Special economic zones, gender and innovations: new evidence from an emerging economy

Abstract

This paper examines how the gender of CEOs affect product and process innovations at the firm level, focusing on how SEZ features affect the gender-innovations nexus. The study employs data containing 328 firms from 2018 to 2021 and panel logit and treatment estimation models. The results show that female CEOs are more likely to lead registered SEZ firms and engage more in product and process innovations. Moreover, while female CEOs contribute positively to product and process innovations, female CEOs in registered SEZ firms and SEZ enclaves tend to have a higher and profound contribution to product and process innovations than their counterparts. These results suggest that national and corporate policymakers should develop and implement corporate and national level policies that (i) limits the glass ceiling barriers in order to propel the contribution of females to innovations and (ii) lower the bureaucratic and complex administrative procedures for operationalizing SEZ firms in order to harness the innovational-enhancing effects of gender inclusiveness.

KEYWORD:

• Gender
• SEZs
• innovations

REVIEWING EDITOR:

• Oyedele Ogundana, Nottingham Trent University, United Kingdom of Great Britain and Northern Ireland

SUBJECTS:

• Business, Management and Accounting
• Economics
• Gender Studies - Social Sciences
• Gender & Development

Introduction

Innovations are important drivers of competitive advantage, growth and prosperity in modern societies for both corporate entities and economies (Chen et al., 2021; Dohse et al., 2019). As such corporate entities, governments and other global institutions spend billions of financial resources to enhance innovation and innovativeness in order to boost corporate, national and global competitiveness, respectively (Dohse et al., 2019). Despite the importance contribution of innovations and innovativeness to corporate and national entities performance, some aspects of innovations are yet to gain the needed attention of academics and policymakers especially in emerging economies in Africa like Ghana where managerial, financial and technological resource allocation for innovational activities are at best scanty. Thus, existent innovation research/studies have mainly focused on where innovation takes place, how innovations take place and who takes part in innovational activities (Dohse et al., 2019; Fagerberg & Srholec, 2008). Interestingly, while in recent times, there have been increasing recognition and focus on gender aspect of innovation (Dohse et al., 2019), this research/study area (gender aspect of innovation) is still an emerging area of study particularly in African emerging economies where innovational resource allocation are little/limited.

While gender aspect of innovation is gaining attention, males are argued to dominate corporate and public positions particularly in African emerging economies (Abor, 2017; Amidu & Abor, 2006; Luh & Kusi, 2023) mostly due to the traditional and cultural roles assigned to females. As a result, policymakers and academics in the last few decades have made strong arguments in favor of gender diversity particularly female inclusion/involvement in both corporate and public leadership (Luh et al., 2022; Luh & Kusi, 2023). Evidently, global policy views the involvement/inclusion of females as a key driver for economic growth and development and inclusiveness and this is reflected in the sustainable development goal 5. Arguably, female inclusion improves and widens the pool of talents and human resource availability (Doldor et al., 2012), present variety in perspectives and opinions/ideas, strengthens the monitoring and supervision functions of leadership (Ain et al., 2021; Cambrea et al., 2020; Chen et al., 2016) and lowers risk appetite/assumption (Belas et al., 2015; Chen et al., 2016; Jia, 2019; Jane Lenard et al., 2014; Yu, 2023). As a result of these important effects of female leaders/female-led firms, the gender based empirical literature is dominated with studies that focused on how female leaders affect firm profitability (Luh et al., 2022; Luh & Kusi, 2023) and risk management (Ertac & Gurdal, 2012; Shropshire et al., 2021) and these are focused on financial institutions particularly banks.

Viewed from somewhat different perspective, few studies document how innovation is influenced by female leaders and or female business owners (see Bauweraerts et al., 2022a; Dohse et al., 2019; Foss et al., 2022; Millward & Freeman, 2002) but such studies in African emerging economies like Ghana are scanty. Thus, how female leaders/female-led firms influence innovations differently from those firms not led by females are limited particularly in emerging economies in Africa. Beyond the scanty ‘female leadership-innovation’ empirical studies in African emerging economies, how economic policy induced firms (specifically special economic zone firms) influence innovations is yet to be documented in the innovation literature for emerging economies in Africa. We argue that economic policy induced firms such as SEZs have stronger adherence to international/global policies such as sustainable development goals 5 (gender equality and women empowerment) (Chaisse & Dimitropoulos, 2021; Chaisse & Ji, 2020; Ramamoorthi, 2012) and have better managerial, financial, technological expertise (Audretsch et al., 2019; Chen & Yang, 2017; Iqbal et al., 2020; Yang et al., 2017) which can provide the right working environment to reinforce the female leadership-innovation nexus. Despite these complementing conditions offered through economic policy induced firms (SEZs) on the ‘female leadership-innovation’ nexus, empirical studies are however silent and yet to document the reinforcing effects of economic policies induced firms (SEZs) on the link between female leadership and on firm innovations.

