Green Entrepreneurship and Inclusive Innovation for Sustainable Creative Businesses: Evidence from the Borobudur Super-Priority Tourism Area

2.1 Green economy

The United Nations Environment Programme (UNEP) states that a green economy is a way of managing the economy that improves human well-being and social equity while reducing risks to the environment and the scarcity of its resources [3]. The green economy has three main principles low-carbon economic growth, efficient use of resources, and social justice. With this approach, economic growth is not only an important factor but also prioritizes sustainability for future generations. In the creative industry, the green economy is an important foundation that will enable businesses to combine creative innovation with environmentally friendly practices. Creative industries rooted in local wisdom and natural resources, such as handicrafts, local cuisine, and performing arts, can be developed using approaches that minimize waste, conserve energy, and support environmental preservation. This model also strengthens social connections among businesses, consumers, and surrounding communities through sustainability values. Creative businesses involving marginalized groups can create broader social impacts and strengthen business social networks [12, 13].

The implementation of a green economy requires synergy between technological innovation, changes in consumption patterns, and policies that support the transition to a sustainable economic system. In the creative industry, this can be achieved through the adoption of clean technology, the use of renewable raw materials, and production systems based on recycling or the circular economy. Technology enables businesses to improve efficiency, reach broader markets, and reduce their carbon footprint through the digitalization of production and distribution processes. Digital platforms such as e-commerce, social media, and business management applications have helped MSMEs survive the crisis and expand their tourism markets. High levels of technology access also contribute to the creation of innovations that are adaptive to environmental challenges. The success of the green economy in this sector heavily depends on the awareness of businesses, as well as support from the government, academia, and consumers [14].

In Indonesia, the concept of green economy is gaining ground in national development policies, particularly in the context of tourism and the creative economy. Priority tourist areas such as Borobudur are encouraged to develop green and inclusive local economic ecosystems. Through this approach, the creative industry not only generates economic value but also serves as a social force for cultural preservation, community empowerment, and sustainable environmental protection. Government support is crucial to the success of programs aimed at transforming the creative industry into a more sustainable one. Policies that are not contextually appropriate or overly administrative can reduce their effectiveness, making the involvement of local stakeholders in policy formulation highly important. Additionally, businesses must have strong networks with stakeholders. Forms of collaboration can include training programs, business incubation, promotion of local products, and integration of business activities into the tourism ecosystem. The better the quality of collaboration among actors, the greater the likelihood of creating an inclusive and sustainable creative business ecosystem [15]. A broad network supported by good managerial skills encourages creative business actors to develop business models that are sensitive to social values and ecological sustainability.

2.2 Green entrepreneurship

Schumpeter's Innovation Theory highlights the important role of entrepreneurship and innovation in driving economic development and organizational success. When it comes to SMEs, this theory only serves to further demonstrate that firms need to innovate by introducing either new products, processes or business models in order to outdo their competition. SMEs often grow by exploiting niche markets and enhancing their ability to innovate quickly, SMEs are one of the preferred and numerous form factor organizations all set to contribute towards creating significant increase in economic activities and jobs [16]. Entrepreneurship, in its basic definition, is the act of creating value by bringing ideas to fruition through the application of an opportunity with some type of innovation and hoping for monetery gain. Entrepreneurship is an attribute of a person that can recognize opportunity, encourage the risks of acting in dynamic environments through innovation and leadership as well as create value [17]. As a result of increasing international awareness in environmental and sustainability issues, this concept has evolved into green entrepreneurship, a form of entrepreneurship that is not only profit-oriented but also considers social and environmental aspects [11]. The approach is based on two central constructs namely green entrepreneurial intention (GEI) and green entrepreneurial orientation (GEO). They are equally important in informing us, as researchers who study the way how environmental values are internalized at the individual, and organizational level in the entrepreneurial process [18].

