The Effect of Green Supply Chain Management Practices on the Sustainability Performance of Turkish Shipyards

Abstract

The study explores the effect of green supply chain management (GSCM) practices on three sustainability dimensions; environmental, social, and economic in Turkish shipyards. The GSCM practices examined are green design, green purchasing, green production, green marketing, environmental management, and recycling. A research model was developed to test the relationship between six GSCM practices and the three sustainability performance dimensions by using a shipyard-level survey. Environmental uncertainty was placed in the model as a moderator variable. The authors utilized a web-based survey. Partial least squares structural equation modeling was used to test the proposed hypotheses. The research revealed that the GSCM practices were positively significant for economic and social performance but not for environmental performance. Environmental uncertainty did not have a moderating effect between GSCM practices and sustainability performances. Quality, Environmental, and Safety Certifications are the first step for GSCM practices but do not mean the company achieved environmental performance positively. The results brought into prominence the GSCM practices in improving the sustainability performance of the Turkish shipyards. While this study was the pioneer in examining the relationship between GSCM and sustainability performance in Turkish shipyards, it enhanced the understanding of GSCM practices and sustainability performances.

1. Introduction

Technological evolutions have caused a transformation towards industrial-based rather than agricultural-based economies. However, boosted production and consumption have yielded excessive waste, environmental pollution, and depletion of natural resources. Consequently, Musmarra, et al. [1] have introduced the notion of sustainable development. It has introduced two questions for enterprises: First, which approaches should the companies follow to attain the desired sustainability status? Second, how should a company demarcate and measure its sustainability performance? Multiple solutions and innovations have been dealing with the environmental concerns of the sustainability concept and its dimensions, including green practices in the supply chain and production processes. The concepts and standards such as “sustainable development goals”, “sustainability performance”, “sustainable ocean principles”, “green supply chain”, and “zero-carbon targets” have been examined in several industries [2,3,4,5,6]. Green Supply Chain Management (GSCM) was a concept created during the 1990s to lessen the environmental impacts of supply chain management [7]. Many enterprises started “going green” in their internal and external procedures to achieve environmental sustainability. Thus, GSCM emanated as a systematic approach and evolved into an integrated aspect of business activities [8]. Samekto and Kristiyanti [9] defined the goal of GSCM as eliminating or minimizing waste (energy, gas emissions, hazardous chemicals, and other wastes) along the supply chain network.

Shipbuilding is one of the oldest industries in human civilization, and it takes place mainly in the open air. The rapid progress and the expansion of the shipbuilding industry have caused many environmental and social problems as a side effect. Ship construction evolved into a complicated compound of art and science, affecting soil and sea. Manufacturing or production, ship repair, servicing, and ship recycling are all distinct industrial processes of shipbuilding. They mainly occur in a shipyard where these operations result in significant pollution, contamination, and emissions into the environment [10]. Shipbuilding requires backup from numerous heavy industries and possible governmental subsidies. Nowadays, shipbuilding is characterized by enhanced globalization, increasing environmental regulations, and fierce competition [11].

Shipyards and their supply chain must protect the environment, respect labor rights, be transparent in their financial management and produce publicly available records. Shipyards are expected to follow international standards and fulfill the expectations of handling negative impacts on the environment, society, and labor resources, even when these do not exist in national law. Constructing greener and more sustainable ships mandated relative sustainability performance goals for the shipyards. In that case, green supply chain management practices can assist the shipbuilding sector in complying with International Maritime Organization (IMO) regulations and achieving current and future sustainability goals [12]. Sustainability goals retain environmental, social, and economic performance supported by necessary policies, decisions, and actions. Academicians and experts have determined several indicators to quantify sustainability performance [13]. The Green Marine Environmental Program (GMEP) was established in 2007 to mitigate the potential environmental impacts caused by the North American maritime industry. The program is voluntary, with a transparent and rigorous certification process. Participants—ship owners, port authorities, seaway companies, terminal facilities, and shipyards—must benchmark their environmental performance using Green Marine’s Self-Evaluation Framework [14]. Strandhagen et al. [4] stated that the shipbuilding industry was under significant economic pressure and needed more efficient solutions to secure economically sustainable operations. Social issues (philanthropy, safety, equity, health and welfare, ethics and human rights), the need for a greener maritime industry, and three dimensions of sustainability pressured the shipbuilding industry. International regulations focus on the industry’s design and operational phases, which carries a ship’s immense life-cycle climate impact. As emissions from the operational phase are reduced, emissions from shipyards during construction, maintenance, and disposal become increasingly important [15]. Klymenko and Lillebrygfjeld Halse [16] investigated how small and medium-sized companies in a regional cluster responded to sustainability development. The findings showed that the companies possess unique knowledge and expertise to be transformational and create a new development path. Ashari et al. [17] determined an Indonesian shipyard’s environmental conditions, environmental management, and environmental management strategies for controlling environmental pollution. The strength, Weakness, Opportunities, and Threat (SWOT) method is used to guide and coordinate environmental management by forming a dedicated department to reduce the unexpected impact of ship repair activities.