Additionally, we take advantage to the difference in the structure, operationalization and practice of SEZ policy in Ghana to offer detailed evidence. Thus, the Ghanaian SEZ architectural policy provides room for non-SEZ registered and SEZ registered firms to operate from within SEZ enclaves while allowing SEZ firms that desire to operate from outside SEZ enclaves to exist. These unique features of the SEZ policy in Ghana offers the opportunity to investigate beyond how female leaders generally impact innovations to further our understanding on how female leaders influence innovations in register SEZ firms, Registered SEZ firm operating outside SEZ enclaves and SEZ firms operating from within SEZ enclaves. Furthermore, unlike most prior studies that sum up product and process innovations, we explore the linkage between female leadership and economic policy induced firms on product and process innovations to offer detailed perspective from an African emerging economy context. Therefore, the three (3) contributions of this study to the literature includes: (i) presenting an African emerging economy perspective to ‘female leadership-innovation’ nexus, (ii) amplifying how economic policy induced firms (SEZs) influence the link between female leadership and innovation in an emerging economy context and (iii) using different indicators of innovations to ascertain detailed understanding on female leader-SEZ-innovation nexus in an African emerging economy. The rest of the paper is organized into brief overview, literature review, methodology, empirical results and discussions and conclusion, policy recommendation and implications

Brief overview of female leader, SEZ economic policy and innovations

Female diversity at the corporate and public levels have become an important issue especially with the emergence of inclusive economic growth and development (Kabeer, 2021; Kim, 2017). From world history perspective, the periods between 1975 and 1995 reports global events and milestones (which include four UN World Conference on women in Mexico (1980), Copenhagen (1985), Nairobi (1990) and Beijing (1995)) that influenced major world decisions/agreement on women rights (Esquivel & Sweetman, 2016; Sen, 2019). Hence, the sustainable development goal 5 which seeks to achieve gender equality and empowering of all women and girls’ projects one of the key global/international agreements. It is argued that, if the world is to achieve sustained economic growth and development, the need to recognize minority and marginalized groups such as women in leadership positions cannot be overlooked and plays a critical role. Hence, this international agreement/goal (sustainable development goal 5) has huge importance in directing policy decisions and resources and reinforces the cry for fighting injustice and marginalization emanating from cultural, traditional and social norms (Agarwal, 2018; Esquivel & Sweetman, 2016; Kim, 2017). As such, multinational/global businesses and corporate entities have strong adherence to such international agreements and makes room for women. Special economic zone (SEZ) firms which are designed to boost local economy and firms through spillover and agglomeration effects can also serve as conduit through which the competence of female leaders can be enhanced, recognized and highlighted (Chaisse & Dimitropoulos, 2021; Chaisse & Ji, 2020). Similarly, given that SEZs have better technology, infrastructure, human resource and technical know-how, they serve innovate

In the corporate setup of Ghana, a number of evidence on the involvement and participation of females in corporate and public sector leadership positions have been documented. For instance, Amidu and Abor (2006) report that female board member for listed firms stood at about 7% as at 2006 representing 18 out of 247 board membership positions. Appiadjei et al. (2017) reported that financial sector firms reported the highest female board representation of 16% while ICT, trading and pharmaceutical listed firms report only 1% and automobile reporting zero female board representation. In recent times, 23% of board seats are held be female and 24% of nonexecutive board position were held be females. Similarly, the financial sector reported the highest female board representation of 38% while females constituted 8%, 15% and 13% of chief executive officers, chief operating officers and chief financial officers.

In the public setup¹, as at 2015 female constituted 29% and 22% of ministers and deputy ministers, respectively. Additionally, as at 2013, 29 and 25% of Supreme Court Judges and High Court judges were females. Around the same time, 24% of chief directors were females while 8.2% and 10.9% of municipal/district chief directors and parliamentarians were females, respectively. While these trends show improvements in female involvement and participation in corporate and public leadership, the trends are indications that females are still underrepresented in Ghana.

In the light of this present study, it observed that out of 328 firms (see Table 1), 12.81% had female CEOs. Within special economic zone firms, 14.14% of CEOs were females while within the non-SEZ firms female CEOs represented 12 .16% of total CEOs. However, in terms of innovativeness, about 56.40% and 45.43% of the 328 firms introduced new products and processes in the operations, respectively. Furthermore, in the SEZ firm sample, about 52.83% and 40.57% of the 106 firms introduced new products and processes in the operations while in the non-SEZ firm sample, about 58.11% and 47.75% of the 222 firms introduced new products and processes in the operations. These trends show that non-SEZ firms turn to be more innovative compared to their SEZ firm counterparts. However, when firm characteristics such as SEZ registration and female CEO statuses are considered jointly, it is observed that while 16.07% and 16.28% of firms that were jointly led by female CEOs and registered as SEZ firms were more product and process innovate compared to their counterpart firms who were led by female CEOs but not registered as SEZ firms. This provides evidence that supports the idea or notion that female CEO led SEZ firms tend to enhance innovativeness better than female led non-SEZ firms.

Table 1. Trend in female CEOs and innovations.

	Full Sample	SEZ Registered Sample	Non-SEZ Sample
Firms	328(100%)	106(32.32%)	222(67.68%)
Female CEO	12.81%	14.15%	12.16%
Female CEOs Engaged in Product Innovation	13.51%	16.07%	12.40%
Female CEOs Engaged in Process Innovation	13.42%	16.28%	12.26%
Product Innovation	56.40%	52.83%	58.11%
Process Innovation	45.43%	40.57%	47.75%

Literature review: theoretical and empirical review

Theoretical review

From a theoretical perspective, theories including gender socialization theory and social role theory argue that and differences and similarities in gender are socially constructed and roles assigned to specific gender is based on distribution of men and women. These theories being psychological theories, provide evidential differences in the roles/behavior of males and females arguing that the roles/behavior of males and females are formed through socialization. Theoretically, it is argued that societies are organized in a manner that makes females exhibit specific roles/behaviors like helpfulness, kindness, sympathy and interpersonal skill while males exhibit specific roles/behaviors like assertiveness, aggressiveness, ambitiousness, forcefulness and competitiveness through socialization (Eagly et al., 2000). While these specific roles/behaviors greatly influence the leadership skills and quality, the differences in the roles of gender as assigned through socialization is deemed to impede the corporate success of females and has come under critical criticism. The limiting effect of the socially assigned roles/behavior for females has induced the request for women empowerment which is focused on giving equitable chance/opportunity to both genders. In line with this, the global sustainable development goal 5 (Achieve gender equality and empower all women and girls) provide evidential claim for indicating that the female gender is vulnerable and disadvantaged and hence a need to empower them to overcome the challenges that are associated with roles assigned to the female gender through socialization.