Specifically, based on the Theory of Planned Behavior [19], GEI refers to a person's intention to establish/expand an environmentally friendly business through the confirmation of attitudes, subjective norms, and perceived behavioral control. Recent longitudinal studies have shown that the stronger the components of the Theory of Planned Behavior, the more it will lead to a significant increase in changing green entrepreneurial behavior [20]. This study measures these intentions in terms of personal desire, anticipation of actual behavior, and readiness to act. Creative business people who have a lot of GEI, are more willing to make new environmentally friendly ideas. In addition, they are more likely to help local communities and create new ways to generate economic value that do not harm the environment [19, 20], which is part of environmentally friendly entrepreneurship, where good business is not only about making money, but also saving the earth, reducing carbon emissions, and treating everyone fairly. In this context, entrepreneurship must focus on issues such as global warming, business waste, or unfair treatment of people. The tendency to start a business is influenced by psychological factors such as beliefs, others' opinions, and how one perceives their ability to act. In the context of green entrepreneurship, this also means caring about sustainability [21]. GEI is affected by many factors as shown by many studies in different contexts: environmental education, personal experience with ecological issues; influence of pro-environmental communities; global exposure to climate change, social justice and environmental degradation [22, 23]. GEI can provide creative business opportunities in the Borobudur tourism area to start new ventures that are flexible, creative, and socially and environmentally beneficial. Business activities suitable for developing green business ideas include traditional food enterprises, local crafts, and community-based ecotourism. According to Rong et al. [8], these businesses are not only rooted in local culture and stories, but also help in the preservation of natural resources and promote the area as an all-round eco-friendly tourist destination.

GEO is a broader concept than the concept of Entrepreneurial Orientation (EO), but it focuses more on the environment. GEO is a way of managing or creating a business that is based on how much attention individuals or companies pay to new ideas, initiating new plans, and their willingness to take risks in environmentally friendly work. How well a company can develop new green innovations for products, ways of working, or jobs that reduce negative impacts on the environment is closely related to GEO. It also shows how ready entrepreneurs are to take risks when using new tools or doing other things that may not provide financial benefits, but have a great opportunity to do good things for society and the environment. The concept is to stop any work that can damage air, water, soil, climate, and make people want to use or buy sustainable products. In green entrepreneurship, GEOs help companies enter the green economy, becoming more competitive, gaining a better reputation with buyers, and creating more sustainable outcomes. Other research also states that GEO is closely related to GEI and can help the development of green business [14]. GEO is proven to assist companies in achieving a better level of sustainability in corporate financial and environmental management [24]. The meta-analysis study of Asad et al. [24] shows that GEO helps in the aspect of corporate social performance so that it can function better and produce green innovation in the company. In the studies of Asad et al. [24] and Anwar et al. [25], stated that GEO is an approach consisting of several components, including willingness to take risks, act proactively, and be innovative. The existence of GEO shows how creative businesses are willing to promote new ideas, be active, and take risks by considering the social, economic, and environmental aspects of the business. In the Borobudur tourism SPD area, this orientation is important to ensure that creative entrepreneurs do not only focus on short-term profits but also build business practices relevant to global issues such as climate change and the preservation of local culture.

2.3 Inclusive innovation ecosystem

As an inclusive innovation ecosystem, it seeks more technology access, plus stakeholder cooperation along with including marginalized sectors so that grassroots-based needs are understood in the most appropriate manners. Yousaf et al. [26] said technology helps enterprises to enhance productivity and widen markets when the production and distribution processes become digitized, so that carbon emissions are minimized. Almtiri et al. [27], which Digital platforms and Social media, along with business management applications have aided to protect creative businesses from being eradicated by the crisis, ended up extending tourism markets and helping broadly in maintaining sustainable innovation in MSMEs. The work of Vashkevich et al. [28] also supports this view as their study on inclusive tourism destinations demonstrates that the involvement of vulnerable groups, for example, disabled people can be a success thanks to sustainable multi-stakeholder collaboration facilitated by digital technology. Moreover, research findings suggest that improved economic-environmental goal alignment of green tourism programs are associated with coordination among “double helix” actors (public-private sector) which leads to increased social legitimation within the green tourism plan [29].

