Green culture, environmental innovativeness, green intellectual capital and circular economy implementation behaviour: A sequential mediation model

Abstract

Although corporations have been propelled to adapt eco-innovation strategies including decent work practices, responsible consumption and production, and transition from linear economy to circular economy model to integrate natural ecosystems, businesses, and waste management, inversely, the actual impact of the circular economy implementation (CEI) in Ghana is yet to be felt. This paper aims to analyse the implications of corporate green culture (CGC) on CEI behaviour, and to design a model that explains the mediating roles of environmental innovativeness (EI) and green intellectual capital (GIC) on the relation between CGC and CEI behaviour from the perspective of a lower resource economy context. Our paper is positioned on positivists’ reasoning and quantitative research approach. A survey-based structured questionnaire has been used to collect data from 300 manufacturing firms in Ghana. Our hypotheses have been tested with the aid of SMART-PLS and structural equation modelling. The indicative results have shown that CGC positively affect CEI behaviour. Moreover, EI and GIC have significant and positive effects on CEI behaviour, Besides, GIC serves as a medium through which CGC indirectly impacts on CEI behaviour. The implications of the study include the emergence of baseline model to encourage CEI behaviour in Ghana and other Sub-Saharan African countries. The paper recommends that the newly developed model be used by practitioners to explain coordinated economic, ecological and social development, and restore ecological environment. Policymakers are encouraged to facilitate the reinforcement of environmental policies and subsequent realization of Sustainable Development Goals.

Keywords:

• Circular economy implementation; corporate green culture
• Environmental innovativeness
• Green intellectual capital

1. Introduction

The United Nations (UN) Agenda 2030 for Sustainable Development have propelled world leaders, policymakers, investors, and academics to upscale commitments to ensure a balance between value for money (economics dimension of development), social justice and respect for human right (social dimension of development), and protection of biodiversity and reduction of pollution (environmental dimension of development) (Alkaraan, 2022; Appiah, 2022; Diéguez-Santana et al., 2021). To this end, there is shift from the generic development to a sustainable development since the latter has the propensity to meet the needs of the present generation without compromising future generations to meet their own needs. Sustainability is becoming increasingly important for companies as we become aware of the environmental and social challenges facing the world. This has led to changes in consumer and business behaviour (Kalemkerian et al., 2022; Kwarteng, Agyenim-Boateng, et al., 2022; Sahu et al., 2022). For organizations to become sustainable, the current economic model must shift from economic priorities to social and environmental priorities. This means that companies and entrepreneurs need to integrate the concept of sustainability into their business models. Many companies have integrated sustainability into their business operations by addressing environmental, social, and economic issues. Companies seeking alternatives are addressing issues such as over-consumption, alternative food and beverage choices, and over-use of resources and sustainable transport (Appiah, Tettevi, et al., 2022; Appiah et al., 2022Davila Delgado & Oyedele, 2020).

Sequel to the above, as the global leaders strive to inculcate responsible consumptions and production in the transition from linear economy due consideration should be given to circular economy. In a linear economy, natural resources are taken, turned into products, and thrown away (Agyemang et al., 2019; Habagil et al., 2020) In contrast, the circular economy model seeks to bridge the gap between production cycles and the natural ecosystems on which people ultimately depend. This means, on the one hand, recycling waste, composting biodegradable waste, or, in the case of treated and non-biodegradable waste, recovering, recycling, and ultimately reusing it. On the other hand, it also means phasing out chemicals (a way to help restore natural systems) and investing in renewable energy. A circular economy is an industrial system whose purpose and design are regenerative or restorative (Bressanelli et al., 2018; Khan et al., 2022; Singh et al., 2022). It replaces the concept of end-of-life with regeneration, switches to renewable energy sources, removes toxic chemicals that inhibit reuse and bioremediation, and seeks to eliminate waste by improving the design of materials, products, systems, and business models. As argued in extant literature (Bocken et al., 2016; Ghisellini et al., 2016; Marrucci et al., 2022) despite the benefits of circular economy in the global shift towards responsible production and consumption particular renewable energy and climate mitigations its implementation has rather been very slow. In the context of lower-middle-income countries such as Ghana little evidence is seen in circular economy initiations, planning, and implementations. Very scanty evidence is available in the area of waste management where the majority of the projects exist on pilot bases. It should be noted that the implementation of circular economy could be macro (national), messo (industry), and micro (firm level). Although corporations have been propelled to adapt eco-innovation strategies, including decent work practices, responsible consumption and production, and transition from linear economy to circular economy model to integrate natural ecosystems, businesses, and waste management; inversely, the actual impact of the circular economy implementation (CEI) in Ghana is yet to be felt (Ada et al., 2022; Kusi-Sarpong et al., 2021; Sarfraz et al., 2022). Pan et al. (2015) found that the scale of incineration, gasification, and anaerobic digestion of waste is considerable. Recently, waste-to-energy (WTE) supply chains have been increasingly established in various industrial parks around the world. The principles for designing WTE supply chains are based on the “5Rs” practices, i.e., reduce, reuse, recycle, recover (energy), and reuse (land). In order to effectively implement a WTE supply chain, policy mechanisms must be able to simultaneously address a range of regulatory, institutional, economic, and technological barriers. For example, strategies for implementing a WTE supply chain should (1) establish political and government accountability, (2) address externalities, social acceptability, and investor mobilization, (3) provide financial incentives and price subsidies, and (4) implement a comprehensive performance evaluation programme. This paper is focused on micro- or firm-level implementation of circular economy models to enhance sustainable development.