The stakeholders in the shipbuilding sector recognized an inseparable link between environmental and commercial factors. Ships and shipyards that diminish waste and emissions during construction have achieved technical, commercial, and environmental outcomes such as decreasing the consumption of raw materials, water, and energy and benefited cost reductions in production. Because of increasing recognition of the sustainability concept, many shipyards have emphasized environmental norms and certificates on their website, presentations, and reports. In addition, international shipyards concentrated on sustainability topics and published yearly reports to measure their sustainability performance and to represent their achievements [18,19,20].

Consequently, by conducting a case study in Turkish shipyards, the present study explores the effect of green supply chain management (GSCM) practices on three sustainability dimensions, environmental, social, and economic. The authors discovered a need to close the gap in the literature about GSCM practices’ effect on sustainability performance. The outcomes give a clear picture of the sustainability performance level in Turkish shipyards, contribute immensely to the emerging literature, and offer a starting point for future studies.

The study is organized as follows. The Section 2 describes the literature review and reveals the research gap in the shipyards regarding sustainability and green supply chain practices. The Section 3 describes the research model construct, related hypotheses, survey design, sample size constraint, and survey application. The following section includes demographic statistics, reliability and validity analyses, structural model assessment, and common method bias analyses. The implications of the results are reevaluated in Section 5. Section 6 summarizes the findings and proposed further research implications. Section 7 describes the limitation of the study.

2. Literature Review

The second phase includes the search execution. The Boolean operators “and” and “or” were used to combine searches. Two searches have been done in each database. The first search in the Scopus database was with the keywords “sustain” or “green” and “shipyard” and resulted in 151 articles. Using similar keywords but with “shipbuilding” resulted in 287 articles. When the Scopus search combined two searches, the total number of articles was 362. In contrast, the first search in the Web of Sciences database resulted in over one million results with the keywords “sustain” or “green” This result shows how the subject is popular in the research area. However, refining the existing search with “shipyard” or “shipbuilding” decreased the number of relevant articles to 96, which is a definitive proof that there is a gap in this area. Endnote software combined the two subtotal searches, and 458 articles remained.

The deeper analysis started in the third phase. Endnote excluded 84 duplicated articles, and the titles and abstracts of all the remaining articles were checked for eligibility concerning their subject areas. Arts and Humanities, Agricultural and Biological Sciences, Chemistry, Medicine, Chemical Engineering and Biochemistry, Genetics, and Molecular Biology were excluded. There remained 27 articles after eliminating articles without full content.

Different conceptualizations of GSCM practices exist in the literature. While some researchers classify greening inbound, greening production, and greening outbound, others classify internal and external actions. Pinto [21] summarized all of these in

Table 1. GSCM practices.

GSCM Practices	Description
Green design	Design of products for reduced consumption of materials/energy, intend to reduce products’ adverse effects on the environment during its entire life cycle, Design of products for reuse, recycling, recovery of materials, parts, Design the products to be easily set up for the users in the most energy-saving ways, Design for reduction of environmentally hazardous substances, Design for recycling waste, and Design for remanufacturing aimed at returning it to a better condition.
Green purchasing	Choice of suppliers by considering the environmental criteria, buying environment-friendly raw materials, Pressuring supplier(s) to take environmental actions
Green production	Generate minimum waste and reduce environmental pollution, Re-manufacturing and lean production, Cleaner Production, Improved capacity utilization, Lower raw material costs, gain production efficiency, improve their corporate image, increase the number of goods delivered on time, do not use hazardous or restricted materials during manufacturing and minimize waste during production, Substituting toxic inputs with environmentally friendly ones.
Customer environmental collaboration	Collaborate with customers to develop environmental management solutions, collaborate with customers to manage, and reverse flows of materials and packaging.
Internal green management Environmental management	Environmental compliance monitoring and Auditing, Total quality environment management, Pollution prevention plans, Environmental manager, and employee training. Employee incentive programs for environmental suggestions
Reverse logistics Recycling	Recovery of the company’s end-of-life Items, Resale or reuse of used parts or components, Old/obsolete items being used or replaced, Recondition and refurbishing of used parts or components

.

Green supply chain management is challenging to describe because of the broad concept. Many academicians reached different definitions. An increasing number of studies about the relationship between green supply chain management practices and companies’ performance have existed in the literature since 2014. These studies have categorized sustainability performance as environmental, economic, social, and operational performance. In contrast, various GSCM practices such as green purchasing, green packaging, recycling, eco-design, internal environmental management, reverse logistics, green production, green marketing, and environmental collaboration with suppliers and customers have been questioned in these studies. The researchers found negative and positive, primarily partial, relationships between GSCM practices and the company’s performance. The authors reviewed all practices in the literature and considered practices directly related to the shipbuilding industry. For example, environmental training is a part of internal environmental management. Therefore, these two applications converged as environmental management. Practices like green packaging and green transportation have little effect on the shipbuilding process and are omitted from the study.