Involvement of marginalized communities, representing engagement by people and at vulnerable groups, such as women, differently abled, low-income, in the creative business value chain. But as more othering goes on in an increasingly diverse society, inclusion also becomes a business strategy to promote social sustainability and community legitimacy [30]. Managerial capacity describes the ability of business actors to plan, organize, direct, and control resources to achieve business objectives effectively and efficiently. As Heubeck [31] and Roy et al. [32] said, managerial capacity is the capability that business actors have to plan, organize, direct and control resources to achieve business objectives effectively and efficiently. This capacity is interpreted within the framework of green entrepreneurship as a sustainable strategy, inclusive human resources and environmental risks management, respectively. MSMEs internalize green principles via MSME opening up participation to vulnerable groups by the influence of Managers' competencies in planning, organizing and controlling. Quantitative evidence from 216 manufacturing MSMEs in Malaysia shows that dynamic managerial capabilities play a crucial role in mediating the influence of sustainable business practices on three-dimensional sustainability performance (economic, social, and environmental) [33]. Similar implications were found in the MSME strengthening index model at cultural heritage sites, where managerial skills mediate the relationship between green orientation and the adoption of resource efficiency practices [34].

2.4 Sustainability creative business (SCB), and government support (GS)

The performance of sustainable creative businesses in tourist areas is measured using economic indicators (profitability, job growth), environmental indicators (waste reduction, resource efficiency), and social indicators (inclusivity, cultural preservation). The positive relationship between GEO and sustainable creative business (SCB) has been confirmed through a cross-country meta-analysis [35], while the involvement of marginalized communities strengthens market legitimacy and product differentiation (handicrafts, cuisine) in historical tourist areas [36].

Government support plays a role in creating a business climate that supports green economy principles and inclusive innovation [37]. This support can take the form of regulations, fiscal incentives, provision of green infrastructure, facilitation of training, and access to markets and financing [38]. It is expected that different fiscal incentive schemes, green certification, and government technical assistance are helping the speed of this leverages towards the acceleration of green creative industries. Small green industries also benefit from the location of Borobudur tourist area with the program “Green Industry Concept” to improve energy efficiency and market access through digital marketing at all layers of MSMEs (initially increased by 18% within one year) [39]. But other research warns that innovation intentions could be dampened in the presence of too many bureaucratic regulations if it is not complemented with service ease.

2.5 Theory of Planned Behavior (TPB)

Theory of Planned Behavior (TPB) is a theoretical model proposed for explaining how human intentions are formed documenting that behavior following from these intentions can be predicted. The TPB suggests that an individual intention to engage in a behavior is based on three main components: attitude towards the behavior, subjective norms, and perceived behavioral control. All three of these operate at the same time to influence whether an individual acts or does not act in a given situation [11, 40]. Recently, TPB has been used in the green entrepreneurship literature to study the intention of business actors to engage with sustainable (environmentally friendly) practices. Environmental sustainability attitude are the dominant key factor for increasing willingness among business actors to perform green innovation, where subjective norms such as community social pressure or demand from consumers turning into an individual postulated construct at performing and actualizing adherence in behaving based on sustainability principles. Meanwhile, perceived behavioral control: the power of an individual to execute their intentions in practice as a consequence of capacity and resources [11, 40].

The appropriateness of the TPB in how green entrepreneurship behavior can be explained is well documented by some extant studies. Which means that TPB also has potential to predict entrepreneurial intention in situations of sustainability-driven sectors, especially MSMEs actors in developing countries. In the creative industry, the TPB model is very useful for understanding the motivations of business actors in implementing environmentally friendly production, recycling-based innovation, and the adoption of green technology [1]. Implications TPB provides a strong theoretical foundation for designing interventions or behavior-change policies. Effective green entrepreneurship training must foster positive attitudes, strengthen supportive social norms, and enhance entrepreneurs' confidence in their ability to manage sustainable businesses. By understanding the psychological and social factors that shape such intentions, policymakers and facilitators can develop more targeted and impactful strategies for the development of the green creative industry [16, 41].

Methods- Sampling Frame, Inclucion Criteria, and Procedure.

Sampling Frame.

The study targeted 120 creative enterprises located within the Borobudur Super-Priority Destination (crafts, culinary, fashion, and creative services). Inclusion Criteria (purposive). (i) operating within the DSP Borobudur area; (ii) having current or planned green practices (e.g., waste management, eco-friendly materials, energy efficiency); (iii) at least one year of operation; (iv) respondent is the owner or key decision-maker. Pre-survey ‘Green’ Screening. We applied three screener items capturing minimum green practice. Units that did not meet the threshold were not invited to the main survey. The percentage of screened-out non-green units was not documented quantitatively at the time of data collection. To enhance transparency, we provide the sampling flow in Table A1 and explicitly report the final analytical sample (n = 65, ≈54% of the 120-unit frame).