Arguably, while having and or promoting females into public and corporate leadership roles is deemed to be a critical success factors for inclusive growth and development attainment, the establishment of SEZs is also viewed as an important economic force for facilitating the request for inclusive growth and development particularly in emerging and developing economies. Following the global chain value approach of establishing SEZs which is driven by the powerful forces of globalization, specialization in production and access to cheaper and high-quality source of factors of production, SEZs are argued to offer an opportunity for economies particularly emerging and developing economies to be integrated into the global value chain while improving the human resource particularly because SEZs operate under high global standards which offer females the needed and necessary conditions and environment that promote the performance of female leaders working SEZ firms. Theoretically, it is clear from the global value chain approach of SEZ that SEZs operate under international standards and have better working conditions that can lower the limiting social and traditional factors that impede the effective contribution of females to firm performance and innovations. Hence, a theoretical justification that female working in SEZ registered firms are more likely to performance better than the counterparts.

As such, following the theoretical review, it is theoretically evident that although the roles performed males and females are social constructed to the disadvantage of females and SEZs through the global chain value operate under international standards and high quality conditions of service offer a better working environment, it is intuitive and rational to argue that the working conditions and environment of SEZs can offer the needed support that can propel female leaders working SEZ firms to be more innovative compared to their counterparts that are working in non-SEZ firms. However, empirical evidence to substantiate this claim is non-existence especially in the context of emerging economies in Africa. Thus, the existent empirical literature is focused on examining either the effect of females on innovations or SEZ on innovations with no attention paid to how females working in SEZs and SEZ enclaves influence firm innovation performance.

Empirical review

Moving on to the empirical review, this study focuses on two strands of empirical studies namely: (i) female effects on innovations and SEZ effects on innovations. From the literature, while earlier empirical literature employed total factor productivity as a proxy for innovation, recent empirical literature use introduction of new product, service, process, entering new markets, research and development expenditure, patent application, patent granted and patent citations as more effective measures of innovation. These innovations are classified as green, environmental and corporate/industrial innovations. On the side of SEZs, studies have used different types of SEZs which include free-trade zones, free economic zones, bonded logistic parks, urban enterprise zones, development zones and export processing zones to assess the contribution of SEZs on innovations.

Females and innovations

Following the empirical literature, it is observed that existent studies on female effects on innovations are predominately focused on China and Asian economies and can be classified as those that focus on green/environmental innovation and those that focus on corporate/industrial innovations. Focusing on green innovations for instance, Javed et al. (2023) studied how female CEOs influence green innovations while moderating this relationship with other factors including state ownership, regional development, firm size and industry type. Using fixed effect panel data of listed firms from Shanghai and Shenzhen stock exchange markets between 2008 and 2016, their result show that female CEOs contribute more to green innovation compared to their male CEO counterparts. Furthermore, the study reveals that state owned firms, firms located in developed regions, large firms and firms in environmentally sensitive sectors positively and significantly moderate the effect of female CEOs on green innovations compared to their male counterparts working in firms with similar features. More so, Moreno-Ureba et al. (2022) analyze the extent to which board gender diversity influences environmental innovation while considering the moderating effects of CSR meeting, board size, board tenure, board meetings and board independence. The study used fixed and random effect panel models and quantile regression to estimate the effect of board gender diversity on environmental innovation. Using a sample of 250 listed firms on the FTSE, the findings show that environmental innovation is positively and significantly influenced by board gender diversity but can only be possible for firms with higher levels of environmental innovation and when there are at least three women in the boardroom. Additionally, the results show that the prove of the one-size-fits-all approach is not appropriate and sufficient to understanding the role of female directors.

Similarly, Wang et al. (2022) examine the effect of female directors on environmental innovation using stakeholder orientation as a mediating variable. The study used a count data model of the Poisson regression to analyze a sample of 29,974 firm-year observation from 4,129 Chinese listed firms between the periods 2010 and 2020. The study posits that female directors on environmental innovation can be enhanced. Furthermore, stakeholder orientation serves as a mediating link that results for the effect of female directors on innovation. Likewise, Li et al. (2018) investigated the moderating effect of ownership type of female directors on environmental innovation. The study was premised on the upper echelon’s theory, feminist caring theory, and social role theory to examined the moderating influence of ownership type of female directors on environmental innovation. A sample of 688 Chinese listed manufacturing firms were used for the study. Employing the multiple regression models, the study found that when the number of female directors increased, it interns increase environmental innovation. Additionally, when female directors are grouped into independent and non-independent directors, female independent directors had a positive and significant influence on environmental innovation of firms. Howbeit, firms’ environmental innovation is less influenced by female chairpersons

In the light of corporate/industrial innovations (patent application, citation, R&D expenditure, new product, service and process introduction), Tahmooresnejad and Turkina (2022) assess the impact of the collaboration effect of women inventors on their innovation production and innovation impact. The study uses social network analysis panel data models to analyze data of international inventors and patents by the European Patent Office from 1978 to 2019. The results show that women involvement in patenting is not high but the share of female inventors increase over the period (1978–2019) from 1.2% to 8.9%. A panel regression model was used to examine the number of patents and citations rates of the patents at each individual level and the findings show that all investors may benefit from the co-patenting, nevertheless, there is a positive influence greater for males than their female counterparts. Similarly, Iman et al. (2022) investigated the extent to which women exhibit patterns in entrepreneurial behavior as compared to men in the context of post-communist societies. Using a sample of 11,617 firms survey data from 25 Eastern Europe and Central Asian states, the study made 11,000 observations and multiple checks to check the robustness of the data. Regressions analysis was used to analyze the data and the findings posit that female firm owners are likely to be more inclined to introduce novel marketing strategies than their male counterparts. Also, in democratization, female managers in top positions are more likely to increase innovation than those who are not, meanwhile these female managers in top leadership are less likely to invest in research and development unlike their male counterparts and finally, democratizations can eliminate disadvantages in gender and increase innovation among firms.

