Dynamic means changing, evolving in response to or in anticipation of environmental change, and capability is defined as an ability to learn and improve in contrast to an organizational capacity which means holding, accommodating, or receiving knowledge in a restrictive manner. Therefore, Bhupendra and Sangle (2015) define dynamic capabilities as the routines of tasks that when followed are internalized and in the long run become part of organizational capabilities. This is because dynamic capabilities create, combine, and use their resources in new ways; meaning they can synthesize and integrate such routines in unique ways leading to the creation of new knowledge, solution, or configurations (Winter, 2003; Barreto, 2010). So, for example, dynamic capabilities could be adopting new technologies to prevent environmental pollution which will also lead to a competitive edge in the ecological domain. However, for such technologies to flourish, certain micro-level capabilities are required for instance management’s eco-friendly perspective, organizational learning, shared vision, and knowledge through cross-functional integration (Leonidou et al., 2015). Few studies have explored how sustainability can become an organizational capability, enhancing the organizational capacity to mold and innovate toward a sustainable paradigm (Gabler et al., 2015). Also, the resource-based theory is employed by many researchers to understand the organizational response to confront environmental imperatives, However, the dynamic capability framework is better suited for the ever-changing micro and microenvironment of the organization.

However, the work on dynamic capabilities as a dominant perspective in the development of corporate sustainability is scarcely discussed in the literature. Its in-depth discussion is required since such studies can guide corporations to develop required capabilities that can help them to respond successfully to a sustainability challenge arising from their external or internal environment by adjusting their strategies accordingly.

According to Teece (2007), there are three coherent clusters of dynamic capabilities including sensing, seizing, and reconfiguration capabilities. Sensing is the capability that identifies and assesses opportunities for sustainability. The process of mobilizing internal and external resources and competencies to capture value is termed seizing capability. Finally, the continuous resource regeneration and orchestration for aligning the organization’s resources with the evolving business environment is referred to as reconfiguration capability. However, another core capability as discussed by Schreyögg and Kliesch-Eberl (2007) is the monitoring capability which refers to continuously checking the validity of capability in regular intervals to prevent it from being obsolete, and facilitating the organization in gaining flexibility and adaptability. Furthermore, the dynamic sustainable capabilities are supported by micro-foundations like employee skills and competencies, organizational routines, and structure which should be considered as a multidimensional construct. These organizational routines are different ways through which the organization can deploy these sensing, seizing, monitoring, and reconfiguration capabilities.

In terms of theoretical approaches employed to study organizational capabilities, the RBV was the most common theory utilized to study resources, capabilities, and strategic management (Barney et al., 2001). RBV provided an explanation regarding the superior performance of some organizations compared to others. Also, Teece et al. (1997) discussed the dynamic capabilities approach to explicate the organizational competitive advantage in dynamic markets. However, focusing on the internal environment only and excluding the ever-changing external environment proved to be a limitation of Resource-Based Theory. This was addressed by Hart (1995) as he proposed the Natural Resource-Based Theory (NRBV) to include natural and organizational resources and environmental issues in RBV. When considering that these resources are ever-evolving, the NRBV transitions to a dynamic resource-based view. In Hart’s view, challenges caused by the natural environment are among the most influential factors in the new pattern of resource development and Organizational Capabilities (Hart and Dowell, 2011). Moreover, over the years, the economic and social environmental challenges raised in the NRBV have multiplied.

In this regard, it is argued by Annunziata et al. (2018) that sustainability-oriented organizations should identify and develop specific capabilities to enhance their competitive advantage. The literature mentions several strategies like the adoption of proactive environmental strategies, for example, Delmas (2011), successful implementation of environmental management systems and practices (Yu and Ramanathan, 2016; Johnson, 2017; Charan and Murty, 2018), adoption of Corporate Social Responsibility (Choi et al., 2019) and management of green supply chain. Hence, it is pertinent to develop the dynamic capabilities to implement these strategies. Despite the fact that there is a body of knowledge on dynamic sustainable capabilities, there is a recent need for more research in this area. According to Gelhard and Von Delft (2016) the literature has been debating about performance in facing economic, social, and environmental issues.

