The role of leadership in business model innovation: a case of an entrepreneurial firm from India


Purpose
The purpose of this paper is to explore the relationship between leadership and business model innovation (BMI) in an entrepreneurial firm. From the literature, it was found that the role of a leader in BMI was unexplored. A research framework was created which was the replication of the model created showing the relationship between leadership and innovation.


Design/methodology/approach
The qualitative single in-depth case study was used to understand the effects of leadership in BMI. The case of an entrepreneurial firm in the graphic and animation education sector from India was chosen to test the research framework. The leader of Xplora Design Skools was observed closely, and he was interviewed multiple times.


Findings
From the analysis, it was clear that, in this organization, the leader was a trigger for BMI through creating and influencing creativity and innovation in the organization. This case also shows that he was making tangible contribution to the work being done and motivating his employees. These initiatives show his influence on the process or execution of BMI.


Originality/value
This is the first study explores the role of a leader in BMI in an entrepreneurial firm in emerging economy contexts like India.



Introduction
Business model innovation (BMI) has been extensively studied by management researchers and practitioners over the last two decades. BMI is an extension of the business model (BM) but includes questions and concepts that are beyond the traditional BM literature. Though many articles covered BM comparatively, very few articles focusing on BMI have been published in peer reviewed journals (Foss and Saebi, 2016). Mostly researchers (Achtenhagen et al., 2013;Aspara et al., 2013;Deshler and Smith, 2011;Doz and Kosonen, 2010;Enkel and Mezger, 2013;Frankenberger et al., 2013;Günzel and Holm, 2013;Khanagha et al., 2014;Moingeon and Lehmann-Ortega, 2010;Mezger, 2014;Pynnonen et al., 2012;Sosna et al., 2010) focused on BMI as a process and attempted to understand the importance of capabilities, leadership and learning mechanism (Foss and Saebi, 2016). The recent review paper of Foss and Saebi (2016) mentions that only the paper authored by Achtenhagen et al. (2013) out of the 150 peer reviewed papers focuses on leadership in BMI, and only 2 papers were found in the reference list which discussed leadership. Out of these two, one discusses how BMI creates leadership for firms rather than about the role of a leader in the BMI process.
In our literature review, we found only seven studies that talked about the role of a leader or a CEO in BMI. So, it is clearly evident that researchers have not paid attention to the role of a leadership in BMI, and it is an under-researched area. Foss and Saebi (2016) argued the importance of studying what facilitates and hinders BMI in entrepreneurial firms in the review paper on the rise of entrepreneurship and BMI literature. It is also observed that very few studies focus on emerging economy contexts, and it requires the attention of researchers to study emerging economies (Lee et al., 2012). This paper focuses on understanding the role of leadership in BMI. Using the framework created from innovation, the strategy literature and a single case study of an entrepreneurial venture in the multimedia education sector from an emerging economy, India, the paper explores the following questions: RQ1. How does a leader or a CEO affect the BMI?
RQ2. How does a leader or a CEO contribute in initialization and continuation of BMI?
In this research, a deductive approach of case-study method was selected and the entrepreneurial multimedia venture from India and its new BMI in design education were studied. The founder who is a CEO of a firm was also observed closely. The findings from the case give interesting insights. The case study explains the role of a leader in starting or initiating BMI in an entrepreneurial firm and the leader's very high involvement in execution and continuation of developing BMI. The paper has made multiple contributions, first, it focuses on the under research stream of process of BMI as suggested by Foss and Saebi (2016). Theoretically, for the first time, it explains how, where and why a leader gets involved in the BMI process. Empirically, it explains, using a single case of an entrepreneurial firm in the emerging market of India, the role of leadership in BMI.
The rest of the paper is organized as follows. The first section is literature review on BM and BMI, the second section is on developing a research framework, the third section is on research methodology, the fourth section is on analysis and findings, the fifth section is on discussion and implication and the last section is limitations, future research and conclusion.

Literature review
Business model Every company has a BM, whether they articulate it or not. According to Mitchell and Coles (2004, p. 15), "A business model is the combination of who, what, when, how and how much an organization uses to provide its goods and services and develop resources to continue its efforts." Santos et al. (2009, p. 11) defined BM as a "configuration of activities of the organizational units that perform those activities both within and outside the firm designed to create value in the production" (and delivery of a specific product/market set). Chesbrough (2007) identified two important functions of a BM: value creation and value capture. Value creation includes a series of activities, from procuring raw materials to satisfying the final consumer, which creates a net value in terms of new product and service. Value capture includes a series of activities for development and operations. Chesbrough (2007) listed six different functions of a BM: articulating of the value proposition; identifying a useful market segment; defining the structure of the value chain and determining the complementary assets; specifying the revenue generation mechanism; positioning the firm within the value network; and formulating the competitive advantage. Hwang and Christensen (2008) categorized BMs into three types: solution shops, value-adding process businesses and facilitated user networks. Solution shops like law firms, advertising agencies, consulting firms and R&D organizations diagnose and solve unstructured problems. Value-adding process businesses transform inputs of resources into outputs for creating a value. User networks have the same people buy and sell and deliver and receive things to and from each other. Santos et al. (2009) identified four interrelated components in a BM: activities, organizational units, relational and