Moreover, Javaid et al. (2023) investigate the impact of female CEOs (She-E-Os) on firm innovation of listed firms in China while additionally exploring the moderating effect of internal corporate environment on the link between She-E-Os and innovation. The study finally examined whether financial constrain moderates the impact of female CEOs on innovation. Using data of all A-share listed companies on the Shanghai stock exchange covering a period between 2008 and 2017, the study employed battery of estimation models to explore the linkages between female CEOs and innovation. The study reveals that the gender of CEOs improves the level of innovation output through improvements in governance structure. On the other hand, female CEOs have no significant effect on state-owned enterprises innovation activities, highlighting the fact that objectivity is the major goal of state-owned enterprises. Additionally, the study posits that the positive effect of female CEOs have on innovation is weakened in the presence of financial constraints. Also, Wu et al. (2021) examined the effect of female chief technology officers on innovation using 5408 firm-year observations covering periods between 1991 and 2010. Employing a battery of estimations techniques including Poisson, negative and positive binomial models their results show that firms with a stronger innovation-support have more of their female chief technology officers contributing more to innovation than their male counterparties. Furthermore, the study reported that the leadership style of FCTO positively and significantly mediated the link between the presence of FCTO and innovations compared to their male chief technology officer counterparts.

SEZs and innovations

In terms of the contribution/effect of SEZs on green innovations, Xie et al. (2022) examined how national and provincial development zones influenced green innovation using time varying difference in difference models covering periods between 2000 and 2017. The results show that while national development zones significantly enhanced green innovation of listed Chinese firms, provincial development zones marginally inhibit the quality of green innovation. Furthermore, the results show that while supply chain finance, environmental regulations and green finance were reported as the mechanisms through which national development zones exerted beneficial effects on green innovation, the laxity in environmental regulations contributed significantly to the negative effect of provincial development zones on green innovation. Likewise, Nie et al. (2022) investigated the impact of local government competition and development zones on green innovations using Chinese city-level data covering periods between 2000 and 2018. Employing spatial durbin model, the results show that local government competition has significant positive effect on green innovation and this result is especially significant when there is existence of development zones. This finding implies that development zones have enhance significant effects on the link between local government competition and green innovation.

In the context of SEZs and corporate/industrial innovations, Ouyang et al. (2022) explored development zones’ effects on innovations in Shanghai, China with data on manufacturing firms. Using patents applications and research and development as innovation measures and geographical range from Shanghai development zones as measure for development zone treatment, a battery of regression models is employed to estimate the contribution of development zones to innovations. The results suggest that development zones have significant positive effect on innovation implying that development zones offer substantial enhancing effects on innovation. Again, Wu, Liu, et al. examined the effect of SEZs on innovations in China using county level data involving 1,851 counties between 1985 and 2011. The study employs difference in difference methodology and reported that the establishment of SEZs have led to increased innovations represented as patent applications, grants and citations. However, the increase in the SEZs induced innovativeness is higher and significant for high-tech orientated SEZs than economic and technological SEZs. It is reported that the increase in innovation is induced through tax credits, technological subsidies and attraction of more enterprise.

Likewise, Kong et al. (2021) studied the effect of high technology development zones (SEZ) affect innovation of A-rate listed firms between 2005 and 2019 in China. The study employs difference in difference methods to reveal that high technological development zones have significant positive effect on innovation capacity of knowledge-intensive service firms especially for science and technology service firms. Additionally, the results show that when a service firm is located in national high-technological development zones have higher innovation ability. More so, Li and Wang (2019) examined the effect of the contribution of SEZs in the form of development zones on innovations using a case study comprising of 733 state-certified high-tech firms in Nanjing between 2008 and 2012. The results report that a strong spatial disparity in the distribution of innovations proxied as number of patents. Evidently, while suburban districts where national and provincial level development zones exist have become major spaces for innovation, planning of industrial cluster in development zones, place-based investment, suburbanization of universities are three main factors that are behind the learning and innovation processes in Nanjing development zones.

Methodology, data and estimation strategy

In this study, data employed is a panel data collected between 2018 and 2021 by Institute of Statistical, Social and Economic Research (ISSER), University of Ghana. The data collection is financed by Kiel Institute and was conducted yearly for four different rounds (2018, 2019, 2020 and 2021). The data collection was targeted at over 500 SEZ and non-SEZ firms covered four years (2018–2021) although 328 firms (both SEZ and non-SEZ firms) completed and submitted questionnaires. The sampled firms were located in Ashanti, Central, Eastern, Greater Accra Volta and Western regions of Ghana.

To investigate the contribution of female CEOs to product and process innovations while focusing on female CEOs in firms with SEZ features, a number of estimation techniques including treatment estimation techniques and regression models are employed. Specifically, treatment estimations employed include propensity score matching (PSM) and inverse propensity weighted score matching while the regression model employed is the logit regression. Thus, unlike other test of difference estimators (like t-test, anova, manova) which lacks the ability to identify and select observations with similar features from the treated and controlled groups to estimate the difference, the PSM treatment estimation model has the ability to identify observations with similar features from the treated and controlled groups to make comparisons in order to estimate difference. For instance, the PSM has the ability to classify firms into those having female CEOs (treated group) and those not having female CEOs (controlled group) and identify and select from both groups (treated and controlled groups) firms that have similar traits/features. The identification of these two groups with similar traits provides a better comparison and robust estimation of difference. Given the confounding and balancing bias/weakness of the PSM, prior studies recommend the inverse propensity weighted score matching technique (IPW) to adjust for confounding and balancing problems that emerge from the PSM process of selecting comparable observations from the treated and controlled groups (Anstrom & Tsiatis, 2001; King & Nielsen, 2019; Rosenbaum, 1987; Tan, 2006). Additionally, both the PSM and IPW provides the opportunity for add control variable in order to improve the quality and accuracy of the estimated difference when using the PSM and IPW difference estimator strategies. As such following the study of Wu, Liu, et al., control variable including production output, revenue, industry, nationality, region and technological effects are controlled for.