Also, organizations are increasingly considering the environmental issues in their macro-environment which might be the side effect of their production processes on account of tighter regulations by the government or increasing pressure from a wide variety of stakeholders. The literature regarding environmental management is plentiful but not much is known about the organization-centric capabilities which can facilitate to adopt of sustainable environmental management (Grewatsch and Kleindienst, 2018). It was still unclear whether being green is the antecedent of organizational capabilities or whether organizations build certain necessary capabilities which help them to become greener as a consequence. The literature suggests that not only does environmental management increase environmental as well as economic and social performance. Although there is ample literature on environmental management and corporate performance, the role of dynamic sustainable capabilities in explaining their relationship is lacking. Also, there is a lack of integration of social and environmental issues in the theoretical framework and empirical model of previous studies (for instance, Grewatsch and Kleindienst, 2018). In this regard, the importance of corporate sustainability is achieving prominence among organizations. Research has recently studied corporate sustainability from the strategic perspective highlighting that organization’s dynamic sustainable capabilities can have a positive impact on corporate performance (Bocken and Geradts, 2020).

Nevertheless, the relationship between organizational capabilities and sustainability benefits like increased triple bottom line performance remains little explored. For example, Gabler et al. (2015) argue that little is known about how firms determine and use the appropriate resources to maximize the performance of environmental initiatives. Then, the impact of corporate capabilities on corporate performance is addressed by Huang and Huang (2020) but it is limited to the transportation industry and the dynamic nature of internal capabilities is not discussed. Therefore, this study argues the following:

Then, according to Jabbour and Santos (2008) and Iqbal (2018), human resource forms the knowledge base of an organization motivating them to invest further into its people. The greening of human resources refers to the set of policies and bundle of practices that enable the organizational human capital to preserve the copious knowledge resource. It is achieved in an eco-friendly and efficient manner in terms of high productivity and low cost (Masri and Jaaron, 2017; Tang et al., 2018). Greening involves engaging in sustainable practices which apply to all the Human Resource functions like recruitment, selection, performance management, pay and reward, employee training, and development among others (Mustapha et al., 2017). It is done to allocate the scarce resources efficiently meanwhile promoting employee morale and their satisfaction with the job which is precisely environmentalism (Renwick et al., 2013; Zaid et al., 2018).

For instance, through recruitment and selection, the potential candidates for employment are informed about the organizational mission and values for sustainability. The green element is included in their job description as well to highlight the desirability of green practices for the organization. Through employee training and development function, green competencies and skills among the employees are fostered. Green team building, green skills, and learning management systems are established to develop and re-enforce pro-environmental behavior (Hameed et al., 2019, 2020). Then, another function of human resource is motivating employee which can be partly done through green performance management which reward employees based on their actions and behaviors about green issues like responsible resource consumption and similar attitude toward the environment and appropriate reward system.

Furthermore, regarding rewards, intrinsic, public, social, and personalized reward systems along with positive performance appraisals lead to pro-social and proactive behavior among employees resulting in higher sustainability performance (Govindarajulu and Daily, 2004; Jackson and Seo, 2010).

The relation between the GHRM and environmental commitment is discussed by Maung et al. (2016) while GHRM’s impact on environmental performance is highlighted by Daily et al. (2012), Kim et al. (2019), and Haldorai et al. (2022). However, the social pillar of the triple bottom line has been mostly ignored in the literature. This scenario is improving since some studies like the one conducted by Amrutha and Geetha (2019) attempt to bridge this gap between GHRM and the social sustainability of the organizations by including the mediating role of Employee Green Behavior (also used interchangeably with the term of Employee Pro-Environment Behavior) and establishing these links through the theories of Ability, Motivation, Opportunity, and Social Identity. Thus there is a major gap in the literature for exploring the social aspect of sustainability in GHRM literature. The questions like what is the contribution of GHRM practices in achieving the triple bottom line can be further explored by empirical testing of each of the Green Human Resource Practice on health, wellness, and wellbeing of the employees along with the organization’s environmental and financial performance.

Also, among GHRM practices, Khan et al. (2020) found that green assessment and rewards are not effective individually for explaining sustainable performance in the case of the Malaysian manufacturing industry. To test their findings, the following is argued:

Literature on green HRM and sustainability is limited and shows mixed results in the research (Shafaei et al., 2020; Yong et al., 2020). Further, little discussion is made regarding Green HRM practices leading to corporate sustainability within Pakistan (e.g., Awan et al., 2017). The contribution of Green HRM practices in developing a sustainable working environment has been confirmed by researchers like Chams and García-Blandón (2019) and Muisyo et al. (2021). Yet, the association between green HRM and a firm’s ecological performance has overall mixed findings. For instance, Shahzad et al. (2020) discussed the association of Environment Sustainability to Corporate Social Responsibility and Green Innovation in the Pakistani manufacturing industry. Similarly, Abbas (2020) established a positive relationship between Corporate Social Responsibility with corporate green performance in Pakistani manufacturing industries. But these researchers have considered only one facet of corporate sustainability which is the environment. The other dimensions of sustainability have been largely ignored in the literature like society or social sustainability in a particular sector. The past studies have contemplated the importance of integrating the studies of HRM practices and sustainability-related performance of the organization and it is also suggested for future studies by Macke and Genari (2019) to determine the impact of sustainable HRM practices on the psychological and social wellbeing of individuals. It then manifests in the behavior and commitment of the management.