Some reviewed studies referenced in this study are; Zhu and Sarkis [22], which examined the relationships between GSCM and environmental and economic performance and uncovered significant implications. Rao and Holt [23] resolved that GSCM practices enhance competitiveness and the environmental and economic performance of the ISO 14001-certificated leading East Asian companies. Azevedo et al. [24] identified that reverse logistics positively affect efficiency but negatively affect environmental costs. Environmentally friendly packaging influences both efficiency and quality positively. Felício et al. [25] stated that green management and efficiency contributed to controlling the impact of pollution with practical effects on economic sustainability. Küçükoğlu [26] scrutinized Turkish companies that have been listed among ISO 500 companies for the previous three years and have ISO 14001. Green innovation activities are vital to a company’s environmental performance, competitive advantage, and sustainability. Dilşad Güzel [27] observed that supply chain integration affected the implementation of the green supply chain and affected business performance. Praharsi et al. [28] discussed the sustainability of wooden fishing boat production and supply chain activities to support the industrial ecosystem. The regulations on the environmental, social, and economic sectors are necessary to achieve the sustainability of the material supply chain. Rupa and Saif [29] examined the impact of GSCM practices on business performance and the environmental sustainability of Bangladesh. Cost and profit were selected as two critical indicators of business performance. Environmental sustainability comprises waste disposal, resource consumption, and greenhouse gas emission. GSCM practices were found statistically significant on cost, waste disposal, resource consumption, and greenhouse gas emission, but not on profit. Vakili et al. [30] highlighted the negative impacts of energy consumption during shipbuilding and proposed to utilize renewable energy and reduce the environmental impact of shipbuilding.

Dizaji [31] analyzed green marketing strategies and their effects on the performance of food manufacturing companies in Istanbul. Green marketing, a green supply chain, and green innovation are significant for business performance. In another study, Derya Öztürk [32] found that green purchasing, production, and packaging positively affected business performance. Günday [33] wrote that green supply chain practices positively impact business performance in the chemistry sector. Yildiz Çankaya and Sezen [34] studied the effect of green supply chain management practices on three essential elements of corporate sustainability—environmental, economic, and social performance—within 281 different companies, including the automotive and electronics, and chemistry sectors. Green production, transport, and packaging positively impact the three sustainability dimensions, and other practices positively impact one or two dimensions. Davoodi and Sazegari [35] investigated the barriers to implementing sustainable supply chain management practices in an Iranian shipyard using fuzzy and nonlinear ANP methods. They found that knowledge and support had the highest impact, followed by cost and economic condition.

In brief, the literature review revealed the research gap in the shipyards regarding sustainability and green supply chain practices that this study aimed to fulfill and to contribute to the emerging shipyards’ sustainability literature through a case study in the Turkish shipyards. The research structure will be discussed in the next section.

3. Methodology

3.1. Research Model Construct

The authors designed the theory starting from the model in Çankaya [36] study, which included the following constructs for GSCM: “green production”, “green distribution”, “green packaging”, “green marketing”, “green purchasing”, “environmental training”, “internal environmental management”, and “recycling”. Green design” is an essential factor in the shipyard and is added to the model. “Environmental training” is a part of internal environmental management, so the authors converged these two practices into “environmental management”. “Green packaging” and “green distribution” have little effect on the shipbuilding processes, and they were not included in the model. “Green purchasing”, “green production”, “green marketing”, and “green recycling” were included because of the direct impact on the shipbuilding process.

The uncertainty effect in strategic processes comes from management perception [37]. Uncertainties are complicated and unforeseen factors, categorized as technology, supply, and demand uncertainties, reflecting the overall level of Environmental Uncertainty in the simplified supply chain. Demand uncertainty measures the fluctuations and variations level in demand. Supply uncertainty indicates the extent to which suppliers can meet manufacturers’ requirements and produce materials with consistent quality. Technology uncertainty is measured when green technology changes within the industry [38]. Sustainability performance is characterized by environmental, economic, and social performance. Environmental uncertainty is placed in the model to test whether it has an impact or a relationship between GSCM practices and sustainability performance dimensions as a moderator variable. Environmental uncertainty is analyzed if it has a moderating effect between GSCM practices or sustainability performances. Figure 1 shows the proposed research model.

The research hypotheses are listed as follows.

Hypothesis 1.

GSCM practices have a positive impact on environmental performance.

Hypothesis 2.

GSCM practices have a positive impact on economic performance.

Hypothesis 3.

GSCM practices have a positive impact on social performance.