This paper aims to analyse the implications of corporate green culture (CGC) on CEI behaviour, and to design a model that explains the mediating roles of environmental innovativeness and green intellectual capital (GIC) on the relation between CGC and CEI behaviour from the perspective of a lower resource economy context. The novelty of this paper is centred on the fact in the context of lower middle-income countries like Ghana, and there is no study which has tested the mediation roles of environmental innovativeness and GIC on CEI using structural equation modelling (SEM) technique. Moreover, the emergence of the baseline model could be used to facilitate CEI behaviour in Ghana and other Sub-Saharan African countries. Policymakers, practitioners, and media are encouraged to facilitate the reinforcement of environmental policies and subsequent realization of Sustainable Development Goals. Based on the outlined novelty, the paper contributes to the development of circular economy and its implementations in the context of lower-middle country to address improve well-being (SDG3), decent work (SDG8), responsible consumption and production (SDG13), and climate action (SDG13). Moreover, this paper has discovered contextual factors that facilitate the CEI in Ghana. Besides, the mediating variables could be used to identify the indirect but significant contributors to the implementations of circular economy. Theoretically, the theory of natural resources-based view (NRBV) with slight modification has been used to develop the new baseline model. Furthermore, the newly developed baseline model is context specific and could be used for replication studies within the larger lower-middle-income countries with precisions to facilitate the adoption and implementation of circular economy models to save the environment and the biodiversity (Anaruma et al., 2022; D. Qu et al., 2022). The rest of the paper is prepared in five sections as follows: theory and hypotheses development are presented in Section 2, methodology and industry profile in Section 3, results and discussion in Section 4, and conclusions and implications in Section 5.

2. Literature review

The main assumption of paper is that circular economy models are the main driver of economic prospect, environmental conservation, and social justice and inclusion (Geissdoerfer et al., 2017; Li et al., 2022; Liu & Lin, 2020). Thus, the proponents of circular economy aim to rethink economic growth and focus on positive social benefits, overcoming the current industrial model of take it or leave it. This means gradually decoupling economic activity from the final consumption of resources and removing waste from the system. A circular economy model supported by a shift to renewable energy will increase economic, physical, and social capital (Diéguez-Santana et al., 2021). It is based on three principles: elimination of waste and pollution, conservation of products and materials, and restoration of natural systems consistent with the NRBV theory. Hart (1995) asserted that based on the propositions of NRBV, organizational resources can play an important role in creating successful corporate environmental strategies. As shown in Figure 1, this paper examines the mediating roles of environmental innovativeness and GIC in between CGC and CEIs. Thus, that paper further contributes to knowledge stock by investigating the effect of two mediators on the relationship between green culture and CEIs.

Figure 1. A sequential research framework.

