Globalization, gender-inclusive education and structural transformation in Africa

The present study's emphasis on the role that gender-inclusive education plays in the impact of globalization on structural change in Africa is reinforced by four main driving forces, which include (1) the fact that the participation of women in the formal economic sector is significantly lower than that of men in the attendant sector (Tandon and Wegerif, 2013; Ellis et al., 2007; Ramani et al., 2013; Asongu and Odhiambo, 2023; Osinubi and Asongu, 2021); (2) the necessity of giving a human face to globalization, particularly in relation to the inclusion of more women in the formal economy (Jorgenson and Clark, 2010; Nchofoung and Asongu, 2022a); (3) the policy necessity of gender economic inclusion in order to achieve some of the Sustainable Development Goals (SDGs) of the United Nations (UN) (Oostendorp, 2009; Ngono, 2021; Osabuohien et al., 2019) and (4) gaps in the relevant literature that still need to be filled (Tifuh, 2022; Asongu et al., 2020a). The next paragraphs elaborate on these four pillars.

First, according to Asongu et al. (2020a), Africa is the continent with the lowest percentage of women in formal political and economic circles, which is consistent with Osinubi and Asongu (2021). The narrative holds that the majority of women in sub-Saharan Africa (SSA) work in subsistence-oriented and informal occupations such as small trade and agriculture (Food and Agricultural Organization-FAO, 2011; Ramani et al., 2013; Ellis et al., 2007; Tandon and Wegerif, 2013; Uduji and Okolo-Obasi, 2019, 2020; Asongu and Odhiambo, 2018, 2019). Additionally, an increasing amount of research backs up the claim that increasing the number of women in political and economic roles will optimize the use of funds meant for economic growth (Vancil-Leap, 2017; Rice and Barth, 2017; Moras, 2017; Marquez, 2017; Uduji and Okolo-Obasi, 2018; Luo et al., 2017; Uduji et al., 2019). According to the World Bank (2018), the world loses over US$160 trillion in revenue when women are excluded from the official economy. This is essentially because women are 27% less likely than men to be employed in the formal job market (Tifuh, 2022; International Labor Organization, 2018).

Second, the premise that openness dynamics – and, by extension, globalization – need to be given a human face is supported by an increasing number of studies. Globalization is associated with increased exclusive economic growth or development that does not benefit the poor, which makes this significant (Jorgenson and Clark, 2010; Nchofoung and Asongu, 2022a, b). In support of inclusive and sustainable development goals such as economic empowerment for women and gender equality, it is therefore crucial to evaluate how globalization affects gender economic inclusion. This stance is in line with the significance of reaching the fifth sustainable development target (often known as SDG5), which is focused on advancing gender equality.

Third, given the foregoing, the possibility of gender economic inclusion among SDGs is evident because, among other things, exclusive development is becoming more visible in both developing and developed countries as a result of rising levels of globalization, among other factors (Asongu et al., 2020a). To further elucidate this perspective, it has been demonstrated that female economic progress is particularly important in the globalization era (Oostendorp, 2009), particularly in light of the growing prejudice against women in emerging countries, which is partly caused by the consequences of globalization (Ngono, 2021; Osabuohien et al., 2019).

Fourth, the need to address a gap in the body of research on how globalization affects macroeconomic outcomes like inclusive development is what drives this study's positioning. Tifuh (2022) and Asongu et al. (2020a) are the two studies that most closely resemble the current inquiry in the literature. While the former does not use transmission channels, the latter does not take sector analysis into account. By looking at the ways that gender-inclusive education mitigates the impact of globalization on Africa's structural development, the current study complements both studies. Therefore, the study's primary purpose is to investigate how the impact of globalization dynamics on structural transformation in SSA is mitigated by gender-inclusive education. Moreover, the empirical analysis is tailored such that thresholds for complementary policies or policy thresholds of the moderating gender inclusion variables are provided in order to equip policymakers with the relevant actionable thresholds that are essential in the policy orientation of how globalization affects structural transformation in the region.