Hypothesis 4a.

Environmental uncertainty has a positive moderating effect between GSCM practices and environmental performance.

Hypothesis 4b.

Environmental uncertainty has a positive moderating effect between GSCM practices and economic performance.

Hypothesis 4c.

Environmental uncertainty has a positive moderating effect between GSCM practices and social performance.

3.2. Survey Design and Application

The survey instrument was developed based on similar studies mainly applied in Turkey in different sectors like automobile, chemical, electronics, food industry, ISO (Istanbul Chamber of Industry) 500 companies, and hospitals [31,36,38,39,40,41,42,43]. Since the study concerns Turkish shipyards, the questions were in Turkish. The English translation of the questions and their references is presented in Appendix A. The pilot survey was conducted with the executives of seven shipyards through separate video calls. Feedback led to modifying a few sentence structures for better understanding and eliminating ambiguity.

The survey consisted of three main parts. The first part was an introduction letter explaining the aim of the study. The second part consisted of information about the company, including the average workforce, establishment year and location, type of vessels constructed in the shipyard, whether a department on environmental performance exists in the company and its size, the job title of the respondent, and which certificates does the company have. The questions about green supply chain management practices, environmental uncertainty, and sustainable performance were found in the third part. This part included 33 questions, which took around 15 min to fill in. The survey used a 5-Likert scale. Turkish shipyard executives and owners were the target population. The original language of the survey is Turkish. The author added an English translation in Appendix B. Turkish Shipbuilders Association (GISBIR) sent the questionnaire to the shipyards. GISBIR consists of almost all shipyards in Turkey. Therefore, its member list (56 existing shipyards) was the sampling group of this research.

The research continues with the analysis of the higher-order model (HOM) to decrease the number of relationships and to have a more parsimonious and straightforward model. The measurement model assessment starts with selecting an approach to form a higher-order model. The authors selected the disjoint two-stage because of its easy use and the most preferred one in the literature [44,45,46]. The reliability and convergent validity for all constructs were established using Cronbach’s Alpha value.

Next, the structural equation model (SEM) method was preferred for conducting the hypothesis tests of the proposed research model. Statistical analysis has been an essential tool for social science researchers for over a century. Applications of statistical methods have expanded dramatically with the advent of computer hardware and software, particularly in recent years, with widespread access to many more forms due to user-friendly interfaces with technology-delivered knowledge. Researchers initially relied on univariate and bivariate analysis to understand data and relationships. It is increasingly necessary to apply more sophisticated multivariate data analysis methods to comprehend more complex relationships associated with current research directions in the social science disciplines. Multivariate analysis involves the application of statistical methods that simultaneously analyze multiple variables. The variables typically represent measurements associated with individuals, companies, events, activities, situations, etc. The sizes are often obtained from surveys or observations used to collect primary data, but they may also be obtained from databases consisting of secondary data. Exhibit 5.1 displays some of the significant statistical methods of multivariate data analysis. SEM is used to test causal relationships between observed and unobserved (hidden) variables [47].

There are two types of SEM: covariance-based SEM (CB-SEM) and partial least squares SEM (PLS-SEM); also called PLS path modeling). CB-SEM is primarily used to confirm (or reject) theories (i.e., a set of systematic relationships between multiple variables that can be tested empirically). It does this by determining how well a proposed theoretical model can estimate the covariance matrix for a sample data set. In contrast, PLS-SEM is primarily used to develop theories in exploratory research. It does this by focusing on containing the variance in the dependent variables when examining the model.

3.3. Sample Size Constraint

Hair et al. [45] proposed using PLS-SEM when a small population restricts the sample size, the structural model is complex and includes many constructs, indicators, and model relationships, and when distribution issues are a concern. Rigdon [48] stated that the nature of the population justifies the sample size. The shipyards in Turkey formed a confined sector and numbered 56. As the authors had direct contact with each company, there was no need to select a sample, so all were invited, and 40 responded to the survey. This response number enabled the research to generalize the results for the Turkish Shipyard industry. Since the total shipyards numbered 56, the response rate was 71%, which is quite a high rate. Cohen [49] provided a table for sample size recommendation for a Statistical Power of 80% for multiple regression models. According to this table, six independent variables require at least 39 samples to detect R² values of at least 0.75 (with a 5% probability of error and 5% significance level). Regarding population size, survey response rate, and the sample size estimations above, the authors stated that 40 samples are enough to represent the whole target population in the study.