2.1. Corporate green culture (CGC) and circular economy implementation (CEI)

The main proposition of the paper based on the NRBV is that CGC drives circular economy implementation is empirically supported (Fok et al., 2022; Hooi et al., 2022; Luu, 2022; Shahriari et al., 2022). CGC is a modern environmental ideology based on science, policy, and aesthetics, aimed at achieving economic and environmental sustainability. For years now, the company has been incorporating this concept into their corporate social responsibility practices. Companies have realized that this paradigm shift affects market behaviour and thus sales, leading to higher profits (Gurlek & Tuna, 2018; Martelo Landroguez et al., 2018). Green culture is used in this study as a variable that influences CEIs intention. According studies, green culture promotes environmental educational experience through cultural and arts (Liu & Lin, 2020; Pham et al., 2020). Companies that adopt green culture encourage the use of reusable, recyclable, and biodegradable materials. Thus, to be conscious of climate change, green culture is among the factors that companies embrace to improve their capabilities. As such the intent to implement circular economy is to reduce waste and promote the use of renewable energy. Fortunately, green culture is a key driver for such ideology. Therefore, as a way of promoting CEI intentions, green culture is a key factor (Ahmed et al., 2021; Rizvi & Garg, 2021; X. Qu et al., 2021). Accordingly, green culture is seen as cultural shift towards social and environmental activities that addresses social injustice and environmental degradation (Buettner, 2013; Wang & Juo, 2021). Drawing on the arguments above, the paper proposes as follows:

H1:

CGC exerts a significant and positive influence on circular economy implementation.

H2:

CGC exerts a significant and positive influence on environmental innovativeness.

H3:

CGC exerts a significant and positive influence on green intellectual capital.

2.2. Environmental innovativeness (EI) and circular economy implementation

Inferring from prior studies (Golgeci et al., 2022; N. Sharma et al., 2022; Yin & Wang, 2018), this paper has argued that environmental innovation is one of the drivers of CEI behaviour and serves as mediator in between CGC and CEI. Environmental innovation is “the production, introduction, or use (in its development or introduction) of a new product, production process, service, or management or business practice that is new to an organization and that leads to a reduction in environmental risk, pollution, or other negative impacts over the life cycle of resource use (including energy use) compared to suitable alternatives” (Syed et al., 2022; Biscione et al., 2021; Karakaya et al., 2014; Kemp & Pearson, 2007). Environmental innovation or eco-innovation, a particular form of innovation that aims to reduce the impact of products and production processes on the natural environment, has only recently appeared in the innovation literature (Khalil et al., 2022; Ozusaglam, 2012; Russell et al., 2019). Eco-innovativeness boosts the ability to implement eco-friendly practices in the process of production and new product development. Innovative companies make decisions that improve their capacity to take advantage economically. Thus, innovativeness is a strategic means the secure companies with competitive position in all aspects of their practices. In this study environmental innovativeness is consider as variable influencing the implementation intention of circular economy. Environmental innovativeness is companies’ ability to create inventions that are eco-friendly throughout its life span. Hence, eco-innovativeness has a significant influence of circular economy (Bocken et al., 2016; Ly & Tan, 2021; Syed et al., 2022). Thus, inventing new products that take into account waste reduction and promote renewable energy usage is the pioneer of eco-innovativeness and hence promote circular economy intention (et al., 2018; Ünal et al., 2019; Syed et al., 2022). Drawing on the arguments above, the paper proposes as follows:

H4:

EI exerts a significant and positive influence on circular economy implementation.

H5:

EI exerts a significant meditating role in between CGC and circular economy implementation.

2.3. Green intellectual capital (GIC) and circular economy implementation

As showed by the available empirical evidence (Jirakraisiri et al., 2021; Shoaib et al., 2021; Yong et al., 2020), this paper has argued that GIC is one of the drivers of CEI behaviour and serves as a mediator in between CGC and CEI. Abd et al. (2022) define GIC as a set of intangible knowledge assets (skills, competencies, and capabilities) that a company manages in order to promote environmental performance. López-Gamero et al. (2011) define GIC as a set of knowledge that an organization holds to promote environmental management that would enable it gain competitive advantage over competitors. Under GIC intangible assets, including knowledge, skill, and relationships of the company, are considered. These assets are difficult to imitate and hence facilitate companies to improve their environmental management (Chen, 2008; Sheikh, 2022; Xi et al., 2022). These intangible assets make companies innovative and improve their practices to be eco-friendly. Thus, companies with higher GIC motivate their employees to utilize reusable materials and encourage the use of renewable energy that will have a minimal negative influence on the environment (Giudice et al., 2021; Jinru et al., 2021; Wang & Juo, 2021). Under this study GIC is considered as a variable that has a influence on CEI intention. Huang and Kung (2011) argue that “green intellectual capital represents a company’s intangible assets, including knowledge, skills, experience, and innovation in environmental protection.” Also, GIC refers to the use of an organization’s knowledge, skills, capabilities, expertise, and relationships for environmental protection (Amores-Salvadó et al., 2021; W. Ali et al., 2021; Wang & Juo, 2021). Drawing on the arguments above, the paper proposes as follows:

H5:

GIC exerts a significant and positive influence on circular economy implementation.