Furthermore, another study that is close to the present study in the extant literature is Asongu et al. (2024) which has examined the role of globalization in lifelong gender inclusion education for structural transformation. The two main hypotheses are not validated, especially as it pertains to (1) globalization and gender-inclusive lifelong learning distinctly influencing structural transformation and (2) globalization moderating lifelong gender-inclusive education to boost structural transformation. The similarity between the underlying paper and the present is that the same data and methodology are employed, though upon close scrutiny, education variables as well as elements in the conditioning information set are different. Moreover, the present study is distinct from Asongu et al. (2024) in terms of objectives (e.g. net effect of lifelong gender-inclusive education on structural transformation versus net effect of globalization on structural transformation), theoretical underpinnings, testable hypotheses, findings and policy implications. These distinctive features are self-apparent, not least because the distinct objectives which are based on different theoretical underpinnings and testable hypotheses engender different findings and policy implications. For instance, while the tested hypothesis in Asongu et al. (2024) is how globalization moderates lifelong gender-inclusive education for structural transformation, in this study, the tested hypotheses on how gender-inclusive education moderates the incidence of globalization on structural transformation are validated through the computation of net effects and corresponding gender-inclusive education thresholds. Globalization's role in moderating lifelong gender-inclusive education for structural transformation has not been validated in Asongu et al. (2024).

The remaining portions of the study are structured as follows. The discussion of theoretical foundations follows this introduction. While Section 3 presents the data and corresponding methodology, Section 4 discusses the empirical findings. Implications and suggestions for future research are presented in Section 5.

In terms of contextualization, it is important to remember that, in line with the existing literature (Deudibe et al., 2020), economic development is typically associated with substantial variations of labor from agriculture to employment of non-farm nature – a tendency that is for the most part understood within the remit of structural transformation. In essence, according to the narrative, employment that is not within the farming sector is riskier and more linked to less underemployment and better production prospects than the agricultural sector. In essence, Deudibe et al. (2020) maintained that compared to the rest of the world, SSA is substantially lagging in terms of structural transformation. This position has been recently confirmed by Saba et al. (2024).

To put the above mentioned observation in more perspective, it is maintained by Malah Kuete and Asongu (2023) that compared to the other regions of the world, structural transformation in Africa is one of the least. This is essentially because according to the narrative, Africa's proportion in global manufacturing in 2000 was 1.2% compared to 1.1% in 2008, compared to the corresponding proportion in Asia that increased from 13% to 25% during the same period. The underlying position is consistent with Mijiyawa (2017) within the remit of structural change and Asongu et al. (2023) in relation to the total value of all economic sectors. It follows from the underlying idea that structural transformation represents a substantial policy concern, especially as it pertains to achieving corresponding externalities in structural transformation such as, inter alia, poverty reduction and employment. The contextualization informs the need for more scholarly and policy research on outcomes of structural transformation in SSA. Therefore, it is important to evaluate how gender-inclusive education mitigates the impact of globalization on structural transformation in Africa, as suggested in the study's introduction.

The link that exists between globalization and economic prosperity within the remit of infrastructural development can be clarified using two major philosophical schools, specifically the neoliberal and hegemonic schools. This aligns with the literature on the impact of globalization on the results of economic progress (Asongu, 2013; Asongu and Odhiambo, 2023).

First, according to the neoliberal school, there are apparent positive externalities from globalization in the forms of production efficiency, investment opportunities, increased trade and technological advancements in addition to its negative externalities because it can be compared to a process of “creative destruction” (Tsai, 2006; Osinubi and Asongu, 2021). Essentially, the literature holds that with the onset of globalization, people are forced to upgrade their abilities as well as adjust to evolving work environments because of a number of causes, including lower earnings and the potential for old occupations to be replaced with new ones. In light of these implications, women might receive training on how to increase their work prospects from the relevant industries. Therefore, as mentioned in the current research, gender-inclusive education can influence the manner in which globalization affects some macroeconomic outcomes in the recipient economies such as structural transformation. According to Grennes (2003), who argued that the labor market benefits from globalization as well, especially when it comes to differences in labor and supply brought about by the process, the positive externality from training women to be more adapted to the needs and challenges of globalization can ultimately influence how globalization affects structural transformation in the country.

To put the foregoing into better context, domestic policy may be designed so that women are given priority education in order to adapt to opportunities, challenges and restrictions arising from globalization, with the ultimate aim of improving structural transformation in the country. The idea that globalization will increase structural transformation, which is crucial for enhancing economic development outcomes like manufacturing value added, is supported by Rodrik et al. (2004) and Firebaugh (2004), who proposed that industrialization can be consolidated in developing nations via globalization dynamics, when inclusive development policies (i.e. gender economic inclusion) accompany such globalization dynamics.

