Short-run Elasticities Results
Variable	Coefficient	Std. Error	Z-Statistic	Probability
D(IPC)	0.006012 ***	0.002133	2.818278	0.0053
D(AID)	−0.142232 ***	0.047126	−3.018147	0.0029
D(EMP)	0.073367	0.132983	0.551701	0.5818
D(WAG)	−0.474064 ***	0.047725	−9.933172	0.0000
D(IG1)	0.050033 *	0.029839	1.676750	0.0953
D(IG2)	−0.369362 ***	0.049511	−7.460230	0.0000
D(IG1*AID)	−0.003746	0.026788	−0.139847	0.8889
D(IG2*AID)	−0.773046 ***	0.036090	−21.42014	0.0000
ECT (−1)	−0.007640 ***	0.002233	−3.421447	0.0008
Long-run Elasticities Results
Variable	Coefficient	Std. Error	Z-Statistic	Probability
IPC	0.195420 ***	0.072478	2.696278	0.0076
AID	−0.059665	0.041789	−1.427772	0.1550
EMP	−0.072341 ***	0.025876	−2.795651	0.0057
WAG	−0.261381 ***	0.032382	−8.071812	0.0000
IG1	0.062325	0.049583	1.256975	0.2103
IG2	−0.332733 ***	0.067236	−4.948719	0.0000
IG1AID	−0.059401	0.061951	−0.958828	0.3389
IG2AID	−0.518408 ***	0.039839	−13.01265	0.0000
Constant	0.730401 ***	0.044763	16.31689	0.0000

*** and * shows significance @ 0.01, and 0.10, respectively; Source: Authors’ Computation.

shows the CS-ARDL long- and short-run estimation results when the secondary school gender parity index (WE2) is used as the dependent variable. The results show that women’s aid (AID), female wages and salaries (WAG), and the rule of law (IG2) are effective ways to close the gender gap in secondary school in the short run. A one per cent increase in AID, WAG, and IG2 will close the gender gap with 0.14 per cent, 0.47 per cent, and 0.37 per cent, respectively. However, income per capita (IPC) and government effectiveness (IG1) do not close the gap; a one per cent increase in IPC and IG1 promotes WE2 with 0.006 per cent and 0.005 per cent, respectively. Female employment (EMP) and the interaction of IG1 and AID (IG1*AID) are insignificant. The interaction of IG2 and AID (IG2*AID) will close the gender gap by 0.77 per cent in the short run.

In the long run, EMP, WAG, and IG2 will close the gender gap; a one per cent increase in EMP, WAG, and IG2 will lead to a 0.07 per cent, 0.26 per cent, and 0.33 per cent reduction in WE2. Also, IPC does not factor into reducing the gender gap in the long run; a one per cent increase in IPC will promote WE2 with 0.20 per cent. The interaction of IG2 and AID will reduce the gender gap by 0.52 per cent in the long run. The outcome of the wages and salaries in the short and long run and the impact of employment in the long run aligned with the study of Li and Xue (2022). Increasing schooling enrolment at this stage is very important for reducing gender inequality, because it tends to lower sexual activities, childbirth, and early marriage rates. Zulaika et al. (2022) emphasised that adolescent pregnancy is one of the major drivers of girls dropping out of school. Asaleye and Strydom (2022) stressed that there is a gap in economic participation between males and females in African economies; this may be responsible for the insignificant income per capita in the long run and the positive effect on WE2 in the short run. The ECT indicates the error correction model, which shows the model’s stability and will adjust to the equilibrium with a speed of 0.73 per cent annually. Likewise, AID, IG1, and IG1*AID are not significant.

Table 4. CS-ARDL test for long- and short-run estimations for Model 3.

GBI Specification: Model 3
Short-run Elasticities Results
Variable	Coefficient	Std. Error	Z-Statistic	Probability
D(IPC)	0.096267 **	0.047671	2.019409	0.0449
D(AID)	−0.084297 **	0.041839	−2.086496	0.0383
D(EMP)	0.053397	0.033460	1.595824	0.1123
D(WAG)	−0.120828 ***	0.042068	−2.872207	0.0046
D(IG1)	0.000559	0.000936	0.596735	0.5514
D(IG2)	−0.092066 *	0.054125	−1.700982	0.0907
D(IG1*AID)	−0.126935 ***	0.042069	−3.017281	0.0029
D(IG2*AID)	−0.952713 ***	0.364579	−2.613184	0.0097
ECT (-1)	−0.085692 **	0.041669	−2.056505	0.0412
Long-run Elasticities Results
Variable	Coefficient	Std. Error	t-Statistic	Probability
IPC	1.144368 ***	0.291795	3.921826	0.0001
AID	−0.050820 ***	0.033104	−1.535133	0.1265
EMP	−0.080600	0.152281	−0.529286	0.5972
WAG	0.210174	0.190566	1.102896	0.2715
IG1	0.007017	0.011045	0.635358	0.5260
IG2	−0.089945 ***	0.046592	−1.930458	0.0551
IG1*AID	−0.121439 ***	0.041783	−2.906440	0.0041
IG2*AID	−1.035009 ***	0.245925	−4.208639	0.0000
Constant	0.050820	0.033104	1.535133	0.1265