H6:

GIC exerts a significant meditating role in between CGC and circular economy implementation.

3. Research methodology

3.1. Research design and approach

We anchored our research on positivists’ reasoning and quantitative research approach. We argued that positivists believe that modern research should be objective based and should be guided by formulation of hypotheses and empirical testing of such hypotheses. Shukla (2022) argued that the entire research process should be kept simple, clean, and precise, including the core assumptions of the research process. The positivists’ assumption is consistent with quantitative research approach as used in this paper because both require the using mathematical and statistical assumptions in order to attain valid and reliable outcome. Moreover, surveys and quantitative studies are inexplicable connected. As averred by Zikmund et al. (2012), survey research involves the process of posing a social question and translating its form into questions that can be used to obtain the opinion of a sample from which conclusions can be drawn about a larger sample. The survey strategy is used because it usually covers a large number of participants at relatively low cost (Appiah, Possumah, et al., 2021; Appiah, Sedegah, et al., 2021). In terms of time framework, cross-sectional design was used because the data were collected at a single point in time during the survey. It was cost-effective compared to a longitudinal study.

3.2. Population

We collected data manufacturing companies from in Ghana. The population of the paper comprised manufacturing companies in Ghana, including automobile parts manufacturing, food processing factories, wood processing, aluminium smelting, light manufacturing, food processing, cement manufacturing, and small commercial shipbuilding. Appiah (2022) reported that the manufacturing industry in Ghana is evolving at a faster rate and a key contributor of employment through job creation, innovation, human resources development, and Gross Domestic Products (GDP). The food processing, automobile spare parts, and aluminium smelting currently dominate the sector. This industry was chosen for the current study due to its immense contribution to the overall Ghanaian economy, as well as the ease (availability and accessibility) of gathering data from this industry.

3.3. Sampling process

The random sampling method was used. Specifically, stratified sampling technique was used wherein the subsectors were used to form strata. The sample of the study included 245 middle and senior managers of the company. A total of questionnaires was administered through Google form link that received 300 responses. A total of 245 responses were suitable for final analysis, representing 81.7% response rate. Only feedbacks that are valid and complete were retained after the survey data cleaning process that consists of searching for and discarding respondents’ answers that do not meet our targets or do not answer the survey in a meaningful way, such as when respondents answer only part of the questionnaire, when they give ambiguous answers, and/or repeatedly choose the same answer option.

3.4. Constructs measurements and data collection

There are four main constructs which have been used in this paper: CGC, EI, GIC, and CEI intentions. These constructs have been sourced from previous studies with slight modifications. Table 1 presents the names of the constructs, number of items scale of measurement, item sources, and Cronbach alpha scores. Data for paper has been collected using structured questionnaire and five-point Liker’s measurement scale. The highest point on the scale was 5 (strongly agree) and the lowest was 1 (strongly disagree). This paper involved human participation; as a result all appropriate ethical protocols were duly observed. These include informed consent, respect for human rights, protection against risks, and ensuring professional integrity (Appiah, Possumah, et al., 2021; Appiah, 2022c). Before the start of data collection, all participants were duly informed about the purpose of the study. They have been informed that they have the right to withdraw from the study at any time. They were assured that the paper poses no direct threat or psychological harm. Finally, the participants were assured of confidentiality of their responses and anonymity of their identities.