4. Results

4.1. Demographic Statistics

71% of the active GISBIR members responded to the survey. The respondents were management board members, directors, business development managers, HSE or quality managers, and other upper-level managers. The oldest shipyard was established in 1970. Twenty-six shipyards were established after the year 2000, which shows continuous developments in the shipbuilding industry. Shipyards are 28 years old on average. Tuzla Bay region and the Yalova regions contain 74% of existing shipyards. A few yards are in Antalya, Çanakkale, Trabzon, and Bursa. The two main operational activities are reparation and maintenance work and constructing new vessels and see platforms. Turkish shipyards create broad-ranging naval vessels with composite, aluminum, and steel platforms. The ship types range from ferry, cargo (chemical, container, dry cargo) ships, scientific research vessels, fishing vessels, live fish transport vessels, yachts, mega yachts, and various types of naval vessels (corvette, patrol, powerboat, landing craft, coast guard), floating dock, platform support vessels, tugboat, and safety boats.

As seen in

Table 2. Shipyards Certification.

Certificate	Quantity	Percentage
ISO 9001	2	5%
ISO 14001 & ISO 9001	13	30%
ISO 45001 & ISO 14001 & ISO 9001	28	65%

, all the examined shipyards possess active ISO certification; 30% have two certificates, ISO 14001 environmental management system, and ISO 9001 quality management, while 65% of shipyards have all three certifications, including ISO 45001 on occupational health and safety management. Turkish shipyards emphasized the certifications as commercial competence. They took necessary internal and external actions to manage the day-to-day activities and responsibilities systematically, contributing to the business performance.

Fourteen shipyards shared their workforce information. The total average personnel numbered 750 workers. While direct personnel accounted for 200 workers, subcontracted personnel reached 550 workers. This fact shows that Turkish shipyards directly employ primary personnel to sustain the minimum requirements and tasks but outsource tasks to wide-ranging suppliers. The direct workforce and subcontracted workforces are shown in Figure 2. Therefore, the shipyards control their workforce and gain a significant cost advantage with experienced suppliers that provide competitive prices. Similarly, Ecorys’ report about the Competitiveness of the European Shipbuilding Industry depicted that a relatively large part of production is outsourced or subcontracted. The production process can be organized by having low labor costs, generating efficient production processes, or outsourcing to retain the cost advantages, Consulting [50].

4.2. Reliability and Validity

The authors decided on a higher-order model (HOM) to decrease the number of relationships and to have a more parsimonious and straightforward model. In this model, the reflective indicators linked to the lower-order constructs (LOC) are connected to the higher-order constructs (HOC). Green supply chain management (GSCM) and environmental uncertainty (ENVR-UNCER) are connected directly to environmental performance (ENVRPER), economic performance (ECOPER), and social performance (SOCPER).

Table 3. Lower and Higher Order Constructs.

Indicators	LOC	Nomination	HOC
GrDESIGN	Green Design	GSCM	Green Supply Chain Management
GrPURCH	Green Purchasing
GrPROD	Green Production
GrMAR	Green Marketing
GrENVR	Environmental Management
GrRCYC	Recycling
UNDEMAND	Demand Uncertainty	ENVR-UNCER	Environmental Uncertainty
UNTEC	Technology Uncertainty
UNSUPPLY	Supply Uncertainty

represents the higher-order model (HOM) structure and the abbreviations of the indicators used in the rest of the analysis. Green design, green purchasing, green production, green marketing, environmental management, and recycling are nominated as lower-order constructs (LOC) and formed the higher-order construct (HOC) named green supply chain management practices (GSCM). Similarly, demand, technology, and supply uncertainty formed the higher-order construct of environmental uncertainty (ENVR-UNCER).

According to “the disjoint two-stage approach”, lower-order constructs are connected directly to ENVRPER, ECOPER, and SOCPER. After lower-order construct reliability and validity were established in the first stage, the next step was to create a higher-order construct.

The latent variables of the LOCs are copied and pasted into the existing data set as a new indicator variable. Hence, the latent variable of each lower-order construct became a new indicator of the higher-order construct. Since the indicators of LOC and the new indicator of HOC “GSCM and ENVR-UNCER” were reflective, the model is nominated as a reflective-reflective higher-order construct model. Each higher-order construct should satisfy the reliability and convergent validity recommended by Sarstedt and Cheah [46]. GSCM is confirmed as a HOC, whereas ENVR-UNCER is confirmed after extracting the UN-DEMAND indicator. The indicator statistic results are given in

Table 4. Indicator Statistics.

Indicators	Outer Loadings > 0.708		VIF < 5.000
	1st cal.	2nd cal.	1st cal.	2nd cal.
Gr-DESIGN	0.768	0.768	1.806	1.806
Gr-ENVR	0.690	0.687	1.761	1.761
Gr-MAR	0.833	0.834	2.458	2.458
Gr-PROD	0.827	0.828	2.099	2.099
Gr-PURCH	0.775	0.775	2.352	2.352
Gr-RCYC	0.823	0.822	2.624	2.624
UN-DEMAND	0.047	extracted	1.033	extracted
UN-SUPPLY	0.645	0.658	1.047	1.016
UN-TEC	0.833	0.829	1.017	1.016

.

