Green Innovative Work Behavior Toward Net-Zero in the Maritime Industry: The Moderating Roles of Climate Change Perception and Government Subsidies

Abstract

Amid growing international pressure for carbon neutrality, the maritime industry is facing mounting institutional demands for environmental innovation. Drawing on protection motivation theory, this study surveyed 499 employees from 1519 shipping service firms to examine how coercive, mimetic, and normative pressures shape green innovative work behavior. By extending protection motivation theory to a highly regulated maritime context, this study demonstrates that institutional pressures do not directly drive green innovation; instead, they enhance employees’ self-protective motivation, which subsequently fosters eco-innovation. Moreover, these relationships are stronger when firms perceive greater climate risks or receive government subsidies, indicating that contextual conditions amplify the translation of motivation into behavior. Overall, the findings reveal how macro-level institutional forces shape the sustainable transition of maritime services through employee psychology, offering governance-relevant insights for policymakers and firms seeking to promote green innovation.

1. Introduction

Since the twentieth century, rising global environmental awareness has led to increasingly stringent environmental regulations, profoundly reshaping contemporary business practices worldwide [1,2,3,4]. In parallel, the pursuit of “net-zero carbon emissions” has emerged as a global policy agenda, prompting governments to implement regulatory and incentive-based measures aimed at accelerating industrial decarbonization. According to the International Energy Agency (IEA), the transportation sector accounted for 27% of global carbon emissions in 2019; together with the power generation sector, it contributed to more than two-thirds of total global emissions [1]. Among different modes of transportation, the maritime industry is particularly carbon-intensive due to its heavy reliance on hydrocarbon-based fuels, whose combustion releases substantial amounts of CO₂ and poses significant challenges for climate change mitigation [5]. In response, the International Maritime Organization (IMO) has introduced a series of international environmental regulations, including requirements related to energy efficiency, sulfur oxide and nitrogen oxide emissions, ballast water management, and oil spill prevention [6]. As a result, shipping companies are increasingly confronted with intensified decarbonization mandates, heightened public scrutiny, and growing technological and compliance-related pressures [7]. Consequently, green shipping has become widely recognized as a critical strategy for reducing greenhouse gas emissions in the maritime sector [8]. Despite the growing importance of green shipping, existing green innovation research has largely focused on performance outcomes while paying comparatively limited attention to the behavioral and psychological mechanisms through which green innovation is enacted within shipping organizations. This gap is particularly salient in the maritime industry, where innovation is often compliance-driven, regulatory dependence is high, and technological substitutability remains limited. Under such conditions, understanding how institutional pressures are internalized by employees and translated into green innovative work behavior (GIWB) is both theoretically important and practically urgent. In particular, limited attention has been paid to the motivational processes through which institutional pressures are psychologically internalized by employees and translated into discretionary green innovative work behavior within highly regulated maritime organizations.

Human factors account for approximately 70–80% of marine sustainability risks, including transport accidents, chemical cargo leaks, marine debris, and oil spills [9]. These risks have triggered a wave of green transformation across the global shipping industry. For instance, the European Union has explicitly identified corporate sustainability as a key to addressing environmental challenges [10]. Increasing attention has thus been paid to the interplay between organizational policies and employee behaviors—particularly how employees psychologically and behaviorally respond to environmental regulations, which are considered a crucial driver for organizations to create social and environmental values [11]. Meanwhile, prior studies suggest that creating a positive organizational climate can encourage employees to engage in GIWB, facilitating sustainable corporate transformation [12,13,14]. Green innovation has emerged as a viable pathway that integrates ecological sustainability with competitive advantage [15,16]. Moreover, employee responses to their organizations’ environmental initiatives are critical to the ultimate social and ecological impact of such efforts [11].

Although green innovation plays a critical role in advancing organizational sustainability, most existing studies have focused on its relationship with performance outcomes [17,18], while the mechanisms through which green innovation emerges—particularly the individual-level drivers within the maritime industry—remain insufficiently explored. In the context of escalating environmental risks and increasingly stringent climate policies, governments have progressively imposed coercive institutional pressures to compel the maritime sector to pursue green transformation [19]. Green transformation in the maritime industry encompasses multiple dimensions, including green energy adoption, sustainable ports, and green supply chain management, and relies heavily on the integration of institutional, technological, and human resource systems [20]. However, merely emphasizing the importance of green shipping is insufficient to explain why, under conditions of intensive regulation, maritime employees exhibit substantial variation in their engagement in GIWB [21]. This variation suggests that regulatory intensity alone is insufficient to explain green innovative engagement, highlighting the need to examine underlying motivational mechanisms at the individual level.

This study draws on Protection Motivation Theory (PMT), which explains how individuals form motivations and behavioral responses when facing external threats [22,23,24]. It should be noted that PMT primarily emphasizes fear-based and self-protective motivations in response to perceived threats. In contrast, prior green behavior research has also highlighted intrinsic motivation and normative commitment as key drivers of pro-environmental actions. Rather than replacing these perspectives, this study adopts PMT to capture a complementary motivational pathway that is particularly salient in highly regulated and compliance-oriented contexts such as the maritime industry. The maritime industry is characterized by intense regulatory oversight, transnational governance, and limited technological substitutability, where environmental pressures often arise from abstract and distal institutional demands rather than immediate risk exposure [25,26,27]. These features represent an important contextual boundary for the application of PMT. Accordingly, rather than proposing a novel causal mechanism, this study uses the maritime context to examine how institutional pressures operate as distal threat cues that influence green innovative work behavior through self-preservation motive (SPM). Accordingly, GIWB in this study is conceptualized as a motivationally driven form of discretionary work behavior, acknowledging that such behaviors may not always translate directly into long-term or fully institutionalized practices. By incorporating enterprise climate change perception and government subsidy policy as moderators, the study further identifies the conditions under which the PMT-based motivational–behavioral relationship is strengthened or weakened.