Second, in relation to the competing school of thought or theoretical strand, the hegemonic viewpoint is predicated on the idea that the phenomenon of globalization is a covert plan to increase the wealth of the wealthy while lowering the prospects for similar wealth for the poor (Osinubi and Asongu, 2021). Regarding this second school of thought, Petras and Veltmeyer (2001) asserted that the goal of globalization is to increase capital accumulation and competitiveness in open markets, both of which are beneficial when supported by technologically sophisticated countries and international development agencies. According to the extant literature, globalization could lead to crises in living standards worldwide, not least because it has had negative effects on the working class and the labor market (Petras and Veltmeyer, 2001). According to Asongu (2013), globalization has had a significant impact on social democracy, particularly in the area of labor laws, which have undergone a social revolution and turned market capitalism into a social capitalism framework in response to concerns about the need to give globalization a human face. The following are further reasons in the literature supporting this second strand: (1) Tsai (2006) and Smart (2003) who have shown how globalization affects the “market ethos” that totally ignores the welfare of citizens; (2) Scholte (2000) who has discussed how the benefits of globalization are geared toward benefiting the wealthy segment of society and (3) Sirgy et al. (2004) who have discussed the negative externalities of globalization.

While the previous section has substantiated the theoretical nexus between globalization and structural transformation, understanding the value of education that is gender-inclusive in the underlying nexus is the focus of the present section. Hence, the arguments underpinning this conceptual framework are tailored to posit for the relevance of gender-inclusive education in the relationship between globalization and structural transformation. Consistent with the extant literature, how human capital influences structural changes and manufacturing in an economy is contingent on inter alia, the degree of openness and globalization prospects in a nation (Saba et al., 2024; Asongu et al., 2024). According to the narrative, how globalization affects manufacturing value added in a nation obviously depends on the nature of human capital and the quality of the labor, which embody the dimension of gender-inclusive education, not least because the involvement of more or less women in the education sector depends on the externalities from the quality of the labor force and human capital. Hence, the improvement in skills associated with the globalization process depends on how such skills are distributed between men and women in an economy. For instance, some narratives maintain that the exclusion of females in education and the labor force in the era of globalization has led to a substantial loss in manufacturing value added and by extension, gross domestic product (GDP) (Girón and Kazemikhasragh, 2022).

In light of the above, in the era of globalization, improvement in skills and policies of lifelong learning in view of improving favorable externalities from structural transformation considerably depend on how the potential labor force is inclusive in many walks of life, including how female are more involved in education prospects (Saba et al., 2024; Asongu et al., 2024). Accordingly, both inclusive development policies and globalization affect structural transformation and/or structural change in economies and thus the intuitive complementarity of inclusive education and globalization for structural transformation.

In keeping with the purpose of the research, in order to enhance readability and flow, as well as improve insights into the empirical section that follows this section, it is worthwhile to summarize the problem statement in terms of a conceptual framework. In Figure 1, globalization is the channel through which gender inclusion education (i.e. primary, secondary and tertiary) affects structural transformation. Hence, considering that gender-inclusive education is used as a moderating variable, the conceptual framework within the scope of interactive regressions exclusively entails the unconditional and conditional or interactive effect of globalization on structural transformation. Moreover, sub-channels of globalization (i.e. political, economic and social) are provided distinctly in Figure 1 because in the corresponding empirical analysis underpinning the figure, beyond the assessment of the net effect of globalization on structural transformation, the net effects of various globalization components on structural transformation are also assessed.