The reliability and convergent validity for all constructs is established, as seen in

Table 5. Construct Reliability and Convergent Validity.

	Cronbach’s Alpha > 0.600		Composite Reliability > 0.700
	1st cal.	2nd cal.	1st cal.	2nd cal.
ENVR-UNCER	0.286	0.220	0.552	0.715
GSCM	0.878	0.878	0.907	0.907
ECO-PER	0.871	0.871	0.921	0.921
ENVR-PER	0.812	0.812	0.887	0.887
SOC-PER	0.599	0.599	0.778	0.778

. All constructs have a Cronbach’s Alpha value higher than 0.600, except ENVR-UNCER, with a value of 0.220. The composite reliability values of the constructs are higher than the recommended value of 0.700. Similarly, average variance extracted values passed the threshold of 0.500.

Table 6. Fornell and Larcker and HTMT.

	ECO-PER		ENVR-PER		ENVR-UNCER		GSCM		SOC-PER
	FL	HTMT	FL	HTMT	FL	HTMT	FL	HTMT	FL	HTMT
ECO-PER	0.887
ENVR-PER	0.122	0.139	0.851
ENVR-UNCER	0.132	0.375	0.310	0.924	0.914
GSCM	0.367	0.371	0.388	0.327	0.636	1.389	0.715
SOC-PER	0.295	0.556	0.201	0.498	0.517	1.620	0.603	0.806	0.881

shows that Fornell and Larcker criterion shows that each square root of the AVE of the construct is greater than its correlation with all other constructs. Additionally, HTMT values are lower than 0.900 except ENVR-UNCER, that have higher values (0.924–1.389–1.620).

However,

Table 7. Cross Loadings.

Indicators	ECO-PER	ENVR-PER	ENVR-UNCER	GSCM	SOC-PER
EcP1	0.895	0.107	0.212	0.360	0.285
EcP2	0.934	0.111	0.010	0.293	0.303
EcP3	0.844	0.147	−0.005	0.208	0.168
EnP1	0.052	0.742	0.233	0.123	0.032
EnP2	0.188	0.952	0.477	0.297	0.193
EnP3	0.051	0.847	0.282	0.285	0.242
Gr-DESIGN	0.285	0.179	0.602	0.767	0.486
Gr-ENVR	0.351	0.206	0.254	0.684	0.323
Gr-MAR	0.253	0:331	0.577	0.833	0.524
Gr-PROD	0.283	0.371	0.601	0.828	0.597
Gr-PURCH	0.249	0.134	0.461	0.777	0.455
Gr-RCYC	0.135	0.077	0.427	0.823	0.515
SoP2	0.380	0.199	0.521	0.475	0.859
SoP3	0.153	0.150	0.570	0.620	0.904
UN-SUPPLY	−0.020	0.310	0.677	0.408	0.405
UN-TEC	0.129	0.317	0.814	0.544	0.516

shows the cross-loadings of all the indicators and monitors that all the indicators of a particular construct load more strongly onto their own than others, which is a sign of discriminant validity [51]. Hence, discriminant validity for all constructs, including ENVR-UNCER, was established.

4.3. Structural Model Assessment

The final step is to assess the structural model to validate the proposed hypotheses, presented in Figure 3. Additionally, three moderator constructs were constituted. ECO-PER, SOC-PER, and ENVR-PER are defined as the dependent variable, ENVR-UNCER as the moderator variable, and GSCM as the independent variable. The findings presented in

Table 8. Direct Relationship Results.

Hypothesis	β	t Statistics	p Values	Result
H1: GSCM -> ENVR-PER	0.049	0.209	0.835	Insignificant
H2: GSCM -> ECO-PER	0.459	2.299	0.022	Significant
H3: GSCM -> SOC-PER	0.409	2.479	0.013	Significant
H4a: ENVR-UNCER -> GSCM & ENVR-PER	0.120	0.851	0.395	Insignificant
H4b: ENVR-UNCER -> GSCM & ECO-PER	−0.092	0.503	0.615	Insignificant
H4c: ENVR-UNCER -> GSCM & SOC-PER	−0.049	0.408	0.683	Insignificant

shows that the except for GSCM -> ENVR-PER (β = 0.049, t = 0.309, p = 0.835), two other GSCM practices hypothesis are accepted; GSCM -> ECO-PER (β = 0.459, t = 2.299, p = 0.022), and GSCM -> SOC-PER (β = 0.409, t = 2.479, p = 0.013).

The other three sub-hypotheses about the moderator effect of the environmental uncertainty were not found significant. They are ENVR-UNCER -> GSCM & ENVR-PER (β = 0.120, t = 0.851, p = 0.395), ENVR-UNCER -> GSCM & ECO-PER (β = −0.092, t = 0.503, p = 0.615), and ENVR-UNCER -> GSCM & SOC-PER (β = −0.049, t = 0.408, p = 0.683).