Sample Adequacy for PLS-SEM (ISR/GE Summary)

We evaluated the adequacy of n = 65 using the inverse square root and gamma–exponential approaches based on the smallest absolute path coefficient (α = 0.05; power = 0.80). Effects that are near-zero would require large N; our realized n = 65 is sufficient for the small-to-moderate focal effects that were significant in the bootstrap. Detailed computations are provided in Appendix A-1. Although the final response rate was moderate, the achieved sample met both statistical power and theoretical representativeness requirements for PLS-SEM analyses.

The green economy is a new paradigm in development that balances economic growth with environmental preservation and social justice. This study examines the implementation of inclusive innovation ecosystems, the role of green entrepreneurs, and government commitment to green creative enterprises in the Borobudur tourist SPD area. The study is quantitative in nature with a cross-sectional approach. The testing framework uses Structural Equation Modeling–Partial Least Squares (SEM-PLS) because: (a) the model is predictive-exploratory, (b) the sample size is limited, and (c) the constructs contain reflective and formative indicators [65]. The analysis was conducted using Smart PLS 4.0.

The population includes all green economy-based creative industry players in the Borobudur tourist SPD area (120 business units, source: Tourism Office 2024). The sampling technique used purposive sampling with the following inclusion criteria: (a) businesses that have been active for at least two years; (b) produce products/services based on local cultural heritage; (c). Implementing at least one environmentally friendly business practice (such as sustainable materials or waste management). The next criterion is: (a). The 10 × indicator rule-the largest construct (GEO) has six indicators, so ≥ 60 observations are required [65]; (b). Power analysis via GPower 3.1 (α = 0.05; power = 0.80; f² = 0.15; five main predictors) recommends ≥ 55 samples [66].

Using purposive sampling and predetermined sampling criteria, a sample of 65 sustainable creative entrepreneurs was obtained. The functional relationship between exogenous variables and endogenous variables is as follows: exogenous variables include: (a) green entrepreneurship (GEO, GEI, managerial capacity), (b) inclusive innovation ecosystem (access to technology-digitalization, involvement of marginalized communities, stakeholder collaboration), and (c) GPS. The endogenous variable is SCB in the Borobudur tourism SPD area. To address the research objectives and hypotheses, a quantitative approach (SEM-PLS) was used. The empirical research model is shown in Figure 1 and Figure 2.

1.png

Figure 1. Sampling flow

2.png

Figure 2. Empirical model of research

6.1 Convergent validity

Convergent validity measures the extent to which indicators of a construct are highly correlated with each other and consistently represent the latent variable being measured. One of the main criteria for assessing convergent validity is the loading factor value, where an indicator is considered valid if it has a loading factor > 0.50 [67]. The loading factor data is shown in Table 2 (Appendix Table A1).

Based on the results of the convergent validity analysis, all loadings were ≥ 0.70 except for KMP1 (0.589) and PBIK5 (0.655). Both indicators were still above the minimum exploratory threshold of 0.50. Therefore, all constructs in the model met the criteria for convergent validity. These two indicators were retained because their factor loadings were still within the acceptable tolerance range for exploratory research. According to Hair et al. [67], factor loadings above 0.50 can be considered adequate when supported by strong theoretical justification and acceptable overall model validity. Moreover, both KMP1 and PBIK5 demonstrated satisfactory cross-loading values, indicating that each indicator loaded higher on its respective construct than on others. Thus, their inclusion was maintained to preserve the comprehensiveness and theoretical consistency of the measurement model.

6.2 Common method bias (CMB) diagnostics

Harman’s single-factor (unrotated) did not indicate a dominant common factor; the Common Latent Factor (CLF) test produced average Δ-loading < 0.20; the marker-variable correction did not alter the significance of focal paths; and construct-level collinearity is acceptable with all inner VIF values < 3.0 (see Table 3).

We minimized CMB by ensuring anonymity, clarifying there were no right/wrong answers, randomizing items across constructs, and using neutral wording. An unrotated EFA across all indicators showed that no single factor accounted for a majority of variance (Table 2).

Table 2. Validity test results

Adding a common latent factor to the measurement model yielded an average indicator loading difference < 0.20 (Table 2), indicating negligible CMB. Marker Variable Approach. Using a theoretically unrelated marker to adjust correlations left the significance of focal paths unchanged (Table 3). All full collinearity VIF values per latent construct were below 3.3 (Table 3). Together, these diagnostics suggest CMB is not material in our data.