Our study centers on 41 African nations over the years 2004–2021. The corresponding information is gathered from three main sources. The KOF Swiss Economic Institute, the International Labour Organization (ILO), and the World Development Indicators (WDI) of the World Bank are the three primary sources of the data, as indicated in Appendix 2. Accordingly, Appendix 2 of the study contains the complete definitions of the variables together with a disclosure of the sources that correspond to them. According to the body of existing literature on industrialization, the primary outcome variable is manufacturing value added (Asongu and Odhiambo, 2022; Asongu et al., 2023). It is important to remember that, in line with Appendix 2, manufacturing value added (as a percentage of GDP) is a sector's net output after deducting all of its outputs from the equivalent intermediate inputs. In essence, it is measured without subtracting depreciations linked to assets that are fabricated or natural resource degradation and depletion. Moreover, the political, economic and social dimensions of the globalization variables are also in line with the extant globalization literature (Dreher et al., 2008; Asongu et al., 2020a). As a result, a composite measure of globalization is employed that includes social factors (i.e. information flow, interpersonal relationships and cultural proximity), political features (such as the number of foreign embassies, participation in international organizations and the number of international agreements signed by the nation) and economic factors (i.e. investment and trade flows, entailing restrictions on attendant flows).

It is important to highlight that the KOF index of globalization takes into account a wide range of variables, including capital flows and trade openness, while discussing the transmission channels. These also cover cross-border contacts between citizens and interactions between governments (Tifuh, 2022). The index is important because it allows for various compounding of different globalization elements and dimensions, showing for the respective nations the annual level of globalization, according to the associated narrative. It is important to keep in mind that total globalization dynamics have sub-components that are used in this study as potential transmission pathways (Osinubi and Asongu, 2021). Economic globalization is comprised of financial and trade globalization, whereas social globalization is composed of cultural, informational and interpersonal globalization dynamics. The sub-dimensions of political globalization are not yet available, though, because of restrictions on data availability at the time of this study. In order to evaluate the transmission channels, the current study utilizes the composite globalization index, which includes the three primary characteristics of globalization including sub-dimensions of social globalization and economic globalization.

In keeping with the purpose of the research, three main moderators or gender-inclusive education are used, notably primary inclusive education, secondary inclusive education and tertiary inclusive education. The selection of three levels of inclusive education is built on the narrative that various levels of education are important in driving industrialization in a country (Asiedu, 2014). Consistent with Appendix 2, it is important to clarify that (1) the ratio of girls to boys enrolled at the elementary and secondary levels in public and private schools is known as the gender parity index (GPI), and it measures the gross enrollment ratio in primary and secondary education. (2) Secondary (gross) school enrollment GPI, which displays the GPI for secondary education's gross enrollment ratio, is the proportion of secondary-aged girls to boys enrolled in public and private schools. (3) School enrollment, tertiary (gross), GPI, which stands for gross enrollment ratio in tertiary education, shows the proportion of women to men enrolled in public and private post-secondary education. To account for variable omission bias, two primary control variables are employed, in keeping with the current inclusive development research. These are GDP per capita and population growth (Efobi et al., 2019; Asongu et al., 2020b; Asongu and Odhiambo, 2023). The use of limited control variables is worthwhile in GMM-oriented regressions in order to avoid concerns related to instrument proliferation, in accordance with extant literature (Osabuohien and Efobi, 2013). The paragraphs that follow go into further depth about these chosen control variables.

First of all, population growth is predicted to have a negative impact because as a country's population grows, the number of people employed in a given economic sector is predicted to decrease (Osinubi and Asongu, 2021). This is especially true if job creation is not correlated with the growth of the population on a proportional basis.

Second, it is impossible to determine the impact of GDP per capita since the latter depends on how different economic sectors contribute to the corresponding GDP per capita and how the benefits of economic prosperity are spread among them. For instance, certain findings in recent literature (Asongu et al., 2020a) do not support the idea that a growth in GDP might create greater chances for women to participate in the official economy, especially by means of gender-inclusive education. According to research (Tchamyou, 2020; Tchamyou et al., 2019; Bicaba et al., 2017; Asongu et al., 2020a, c), since the economic growth recovery over the last decade in SSA has not been tapped into to promote equitable development outcomes, it is expected that economic growth is adversely correlated with the outcome variable taken into consideration in this study.

The list of countries included in the study is disclosed in Appendix 1. The variables are defined in this section of Appendix 2, while Appendix 3 contains the summary statistics. Appendix 4 has the related correlation matrix.