Beyond the estimation of difference using PSM and IPW, the study employs logit regressions to provide further understanding on how female CEOs affect product and process innovations in SEZ firms operating outside SEZ enclaves, SEZ firms operating inside SEZ enclaves and non-SEZ firms operating outside SEZ enclaves. Also, the logit regression technique is to provide an alternative for overcoming the misbalancing effect of the propensity score matching process and procedure (see King & Nielsen, 2019). The logit model is represented as: (1) $$\text{\hspace{0.17em}Log\hspace{0.17em}}\left(P/1-P\right)$$(1) where P{Y = 1} is the probability of success or presence of an outcome (female CEO). Specifically, while the study follows the study of Wu et al. (2021) which focused on studying the effect of SEZs on innovations in China, this present study models innovation as follows: (2) $$\mathit{\text{INNO}}{V}_{i,t}={\varphi }_0+{\varphi }_1\hspace{0.17em}\mathit{\text{CEOFE}}{M}_i+{\displaystyle \sum }_{j=2}{\varphi }_j{X}_{\mathit{\text{it}}}+{\epsilon }_{i,t}$$(2) (3) $$\mathit{\text{INNO}}{V}_{i,t}={\varphi }_0+{\varphi }_1\mathit{\text{CEOFEMSEZRi}}{n}_i\hspace{0.17em}+{\varphi }_2\mathit{\text{CEOFEMSEZRou}}{t}_i+{\varphi }_3\mathit{\text{CEOnonFEMSEZRou}}{t}_i{}_i+{\displaystyle \sum }_{j=4}{\varphi }_j{X}_{\mathit{\text{it}}}+{\epsilon }_{i,t}$$(3)

From the above equations INNOV represents product and process innovations, CEOFEM represents female CEOs and X is a vector of control variables including productivity, revenue, industry, nationality, region and technological effects. Specifically, in Equation (3)(3) $$\mathit{\text{INNO}}{V}_{i,t}={\varphi }_0+{\varphi }_1\mathit{\text{CEOFEMSEZRi}}{n}_i\hspace{0.17em}+{\varphi }_2\mathit{\text{CEOFEMSEZRou}}{t}_i+{\varphi }_3\mathit{\text{CEOnonFEMSEZRou}}{t}_i{}_i+{\displaystyle \sum }_{j=4}{\varphi }_j{X}_{\mathit{\text{it}}}+{\epsilon }_{i,t}$$(3) where the study attempts to estimate the effect of female CEOs in firms with different SEZ dynamics, CEOFEMin represents female CEOs working registered SEZs operating inside SEZ enclaves, CEOFEMout represents female CEOs working registered SEZs operating outside SEZ enclaves and CEOnonFEMout represent female CEOs operating in non-SEZ registered firms outside SEZ enclaves. For Equation (1)(1) $$\text{\hspace{0.17em}Log\hspace{0.17em}}\left(P/1-P\right)$$(1) , we attempt to estimate the effect of female CEOs in the full sample, firms registered as SEZ firms, firms registered as SEZ firms and operating from inside SEZ enclaves and firms registered as SEZ firms and operating from outside SEZ enclaves. These estimations are done to enhance our understanding on how female CEOs ­contribution to innovations under different SEZ setups. In Equation (3)(3) $$\mathit{\text{INNO}}{V}_{i,t}={\varphi }_0+{\varphi }_1\mathit{\text{CEOFEMSEZRi}}{n}_i\hspace{0.17em}+{\varphi }_2\mathit{\text{CEOFEMSEZRou}}{t}_i+{\varphi }_3\mathit{\text{CEOnonFEMSEZRou}}{t}_i{}_i+{\displaystyle \sum }_{j=4}{\varphi }_j{X}_{\mathit{\text{it}}}+{\epsilon }_{i,t}$$(3) , however, we attempt to estimate the effect of female CEOs in different SEZ setups in the same regression to provide more and detailed information on the link between female CEOs and SEZs on innovations in one regression model.

The dependent variable for this study is firm level product and process innovations. It is measured as a dummy variable which assumes a value of 1 if a firm introduced new and or upgraded its product and or process and 0 otherwise for the years under review (Capitanio et al., 2010; Diamond & Vangen, 2017; Kahn, 2018). The variable of interest, female CEOs, is also measured using a dummy variable that assumes a value of 1 if a firm is led by a female CEO and 0 otherwise. The expectation is that female CEOs working under SEZ setups would have higher and significant effect on innovations compared to their female CEO counterparts who work in non-SEZ firms. The argument is that SEZ firms adhere to international/global standards, have better managerial, financial, technological and corporate environment which can promote female CEOs to positively contribute to innovations compared to the counterpart female CEOs operating in non-SEZ firms. In terms of the control variables, log of real production output is employed and is expected to contribute positively to innovations because production output may induce innovative ways of improving the production process or improving the product produced (Wu et al., 2021). Similarly, log of real revenue is expected to lower innovativeness because firms the generate more revenue have lower incentive to innovate and a negative effect of revenue on innovations. Additionally, industry, nationality, region and technological effects are controlled for in the models to enhance the quality and accuracy of the results.