Table 1. Constructs’ Sources, Measurements and Reliability Scores

Constructs name	Items	Type of scale	Sources	Cronbach’s alpha
Corporate green culture	5	5-point Likert	Shahriari et al. (2022)	0.93
Environmental innovativeness	3	5-point Likert	Liao et al. (2022	0.96
Green intellectual capital	4	5-point Likert	Issa et al. (2022)	0.93
Circular economy implementation behaviour	5	5-point Likert	Wamba et al. (2023) Kusi-Sarpong et al. (2021), Li et al. (2022)	0.95

3.4.1. Authors’ compilation

3.4.1.1. Data analysis techniques

The hypotheses of the paper have been tested with the aid of SMART-PLS version 3.9 and structural equation modelling (SEM) techniques. PLS is more suitable for models with multiple structures and models with multiple structure sequences. The advantage of PLS-SEM is the fact that it supports modelling development especially those at the earliest stage of development such as green culture and CEI at the micro level of development (Hair et al., 2017). The approach involves two main steps, namely measurement model and structural model. While the latter is used to validate the instruments of the study, the former is used to test model’s hypotheses. To assess the measurement model, we investigated the construct validity which usually comes in two forms, notably convergent and discriminant. Convergent validity has been measured in this paper through factor loadings, composite reliability, and average extracted variance (AVE). Moreover, the discriminant validity was measured through the square of AVE estimates in comparison with the correction matrix coefficients. Concerning the structural model evaluation, the hypotheses of the model have been tested and assessed using t-values and beta values. Any relationship with t-value below 1.96 was rejected.

4. Results

4.1. Participants’ profile

As indicated in the demographic table (see Appendix A), slightly above half (58.8%) participants were males and the remaining 41.2% were females. Moving on, 32.7% of the participants were aged between 25–29 years, 19.2% were aged between 30–34 years, 15.9% were aged between 35–39 years, 13.1% were aged between 18–24 years, 12.7% were aged 40–44 years, 2.4% were aged between 45–49 years, and 55+ each and the least 1.6% were aged between 50–54 years. Besides, 38.0% of the participants were degree/HND holders, 25.7% were master’s degree holder, 18.0% were secondary school graduates, 13.5% were other degree holders, and 4.9% were primary school leavers. Again, slightly below half (48.6%) of the participants have worked for below 5 years, 24.1% have worked between 5–9 years, 15.1% of the participants have worked between 10–14 years, 9.0% of the participants have worked between 15–19 years, and finally, 3.3% have worked above 20 years. Also, slightly below half (43.7%) of the participating manufacturing firms were located in the urban areas, 31.0% were located in the rural areas, and the least 25.3% were located in the peri-urban areas.

4.2. Descriptive statistics, multicollinearity test, and factor loadings

The results have shown (see Table 2) that the mean value of the model ranged between 4.22 and 4.59 which implied that majority of the participants had rated the items on the high size of the Likert’s scale, e.g., agree and strongly agree. Likewise, the standard deviation values ranged from 0.455 to 0.489 which similarly implied that the were no significant degrees of variations among participants choices on the 5-point scale, e.g., all the standard deviation values were less than 1 (SD < 1). The variance inflation factor (VIF) scores were estimated to test for the presence of multicollinearity issues in the model. The results have showed that all the VIF scores were lesser than 5, which indicates that there was negligible amount of multilinearity issues in the model, e.g., VIF < 5. The individual factor loadings ranged between 0.88 and 0.98 which exceeded the minimum required 0.7 score.

Table 2. Descriptive statistics, multicollinearity, and factor loadings

Variable	Indicator	Factor loadings	VIF	Mean value	Standard deviation	Min	Max
Circular economy implementation intention	CIRCULAR1	0.92	4.59	4.707	0.455	4.000	5.000
	CIRCULAR2	0.90	3.51	4.634	0.482	4.000	5.000
	CIRCULAR3	0.88	1.52	4.671	0.470	4.000	5.000
	CIRCULAR4	0.94	3.92	4.598	0.490	4.000	5.000
	CIRCULAR5	0.90	4.26	4.555	0.497	4.000	5.000
Environmental innovativeness	ENVINNO1	0.95	3.26	4.402	0.730	3.000	5.000
	ENVINNO2	0.98	2.68	4.439	0.734	3.000	5.000
	ENVINNO3	0.95	3.43	4.439	0.682	3.000	5.000
Green culture	GC1	0.86	3.65	4.591	0.492	4.000	5.000
	GC2	0.94	4.05	4.628	0.483	4.000	5.000
	GC3	0.84	3.60	4.445	0.497	4.000	5.000
	GC4	0.97	2.08	4.591	0.492	4.000	5.000
	GC5	0.78	2.96	4.396	0.489	4.000	5.000
Green intellectual capital	GIC1	0.91	4.13	4.262	0.583	3.000	5.000
	GIC2	0.94	4.99	4.445	0.627	3.000	5.000
	GIC3	0.85	3.81	4.220	0.564	3.000	5.000
	GIC4	0.95	4.42	4.360	0.614	3.000	5.000