4.4. Common Method Bias

Kock [52] wrote that the occurrence of VIF greater than 3.3 is proposed as an indication of pathological collinearity and also as an indication that a model may be contaminated by common method bias. Therefore, if all VIFs resulting from a full collinearity test are equal to or lower than 3.3, the model can be considered free of common method bias. After creating a dummy variable using a random function in Excel and adding it as a latent dependent variable into the existing model, all other constructs became 50 independent variables, and the PLS algorithm was calculated. The inner VIF values in

Table 9. Inner VIF values.

	RAND
ECO-PER	0.892
ENVR-PER	0.134
ENVR-UNCER	0.108
GSCM	0.329
RAND
SOC-PER	0.344

are below the 3.3 thresholds, which states there is no common method bias in the model.

5. Discussion

GSCM practices have a positive impact on social and economic performance, whereas having no significant impact on environmental performance, although Turkish shipyards have a 95% ISO 14001 certification ratio. This incident demonstrates that certification implementations and GSCM practices mainly cater to social and economic concerns rather than environmental concerns. In addition, these certificates might be a penetration tool to the international market from the executives’ perspective. Environmental performance awareness is inadequate in Turkish shipyards because most producers’ priorities in developing countries like Turkey are to improve their economic situation and avoid economic risks [34]. A certificate is the first step for GSCM practices, but it does not mean the company achieved environmental performance positively. The certification is insufficient to justify that the companies are environmentally protected and approved using GSCM. The qualified certificate and standard only prove what they say to apply. Therefore, this incident is not unique to the shipbuilding industry, where other studies in further sectors have different outcomes. For example, Dizaji [31] stated that green marketing mix, green innovation, and green supply chain were significant, whereas green brand was insignificant for food industry performance. In the other study, Çankaya [36] found that green production, green packaging, green marketing, recycling, and internal environmental management positively affect environmental performance. However, green purchasing and environmental training do not have any effect. This incident might be a deficiency in environmental education. Although it contributes to how to do a job better, faster, and less costly in environmental issues, it is not enough to increase environmental performance. Turkish enterprises apply green education superficially, which is insufficient to improve environmental performance. Green et al. [40] found that green purchasing does not significantly impact environmental performance. In addition, Eltayeb et al. [41] studied three GSCM practices. Eco-design is the only practice that positively affects environmental outcomes, whereas green purchasing, green education, and reverse logistics do not have an effect.

The main difference between this study and other studies in the literature is that the hypotheses for GSCM practices converged in this analysis due to the sample size and model complexity. The result would be different if each GSCM practice were distinctly present in the structural model for hypotheses testing. Younis et al. [53] studied eco-design, green purchasing, environmental corporation, and reverse logistics as GSCM practices and their impact on corporate performance, defined as economic, operational, environmental, and social performance in UAE companies. The analysis showed an insignificant relationship between GSCM practices and environmental performance, although 48% of sample companies had ISO 14001 certification. Younis et al. [53] explained in this incident that companies in the UAE implemented GSCM practices without assessing which practices deliver the best outcome. The GSCM implementation, especially on external activities, is moderate except for internal environmental initiatives. GSCM practices affect firms’ financial/market performance and customer satisfaction but are not significant to environmental performance [54]. The result comparison between the current work and other studies in the literature is presented in

Table 10. Research Results.

Authors	Research Subject	Sector	Result
Existing study	Six GSCM practices’ (green production, green marketing, green purchasing, green design, recycling) effects on three sustainability performances (environmental, social, economic)	Turkish shipyards	GSCM significantly affect social and economic but not environmental performance
Yildiz Çankaya and Sezen [34]	Seven GSCM practices’ (green purchasing, green manufacturing, green packaging and distribution, green marketing, investment recovery, internal environmental management, and environmental education) effects on three sustainability performances (environmental, social, economic)	Automotive, electronics, and chemical sectors	5 of 7 GSCM practices effects positively environmental, 3 of 7 GSCM practices effects positively economic, 4 of 7 GSCM practices effects positively social performance
Dizaji [31]	Four green marketing strategies’ (green marketing mix, green supply chain, green innovation, green brand) effects on the food industry’s performance	Food sector	Except for green brands, all three of marketing strategies positively affect food sector performance
Green et al. [40]	Six GSCM practices’ (internal environmental management, green information system, green purchasing, cooperation with customers, eco-design, and investment recovery) effects on the environmental, operational, and organizational performance	Manufacturing managers	Green purchasing significantly affects economic performance but not environmental performance. Cooperation with customers affects environmental but not economic performance. Investment recovery affects environmental performance but not economic performance.
Eltayeb et al. [42]	Green Supply Chain Initiatives’ (Eco-Design, Green Purchasing, Reverse Logistics) effects on the environmental, economic, cost reduction, and intangible outcomes	Malaysian certified companies	Only eco-design significantly affects environmental, economic outcomes and intangible outcomes. Eco-design and reverse logistic effects positively cost reduction outcome.
Younis et al. [53]	Four GSCM practices’ (eco-design, green purchasing, environmental cooperation, and reverse logistics) effects on the four dimensions of corporate performance (operational performance, environmental performance, economic performance, and social performance)	all ISO 14001-certified and non-certified manufacturing firms in the UAE	None of the four GSCM practices were found to have any impact on the environmental performance, green purchasing, and environmental cooperation effects significantly on the operational performance. Only green purchasing effects significantly economic performance, and only reverse logistics practice effects significantly social performance.
Rahman et al. [54]	GSCM Practices’ (green purchasing, investment recovery, eco-design, reverse logistics, cooperation with the customer, and internal environmental practices) effect on environmental and firm performance	EMS ISO 14001-certified manufacturing firms registered as members of the Federation of Malaysian Manufacturers	GSCM practices significantly affect firm performance but are not significant for environmental performance.