Table 2 presents a summary of the all the variables employed in this study.

Table 2. Summary of variable employed.

Variable	Measurement	Symbol	Expected Sign
Product Innovation	Dummy variable which assumes a value 1 if a firm introduced new or upgraded its product	Product_Innov
Process Innovation	Dummy variable which assumes a value 1 if a firm introduced new or upgraded its process	Process_Innov
Female CEO	Dummy variable which assumes a value 1 if the CEO of a firm is a female	CEOFEM	+
Female CEO in Registered SEZ firms	Dummy variable which assumes a value 1 if the CEO of a SEZ registered firm is a female	CEOFEMSEZin	+
Female CEO in Registered SEZ firms and operating from SEZ enclaves	Dummy variable which assumes a value 1 if the CEO of a SEZ registered firm is a female and operates from an SEZ enclave	CEOFEMSEZout	+
Female CEO in Registered SEZ firms and operating from nonSEZ enclaves	Dummy variable which assumes a value 1 if the CEO of a SEZ registered firm is a female and operates outside an SEZ enclave	CEOFEMSEZin	+
Production Output	Natural log of production output		+
Revenue	Natural log of revenue	LnRev	–

Empirical results and discussions

Table 3 reports the summary statistics and is basically used to screen for outliers. Using the mean, minimum, maximum and standard deviation values, there is no evidence of outliers (see Lee et al., 2008; Rousseeuw & Hubert, 2011). In Table 3, it is observed that during 56.4% and 45.4% of the firms in the study were engaged in product and process innovations, respectively. This implies that on the average 56.4% and 45.4% of firms either introduced new or upgraded products and or introduced new or upgraded their processes. In the context of female CEOs, we observe that 12.8% of the firms during the four-year period had female CEOs. Specifically, majority constituting 8.2% of these female CEOs led non-SEZ firms operating outside SEZ enclaves while 4% and 0.6% of these female CEOs led SEZ registered firms operating outside SEZ enclaves and SEZ firm registered firms operating inside SEZ enclaves. These provide an indication that SEZ firms have less females as CEOs compared to the non-SEZ firm counterparts. In line with multicollinearity, the pairwise correlation (see Table 4) and variance inflation factor (VIF) (see Table 3) are employed to detect problem of multicollinearity. Following prior literature (Kim, 2019; Shrestha, 2020) rely on VIF value of 10 to detect multicollinearity threshold, there is no evidence of multicollinearity.

Table 3. Summary statistics.

Variable	Obs	Mean	Std. Dev	Min	Max	VIF
product innov	1312	.564	.496	0	1	–
process innov	1312	.454	.498	0	1	–
ceofem	1312	.128	.334	0	1	1.05
fceoSEZin	1312	.006	.078	0	1	1.00
fceoNonSEZout	1312	.082	.275	0	1	1.03
fceoSEZout	1312	.04	.195	0	1	1.01
lnY	970	12.508	2.944	.463	19.165	5.09
lnrrevenue	1064	12.624	2.871	−.967	20.465	5.05

Table 4. Pairwise correlations.

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
(1) product_innov	1.000
(2) process_innov	0.802*	1.000
(3) ceofem	0.024	0.017	1.000
(4) fceoSEZin	−0.010	0.007	0.204*	1.000
(5) fceoNonSEZout	0.017	0.016	0.782*	−0.023	1.000
(6) fceoSEZout	0.021	0.003	0.530*	−0.016	−0.061*	1.000
(7) lnY	−0.022	−0.012	−0.084*	−0.013	−0.121*	0.043	1.000
(8) lnrrevenue	−0.009	−0.024	0.010	−0.015	−0.043	0.083*	0.893*	1.000

*** p < 0.01, **p < 0.05, *p < 0.1.

The main results for this study are reported in Tables 5–7. Tables 5 and 6 present the treatment estimation model results (inverse propensity weighted score matching and propensity score matching) while Table 7 reports on the results from the logit regression model. In Tables 5 and 6, 8 models are reported and the first 4 set models in each table reports on the effects of having female CEOs under different SEZ setups on product innovation while the second 4 set models in each table reports on the effects of having female CEOs under different SEZ setups on process innovation. Similarly, Table 7 reports 10 logit models and the first 5 set logit models focuses on how female CEOs in different types of SEZ firms on product innovation and the second 5 set of logit models reports on how of female CEOs in different SEZ firm setups influence process innovation.

Table 5. Treatment estimation of effect of female CEOs on innovations - IPW.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	Product Innovation				Process Innovation
VARIABLES	Full Sample	Registered SEZ Firms	Registered SZEs Outside Enclave	Registered SEZ firms Inside Enclave	Full Sample	Registered SEZ Firms	Registered SZEs Outside Enclave	Registered SEZ firms Inside Enclave
r1vs0.ceofem	0.097*	0.223***	0.256***	0.560***	0.115**	0.316***	0.326***	0.674***
	(0.050)	(0.072)	(0.065)	(0.073)	(0.053)	(0.074)	(0.076)	(0.062)
Observations	952	326	260	66	952	326	260	66

Note: Log of production output, log of revenue, nationality, industry, region and technological effects are employed as control variables in all the models.

Standard errors in parentheses – ***p < 0.01, **p < 0.05, *p < 0.1.

Table 6. Treatment estimation of effect of female CEOs on innovations - PSM.

	(9)	(10)	(11)	(12)	(13)	(14)	(15)	(16)
	Product Innovation				Process Innovation
VARIABLES	Full Sample	Registered SEZ Firms	Registered SZEs Outside Enclave	Registered SEZ firms Inside Enclave	Full Sample	Registered SEZ Firms	Registered SZEs Outside Enclave	Registered SEZ firms Inside Enclave
r1vs0.ceofem	−0.001	0.328***	0.242***	0.561***	0.077	0.448***	0.319***	0.727***
	(0.064)	(0.050)	(0.060)	(0.062)	(0.065)	(0.050)	(0.066)	(0.054)
Observations	952	326	260	66	952	326	260	66

Note: Log of production output, log of revenue, nationality, industry, region and technological effects are employed as control variables in all the models.