4.3. Measurement model

As mentioned earlier, convergent validity has been measured in this paper through factor loadings, composite reliability, and AVE. The analysis showed that all the indicators had factor loadings greater than 0.7 as shown in Table 2. Again, composite reliability (CR) estimates ranged between 0.94–0.97 which is higher than the minimum acceptable 0.70 score. AVE scores ranged between 0.78–0.92 which is above 0.50. again, CA scores were above 0.70 indicating acceptable convergent validity and within-data consistency (Chin, 1998) Cronbach’s alpha (Hair et al., 2017) as shown in Table 2. Moreover, the discriminant validity was measured through Heterotrait-Monotrait Ratio (HTMT) as suggested by Henseler et al. (2015) as well as the cross-items loadings. Using this ratio, the inter-construct correlation should be less than 0.90. As detailed in Table 3 the correlation estimates ranged between 0.15–0.55 which is below 0.90. Therefore, the model is acceptable for discriminant validity. Moreover, items loadings are far higher than the cross-items loadings as seen in Table 4.

Table 3. Convergent validity

	CA	CR	AVE	CEII	EI	CGC	GIC
Circular economy implementation intention	0.95	0.96	0.83	0.91
Environmental innovativeness	0.96	0.97	0.92	0.21	0.96
Corporate green culture	0.93	0.94	0.78	0.26	0.14	0.88
Green intellectual capital	0.93	0.97	0.83	0.34	0.02	0.55	0.91

Note: Italic values in the diagonal row are square roots of the AVE.

Table 4. Discriminant validity (Heterotrait-Monotrait Ratio)

	CEII	EI	CGC	GIC
Circular economy implementation intention
Environmental innovativeness	0.21
Corporate green culture	0.24	0.15
Green intellectual capital	0.35	0.15	0.55

Note: CA=Cronbach alpha; CR=composite reliability; AVE=average variance extracted; CEII=circular economy implementation intention; EI=environmental innovativeness; CGC= corporate green culture; GIC=green intellectual capital.

4.4. Structural model

The predictive strength of the model has been assessed using the R² values while the predictive relevance been assessed using the Q² values as showed in Table 7. From the model the R² values ranged between 0.16–0.30 which suggest that the independent variables in the model are able to explain between 16–30% changes in CEI behaviour. The Q² values have been used to validate the predictive relevance of the model as shown in Table 6. The results have shown that the Q² values are greater than zero, implying that there is enough evidence to support the predictive relevance of the model despite the relative week R² values as shown in Figure 2. The results in Table 5 and 6 show that CGC has positive and significant (beta = 0.216, t = 4.293) effect on CEII, CGC has positive and but insignificant effect (beta = 0.143, t = 1.627) on EI, CGC has positive and significant effect (beta = 0.548, t = 13.901) on GIC. Moreover, EI has significant and positive effect (beta = 0.207, t = 2.725) on CEII, GIC has significant and positive effect (beta = 0.341, t = 4.695) on CEII. Also, the results have shown that GIC significantly mediate the relationship (beta = 0.187, t = 4.158) between CGC and CEII. Meanwhile environmental innovativeness failed to mediate the relationship (beta = 0.03, t = 1.279) between CGC and CEII. Figure 3 shows the path-coeffients, T-values and P-values.

Figure 2. Path – coefficients and R² values.

Figure 3. Path – coefficients, P-values and T-values.

Table 7. Construct cross-validated redundancy

	SSO	SSE	Q^2 (=1-SSE/SSO)
Circular economy implementation intention	820.00	721.39	0.12
Environmental innovativeness	492.00	482.71	0.02
Corporate green culture	820.00	820.00
Green intellectual capital	656.00	501.86	0.23