The present study uses the generalized method of moments (GMM) empirical strategy as an estimating technique, which is consistent with the motivational aspects described in the introduction, particularly as it relates departing from the study by Tifuh (2022) which has focused on fixed-effect regressions. Additionally, elements such as the underlying methodology's compatibility with the data structure as well as certain robustness components of the methods that are also apparent are covered in the paragraphs that follow. The emphasis of the study is on 41 nations during the time frame 2004 to 2021. It follows that the number of countries is more than the number of years in each period. This is a primary condition for the employment of the GMM empirical technique. Second, the outcome variable seems to be persistent, especially considering that the correlation between the outcome variable's first lag and level series is greater than 0.800, which is the standard recommendation made by the literature on the application of GMM as an estimation technique (Tchamyou, 2019; Efobi et al., 2019). Third, the estimation technique takes into account some aspects of endogeneity, comprising (1) time or year effects, which take into account unobserved variability and cross-sectional dependency, and (2) an instrumentation process, which takes into consideration the simultaneity or reverse causality dimension of endogeneity. It is crucial to consider these endogeneity components in order to provide reliable results. The estimation approach for the relevance of gender-inclusive education in globalization dynamics to influence structural transformation is summarized in the following equations at levels (1) and first difference (2).

where $MVA$ denotes manufacturing value added; $∅0$ is the constant; Edu represents gender-inclusive education; $Gblo$ stands for a globalization dynamic (i.e. total globalization, social globalization, political globalization, trade globalization, cultural globalization, interpersonal globalization, cultural globalization and informational globalization); Inter denotes the interaction between inclusive education and a globalization dynamic; $W$ represents the vector of control variables (Population and GDP per capita); $τ$ is the unit coefficient of autoregression given that a lagged year is sufficient to capture past information; $ωt$ is our time-specific constant; $φi$ denotes the country-specific effect and $εit$ represents the error term. According to Roodman (2009), the GMM analytical technique used in the current study is based on forward orthogonal deviations. As a result, it is an enhanced version of the Arellano and Bover (1995) method that has been developed to produce findings with more reliable estimated coefficients than more conventional difference and system GMM approaches (Boateng et al., 2018; Efobi et al., 2019; Tchamyou et al., 2019).

Insights into features related to identification, simultaneity and exclusion restrictions that are particularly relevant in a worthwhile specification are important in a robust GMM method. The keywords underpinning this section are elaborated in turn in what follows (i.e. concepts of identification, simultaneity and exclusion restriction). The framework for identification, which is the first step, clarifies the three key ideas that form the basis of the specification framework: the dependent variable, the endogenous explanatory or predefined variable and the strictly exogenous variable. The outcome variable for this study is manufacturing value added which is a proxy for structural transformation, according to the study's motivation as well as disclosed aspects in the data section. The predefined or endogenous explanatory factors are all independent variables of interest, which is consistent with the existent GMM-centric research (Tchamyou and Asongu, 2017). According to the extant GMM-centric literature, these endogenous explanatory variables constitute population and GDP per capita which are both considered control variables as well as inclusive education and globalization dynamics, which are the independent variables of interest (Meniago and Asongu, 2018). Years are regarded as strictly exogenous variables in the study since; in accordance with Roodman (2009), they are unlikely to become endogenous after a first difference. It follows that years which are also used to compensate for cross-sectional dependency and the unobserved heterogeneity are acknowledged as exhibiting strict exogeneity.

Second, focusing on the simultaneity aspect of this section, it is important to highlight how instrumental variables employing forward differences are employed to deal with the problem of reverse causation (i.e. both the independent variable of interest and the outcome variable affecting one another). In order to remove fixed effects that may potentially skew estimated coefficients due to the connection between fixed effects and the lagged dependent variable, Helmert transformations are used. Considering the fundamental transformation that permits forwarding differenced and lag observations to have parallel or orthogonal conditions (Love and Zicchino, 2006; Arellano and Bover, 1995; Roodman, 2009), such a procedure for reducing fixed effects is compatible with the associated research on the topic (Tchamyou and Asongu, 2017).

Third, keeping with the topics covered in the first strand of this section and focusing on the exclusion restriction hypothesis, the Difference in Hansen Test (DHT) is used to evaluate the supposition that the identified strictly exogenous variables influence the outcome variable only through the exogenous dimension of the endogenous explaining variables. The null hypotheses of Hansen/Sargan tests should not be rejected so that the instruments that are identified explain the manufacturing value added, exclusively via the globalization mechanisms that are identified, according to the information criteria for the validity of instruments, which are consistent with more conventional GMM-centric approaches (Amavilah et al., 2017).