***, ** and * shows significance @ 0.01, 0.05, and 0.10, respectively; Source: Authors’ Computation.

shows the CS-ARDL long- and short-run estimation result when the female genital mutilation prevalence rate (GBI) is used as the dependent variable. The result shows that women’s aid (AID), female wages and salaries (WAG), and the rule of law (IG2) play an important role in reducing GBI in the short run. A percentage increase in AID, WAG, and IG2 will cause a decline of 0.08 per cent, 0.12 per cent, and 0.09, respectively, in GBI. Surprisingly, income per capita (IPC) has a positive relationship with GBI in the short run, which means a one per cent surge in IPC will promote 0.10 per cent in GBI. Female employment (FEM) and government effectiveness (IGI) are insignificant. The interactions of IG1 and IG2 on GBI show that an increase of one per cent will lead to 0.13 per cent and 0.95 per cent, respectively, in GBI. In the long run, AID and IG2 will reduce GBI; a one per cent increase in AID, EMP, and IG2 will lead to a 0.05 per cent and 0.09 per cent reduction in GBI. Also, EMP, WAG, and IGI are not significant.

The outcome of the rule of law and women’s aid in the short and long run conforms with the study of Jacobson et al. (2023). The contributing role of women’s aid and female wages and salaries in reducing the female genital mutilation prevalence rate are in line with studies by Williams-Breault (2018), Doucet et al. (2020), and Ameyaw et al. (2021). Financial assistance and sufficient economic control through aid, employment, and wages can reduce the female genital mutilation prevalence rate. Also, women with low decision-making abilities may be involved in the genital mutilation of their daughters. The interactions of IG1 and IG2 with AID will reduce GBI by 0.12 per cent and 1.03 per cent, respectively. The ECT indicates the error correction model, which shows the model’s stability, and will adjust to the equilibrium with a speed of 0.09 per cent annually.

4.3. VECM Estimation Results

Table 5. Variance decomposition results.

WE1 Specification: Model 1
Period	AID	IPC	IG1	IG2	WE1	WAG	EMP
2	95.96037	0.399035	0.369654	2.460899	0.027452	0.717373	0.065214
4	91.21851	0.813799	0.900115	5.141907	0.112719	1.756784	0.056169
6	90.03419	0.769043	1.391009	5.644647	0.207269	1.881828	0.072010
8	89.41985	0.756280	1.639096	5.942294	0.219629	1.920238	0.102609
10	89.01379	0.741266	1.774976	6.147428	0.226860	1.967023	0.128660
WE2 Specification: Model 2
Period	AID	IPC	IG1	IG2	WE2	WAG	EMP
2	95.40190	0.482270	0.298900	2.543553	0.244220	0.964446	0.064708
4	90.63435	0.915922	0.713618	5.195303	0.260254	2.216325	0.064224
6	89.65575	0.886859	1.114127	5.649338	0.202105	2.400296	0.091523
8	88.98479	0.905719	1.297271	5.978601	0.162514	2.539631	0.131471
10	88.54363	0.905109	1.399759	6.196317	0.136422	2.648396	0.170368
GBI Specification: Model 3
Period	AID	IPC	IG1	IG2	GBI	WAG	EMP
2	95.25990	0.511234	0.398023	2.536498	0.275759	0.929743	0.088839
4	90.54686	0.851250	0.877563	5.221692	0.413954	2.011703	0.076977
6	89.54657	0.757104	1.247263	5.731181	0.509374	2.129747	0.078762
8	88.91262	0.737059	1.427807	6.061713	0.527286	2.241932	0.091585
10	88.51877	0.716382	1.531452	6.270587	0.533333	2.320709	0.108772

Source: Authors’ Computation.

presents the forecast error shock of women’s foreign aid on the three specifications used in this study, that is, female children out of primary school (WE1), the secondary school gender parity index (WE2), and the female genital mutilation prevalence rate (GBI). In the three specifications, the forecasting error shock of AID shows more variation in the rule of law (IG2), followed by income per capita (IPC). Also, for specification Two (with the inclusion of WE2), the forecast error shock of AID has the least effect on government effectiveness with the periods of 2 to 4 and on female employment with the periods of 8 to 10. On Specifications One (with the inclusion of WE1) and Three (with the inclusion of GBI), the least variation from AID shock was observed in female employment (EMP).