Table 5. Cross-loadings

	Circular economy implementation intention	Environmental innovativeness	Corporate green culture	Green intellectual capital
CIRCULAR1	0.92	0.16	0.17	0.32
CIRCULAR2	0.90	0.03	0.06	0.24
CIRCULAR3	0.88	0.09	0.11	0.28
CIRCULAR4	0.94	0.26	0.29	0.35
CIRCULAR5	0.90	0.31	0.40	0.32
ENVINNO1	0.25	0.95	0.08	−0.01
ENVINNO2	0.23	0.98	0.11	0.03
ENVINNO3	0.13	0.95	0.24	0.03
GC1	0.19	0.05	0.86	0.45
GC2	0.26	0.18	0.94	0.63
GC3	0.10	0.19	0.84	0.36
GC4	0.36	0.12	0.97	0.53
GC5	0.18	0.06	0.78	0.34
GIC1	0.34	−0.13	0.37	0.91
GIC2	0.31	0.13	0.57	0.94
GIC3	0.18	−0.15	0.39	0.85
GIC4	0.38	0.11	0.60	0.95

Table 6. Path – coefficient and hypotheses testing

Hypotheses paths	Beta	Mean	Std. deviation	T-statistics	Decision
Direct effect
H1: CGC -> CEII	0.216	0.218	0.050	4.293	Supported
H2: CGC -> EI	0.143	0.144	0.088	1.627	Unsupported
H3: CGC -> GIC	0.548	0.549	0.039	13.901	Supported
H4: EI -> CEII	0.207	0.211	0.076	2.725	Supported
H5: GIC -> CEII	0.341	0.342	0.073	4.695	Supported
Mediation (indirect) effects
H6: CGC -> GIC -> CEII	0.187	0.189	0.045	4.158	Supported
H7: CGC -> EI -> CEII	0.030	0.029	0.023	1.279	Unsupported

Note: CA=Cronbach alpha; CR=composite reliability; AVE=average variance extracted; CEII=circular economy implementation intention; EI=environmental innovativeness; CGC= corporate green culture; GIC=green intellectual capital.

4.5. Discussion

As part of the efforts to develop a baseline model to explain coordinated economic, ecological, and social development, improve resource use efficiency, and restore ecological environment, this paper was conducted. Specifically, the paper has analysed the implications of CGC on CEIs behaviour, and designed a model that explains the mediating roles of environmental innovativeness and GIC in between CGC and CEI behaviour with a concentration on lower-middle income economy. The paper has found that green culture positively affects CEI behaviour among manufacturing companies which is largely consistent with the first proposition of the paper which states that CGC drives CEI (Rizvi & Garg, 2021; Shahriari et al., 2022). CGC is scientifically and policy based that is aimed at achieving economic and environmental sustainability. For years now, the company has been incorporating this concept into their corporate social responsibility practices. Companies have realized that this paradigm shift affects market behaviour and thus sales, leading to higher profits (Martelo Landroguez et al., 2018). Moreover, green intellectual capital has a significant effect on CEI behaviour which is in agreement with prior studies (Syed et al., 2022; Khalil et al., 2022). The paper argued that GIC is one of the drivers of CEI behaviour and serves as mediator in between CGC and CEI (Amores-Salvadó et al., 2021; Liao et al., 2021; W. Ali et al., 2021). Again, the paper has revealed that GIC serves as a medium through which green culture could indirectly impact on CEI behaviour. As showed by the available empirical evidence (Jiao et al., 2022; Jirakraisiri et al., 2021; Sheikh, 2022; Xi et al., 2022), consistent with prior studies, Abd et al. (2022) asserted that GIC provides a set of intangible knowledge assets (skills, competencies, and capabilities) that a company manages in order to promote environmental performance. Moreover, López-Gamero et al. (2011) argued that GIC could be used to promote environmental management that would enable it gain competitive advantage over competitors.

5. Conclusion, implications, and limitations

5.1. Conclusion

This paper was undertaken to analyse the impact of CGC on CEI behaviour and to design a model that explains the mediating roles of environmental innovativeness and GIC on the relation between CGC and CEI behaviour from the perspective of a lower resource economy context. Our paper is positioned on positivists’ reasoning and quantitative research approach. The paper has found that CGC positively affects CEI behaviour. Moreover, environmental innovativeness and GIC have significant and positive effects on CEI behaviour. Besides, GIC serves as a medium through which CGC indirectly impacts on CEI behaviour. The paper concludes that CGC is a significant determinant of CEI behaviour. Moreover, CEI behaviour and environmental innovativeness serve as a medium to enhance CEI behaviour in the context of a lower middle-income economy.