Table 3. Construct-level collinearty (Inner VIF)

6.3 Construct reliability and validity

Construct validity can be seen from the Average Variance Extracted (AVE) value, with a minimum limit of 0.50, which indicates that the latent variable is able to explain more than 50% of the variance of its indicators. Construct reliability is evaluated using three main measures: Cronbach’s Alpha, rho_A, and Composite Reliability, all of which must be above 0.70 to meet the criteria for internal reliability [68-70] (Table 4).

Table 4. Construct validity and reliability

Based on the test results, both construct validity and construct reliability have been fulfilled, so that the instruments used in this model are valid and reliable [68-70].

6.4 Discriminant validity

GEO Specification (Reflective) and Robustness. We modeled GEO as reflective because indicators (innovativeness, proactiveness, risk-taking) are manifestations of the underlying strategic posture; changes in the latent trait should affect all indicators in the same direction and indicators are interchangeable. Support comes from high outer loadings, CR/ρA/α above thresholds, AVE > 0.50, and outer VIF < 5. As a robustness check, we also estimated a second-order reflective–reflective GEO (INNO, PROAC, RISK) and found that key structural conclusions remain unchanged (Table 4 and Appendix Table A2).

Discriminant Validity (GEI–GEO). HTMT_GE I–GEO = 0.983 (> 0.90) reflects close conceptual relatedness (intention vs. orientation) among MSMEs where intention-to-practice translation is rapid. We retain both constructs due to their non-redundant roles (GEI as intention; GEO as strategic posture) and provide robustness via the second-order GEO, cross-loading inspection, and VIF checks; we acknowledge this as a limitation and suggest more distinctive indicators or longitudinal designs (Appendix Table A3).

MI Indicator Allocation. We removed item KMP1 (managerial) from MI, as it reflects managerial capacity rather than inclusion. Retained MI indicators focus on access to information/markets, network participation, and benefit sharing. Post-adjustment reliability and validity remain adequate (see revised Table 2 and Appendix Table A2).

According to Henseler et al. [71], a good Heterotrait-Monotrait Ratio of Correlations (HTMT) value should be below 0.90 to ensure that the constructs in the model do not overlap and measure different concepts (Table 5 and Appendix Table A4).

Based on the HTMT test results in Table 5, all construct pairs met the recommended threshold of < 0.90 for discriminant validity, except for the GEI and GEO pairs, which reached 0.983. This value can be explained by the high conceptual fit between “intent” and “orientation” toward green practices in creative business activities at Borobudur. Previous studies have confirmed that, at the micro and small scales, GEIs are often directly manifested in operational behavior without a long time lag, leading respondents to view the two as a single entity [72, 73]. A meta-analysis by Schlaegel and Koenig [74], also shows a high correlation-even above 0.70-between entrepreneurial intention and subsequent behavior, particularly in the context of sustainability-oriented entrepreneurship. Thus, the high GEI–GEO correlation reflects the reality in the field rather than a weakness of the instrument.

6.5 Inner model analysis

The R Square (R²) value is used to measure the predictive ability of the structural model in explaining the variance of the dependent (endogenous) variable. According to Chin [75], an R² value of ≥ 0.67 is considered strong, around 0.33 - < 0.67 is considered moderate, and < 0.33 is considered weak.

Table 5. Heterotrait-monotrait ratio (HTMT)

Table 6. R² values

The results of the analysis in Table 6, show that the model has good predictability for endogenous variables. This research model is also able to estimate and explain the sustainability performance of creative MSMEs in the Borobudur Tourism SPD area. This is evidenced by the values of 0.635 (GEO), 0.549 (MI), and 0.543 (SCB) in Stone-Geisser Q2 obtained through the benchmark test. All these Q² values exceed the high predictive relevance threshold of 0.35 [67]. Through PLS-Predict, it shows positive Q²predict values and smaller RMSE-PLS than the reference linear model, so out-of sample data can be predicted in this model (Table 7) [76].