According to Asongu and le Roux (2019), there are four basic sources of endogeneity: (1) heterogeneity that is unobserved; (2) bias resulting from variable omission; (3) errors in the measurement of variables and (4) simultaneity or reverse causality. The proposed empirical technique and analytical approach are designed to correct the first and fourth endogeneity problems because it is challenging for a single estimating approach to address all endogeneity concerns. As a result of taking into account country-fixed effects, by definition, the latter considers the unobserved heterogeneity. Moreover, the instrumentation process is tailored to address the concerns of simultaneity or reverse causality during the estimation process using internal instruments.

Tables 1–7 in this section present the empirical findings. The main findings and corresponding linkages pertaining to gender-inclusive primary education are disclosed in Tables 1 and 2 in Section 4.1, while the corresponding findings related to gender-inclusive secondary education are provided in Tables 3 and 4. The attendant results focusing on nexuses between globalization, tertiary inclusive education and structural transformation as well as corresponding channels are provided in Tables 5–7.

This section focuses on how gender-inclusive primary education moderates the incidence of globalization (i.e. overall, social, economic, political) on structural transformation, as shown in Table 1, while the corresponding social globalization (i.e. interpersonal, informational and cultural) channel is shown in Table 2. It is important to remember that the other globalization channels are not considered apart from the social globalization channel owing to data availability constraints and/or models that did not pass post-estimation diagnostic tests ^[1]. Hence, owing to the space constraints, for the most part, only models from which relevant policy implications can emerge are disclosed in this section (and in subsequent sections). In other words, it is important to remember that although Table 1 presents the primary conclusions on the overall dynamics of globalization (i.e. total globalization, social globalization, economic globalization and political globalization), the corresponding channels of social globalization are engaged in Table 2 in light of the significance of the social globalization channel in Table 1. It follows that the channels of other globalization dynamics (i.e. economic and political) are not reported owing to their non-significance and space constraint.

In order to assess the investigated hypotheses, two main information criteria are worthwhile: (1) the estimated GMM models should be valid on the one hand and (2) on the other, the corresponding information criteria for calculating the net effects and/or thresholds must also be met. In essence, it is worthwhile to articulate that in order to avoid the downfalls of interactive regressions mentioned in Brambor et al. (2006), total effects and/or thresholds are computed so that the corresponding interactive estimated coefficients are not interpreted as in linear additive models. To put this computation framework in more perspective, in the first specification or second column of Table 2, the net effect from the role of inclusive primary and secondary education in moderating interpersonal globalization to influence structural transformation is 0.062 ([−0.410 × 0.919] + [0.439]). In the corresponding specification, 0.919 is the mean value of primary and secondary inclusive education, 0.439 is the unconditional effect interpersonal globalization on manufacturing value added, while −0.410 is the conditional or interactive impact of interpersonal globalization on manufacturing value added. Some net impacts are not calculated because at least one estimated coefficient required for the related computation is not significant, as explained in the accompanying table's footnote. It is worthwhile to note that the underlying computation is consistent with contemporary literature (Nchofoung and Asongu, 2022a, b; Nchofoung et al., 2021, 2022) on the interactive regressions, especially as it pertains to the importance of avoiding pitfalls of interactive regressions documented in Brambor et al. (2006).

Still in accordance with the corresponding literature on interactive regressions, while the net effect used in the underlying example in the previous paragraph is positive, the corresponding conditional or interactive incidence is negative. The implication here is that while primary and secondary inclusive education moderates interpersonal globalization in order to engender an overall positive effect on manufacturing value added, there are maximum levels of the primary and secondary gender-inclusive education index that once attained, the total effect changes from positive to negative. Hence, at the corresponding thresholds, complementary policies are worthwhile in order to maintain the overall positive effect from the moderating role of primary and secondary inclusive education in the incidence of interpersonal globalization on manufacturing value added. To have a better understanding of the underlying insight, based on the previous example, the primary and secondary inclusive education threshold for complementary policy is 1.070 (0.439/0.410).