.

Environmental uncertainty did neither have a moderating effect between GSCM practices nor sustainability performances. The moderator effect of environmental uncertainty was found only between eco-process innovation and sustainability dimensions. Yildiz Çankaya and Sezen [55] found that the moderator effect of environmental uncertainty was found only between eco-process innovation and sustainability dimensions. Therefore, moderator effects may be explored to comprehend better GSCM practices’ impact on the sustainability performance dimensions if each GSCM practice was analyzed separately. Environmental uncertainty reflects the sense of doubt experienced by managers when facing the problem of predicting future competitive conditions. The environmental uncertainty concept has drawn the attention of researchers who have mainly focused on how it affects corporate strategy. The relationships companies maintain with other economic actors to ensure access to resources, and the prevailing instability in such relationships account for the significance of environmental uncertainty in corporate strategy. The impact of uncertainty in strategic processes does not come about impersonally, nor does it emerge from complex dependence relationships, but instead through management perception, which has led to much research addressing management perception of uncertainty, highlighting this variable’s subjective nature. Despite this kind of uncertainty being described as objective, it should be remembered that it also derives from perception developed by an outside observer (whether a researcher, expert, or analyst). Therefore, the results of hypotheses 4a-4b-4c supported the concept of environmental uncertainty not being a moderator in shipyards once again. Santos Álvarez and García Merino [37] have also stated that it may therefore be concluded that in both cases, the evaluation of uncertainty depends on individual perception, which aims to mirror the difficulties individuals face when anticipating future conditions in a business environment.

6. Conclusions

This study highlighted GSCM practices presently adopted by Turkish shipyards and their impact on three sustainability performances; environmental, social, and economic. The GSCM practices “green design”, “green purchasing”, “green production”, “green marketing”, “environmental management”, and “recycling” were used in the model. The impact of GSCM practices was significant on social and economic performance. Nevertheless, the environmental performance was not significantly affected by these practices within the Turkish shipyards, even though 95% of the shipyards obtained ISO 14001 certification. This level reveals a current precarious situation; the respondents might see the GSCM practices as a prerequisite for the customers, a necessity of a certificate or classification institution, and an instrument for competitive advantage rather than to achieve some internal and external environmental objectives. Turkish shipyards focused on economic and social performance to sustain their presence in the highly competitive sector. Shipyards are interested in their corporate image and the social commitment to more resources to implement GSCM practices regarding market regulations and trends. The environmental concerns were a supplementary issue instead of a contributing factor. However, different outcomes of implementing GSCM practices in other sectors of the Turkish economy have been noted.

The environmental uncertainty did not have a moderating effect on the relationship between GSCM practices and sustainability performances. Since this moderating effect is not present, the strength of the relationship between GSCM practices and sustainability performance is assumed to be constant.

This study contributes to the development of GSCM theory by investigating the relationships between GSCM practices and corporate sustainability performances. The impact of the seven GSCM dimensions on the three dimensions of sustainability performance (environmental, economic, and social) was examined. This will enable managers to identify appropriate GSCM practices to strengthen the performance areas that need improvement. This study’s most outstanding contribution to the literature is that it is the first study focusing on the sustainability and GSCM issue in the Turkish shipbuilding sector.

The authors propose to apply the same model to the shipyard suppliers that contain a broad range of food, maintenance-repair, cabin, machinery, and spare parts companies. Since there are rare studies about the shipyards and sustainability, more studies are required to reveal the factors that impact the shipyards’ sustainability assessment.

7. Limitations

The study limited itself to the limited number of shipyards in Turkey. COVID-19 pandemic measurements contributed to using web-based surveys instead of face-to-face interviews. Time constraint was also one of the limitations of this study because the shipyards responded late. Therefore, data collection took a long time. The need for more funding is limited to hiring professional interviewers.