Against this backdrop, we propose an integrative research model to deepen the understanding of how institutional pressures activate individual-level motivation, leading to pro-environmental behaviors that enhance corporate environmental governance and sustainability responsibility. This framework also offers insights relevant to the United Nations Sustainable Development Goals (SDGs), particularly Goal 14: “Life Below Water” and Goal 13: “Climate Action.” By promoting GIWB among maritime employees, this study aims to fill existing research gaps, reduce marine pollution, improve eco-efficiency in shipping operations, and encourage environmentally responsible business models. The following sections introduce the theoretical foundations of each variable and present the study’s hypotheses, followed by an empirical analysis. We then discuss the results, draw conclusions, and provide recommendations for future research. This discussion expands the applicability of PMT within the domain of green management and sustainable ocean development.

2. Literature Review

2.1. Relationship Between Institutional Pressures and Self-Preservation Motive

Amid rising global environmental protection awareness, corporate organizations are increasingly subjected to multiple forms of external institutional pressures that shape their responses to sustainability and environmental challenges. Institutional pressures—including coercive, normative, and mimetic pressures—originate from the broader regulatory and social environment and influence internal organizational arrangements such as managerial structures, decision-making processes, and compliance practices [28]. To maintain legitimacy and ensure organizational survival, firms often adjust their strategies and operational orientations in response to these pressures [29].

Institutional theory distinguishes among coercive pressures arising from formal regulations, normative pressures derived from social expectations and professional standards, and mimetic pressures reflecting imitation of leading firms under uncertainty [28]. While prior research has primarily examined the effects of these pressures on organizational-level outcomes such as regulatory compliance and environmental strategy [30,31], relatively little is known about how institutional pressures are cognitively interpreted by employees and translated into individual-level psychological mechanisms [32]. PMT provides a useful framework for explaining this micro-level translation process. PMT suggests that perceived external threats trigger threat appraisal and coping appraisal, which together activate SPM and motivate protective action [22,33]. Although PMT has been widely applied in health and environmental behavior research [34], it has rarely been integrated with institutional theory to explain how abstract institutional pressures shape employee motivation. In the context of the maritime industry, where environmental regulations are stringent and compliance failures carry substantial consequences, institutional pressures are likely to be perceived as salient threats, thereby activating employees’ SPM. Accordingly, coercive, normative, and mimetic pressures are expected to positively influence SPM.

Accordingly, this study argues that when confronted with various forms of institutional pressures, employees in the maritime industry may perceive such pressures as threats to their organizational or personal environments, stimulating their SPM. Based on this theoretical rationale, the following hypotheses are proposed:

H1: 

Coercive pressures have a positive effect on self-preservation motive.

H2: 

Normative pressures have a positive effect on self-preservation motive.

H3: 

Mimetic pressures have a positive effect on self-preservation motive.

2.2. Relationship Between Self-Preservation Motive and Green Innovative Work Behavior

Innovative work behavior refers to employees’ deliberate efforts to generate, promote, and implement new ideas, processes, or solutions within their work roles and has been widely recognized as a key driver of organizational adaptability and long-term competitiveness [35]. In the sustainability context, GIWB extends this concept by emphasizing environmentally beneficial innovations, including the generation and implementation of eco-friendly ideas and practices [36]. Prior research has predominantly highlighted the positive outcomes of GIWB, demonstrating its contributions to organizational performance, competitive advantage, and environmental improvement [37].

In addition, most GIWB studies have focused on manufacturing, energy, or general service sectors, offering limited insight into highly regulated industries such as maritime shipping, where green innovation is shaped by stringent regulations, technological constraints, and compliance-oriented institutional environments [38,39,40]. Consequently, the individual-level drivers of GIWB in the maritime context remain underexplored. To address this gap, the present study adopts an individual-level perspective to examine how institutional pressures are internalized as psychological motivation and subsequently translated into GIWB among maritime employees.

Drawing on PMT [22,41,42], individuals may respond to environmental threats through affective reactions such as fear or anxiety, which activate a SPM and encourage protective behaviors aimed at mitigating potential risks or losses. In the maritime industry, environmental regulations and climate-related risks often function as distal and abstract threat cues rather than immediate hazards. Accordingly, this study conceptualizes SPM as a central mediating mechanism through which institutional pressures are translated into proactive green innovation, rather than assuming a direct fear–behavior relationship. Therefore, this study proposes the following hypothesis:

H4: 

Self-preservation motive has a positive effect on green innovative work behavior.

When organizations face increasingly severe environmental challenges, employees are exposed to heightened perceptions of environmental risk and uncertainty. According to PMT, these perceptions activate threat appraisal and coping appraisal processes, which together give rise to SPM [23,43]. External contextual factors such as regulatory requirements, stakeholder expectations, competitive dynamics, and environmental norms can be conceptualized as institutional pressures that shape how employees perceive environmental threats [44,45]. Rather than directly triggering discretionary behaviors, coercive, normative, and mimetic pressures intensify employees’ threat appraisal and reinforce SPM, which functions as a key psychological mechanism translating external pressures into internal motivation.

GIWB represents a proactive coping response through which employees generate, promote, and implement environmentally beneficial ideas [46,47]. By engaging in GIWB, employees seek to mitigate perceived environmental risks while simultaneously enhancing organizational sustainability performance [48,49]. In this sense, SPM serves as a motivational conduit linking institutional pressures to GIWB. Accordingly, SPM is expected to mediate the relationships between coercive, normative, and mimetic pressures and green innovative work behavior.

Therefore, when maritime industry professionals face pressure from government regulations, public expectations regarding corporate ethics and sustainability, and peer influence from industry leaders promoting sustainability practices, their SPM is likely to be stimulated. This, in turn, leads to the adoption of GIWB as a concrete response to mitigate environmental risks and satisfy external demands. Based on the above, the following hypotheses are proposed:

H5a: 

Self-preservation motive mediates the relationship between coercive pressures and green innovative work behavior.

H5b: 

Self-preservation motive mediates the relationship between normative pressures and green innovative work behavior.

H5c: 

Self-preservation motive mediates the relationship between mimetic pressures and green innovative work behavior.

2.3. Moderating Effects of Enterprise Climate Change Perceptions and Government Subsidy Policy

2.3.1. Moderating Effect of Enterprise Climate Change Perceptions

As climate change increasingly threatens global ecosystems and human health, public and organizational perceptions of climate risks have become critical antecedents of environmental action [50,51]. ECCP reflects an organization’s understanding of the causes, consequences, and temporal dynamics of climate change and has been widely recognized as an important psychological precursor to climate-related responses [52,53].