In light of the aforementioned, the following findings are clear in Tables 1 and 2, especially as it refers to the importance of primary and secondary inclusive education in influencing globalization for manufacturing added value. First, primary and secondary inclusive education does not significantly moderate overall globalization and corresponding social, economic and political globalization dynamics for a significant effect on manufacturing value added (i.e. in Table 1). Second, primary and secondary inclusive education moderates interpersonal globalization and cultural globalization channels for overall positive net effects on manufacturing value added with corresponding thresholds for complementary policies of 1.070 and 0.984, respectively. The corresponding thresholds have statistical validity, make economic sense and are valuable for policy since; according to the summary data, they fall between the attendant moderating variable's lowest and maximum values. Put otherwise, 1.070 and 0.984 are situated between 0.599 (i.e. minimum) and 1.176 (i.e. maximum) in the summary statistics in Appendix 2.

This section provides findings on linkages between globalization, inclusive secondary education and structural change in Tables 3 and 4. Table 3 focuses on how gender-inclusive secondary education moderates the incidence of globalization (i.e. overall, social, economic, political) on structural transformation, while the corresponding social globalization (i.e. interpersonal, informational and cultural) channel is shown in Table 4. Accordingly, Table 3 presents the primary results, although Table 4 discloses results related to corresponding channels. The information criteria used for the assessment of findings in the corresponding tables are similar to the information criteria used in the assessment of results provided in Tables 1 and 2. Based on the considered information criteria, results from Tables 3 and 4 support the following conclusion. First, secondary inclusive education does not significantly moderate overall globalization and corresponding social, economic and political globalization dynamics for a significant effect on manufacturing value added (i.e. in Table 3). Second, with respect to mechanisms from the social globalization channel, inclusive secondary school education moderates interpersonal globalization to engender an overall positive effect on manufacturing value added and the corresponding threshold for complementary policy is 1.284 (secondary school enrolment, gender parity index) which is higher than the maximum corresponding value of the moderator in the summary statistics.

This section provides findings on linkages between globalization, inclusive tertiary education and manufacturing value added in Tables 5–7. Table 5 focuses on how gender-inclusive tertiary education moderates the incidence of globalization (i.e. overall, social, economic, political) on structural transformation, while the corresponding economic globalization (i.e. trade and financial) channel is assessed in Table 6. Moreover, the social globalization (i.e. interpersonal, informational and cultural) channel is examined in Table 7. The main findings are shown in Table 5, Table 6 reports findings related to the mechanism of economic globalization, while Table 7 shows results pertaining to the channels of social globalization. Consistent with the previous narrative, the information criteria used for the validation of the findings as well as assessing the underlying hypotheses being investigated are maintained. Tables 5–7 allow for the establishment of the following conclusions. First, inclusive tertiary moderate globalization includes social globalization, political globalization and total globalization to create a favorable overall impact on manufacturing value added and the corresponding thresholds for complementary policies are 1.320, 1.117 and 0.821 (tertiary school enrolment, gender parity index), for the impact of globalization in its overall, social globalization and political globalization, respectively. Given that they fall inside the statistical range, the thresholds have policy relevance. Second, from the channels of economic globalization, inclusive tertiary education moderates financial globalization for a positive net effect on the outcome variable with a corresponding threshold for complementary policy of 0.734 (tertiary school enrolment, gender parity index) which is within the statistical range.

Third, from the channels of social globalization, inclusive tertiary education moderates (1) interpersonal globalization for a favorable net effect on the outcome variable with a corresponding threshold for complementary policy of 1.138 (tertiary school enrolment, gender parity index) and (2) cultural globalization for a positive impact on the dependent variable, with a corresponding threshold for complementary policy of 0.981. The computed thresholds make economic senses because they are within the statistical range.

Concerning the nexus of the results with the existing body of literature, it is worthwhile to note that the findings are broadly consistent with the strand of literature positing that globalization has had positive effects in terms of gender inclusion, which obviously leads to other favorable externalities in the economy (Islam and Muzi, 2022; Dinkelman and Ngai, 2022; Thaddeus et al., 2022).