Standard errors in parentheses – ***p < 0.01, **p < 0.05, *p < 0.1.

Table 7. Effect of female CEOs on innovations - logit.

	(17)	(18)	(19)	(20)	(21)	(22)	(23)	(24)	(25)	(26)
	Product Innovation					Process Innovation
VARIABLES	Full Sample	Registered SEZ Firms	Registered SZEs Outside Enclave	Registered SEZ firms Inside Enclave	All	Full Sample	Registered SEZ Firms	Registered SZEs Outside Enclave	Registered SEZ firms Inside Enclave	All
ceofem	0.660***	1.093***	1.244***	1.943**		0.660***	1.093***	1.244***	1.943**
	(0.201)	(0.371)	(0.414)	(0.820)		(0.201)	(0.371)	(0.414)	(0.820)
fceoSEZin					−0.558					0.072
					(0.625)					(0.684)
fceoSEZout					0.827**					1.128***
					(0.412)					(0.386)
fceoNonSEZout					0.406*					0.561**
					(0.240)					(0.237)
lnY	0.099*	0.204*	0.221*	−0.126	0.029	0.099*	0.204*	0.221*	−0.126	0.099*
	(0.057)	(0.120)	(0.122)	(1.361)	(0.053)	(0.057)	(0.120)	(0.122)	(1.361)	(0.058)
lnrrevenue	−0.116**	−0.255**	−0.231*	0.057	−0.049	−0.116**	−0.255**	−0.231*	0.057	−0.120**
	(0.059)	(0.126)	(0.129)	(1.259)	(0.054)	(0.059)	(0.126)	(0.129)	(1.259)	(0.059)
Constant	16.894	−12.225	−38.254	−36.542	20.534	16.894	−12.225	−38.254	−36.542	15.462
	(119.911)	(211.287)	(234.411)	(954.729)	(119.117)	(119.911)	(211.287)	(234.411)	(954.729)	(120.077)
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Sector Effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Nationality Effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Regional Effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	952	326	260	66	952	952	326	260	66	952

Note: Nationality, industry, region and technological effects are employed as control variables in all the models.

Standard errors in parentheses – ***p < 0.01, **p < 0.05, *p < 0.1.

Contribution of female CEOs to product and process innovation

From the treatment estimation model results (IPW and PSM) particularly the IPW results, we observe that female CEOs contribution positively and significantly towards both product and process innovations. Specifically, in Tables 5 (see models 1 & 5) and 7(models 9 and 14), female CEOs are reported to be more likely to have higher contribution to product and process innovations by 0.097 and 0.115 units more than their counterparts implying that female CEOs in general contribute positively to product and process innovations. Thus, we argue that female leaders bring new/fresh and well-managed expertise ideas to the male dominated corporate leadership cohort (Chen et al., 2005, 2018; Hillman et al., 2002; Huse & Grethe Solberg, 2006; Miller & del Carmen Triana, 2009), are more inclined towards activities such as R&D, patent application and citation (Javaid et al., 2023; Saggese et al., 2021; Wu et al., 2021) and increase efficiency in decision making by offering more innovative and effective choices (Galia et al., 2015; Na & Shin, 2019). These established attributes of female leaders provide justification for the finding of this study where female CEOs contribute more significantly to product and process innovations.

Contribution of female CEOs to product and process innovations under different SEZ setups

Following the results from in Tables 5–7, we observe that female CEOs operating/working under different SEZ setups/types have varying significant effects on innovations. We find that, more intense the SEZ setups, the higher the contribution of female CEOs to product and process innovations. Specifically, from models 2 and 6 (Table 5) models 10 and 14 (Table 6) and models 18 and 23 (Table 7), it is evident that female CEOs led firm that are registered as SEZ firms reported higher and significant contributions to both product and process innovations compared to the full sample where no SEZ firm effects are considered. Similarly, in models 3 and 7 (Table 5) models 11 and 15 (Table 6) and models 19 and 24 (Table 7), it is observed that female CEOs in registered SEZ firms but operating from outside SEZ enclaves tend to contribute significantly higher than those female CEOs who led firms that have only registered as SEZ firms. Furthermore, we observe similar trends in models 4 and 8 (Table 5) models 12 and 16 (Table 6) and models 20 and 25 (Table 7), females CEOs in registered SEZ firms and operating from SEZ enclaves tend to contribute significantly higher than those female CEOs who led firms that have only registered as SEZ firms and those who solely work from SEZ enclaves.

These results magnify and emphasize the unique importance of SEZ firms in empowering female leaders to overcome the unseen barriers that impede the smooth and successful contribution of female CEOs to innovations. Arguably, prior literature has shown that SEZ firms are well-resourced (financially and technically), following international standards and provide gender-oriented policies that support, protect and enhance the females in the quest to contribute product and process innovations. Clearly, economic policy induced setups such as SEZ firms are critical and important in providing the requisite corporate environment, conditions and capacities that propels/enhances the meaningful contribution of females to firm innovativeness. In terms of the control variables, while production output and revenues positively and negatively influence both firm product and process innovativeness and are largely consistent with the expectation, nationality, industry, regional and technological/year effects are controlled for across all the models. Hence, the results are robust and resilient to nationality, industry regional and technological effects.