Nevertheless, prior studies consistently report an awareness–behavior gap, even in contexts with high levels of environmental knowledge [54]. This gap is often explained by the gradual and cumulative nature of climate change and by temporal discounting that weakens incentives for immediate action [55,56,57,58]. These insights indicate that climate awareness alone is insufficient to ensure proactive environmental behavior. At the same time, climate change perception has been widely conceptualized as a critical “psychological gateway” that enables individuals to recognize environmental risks as personally and organizationally relevant, thereby creating the preconditions for subsequent adaptive responses [59].

Within the PMT framework, ECCP does not directly induce behavior but instead shapes how individuals interpret environmental threats and evaluate the necessity of protective responses. Although PMT has been criticized for its emphasis on fear-based motivation, ECCP provides a cognitive context that renders abstract and distal climate risks more salient within organizational settings.

In the maritime industry, where environmental pressures are largely institutional and regulatory in nature, ECCP strengthens the translation of employees’ SPM into GIWB by increasing the perceived relevance and urgency of environmental threats. Accordingly, ECCP is conceptualized as a key boundary condition that amplifies the motivational–behavioral linkage proposed by PMT, rather than as an independent driver of green behavior. We propose the following hypotheses accordingly:

H6a: 

Enterprise climate change perceptions moderate the mediating relationship between normative pressures and green innovative work behavior through self-preservation motive. This relationship is stronger when perceptions of climate change are higher and weaker when they are lower.

H6b: 

Enterprise climate change perceptions moderate the mediating relationship between coercive pressures and green innovative work behavior through self-preservation motive. This relationship is stronger when perceptions of climate change are higher and weaker when they are lower.

H6c: 

Enterprise climate change perceptions moderate the mediating relationship between mimetic pressures and green innovative work behavior through self-preservation motive. This relationship is stronger when perceptions of climate change are higher and weaker when they are lower.

2.3.2. Moderating Effect of Government Subsidy Policy

The global promotion of carbon neutrality has accelerated green transformation across sectors, leading governments and international organizations to adopt a range of regulatory and incentive-based policy instruments. Prior research in agriculture, transportation, and energy consistently indicates that government subsidies can lower economic barriers and enhance positive evaluations of green practices, thereby facilitating environmentally responsible behavior [1,60,61,62,63].

Nevertheless, much of this literature emphasizes policy effectiveness in terms of observable outcomes, with limited attention to the psychological mechanisms through which subsidies influence individual behavior. From a PMT perspective, subsidies do not directly induce action; instead, they shape individuals’ appraisal of coping efficacy by signaling the availability and feasibility of effective responses to environmental threats. Studies further suggest that subsidies improve supply chain performance and stimulate sustainable innovation by reducing perceived risks and enhancing access to critical resources [64,65,66]. These findings imply that government subsidies function as institutional resources that strengthen beliefs about the manageability and effectiveness of green actions.

In the maritime industry, where environmental pressures are largely institutional and compliance-driven, government subsidies are unlikely to independently motivate GIWB. Rather, by reducing cost burdens and uncertainty, subsidies reinforce the translation of employees’ SPM into discretionary green innovative work behavior. Accordingly, government subsidy policy is conceptualized as a boundary condition that amplifies the motivational–behavioral linkage predicted by PMT. Therefore, we propose the following hypotheses:

H7a: 

Government subsidy policy moderates the mediating relationship between normative pressures and green innovative work behavior through self-preservation motive. This relationship is stronger when the level of subsidies is higher and weaker when it is lower.

H7b: 

Government subsidy policy moderates the mediating relationship between coercive pressures and green Innovative work behavior through self-preservation motive. This relationship is stronger when the level of subsidies is higher and weaker when it is lower.

H7c: 

Government subsidy policy moderates the mediating relationship between mimetic pressures and green innovative work behavior through self-preservation motive. This relationship is stronger when the level of subsidies is higher and weaker when it is lower.

Based on this study’s research motivation, objectives, and theoretical hypotheses, we adopt PMT to construct an integrated research framework that models GIWB among maritime industry employees. The framework is designed to elucidate the relationships among key variables, including normative, coercive, and mimetic pressures, SPM, GIWB, ECCP, and GSP (Figure 1).

3. Research Methodology

3.1. Sample and Questionnaire Administration Procedure

Data were collected between April and June 2025, corresponding to the announcement period of the IMO’s most recent carbon reduction regulations. The study population comprised managerial personnel from maritime freight forwarders, container terminal operators, and shipping agencies listed in the publicly available database of the Taiwan Maritime and Port Bureau. Because organizational consent was required and survey links had to be disseminated through internal corporate contact points, this study employed a voluntary non-probability sampling approach (voluntary response sampling). This sampling approach is commonly adopted in organizational and industry-based survey research where access to decision-makers is constrained, and it is considered appropriate for theory testing rather than population parameter estimation. Such a sampling strategy may give rise to response bias, as firms with stronger sustainability orientations or a greater willingness to engage in academic research may be more inclined to participate. To test the proposed hypotheses, data were collected using a structured questionnaire administered via the online survey platform SurveyCake.

To ensure data quality and achieve an adequate response rate, a multi-stage sampling and questionnaire distribution procedure was implemented. First, the research team compiled contact information for potential firms from publicly accessible government websites and distributed research invitation emails. Only organizations that explicitly expressed willingness to participate were provided with the formal online survey link. Second, the survey was forwarded by designated internal corporate contacts to eligible managerial respondents. The cover page of the questionnaire clearly stated that participation was anonymous, that the data would be used solely for academic research purposes, and that no personally or organizationally identifiable information would be collected. In addition, respondents were informed that there were no right or wrong answers and that the survey aimed solely to capture their genuine perceptions. Anonymous participation and de-identified data handling procedures were adopted to mitigate the risks of social desirability bias and common method bias. Finally, to enhance response willingness, a small incentive was offered: respondents who completed the questionnaire were able to redeem a coffee voucher at a local convenience store via a designated link. Nevertheless, the study acknowledges that the use of incentives may disproportionately encourage participation among certain groups (e.g., individuals more inclined to complete surveys or more sensitive to sustainability-related issues). Accordingly, this potential source of response bias is explicitly recognized and discussed in the study’s limitations.