5. Conclusions and Policy Recommendations

Gender-based violence (GBV) has adversely affected both men and women. Still, women are significantly more vulnerable, because violence reinforces and amplifies pre-existing gender differences in involvement in the economy, politics, and society. Due to greater economic equality and women’s empowerment, there has been a decrease in GBV in developed economies. However, discrimination against women and girls is still more common in African countries. The characteristics and cultures of African nations explain why these attempts have failed in the fight against GBV. However, there are significant campaigns and initiatives to combat GBV. Despite these efforts, about 30 per cent of women experience intimate partner abuse, with Southeast Asia and sub-Saharan Africa seeing the worst rates. More so, it suggests that women’s low economic independence and educational achievement are related to GBV against women. Considering this, this study used a cross-sectional auto-regressive distributed lag (CS-ARDL) and a vector error correction model (VECM) to investigate the short- and long-run implications of foreign aid and institutional quality on women’s empowerment and GBV, the mediating role of institutional quality and foreign aid on women’s empowerment and GBV, and finally, the shock effect of foreign aid on women’s empowerment and GBV.

This study’s main results are as follows. The CS-ARDL results shows that when female children out of school is used as the dependent variable in the first specification, the rule of law plays a key role in reducing the number of female children out of primary school; this finding lends credence to the free education that is practised in most African countries and the right to basic education, which is protected by international laws such as the Universal Declaration of Human Rights; the UNESCO Convention against Discrimination in Education; the International Covenant on Social, Economic, and Cultural Rights; and the Convention on the Rights of the Child. The interaction between the rule of law and women’s foreign aid also reduces the number of female children out of primary school. In the second specification, when the secondary school gender parity index is used as the dependent variable, increasing women’s aid and the rule of law are effective ways to close the gender gap in secondary school in the short run. While in the long run, the rule of law will reduce the gender gap in secondary school, and women’s aid will be insignificant.

In the third specification, when the female genital mutilation prevalence rate is used as the dependent variable, women’s aid and the rule of law are important in the reduction of the female genital mutilation prevalence rate in the short and long run; this means that financial assistance and sufficient economic control through aid can reduce the female genital mutilation prevalence rate. Also, women with low decision-making abilities may be involved in the genital mutilation of their daughters. The interactions of government effectiveness and the rule of law with women’s aid will reduce the female genital mutilation prevalence rate in the long run. In the long run, the interaction of government effectiveness and women’s aid is insignificant in all three specifications. The VECM results show that the forecasting error shock of foreign aid shows more variation in the rule of law than any other variable, followed by income per capita.

This study recommends that, while enhancing the legal structure is very important in Africa, given the low coefficient values from the findings to reduce discrimination against the right to education, more should be done to increase the enrolment of female students through the maximisation of women’s aid to achieve the short- and long-term objectives of reducing gender violence. A lack of funding and other resources may restrict the complete implementation of the right to quality education. Female genital mutilation may be prevented with appropriate preventative measures; while the rule of law shows significant improvement in the short and long run, government effectiveness is insignificant. Also, women’s aid reduces female genital mutilation at a low rate. More resources are required to stop female genital mutilation, minimise suffering for young girls, and remove expenses to care for themselves after the act. Also, there is a need to review current policies, programs, and laws to create awareness and more protection against this act or practice. Promoting women’s empowerment in education and closing the gender gap in achieving sufficient economic power through financial aid is another vital strategy that improves women’s ability to make crucial decisions, like giving up female genital mutilation and increasing their voices in society against cultural beliefs and child discrimination. Finally, the results of this study call for strong enhanced government support and funding to end the practice of female genital mutilation and discrimination against young females’ education in the short and long run.

Notably, despite the comprehensive nature of our study on foreign aid and institutional quality’s impact on gender-based violence, a limitation arises from the unavailability of macro-level data on physical assault in marriages; this absence restricts the direct analysis of this subject. Therefore, we suggest that future research should note this to further enhance the efficacy of preventive policies.