Robustness checks

In this paper, we check for robustness by (i) employing two different estimation models, (ii) using two different innovation variables as dependent variables, (iii) controlling for year, sectoral, nationality and regional effects and (iv) using logged variables to control for outliers. Thus, we first employ treatment effect models and logit regression models to check for accuracy, reliability and consistency in the results.

In terms of the treatment effects, both PSM and IPW were employed to enable the study to estimate the accurate differences between female-led SEZ firms and female-led non-SEZ firms by observing observation from treated and controlled groups to estimate the difference. This ensures that the estimated difference is attributable to the intervention (Female CEO). Although the PSM and IPW estimates the difference in the treated and controlled groups, the IPW has an extra advantage of overcoming confounding and balancing problems that emerge from the PSM process of selecting comparable observations from the treated and controlled groups.

Hence, the estimated differences between female-led SEZ and female-led nonSEZ firms are accurate, reliable and consistent. In terms of the logit regressions, while it offers an alternative means of estimating the effect of the female-led SEZ firms on innovations, it enables us to estimate this effect more accurately given that the dependent variable (product/process innovation) is measured using dummy variables. Beyond models, we measure innovations using two different proxies including product and process innovations. The use of these two variables improves understanding and further serves as alternative measure to check for consistency and reliability in the results and findings.

Furthermore, to ensure the results and findings are not driven by year, sectoral, nationality and regional locations of the firms, year, sectoral, nationality and regional effects are controlled for. This ensures that the results and finding are not biased by the year, sectoral, nationality and regional effects of the firms in the study. More so, the study relies on prior studies in modelling and selecting the variables used. Additionally, the control variables are logged to control for the potential adverse effects of outliers while the pairwise correlation is used to screen and detect multicollinearity. Given the above controls and robustness measures employed, the results and findings are accurate, reliable and fit for generalization.

Conclusions, policy implications and recommendations

Economic analysts and policymakers are interested in enhancing the inclusive growth, development and welfare of their citizenry. As such, the attainment of gender equality and empowering all women and girls which is sustainable development goal 5 has emerged as a strong pillar for achieving gender inclusive growth and development. Interestingly, lack of productive resources coupled with limited government resources impede the attainment of inclusive growth and development and economic policies including special economic zones (SEZs) have emerged as critical tools for attracting productive resources to enhance the attainment of inclusive growth and development. It is argued that SEZ firms are well-resources, adhere to international gender affirmative policies, have better mentoring and training practices and working conditions which can provide the needed and necessary working environment that can influence the contribution of female CEOs to innovations for workplace inclusivity. Hence, this study attempts to investigate how female leadership, specifically female CEOs, impact firm level innovations under different operationalization structures/features of special economic zones as economic policy in Ghana.

Obtaining primary data on 328 firms between 2018 and 2021, the study employed treatment estimations and regression models to ascertain how female CEOs influence product and process innovations under different SEZ setups. The results reveal that on the average, registered SEZ firms tend to have more female CEOs and have their female CEOs engaging more in product and process innovations. Similarly, while female CEOs generally contribute positively and significantly to both product and process innovations, it is evident that the contribution of female CEOs in SEZ registered firms and or operating from SEZ enclaves/zones is higher and more profound on both product and process innovations compared to their counterparts leading non-SEZ firms. These results are consistent across the different estimations techniques and robust to difference across sectors, nationalities, regions and years.

These results and findings have important implications and recommendations for gender and SEZ policies particularly in emerging economies. First, corporate managers and policymakers are encouraged to implement corporate and national policies that lower/eliminate the glass ceiling barriers that impede the progression of females to leadership positions given that female leaders can meaningfully and significantly contribute to innovations. Similarly, economic policymakers must endeavor work on reducing the bureaucratic and complex administrative procedures and processes for registering as SEZ firm and operating in SEZ enclaves to attract more SEZ firms since such SEZ firms and enclaves are good channels through which gender inclusiveness and innovativeness can be encouraged for inclusive growth and development. In terms of future research direction, policy and academic researchers are encouraged replicate this study to confirm and explore additional channels through which female leaderships can improve firm level performance.

Author contributions statement

Baah Aye Kusi was involved in the drafting of the paper, conceptionalization, designing, analysis and interpretation of the data. Charles Ackah and Robert Osei were involved in the conceptionalization and designing, revising of the manuscript critically for intellectual content and the final approval of the version to be published.

Availability of data and materials

Data will be made available on request

Acknowledgement

We are grateful to The Kiel Institute for the World Economy for financing the data collection for the paper and also their invaluable suggestions and comments which have greatly improved the quality of the paper.

Disclosure statement

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

Additional information

Funding

Funding has been received for the data collection process from The Kiel Institute for the World Economy.

Notes on contributors

Charles Godfred Ackah

Charles Godfred Ackah is an Associate Professor and research fellow at the Institute of Statistical, Social and Economic Research at the University of Ghana and is interested in applied trade and trade policy and analysis, poverty and distribution impacts of economic policies, labour market adjustment, and consumption and demand behaviour.

Robert Darko Osei

Robert Darko Osei is an Associate Professor in the Institute of Statistical, Social and Economic Research (ISSER), University of Ghana, Legon, and also the Dean for the School of Graduate Studies at the University of Ghana. Robert has published widely in edited volumes and top international journals.

Baah Aye Kusi

Baah Aye Kusi holds Doctor and Master of Philosophy degrees in Finance and obtained from University of Ghana. He is also a Chartered Financial Economist of the Association of Certified Chartered Economists (ACCE) and a senior lecturer at University of Ghana Business School, Ghana. Baah has over eleven years of experience in teaching and researching in finance, banking, insurance and economics at University of Education, Central University, Ghana Institute of Journalism, University of Ghana, Valley View University, Almond Institute (now Accra Business School), and Blue Crest University College.

Notes

1 GHA103987.pdf (ilo.org).