A total of 1519 maritime firms were contacted to assess their willingness to participate in the survey. After eliminating responses with inconsistencies or incomplete data, 499 valid responses were obtained, yielding an effective response rate of 32.85%. This response rate is comparable to, or higher than, those reported in prior organizational survey studies involving managerial respondents in regulated industries. The demographic profile of the respondents is as follows. Regarding gender, respondents were predominantly male (84.5%). Most respondents were between 31 and 40 years old (51.3%). Regarding education level, the majority held at least a bachelor’s degree (60.8%). For years of employment, over one-third of respondents had over six years of experience in the shipping service industry (33.9%). Regarding tenure with current mid-level supervisor, most had worked with their current supervisor for three to four years (51.6%). This gender composition is consistent with the industry reality that senior management positions in the shipping sector are predominantly held by men; nevertheless, it may constrain the generalizability of the findings to contexts characterized by different gender distributions.

3.2. Measurement Instruments

This study employed a structured questionnaire encompassing seven constructs: normative pressures, coercive pressures, mimetic pressures, SPM, GIWB, ECCP, and GSP. All constructs were measured using validated scales adapted from previous studies. To ensure the semantic accuracy and contextual relevance of the translated items, the questionnaire underwent a rigorous review process with academic experts from relevant fields. The final version of the questionnaire items is presented in the Appendix A.

The scales for normative, coercive, and mimetic pressures were adapted from Liang et al. [67] and Colwell and Joshi [32]. The normative pressure scale includes five items (Cronbach’s α = 0.917), the coercive pressure scale comprises four items (Cronbach’s α = 0.942), and the mimetic pressure scale comprises four items (Cronbach’s α = 0.882). This study places greater emphasis on certain dimensions of mimetic pressure—particularly technological imitation—while offering more limited coverage of strategic and managerial imitation. In addition, some item wordings may exhibit semantic proximity to normative pressure. To address these concerns, the measurement model assessment went beyond the heterotrait–monotrait (HTMT) ratio by also evaluating construct discriminant validity using the fit indices of a seven-factor model and the criteria proposed by Fornell and Larcker. Furthermore, potential limitations related to the content validity of the measurement scales, along with directions for future refinement, are explicitly acknowledged in the study’s limitations section. The SPM scale was derived from Trumbo and Harper [68] and Orchowski et al. [69] and includes nine items, with a Cronbach’s α of 0.974. To assess GIWB, six items were adapted from Scott and Bruce [70] and Yan et al. [71], resulting in high internal consistency (Cronbach’s α = 0.953). The ECCP scale was based on Van Valkengoed et al. [72] and comprises 13 items, demonstrating excellent reliability (Cronbach’s α = 0.980). Finally, the GSP construct was measured using a three-item scale adapted from Chang [1], with a Cronbach’s α of 0.883. Because Chang’s [1] original study was conducted in the context of electric vehicle–related behaviors, its core measurement items focus on financial incentives and policy support, which are conceptually aligned with the subsidy logic underlying the shipping industry’s green transition. Accordingly, we performed context-specific semantic adaptations to fit the maritime setting and operationalized subsidies and tax incentives as a single construct representing external incentive intensity. In addition, the mimetic pressure scale was extended to incorporate managerial and strategic dimensions, thereby ensuring coverage beyond purely technological imitation.

Aligning with psychometric recommendations, all items were measured using a 7-point Likert scale. Responses ranged from 1 (strongly disagree) to 7 (strongly agree). This scale format was selected because higher-point Likert scales are better suited to capture subtle attitudinal differences [73], offering greater measurement precision than three- and five-point scales [74]. Higher scores indicate stronger agreement or perceptions regarding the construct being measured; lower scores indicate weaker agreement or perceptions.

4. Results

This study follows the standard analytical procedures of structural equation modeling (SEM). Descriptive statistics and measurement model assessments were first conducted to establish the reliability, validity, and discriminant validity of each construct, thereby ensuring that the measurement instruments accurately capture maritime employees’ cognitive representations of institutional pressures, psychological motivation, and GIWB. Upon confirming the adequacy of the measurement model, subsequent structural model estimation and path analyses were appropriately employed to examine the causal relationships among the constructs and to address the study’s central research questions: how institutional pressures influence maritime employees’ actual green behaviors through psychological mechanisms and under which conditions these effects are amplified or attenuated. Given the cross-sectional survey design, the estimated relationships should be interpreted as associative rather than strictly causal. Although structural equation modeling allows for the simultaneous estimation of complex relationships, potential endogeneity among the variables cannot be fully ruled out.

4.1. Descriptive Statistics and Correlation Analysis

Table 1. Pearson’s correlation coefficients and HTMT analysis among study constructs.

Constructs		1	2	3	4	5	6	7
1	Normative Pressures	(0.868)	0.365	0.066	0.594	0.458	0.100	0.203
2	Coercive Pressures	0.339 **	(0.924)	0.094	0.317	0.246	0.022	0.452
3	Mimetic Pressures	−0.083	−0.060	(0.861)	0.197	0.168	0.085	0.056
4	Self-preservation Motive	0.299 **	0.569 **	0.183 **	(0.909)	0.672	0.233	0.194
5	Green Innovative Work Behavior	0.23 **	0.434 **	0.155 **	0.647 **	(0.90)	0.274	0.084
6	Enterprise Climate Change Perceptions	−0.020	−0.097 *	−0.079	−0.229 **	−0.266 **	(0.898)	0.035
7	Government Subsidy Policy	−0.431 **	−202 **	0.049	−0.192 **	−0.086	0.037	(0.90)

Note. (1) Sample size = 499; (2) diagonal values in bold represent the square roots of the average variance extracted (AVE); (3) lower triangle shows Pearson’s correlation coefficients; upper triangle represents HTMT values; (4) *: p < 0.05, **: p < 0.01.

presents the means, standard deviations, and inter-variable correlation coefficients. The square roots of the average variance extracted (AVE) are displayed on the diagonal and were all greater than the corresponding inter-construct correlations in the lower triangle of the matrix. This result confirms the presence of adequate discriminant validity, following the criterion proposed by Fornell and Larcker [75]. To further strengthen the robustness of the discriminant validity assessment, this study employed the HTMT (Heterotrait–Monotrait ratio) criterion. The values in the upper triangle of the matrix were all below the conservative threshold of 0.85, as recommended by Hu and Bentler [76] and Henseler et al. [77] These findings provide additional evidence supporting discriminant validity among all constructs.