5.1.1. Theoretical implications

The novelty of this paper is centred on the fact in the context of lower middle-income countries like Ghana, there is no study which has tested the mediation roles of environmental innovativeness and GIC on CEI using SEM technique. Moreover, the emergence of the baseline model could be used to facilitate CEI behaviour in Ghana and other Sub-Saharan African countries. The newly developed model could be used to explain coordinated economic, ecological, and social development, improve resource use efficiency, and restore ecological environment. The NRBV with slight modification has been used to developthe new baseline model. This model is context specific and could be used for replication studies within the larger lower-middle income countries with precisions to facilitate the adoption and implementation of circular economy models to save the environment and the biodiversity. Moreover, the paper further contributes to empirical knowledge stock by investigating the effect of two mediators on the relationship between green culture and CEIs which have remained fuzzy in the Ghanaian economy where the current study is focused.

5.2. Practical implications

Policymakers, practitioners and media are encouraged to facilitate the reinforcement of environmental policies and subsequent realization of Sustainable Development Goals. This paper contributes the development of circular economy and its implementations in the context of lower-middle country to address improve well-being (SDG3), decent work (SDG8), responsible consumption and production (SDG13), and climate action (SDG13). The mediators will allow investors and policymakers to identify other indirect but significant contributors to the implementations of circular economy. These results could be applied to facilitate investment decisions with respect to the adoption and implementation of circular economy models in the context of lower-middle country to address improve well-being, decent work, responsible consumption and production, and climate action. Moreover, the Ghana Environmental Protection Agency could collaborate with Ghana Enterprise Agency (GEA) to educate firm owners, managers, and investors on the need to adapt the circular at all levels of operations to reduce pollution, wastage, and degradation.

5.3. Limitations of the study

The present study has been conducted to analyse the relationship between CGC and CEI behaviour. Future studies should consider situational variables such as leadership commitment and green human resources management that could serve as moderators, and mediators. It is suggested that future studies should consider low-income countries or better still do a comparative study between the low- and middle-income countries with respect to CEIs. Again, the current study adopted deductive reasoning and quantitative approach. It is highly recommended that future studies focus on different research approach including qualitative and inductive. Moreover, different population should be used in future studies since the present study focused entirely on manufacturing companies in Ghana.

Disclosure statement

No potential conflict of interest was reported by the authors.

Appendix A

Survey instrument

Demographics information

Table

Variables	Frequency	Percentage
Gender
Male	144	58.8
Female	101	41.2
Age (years)
18–24	32	13.1
25–29	80	32.7
30–34	47	19.2
35–39	39	15.9
40–44	31	12.7
45–49	6	2.4
50–54	4	1.6
55+	6	2.4
Education
Master’s degree	63	25.7
Degree/HND	93	38.0
Secondary	44	18.0
Primary	12	4.9
Others	33	13.5
Number of years worked
<5	119	48.6
5–9	59	24.1
10–14	37	15.1
15–19	22	9.0
20+	8	3.3
Firm Location
Rural areas	76	31.0
Urban areas	107	43.7
Peri-urban areas	62	25.3

Table

Constructs	Sources	Measurement scale
Corporate green culture The company I’m working for has many environmental concerns. The company I’m working for is striving to protect the environment. The company I’m working for observes relevant environmental laws and regulations. The company I’m working for makes some changes in its management practices to protect the environment. The company I’m working for is environmentally accountable.	Shahriari et al. (2022)	5 – strongly agree 4 – agree 3 – neutral 2 – disagree 1 – strongly disagree
Environmental innovativeness Our organization uses less or non-polluting/toxic materials. Our organization improves environmentally friendly packaging for existing and new products. Our organization recovers end-of-life products and recycling.	Liao et al. (2022)	5- Strongly Agree 4- Agree 3- Neutral 2- Disagree 1-Strongly Disagree
Green intellectual capital The company awards financial bonuses for pro-environmental achievements. The company provides flexible working hours in order to undertake pro-ecological activities. The company evaluates employees’ ecological competencies. The company invests in environmental training for employees.	Issa et al. (2022)	5 – strongly agree 4 – agree 3 – neutral 2 – disagree 1 – strongly disagree
Circular economy implementation behaviour Our company recycles waste products from consumers. Our company collects waste during production. Our company reprocesses the waste products recovered. Our company reuses raw materials to produce similar products. Our company is committed to reducing energy consumption.	Wamba et al. (2023) Kusi-Sarpong et al. (2021),	5 – strongly agree 4 – agree 3 – neutral 2 – disagree 1 – strongly disagree