The current study aims to evaluate the ways in which the impact of globalization on structural change in Africa is mitigated by gender-inclusive education. The empirical data are based on the interactive GMM, and the focus is on 41 nations from 2004 to 2021. The following conclusions are established. First, some positive net effects are established, which implies that gender-inclusive education broadly moderates globalization as well as the corresponding channels to engender an overall positive effect on structural transformation in the sampled countries. These positive effects primarily manifest through specific globalization channels (e.g. interpersonal globalization, cultural globalization and financial globalization) and at higher education levels (tertiary education). However, given the negative conditional or interacting consequences, thresholds for complementing policies are suggested. These thresholds for complementary policies are maximum levels of gender-inclusive education that when attained, complementary measures are needed in order to sustain the positive role of gender-inclusive education in moderating globalization for a generally favorable impact on structural transformation. The following subsection discusses the ramifications for policy.

The first implication for policy is that inclusive globalization policies are likely to engender rewarding outcomes in terms of industrialization particularly in relation to enhancing manufacturing value added. This aligns with growing calls by scholars and policymakers for globalization to be given an African face (Seoane, 2022; Verkhovets and Karaoğuz, 2022), not least because traditionally, most countries in Africa have been constrained to produce what they do not consume and consume what they do not produce (Chipato, 2023; Byiers and Woolfrey, 2023). It follows that the transformation of natural resources and agricultural produce into finished and semi-finished forms is particularly worthwhile if complemented with gender-inclusive education policies. However, as shown from the findings, gender-inclusive education is a prerequisite, but not a sufficient condition, for globalization dynamics to positively affect industrialization in the sampled countries.

Second, as shown in the findings, at certain thresholds of gender-inclusive education, gender-inclusive policies must be supplemented by further regulations so that the corresponding incidence on globalization leads to an improvement in manufacturing value added. Accordingly, these thresholds for complementary policies are intuitive and make economic sense, not least because most of the computed thresholds are approaching the maximum levels of the gender-inclusive educational indexes apparent in the summary statistics. It follows that policymakers should work towards establishing the empirical relevance of complementary inclusive education with the relevant measures that can drive the positive incidence of globalization on manufacturing value added further. Such policies can entail, inter alia, targeted female financial literacy and entrepreneurship programs that enable the female gender to not only be educated but equally be equipped with adequate financial and business tools needed to take advantage of industrial opportunities offered by globalization dynamics.

Third, the favorable influence on gender-inclusive education on manufacturing added value via globalization is contingent on the aggregate globalization proxy (i.e. total, political, economic or social) as well as on the corresponding channels of the attendant aggregate globalization dynamics. It follows that the policy recommendations prescribed in the previous two paragraphs should not be blanket but contingent on the considered globalization channels. Hence, policymakers who simply acknowledge (i.e. form this study) that gender-inclusive education moderates globalization for a positive effect on industrialization can be getting their dynamics badly wrong, not least because as we have established, there are contingencies in the nexuses and thus, specific policy prescription and implementation should be substantiated with empirical validity.

In terms of theoretical implications, the present study has reconciled both the endogenous growth theory and the exogenous growth theory, not least because according to the findings, economic development by means of structural transformation can be induced by globalization (i.e. an external force) and gender-inclusive education (i.e. as a result of an internal process). Hence, the exogenous growth theory is confirmed by the relevance of globalization, while the endogenous growth theory is confirmed by the importance of human capital in driving structural transformation.

The study apparently also allows space for future research, especially in view of understanding how other inclusive development moderators can influence the incidence of globalization dynamics on structural transformation in the sampled region and by extension, on other developing countries. Moreover, this empirical analysis can be considered within the remit of other developing countries, not least because the findings established within the sampled region cannot be extended to other developing countries unless such extension is substantiated with empirical validity. Furthermore, future research should take into consideration country-specific studies with the relevant estimation techniques for more country-specific policy implications, considering the estimation technique's apparent flaw of not being designed to account for country-fixed effects in order to mitigate endogeneity concerns. In addressing these issues in future research, incorporating comparative frameworks across regions and countries can enhance understanding of the nexuses among globalization, gender-inclusive education and structural transformation, especially when factors such as technological advancements and international aid are involved for a more holistic view of the subject. These underlying future research directions obviously build on limitations of the present study. Furthermore, the findings and corresponding policy implications are exclusively relevant to the context of Africa. Hence, generalization should be taken with caution. Accordingly, the study is based on 41 nations for which information was accessible at the time of the study. Therefore, extension of the findings to other regions of the world should be subject to empirical validity; this serves as a path for further investigation.

All authors read and approved the final version.

The supplementary material for this article can be found online.