Moreover, a recent study conducted by Rehman et al. (2021) on 244 Malaysian large manufacturing corporations revealed the lack of association between GHRM and sustainability performance. Consequently, researchers like De Stefano et al. (2018). acknowledge the importance of further examining the relationship between GHRM and corporate social sustainability performance. The paucity of research in this area serves as the motivation to address sustainability performance in the current study seeking empirical evidence of a significant association between GHRM and sustainable performance as conceptualized in Figure 1.

In terms of the target population, discussion regarding Pakistan’s manufacturing sector is paramount since it is the second-largest contributor to the country’s GDP, the first being the agriculture sector. Therefore, the manufacturing company’s sustainability status or performance is pivotal to study since it can play a significant role in steering Pakistan on the road toward sustainable development (Ikram et al., 2019). Hence, the impact of two macro-level components of management is investigated on two levels of Corporate sustainability performance. Corporate Sustainable Performance, as a result, leads to achieving a competitive edge through balancing improved performance as well as demonstrating a responsible attitude toward society and the environment. This research investigates the relationship between the GHRM, dynamic sustainable capabilities, and corporate sustainability performance variables with each other to test a predictive Corporate Sustainability Performance framework from the perspective of human resource management. Corporate Sustainable Performance has the potential to provide a competitive edge to a highly lucrative industry in Pakistan.

Further, SECP registered and PSX listed organizations from the manufacturing sector make the sampling frame for this research. Six large-scale industries based on their contribution to the GDP, their export percentages, and their contribution to the employment generation are selected and purpose sampling is done (Economic Survey of Pakistan, 2019; OECD, 2020). For instance, textile has the highest contribution to the country’s economy, and the COVID-19 pandemic has presented a unique opportunity; due to the shutdown of China, which was the single largest textile-related exporter, the Pakistani textile industry is working at full capacity to replace China in the global market. Also, it has nearly achieved its export target of $24–25 billion (Siddique, 2020).

The respondents were employees from large-scale manufacturing companies including men and women with 2 years and above experience. These companies belonged to industries like textiles, pharmaceuticals, automobiles, chemicals, food and beverages, and coke and petroleum products from major cities of Pakistan including Karachi, Hyderabad, Lahore, Faisalabad, and Quetta.

In this study, the items used in the survey were adapted from existing research to fit the context of Corporate sustainable performance and a seven-point Likert scale is used ranging from strongly disagree to strongly agree. Items for GHRM are adapted from the study by Tang et al. (2018). The dynamic sustainable capabilities scale is adapted from Wang and Ahmed (2007) and Shang et al. (2020). Finally, Corporate sustainable performance is defined as the performance of a corporation in all the dimensions or parameters of sustainability including environmental, social, and economic (Schaltegger and Wagner, 2006). Scale adapted from Wijethilake (2017) and Shang et al. (2020). The questionnaire items are shown in Figure 2.

The relationship between latent constructs and their indicators is examined through the measurement model and it is commonly employed to determine the inter-relationship patterns among the constructs of a conceptual model. To analyze and establish the causal associations among several constructs, a good measurement model is required. The first step in the analysis process is to check the reliability and validity of a measurement model before conducting the tests for the structural model. The measurement model needs to have acceptable levels of reliability and validity.

The reliability is equal to the squared correlation between the true construct and the construct scores in absence of systematic error. In this research, three types of reliability measures are determined. According to Dijkstra and Henseler (2015), the most important among them is ρA. Then, composite reliability (ρc) and Cronbach’s alpha are also calculated over the measure of composite reliability Jöreskog’s ρ, ω, or ρc, and Cronbach’s alpha is also calculated. In particular, the true reliability is under-estimated by Cronbach’s alpha and therefore, should only be considered as a lower boundary or minimum criteria of the reliability.

Table 2 shows all three reliabilities of the constructs used in this research. All the values of Dijkstra–Henseler’s rho (ρ_A), Jöreskog’s rho (ρ_c), and Cronbach’s alpha exceed the threshold level of 0.7, which represent that the constructs adapted for this research are highly reliable (Henseler et al., 2014).