The results indicate that all study constructs exhibit adequate statistical reliability and validity, thereby providing a solid foundation for subsequent confirmatory factor analysis and structural model testing. This ensures that inferences regarding the effects of institutional pressures and psychological mechanisms on green behavior in the maritime industry are not confounded by measurement error.

4.2. Confirmatory Factor Analysis (CFA)

The results of the confirmatory factor analysis indicate that all factor loadings, composite reliability (CR), and average variance extracted (AVE) for the measured constructs exceeded the recommended thresholds proposed by Fornell and Larcker [75], demonstrating adequate convergent validity (see Appendix A for details). Regarding discriminant validity, Table 1 shows that all HTMT values in the upper triangle of the correlation matrix were below the conservative threshold of 0.85 [76,77]. The square roots of the AVEs (diagonal values) were greater than the inter-construct correlations (lower triangle), further supporting discriminant validity [75].

Table 2. Results of confirmatory factor analysis.

Model	${\mathit{x}}^2$	df	${\mathit{x}}^2$/df	$△{\mathit{x}}^2$	△df	$△{\mathit{x}}^2$/△df	CFI	TLI	RMSEA	SRMR
One-factor Model	16,526.324	902	18.322	-	-	-	0.348	0.316	0.187	0.2332
Two-factor Model	9975.669	901	11.072	−6550.655	−1	6550.655	0.621	0.602	0.142	0.192
Three-factor Model	8821.166	899	9.812	−7705.158	−3	2568.386	0.669	0.652	0.133	0.1987
Four-factor Model	5809.478	896	6.484	−10,716.846	−6	1786.141	0.795	0.783	0.105	0.133
Five-factor Model	4737.649	892	5.311	−11,788.675	−10	1178.868	0.84	0.83	0.093	0.0972
Six-factor Model	2939.749	887	3.314	−13,586.575	−15	905.772	0.914	0.909	0.068	0.0715
Seven-factor Model	2007.98	881	2.279	−14,518.344	−21	691.35	0.953	0.949	0.051	0.028

Note. (1) One-factor model: all constructs—coercive pressures, normative pressures, mimetic pressures, SPM, GIWB, ECCP, and GSP—are combined into a single factor; (2) two-factor model: (a) coercive pressures, normative pressures, mimetic pressures, and SPM are combined; (b) GIWB, ECCP, and GSP are combined into the second factor; (3) three-factor model: (a) coercive pressures, normative pressures, and mimetic pressures are combined; (b) SPM; (c) GIWB, ECCP, and GSP are grouped as one factor; (4) four-factor model: same structure as the three-factor model but with one additional distinction (3); (5) five-factor model: (a) coercive pressures, normative pressures, and mimetic pressures are combined; (b) SPM; (c) GIWB; (d) ECCP; (e) GSP; (6) six-factor model: (a) coercive pressures; (b) normative pressures; (c) mimetic pressures; (d) SPM; (e) GIWB; (f) ECCP and GSP are combined; (7) seven-factor model: all constructs are treated as independent latent variables—coercive pressures, normative pressures, mimetic pressures, SPM, GIWB, ECCP, and GSP; (8) Model fit indices were estimated using AMOS 26 statistical software.

presents the model fit indices for the seven-factor model. The model demonstrates an acceptable overall fit (χ²/df = 2.279, CFI = 0.953, TLI = 0.949, SRMR = 0.051, RMSEA = 0.028). The low RMSEA value indicates a close approximate fit of the measurement model to the observed data. Given the relatively large sample size and the complexity of the multi-factor model, RMSEA values below conventional cutoffs are not uncommon and do not necessarily indicate overfitting. Therefore, the overall pattern of fit indices should be interpreted jointly, suggesting that the measurement model exhibits a well-balanced and robust fit. Chi-square difference tests showed that the seven-factor model outperformed all competing alternative models. This result indicates that the measurement structure is statistically superior and supports construct distinctiveness. To examine common method variance (CMV), we conducted Harman’s single-factor test. The total explained variance across all factors was 80.783%, with the first factor accounting for only 32.704% of the variance. These results fall below the recommended cutoff, indicating no serious CMV concern in this study [78,79].

Taken together, the results indicate that the measurement model satisfies established empirical standards with respect to reliability, convergent validity, discriminant validity, and controls for common method bias. This provides a sound methodological foundation for subsequent structural equation modeling and moderation–mediation analyses to rigorously examine the relationships among institutional pressures, psychological motivation, and green innovative work behavior, thereby addressing the practical challenges faced by the maritime industry in advancing sustainable transformation.

4.3. Hypothesis Testing

This study employed AMOS 26 to test the direct and indirect effects of the proposed research model. Specifically, bootstrap resampling with 5000 iterations and a 95% confidence interval was applied using percentile bootstrap and bias-corrected percentile bootstrap methods to examine the three mediation models [80]. An effect was considered statistically significant if the confidence interval did not include zero, aligning with the established criteria [81].

The results in

Table 3. Standardized direct, indirect, and total effects of the hypothesized model.

		Point Estimate	Product of Coefficients		Bootstrapping
					Percentile 95% CI		Bias-Corrected Percentile 99% CI
			S.E.	Z	Lower	Upper	Lower	Upper
Direct Effect
Normative Pressures → Self-preservation Motive		0.197	0.056	3.5179 ***	0.108	0.291	0.114	0.297
Coercive Pressures → Self-preservation Motive		0.661	0.053	12.4717 ***	0.574	0.748	0.575	0.75
Mimetic Pressures → Self-preservation Motive		0.24	0.07	3.4286 ***	0.119	0.349	0.123	0.352
Normative Pressures → Green Innovative Work Behavior		0.042	0.051	0.824	−0.042	0.125	−0.038	0.129
Self-preservation Motive → Green Innovative Work Behavior		0.661	0.057	11.596 ***	0.569	0.754	0.57	0.754
Coercive Pressures → Green Innovative Work Behavior		0.133	0.071	1.873	0.011	0.249	0.007	0.244
Mimetic Pressures → Green Innovative Work Behavior		0.072	0.05	1.44	−0.012	0.15	−0.012	0.15
Indirect Effect
Normative Pressures → Self-preservation Motive → Green Innovative Work Behavior		0.13	0.039	3.33 ***	0.07	0.2	0.073	0.204
Coercive Pressures → Self-preservation Motive → Green Innovative Work Behavior		0.437	0.049	8.918 ***	0.359	0.521	0.364	0.53
Mimetic Pressures → Self-preservation Motive → Green Innovative Work Behavior		0.159	0.05	3.18 **	0.073	0.24	0.078	0.244
Total Effect		2.7	0.187	14.598 ***	2.413	3.027	2.429	3.036