6.6 Path coefficient analysis

Interpretation of research results is important in determining the relative contribution of each construct in explaining the phenomenon under study and evaluating hypotheses. Through path coefficient analysis, the direction and strength of the relationship between the independent and dependent variables in the study can be determined. In addition, a significant effect is obtained through the T statistical significance test and the p value, provided that the p value must be less than 0.05 (Figure 3).

The estimation results show that most of the paths in the model have a significant effect on the dependent variable. The path GEI (X5) → GEO (Z2) has the strongest and most significant influence with a coefficient of 0.829 (p < 0.001), followed by MI (Z1) → SCB (Y) at 0.506, and MC (X4) → MI (Z1) at 0.589. Meanwhile, the effect of AT (X1) on SCB (Y) is significant with a coefficient of 0.211 (p = 0.042). However, there are two statistically insignificant relationships, namely between GPS (X3) and SC (X2) on SCB (Y) with p-values of 0.590 and 0.900, respectively.

These findings indicate that most constructs contribute to the model, but not all relationships statistically provide a significant direct effect. The Q² values obtained from blindfolding confirm substantial predictive power for all three endogenous constructs (Q² = 0.308–0.486). The PLSpredict evaluation shows that all indicators have a lower RMSE_PLS than the linear model, indicating strong out-of-sample predictive power. Effect-size analysis (f²) reveals the greatest influence is on the GEI → GEO path (f² = 0.87; large) and MI → SCB (f² = 0.38; large), while AT → SCB is only small–medium (f² = 0.05) (Appendix Table A5).

3.png

Figure 3. Results of path coefficient analysis

Table 7. Path coefficients

These findings also confirm the dominance of cognitive-strategic mechanisms and social inclusion over mere access to technology. Bootstrap mediation tests show that GEI enhances SCB through two channels: (a) the social channel GEI → MI → SCB (β = 0.180; p < 0.001); and (b) the strategic channel GEI → GEO → SCB (β = 0.296; p < 0.001). Since the direct path GEI → SCB is not significant, these two paths indicate full-partial mediation that validates the “green-inclusive entrepreneurship” framework (Appendix Table A6).

The f² analysis reveals varying levels of effect size among the studied constructs. Green Entrepreneurial Intention (X5) shows a very large effect (f² = 0.829) on Green Entrepreneurial Orientation (Z2), indicating that entrepreneurs’ green intentions strongly influence their orientation toward sustainable practices. Similarly, Managerial Capacity and Training (X4) and Marginalized Inclusion (Z1) exert large effects (f² = 0.589 and 0.506, respectively) on Sustainable Development in the Creative Industry (Y), highlighting the critical roles of managerial competence and inclusive participation. Green Entrepreneurial Orientation (Z2) also demonstrates a large effect (f² = 0.357) on sustainability outcomes, while Access to Technology and Digitalization (X1) contributes a moderate effect (f² = 0.211). Conversely, Government Policy Support (X3) and Stakeholder Collaboration (X2) show minimal or negative effects (f² = -0.055 and -0.016), suggesting that policy frameworks and partnerships have yet to translate into substantial sustainability impacts within the creative industry context (Appendix Table A7).

This study confirms that GEI through GEO and MI are the main determinants of SCB performance in MSMEs, especially creative businesses in the Borobudur tourism SPD area. Access to digital technology (AT) and managerial capacity (MC) act as additional levers. Meanwhile, stakeholder collaboration (SC) and government support (GS) are still hampered by ceremonial governance.

Theoretically, this research integrates the Theory of Planned Behavior (TPB) and the inclusive innovation framework, introducing the social-ecological dual mediation mechanism and highlighting MC as a lever for inclusion, expanding the repertoire of green entrepreneurship in emerging tourism destinations. Practically, the findings emphasize the need for: (i) digital and managerial upskilling programs that transform green intentions into operational SOPs, and (ii) concise performance-based green certification to replace ceremonial collaboration initiatives.

The practical implication is that MSMEs players need to establish key performance indicator-based green SOPs and systematically involve vulnerable groups. The government should transform environmental certification into a results-based one-stop service and link fiscal incentives to waste reduction achievements. Digital platforms should provide IoT-based carbon footprint dashboards to make green performance measurable and marketable.

Future research should test the moderation of policy quality and financial capability, and assess the effectiveness of digital green voucher schemes in accelerating green technology adoption. Thus, the green-inclusive entrepreneurship agenda can move from the conceptual realm to measurable policy protocols at the national level.