Convergent validity is calculated for the constructs to ensure that the items of the same construct should be related to each other as well as the construct. On the other hand, discriminant validity helps to ensure whether items from different constructs are different from each other. The average variance extracted (AVE) is the most common measure of convergent validity whereas the Hetero-trait mono-trait ratio of correlations (HTMT) measures the discriminant validity.

Table 3 exhibits that the convergent validity of each construct exceeds the standard value, i.e., 0.5, confirming the validity of the constructs in the instrument.

Since the reliability statistics are high, therefore, item if the deleted option is not availed in this case. The next step is to move the pertinent data to SmartPLS for predictive model measurement and analysis. The structural model (representing relationships among constructs) has three constructs. Among them, the latent variables were GHRM Practices (GRS and GPR), Dynamic sustainable capabilities (MC and RC), and Corporate sustainable performance (ENP and SP). Based on the hypotheses generated in the previous phase, all the constructs are treated as first-order and reflective in nature (Hair et al., 2017; Sarstedt et al., 2019). The measurement model has twenty-seven indicators, where all items were directly measured in the research sample (reflective in which the construct defines the indicator variables) as conducted by Annunziata et al. (2018) and Zaid et al. (2018) for CSP. Also, organizational capabilities have reflective measurements similar to the study conducted by Huang and Huang (2020). Table 5 shows the outer loadings of the measurement model.

The paths between constructs are exhibited as standardized coefficients. In Table 6, there are two target constructs also known as endogenous constructs whereas the rest of the constructs in the model are categorized as predictors or exogenous. The graphical representation of R squared is provided in Figure 5. As suggested by Cohen (1992) in social science research, R-square value of 0.12 or below indicates low, between 0.13 and 0.25 values indicate medium, 0.26 or above, and above values indicate high effect size.

The R square for Corporate sustainable performance is 15.2% meaning 15.2% of the variance in CSP can be successfully explained by these exogenous variables, namely, GHRM and Dynamic sustainable capabilities.

The Q square measure is used to determine the predictive relevance of the reflective construct in the PLS-SEM model. This statistic is highly relevant since this research model is essentially predictive and has all reflective constructs and indicators. In this case, the construct cross-validated redundancy approach is employed (Hair et al., 2014) because of its suitability in its evaluation. It uses components of the structural, and path model as well as anticipated excluded data points with the following results.

As depicted in Table 7, the value for Corporate sustainability performance meets the criteria; it is significant with a value of greater than zero, however, its predictive accuracy is weak.

Then, effect size or f square needs to be measured which provides the contribution of a variable to the predictive relevance of the concerned construct. It is checked to employ the same threshold values, meaning that according to Hair et al. (2017), f square values higher than 0.35 is considered as having a large effect size, 0.15–0.35 are medium, whereas value lying between 0.02 and 0.15 is small. As depicted in Table 8, the independent variables’ DSC and GHRM effect size on the dependent variable (CSP) is small.

After substantiating the explanatory and predictive powers of our model, the researchers proceeded toward the final steps of testing the extent to which various factors affect Corporate sustainability performance through path analyses. A path model was tested that related two independent variables including GHRM and Dynamic sustainable capabilities with Corporate sustainable performance (Hair et al., 2019). It is shown in Figure 6.

Consistent with Ha1, overall Green HRM was found to significantly impact Corporate sustainable performance CSP (β = 0.198, p = 0.00). Hence the first alternate hypothesis was accepted. Ha2 stated that DSC on the whole has a significant impact on Corporate sustainable performance. It was found to be substantiated. (β = 0.257, p = 0.00) as shown in Table 9.

Further, the sub-hypotheses for Ha1 and Ha2 were evaluated in Table 10 with the following result:

The results of the bootstrapping are provided in Table 10 which shows that among the eight hypotheses, three are empirically supported the remaining five hypotheses could not be supported and the hypothesized relationships among the exogenous constructs and with the endogenous construct could not be established for those hypotheses. Moreover, the direct effect has been validated through the two-tailed test with a 0.05 significance level. The above table reveals that green recruitment and selection has a positive and significant impact on both corporate social and environmentally sustainable performance. Further, among the dynamic sustainable capabilities, monitoring capability has a significantly positive impact on corporate sustainable social performance.

A common problem in research that can compromise its rigor is the presence of common method bias in the research instrument which can become a source of measurement errors (Ardura and Artola, 2020). Extraction sums of squared loadings values below 50% in Table 11 indicate an absence of the common method bias according to Jordan and Troth (2020) and hence add credibility and rigor to our findings.