Note: (1) Standardized estimating of 5000 bootstrap samples; (2) ***: Z > 3.29, **: Z > 2.58.

can be summarized as follows: 1. Standardized direct effects: (1) Normative Pressures → SPM = 0.197 (Z = 3.5179 ***, p < 0.001); (2) Coercive Pressures → SPM = 0.661 (Z = 12.4717 ***, p < 0.001); (3) Mimetic Pressures → SPM = 0.24 (Z = 3.4286 ***, p < 0.001); (4) SPM → GIWB = 0.661 (Z = 11.5965 ***, p < 0.001); (5) Normative Pressures → GIWB = 0.042 (Z = 0.824, p > 0.05); (6) Coercive Pressures → GIWB = 0.133 (Z = 1.873, p > 0.05); (7) Mimetic Pressures → GIWB = 0.072 (Z = 1.44, p > 0.05). 2. Standardized indirect effects: (1) Normative Pressures → SPM → GIWB = 0.13 (Z = 3.333, p < 0.001); (2) Coercive Pressures → SPM → GIWB = 0.437 (Z = 8.918, p < 0.001); (3) Mimetic Pressures → SPM → GIWB = 0.159 (Z = 3.18, p < 0.01). 3. Standardized total effects: 2.73 (Z = 14.598 ***, p < 0.001). The empirical results support H1–H4 and the mediating H5a–c.

This study applied Model 6 of Hayes’s [79] PROCESS macro to examine H6a–c and H7a–c. After controlling for gender, age, and educational background, the results indicate the following (

Table 4. Results analysis table.

		1		2		3
		Self-Preservation Motive	Green Innovative Work Behavior	Self-Preservation Motive	Green Innovative Work Behavior	Self-Preservation Motive	Green Innovative Work Behavior
		Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Control Variables
Gender		0.2082 (0.1303) [−0.0478, 0.4642]	−0.1472 (0.1139) [−0.3710, 0.0765]	0.0837 (0.1118) [−0.1360, 0.3034]	−0.1680 (0.1128) [−0.3896, 0.0536]	0.1092 (0.1341) [−0.1542, 0.3726]	−0.1707 (0.1136) [−0.3938, 0.0524]
Age		−0.0068 (0.0668) [−0.1381, 0.1245]	0.0004 (0.0583) [−0.1141, 0.150]	−0.0961 (0.0577) [−0.2095, 0.0173]	−0.0181 (0.0585) [−0.1331,0.0968]	−0.0130 (0.0689) [−0.1485, 0.1224]	−0.0010 (0.0585) [−0.1159,0.1138]
Education		−0.0004 (0.0755) * [−0.1488, 0.1480]	0.0503 (0.0661) * [−0.0795, 0.1802]	−0.0245 (0.0648) [−0.1518, 0.1028]	0.0428 (0.0658) [−0.0864, 0.1721]	0.0588 (0.0777) [−0.2114, 0.0939]	0.0431 (0.0662) [−0.0870, 0.1732]
Independent Variables
Normative Pressures		0.3617(70.0652) *** [20.611, 0.4623]	0.0864 (0.0511) [−0.0139, 10.867]
Coercive Pressures				0.6663 (0.0430) *** [0.5818, 0.7509]	0.1315 (0.0536) * [0.0261, 0.2369]
Mimetic Pressures						0.1758 (0.0426) *** [0.0921, 0.2595]	0.0170 (0.0377) [−0.0570, 0.0910]
Mediating Variable
Self-preservation Motive			0.6101 (0.0456) *** [0.5205, 0.6997]		0.5654 (0.0510) *** [0.4652, 0.6656]		0.6255 (0.0447) *** [0.5376, 0.7134]
Interference Variable
Enterprise Climate Change Perceptions			−0.1181 (0.0398) ** [−0.1964, −0.0398]		−0.1190 (0.0397) ** [−0.1970, −0.0410]		−0.1148 (0.0399) ** [−0.1932, −0.0364]
Government Subsidy Policy			0.0548 (0.0341) [−0.0122, 0.1218]		0.0406 (0.0314) [−0.0212, 0.1024]		0.0830 (0.0334) * [0.0174, 0.1487]
Interaction
Self-preservation Motive×Enterprise Climate Change Perceptions			0.114 (0.0327) *** [0.0471, 0.1756]		0.1102 (0.0326) ** [0.0462, 0.1742]		0.1087 (0.0334) *** [0.0430, 0.1744]
Self-preservation Motive×Government Subsidy Policy			0.0784 (0.03333) * [0.0130, 0.1437]		0.0714 (0.0334) * [0.0058, 0.1369]		0.0830 (0.0334) * [0.0174, 0.1487]
R *		0.0943	0.4588	0.3286	0.4622	0.0360	0.4558
F		120.8617 ***	460.0557 ***	600.4541 ***	460.6978 ***	40.6164 ***	450.5129 ***

Notes: (1) n = 499; (2) * p < 0.05; ** p < 0.01; *** p < 0.001; (3) Model 2: H6a, H7a, Model 4: H6b, H6b, Model 6: H7c, H7c; (4) The value is the standard error.

). In Model 2, both ECCP (β = 0.0114, 95% CI [0.0471, 0.1756], p < 0.001) and GSP (β = 0.0784, 95% CI [0.0130, 0.1437], p < 0.05) had significant and positive moderating effects on the relationship between normative pressures and GIWB through SPM.

In Model 4, moderating effects were also evident in the pathway from coercive pressures to GIWB via SPM, with ECCP (β = 0.1102, 95% CI [0.0462, 0.1742], p < 0.01) and GSP (β = 0.0714, 95% CI [0.0058, 0.1369], p < 0.05) demonstrating significant and positive moderating effect. Similarly, in Model 6, ECCP (β = 0.1087, 95% CI [0.0430, 0.1744], p < 0.001) and GSP (β = 0.0830, 95% CI [0.0174, 0.1487], p < 0.05) were shown to significantly and positively moderate the indirect effect of mimetic pressures on GIWB via SPM.

To further test H6a–c and H7a–c, this study employed a moderated mediation analysis using Model 16 of Hayes’s [82] PROCESS macro. After statistically controlling for gender, age, and educational background, the results from Table 4 (Models 2, 4, and 6) demonstrate that the model explained a substantial portion of the variance in GIWB. The explanatory power of each model was significant and robust: R² = 0.4588, F(9, 489) = 46.0557, p < 0.001 for Model 2; R² = 0.4622, F(9, 489) = 46.6978, p < 0.001 for Model 4; R² = 0.4558, F(9, 489) = 45.5129, p < 0.001 for Model 6.

Further analysis revealed that ECCP and GSP had significant positive moderating effects on the indirect relationships between institutional pressures and GIWB via SPM. Specifically, ECCP significantly moderated the mediating path from normative pressures to GIWB through SPM (β = 0.011, SE = 0.032, p < 0.001), from coercive pressures (β = 0.110, SE = 0.032, p < 0.01), and from mimetic pressures (β = 0.108, SE = 0.032, p < 0.001). Similarly, GSP showed significant moderating effects in the same mediation chains: for normative pressures (β = 0.078, SE = 0.033, p < 0.05), coercive pressures (β = 0.071, SE = 0.033, p < 0.05), and mimetic pressures (β = 0.083, SE = 0.033, p < 0.05).

The indices of moderated mediation were statistically significant across all tested pathways, providing consistent support for the hypothesized conditional indirect effects. The confidence intervals for all indices excluded zero, indicating that the indirect effects of institutional pressures on green innovative work behavior via SPM systematically varied depending on the proposed moderators.

Following Aiken and West [83], we constructed interaction plots to illustrate the effects of high versus low levels of ECCP and GSP on the relationship between SPM and GIWB. Using normative pressures as an example, the interaction plots (Figure 2 and Figure 3) indicate that higher levels of ECCP strengthened the positive relationship between SPM and GIWB. Moreover, higher levels of GSP enhanced the positive association between SPM and GIWB. These findings align with the theoretical expectations, providing empirical support for H6a–c and H7a–H7c.

5. Conclusions and Implications

This study is situated within the maritime industry context and examines how institutional pressures shape employees’ GIWB through psychological motivational mechanisms while further highlighting the critical role of contextual factors in this transformation process. As an island economy surrounded by sea and highly dependent on maritime trade, Taiwan’s shipping industry represents not only a central pillar of economic activity but also a key arena for advancing industrial sustainability and green transformation. The empirical findings demonstrate that the effective promotion of green shipping cannot rely solely on external institutional mandates; rather, it hinges on whether such pressures are successfully internalized as behavioral motivation at the employee level.

Drawing on PMT as the core theoretical framework, this study provides empirical evidence that institutional pressures do not directly drive GIWB. Instead, their behavioral effects operate through a psychological transformation mechanism—SPM. This finding extends and refines the PMT literature by suggesting that, in highly institutionalized and professionalized industry settings, external pressures alone are insufficient to induce behavioral change. Protective motivation is effectively activated only when employees interpret institutional requirements as salient threats related to their roles, responsibilities, and risk exposure. Specifically, the results show that normative, coercive, and mimetic pressures each significantly enhance employees’ SPM, which fully mediates the relationships between these institutional pressures and GIWB. This pattern indicates that the influence of institutional pressure on green behavior is neither linear nor immediate but rather an indirect process that materializes only after psychological internalization.

Notably, the direct effects of institutional pressures on GIWB are not statistically significant. In contrast to prior studies suggesting that institutional pressure can directly stimulate pro-environmental behavior, the present findings reveal a different underlying mechanism. Within the professionally oriented and highly regulated maritime industry, employees may have become accustomed to regulatory requirements and industry standards, causing institutional demands to be perceived as routine obligations rather than sources of behavioral motivation. Given the absent psychological processes of threat appraisal and coping appraisal, institutional pressure alone is therefore unlikely to translate into proactive green innovative behavior. This insight not only underscores the limitations of the “outside-in” logic commonly emphasized in institutional theory but also reinforces the explanatory value of PMT in accounting for institutional–behavioral disconnects.

Moreover, the findings indicate that when organizations provide clear and credible supportive institutional arrangements, employees are more likely to convert protective motivation into concrete action. From a PMT perspective, ECCP and GSP not only reduce perceived behavioral costs but also enhance employees’ evaluations of coping efficacy and self-efficacy, thereby amplifying the behavioral consequences of protective motivation. By doing so, this study extends PMT’s conceptualization of coping appraisal to the organizational level, demonstrating that institutional support itself constitutes a critical condition for the realization of psychologically driven green behavior. The following section outlines this study’s theoretical contributions, managerial implications, research limitations, and future research directions.

5.1. Theoretical Contributions

5.1.1. Extending the Application of Protection Motivation Theory

This study expands the application scope of PMT beyond its traditional focus on health behaviors and disaster response [33,34]. We extend this theory into the domain of organizational sustainability—specifically, the GIWB within the high-polluting, high-carbon maritime industry. Empirical findings confirm that when maritime employees perceive institutional environmental pressures, it triggers their SPM, leading to environmentally beneficial innovative actions. This finding supports the applicability of PMT in organizational behavior and sustainable management. Moreover, by incorporating ECCP and GSP as moderating variables, the study demonstrates how external institutional and internal cognitive factors can amplify the behavioral translation of protective motivation. This echoes the assertion by Van Valkengoed et al. [55] and Corner et al. [53] that climate awareness functions as a psychological “gateway” to behavioral change. Moreover, this study highlights the explanatory advantages of PMT in contexts characterized by high regulatory intensity and elevated risk. Compared with theoretical frameworks that primarily emphasize subjective norms or perceived behavioral control, PMT is better suited to capturing how institutional pressures function as threat cues and how individuals, through processes of threat appraisal and coping appraisal, translate risk perceptions into motivational states and subsequent action.

5.1.2. Empirically Validating the Mechanisms of GIWB in the Maritime Sector

Grounded in PMT, this study empirically examines how institutional pressures shape GIWB among personnel in the maritime industry. The findings indicate that coercive, normative, and mimetic pressures each exert a significant positive effect on SPM. In turn, SPM promotes GIWB and serves as a key mediating mechanism linking institutional pressures to employees’ green innovative behaviors. Moreover, ECCP and GSP exhibit significant moderating effects, such that employees are more likely to translate protective motivation into actual green actions when climate awareness is heightened or policy incentives are stronger [53,59].

5.1.3. Enriching the Literature on Green Shipping

Amid a global shift from fossil fuel-based transportation to green and zero-carbon alternatives, the maritime industry plays an essential role in energy transition and sustainable development. This study centers on GIWB among maritime personnel and positions ECCP and GSP as boundary conditions. By doing so, it elucidates how external institutional forces and policy instruments amplify the psychological–behavioral transformation pathway from institutional pressure to SPM and, ultimately, to GIWB within the shipping context. In this way, the study offers a more microfoundational perspective on the mechanisms underlying green shipping behavior.

5.2. Managerial Implications

Based on the empirical results, this study proposes three levels of managerial implications to advance green shipping and achieve the goal of net-zero carbon emissions for sustainable development.

5.2.1. Formulating Subsidy Policies to Encourage Green Innovative Work Behavior

The study finds that government subsidies significantly moderate the effect of the SPM on GIWB. Government subsidies, as a policy instrument, can reduce the cost thresholds associated with the adoption of green actions by both organizations and employees, thereby increasing the likelihood that motivational drivers are translated into actual behavior. Policymakers should therefore design more targeted and differentiated subsidy schemes to alleviate barriers to green action and to enhance organizational engagement in sustainability initiatives.

5.2.2. Enhancing Enterprise Climate Change Perceptions

The findings also show that ECCPs significantly strengthen the relationship between SPM and GIWB. Shipping firms can embed environmental sustainability into their organizational culture through internal climate risk education, goal-setting practices, and structured communication mechanisms. By strengthening employees’ understanding of climate-related risks and their sense of responsibility, such initiatives can facilitate the translation of sustainability intentions into concrete behavioral implementation.

5.2.3. Establishing Green Institutional Norms and Peer Learning Mechanisms to Internalize Pro-Environmental Behavior

This study’s empirical evidence indicates that normative, coercive, and mimetic pressures significantly enhance maritime employees’ SPM. Firms can reinforce coercive pressures by establishing consistent green operational standards and auditing systems while simultaneously strengthening positive mimetic pressures through benchmarking practices and inter-firm knowledge exchange. Together, these mechanisms can encourage employees to internalize sustainability values and foster the emergence of voluntary green behaviors.

5.3. Research Limitations and Future Research Directions

This study is subject to several limitations that provide valuable directions for future research.

First, this study adopted a cross-sectional design, which allows for the depiction of psychological and behavioral states at a specific point in time. However, such a design limits the ability to rigorously establish the causal direction and temporal sequencing among institutional pressures, SPM, and GIWB, which are theoretically conceptualized in PMT as dynamic and evolving processes. It also constrains the capacity to rule out potential reverse causality, such as the possibility that engagement in GIWB may, in turn, reinforce perceived institutional pressures or motivational states. Future research is therefore encouraged to employ longitudinal or multi-wave research designs—for example, by collecting measures of institutional pressures, SPM, and GIWB across multiple time points—to more closely capture the dynamic and adaptive processes emphasized by PMT.

Second, the sample in this study was drawn from maritime-related firms registered with the Taiwan Maritime and Port Authority. The disproportionately high proportion of male respondents (84.5%) may limit the generalizability of the findings across organizational contexts characterized by different workforce compositions and regulatory regimes, such as variations between the European Union Emissions Trading System and policy frameworks in Asian economies. Future research is encouraged to incorporate cross-national samples and conduct subgroup analyses based on gender composition, policy environments, or levels of institutional stringency in order to strengthen external validity.

Third, this study relied on a voluntary participation recruitment strategy incentivized by coffee vouchers, which may introduce response bias and social desirability bias. Such biases could influence the self-reported levels and observed relationships of GIWB and may contribute to an intention–behavior gap, whereby motivationally driven behavioral intentions do not fully translate into sustained or observable behaviors. This concern is particularly relevant given that PMT emphasizes motivational responses to perceived threats rather than long-term behavioral enactment. Future research is encouraged to triangulate self-reported measures with objective indicators (e.g., environmental audits, energy-saving and emission-reduction performance) or multi-source assessments (e.g., supervisor and peer ratings) and to conduct nonresponse bias analyses to further mitigate concerns regarding self-report and sampling biases.

Fourth, although this study implemented procedural remedies such as anonymity and confidentiality, and conducted method factor analyses in addition to Harman’s single-factor test, common method variance (CMV) cannot be entirely ruled out in a self-reported research design. This limitation may result in inflated estimates of the observed relationships, and thus the effect sizes should be interpreted with caution. Future research may mitigate CMV concerns by employing multi-source data, temporally separated measurements, or more rigorous CMV modeling techniques, thereby strengthening internal validity.

Fifth, with respect to measurement and model specification, future studies may extend the operationalization of mimetic pressure to more explicitly capture technological, strategic, and managerial imitation while reducing potential semantic overlap with normative pressure. Moreover, government policy instruments could be further disaggregated into distinct dimensions—such as subsidies and tax-based incentives—to examine differentiated moderating effects. Finally, incorporating key organizational-level control variables (e.g., firm size, green technological capability, and industry competitive intensity) would help alleviate omitted variable bias and enhance the robustness and precision of empirical inferences. In addition, potential endogeneity concerns cannot be fully excluded in the current research design. Reciprocal relationships or omitted variables may simultaneously influence institutional pressures, self-preservation motivation, and green innovative work behavior. Although the present study focuses on theory testing rather than causal identification, future research may employ instrumental variable approaches, longitudinal designs, or quasi-experimental methods to more rigorously address endogeneity and strengthen causal inference.