71
Role of leadership in business model innovation transactional linkages and governance mechanisms. They further stated that these four components help organizations create value proposition or create a product and service for customers, so they can use them effectively, conveniently and affordably. Johnson et al. (2008) explained a framework which is called a Four-Box Model that focuses on customer value proposition, the profit formula, key processes and key resources. The company arranges and allocates tangible and intangible resources and creates value proposition for customers. The company also performs processes for creating value proposition. In turn, the company generates a profit formula. He divided this profit formula into four buckets: revenue model, cost structure, margin model and resource velocity. Sekhar Tankhiwale (2009) on similar lines described a BM which creates an organization's value proposition to various markets and customers. It also considers capabilities, partner network (supply chain) and distribution channels required for creating, marketing and delivering this value. A BM not only considers revenue generation mechanism but also takes care of managing cost structure effectively. He identified nine components of a BMcore capabilities, partner network, value configuration, value proposition, cost structure, customer relationship, distribution channel, target customers and revenue stream. Core capabilities, partner network and value configuration are supply side components which affect cost structure and value proposition. Demand side components are customer relationship, distribution channel and target customers which affect value proposition as well as revenue stream of firms. According to Coleman (2009), new BM requires two capabilities, first developing innovating structures and processes and, second, creating workplace cultures for fostering the capability to build effective relationships. Hence, it is clear that many researchers define BM differently and explain different dimensions and aspects of BM connecting with different theoretical lenses.

Business model innovation
According to Mitchell and Coles (2004), BMI is a novel replacement of the existing BM that provides product or service offerings to customers and end users that were not previously available. Marcides (1997Marcides ( , 1998Marcides ( , 2006 defined BMI as the redefinition of existing product, service or process and the process of providing and marketing to customers. On similar lines, Santos et al. (2009) considered BMI a reconfiguration of activities in the existing BM of a firm that is new to the product/service market in which the firm competes. In Coleman (2009), a BM tries to create a value to the current and potential customers by combining activities undertaken by the organization, its partners and suppliers (Davis and Spekman, 2004). Mitchell and Coles (2004) stated that BMI often finds ways to provide new benefits at lower costs, enhancing circumstances for themselves while providing more for their customers. It creates these benefits through adding complementary products and services that build on what you already provide, learn from best practice experiences and share the lessons with customers, combine solutions to help customers grow the market faster, adjusting prices to encourage more purchases, fixed price for unlimited entry, prices that lower operating costs, and cost reductions. These actions also stimulate growth, help lower costs and improve service, help customers find their own lower cost improvements.
According to Santos et al. (2009), BMI is not discoveries and considering that it depends on technological innovation is a mistake. It is also not a new product or market introduction. BMI is a new strategic path for a Free Standing Business (BFS) as well as a unit residing multi-business corporation (BUC). The only cost of BMI is the cost of organizational change, not investing in new product/service, R&D, technology acquisition, etc. So, BMI is less expensive and less risky.
They further stated that BMI involves four forms of reconfiguration activities: relinking, repartitioning, relocating and reactivating. Relinking is an alteration in the connections between organizational units currently performing activities. Repartitioning is an alteration in the physical, cultural and institutional boundaries of the organizational units currently 72 NEJE 22,2 in innovation is not only motivating creative people but creating an environment to foster intrinsic motivation among them. The individual characteristics, such as cognitive complexity, self-efficacy and self-monitoring may allow a single leader to effectively execute the multiple-competing roles for creative ventures. Sternberg et al. (2003) argued that the leaders' motivation to innovate and the kind of innovations they are willing to pursue depend on strategic choices made by leaders based on their perceptions of environmental risks and opportunities. However, Geletkanycz and Hambrick (1997) found that outside sources of knowledge are often critical to the innovation process and which can be leveraged through the external ties of top executives. These ties could be intra-industry as well as extra-industry which not only affect organization for strategic choices and performance but also foster innovation. Incentives and bonus also influence innovation and invention activities in the organization (Makri et al., 2006).
From the above arguments and literature review, a research framework or a model was created that shows factors related to leadership and their relation with the innovation in organization. As shown below, the framework includes leader's individual characteristics, technical and professional expertise, strategic choices made by leaders, creative skills, external ties, and incentives and bonus, all factors considered to affect leadership for innovation in organization ( Figure 1).
Role of leadership business model innovation. Different researchers define leadership differently, and for this study, we taken a broad definition of leadership. The leadership is a force which leads the organization, which could be an individual, an owner of the organization, the CEO or TMT. Chesbrough (2007, p.16) stated that, "in my observations, many organizations have a business model innovation leadership gap." He believed that BM innovation requires involvement of top leadership, not only of chief technology officer and others. Chesbrough (2007) stated that the role of CEO is very critical for creating a value for customers. Different functional managers and officers have different functional and departmental responsibilities, same is the situation with general managers and division presidents as they need to take care of financial performance of a business unit, but no one except the CEO or a leader can drive the entire business and change or reconfigure BM (Chesbrough, 2007). Chesbrough (2007) explained that top managers mostly work within the current BM. They work by exploiting the advantage of the current and established BM and are not comfortable changing the existing BM because changing and bringing new BM requires more efforts and more data to justify its consideration.  (Mumford and Licuanam, 2004;Geletkanycz and Hambrick;Makri et al., 2006 Motivating people and group

Influence creativity and Innovation and Idea Generation
Make tangible contribution to the work being done Evaluation of creative ideas of followers Role of leadership in business model innovation Coleman (2009) found the importance of Australian businesses as part of their wider innovation strategies, building their capacity to enter into successful relationships with organizations in China, India and other emerging economic powers in Asia, a commitment to both education and providing personnel with the practical experience of developing relationships in Asian nations. In this kind of set up, the role of a leader or CEO becomes very important. According to Santos et al. (2009), BUC managers rarely have the autonomy to alter their existing BMs without the involvement, or even perhaps the permission, of the corporate center. When the corporation has created tight operating linkages among the various units, each individual BUC becomes further constrained, which renders BMI more difficult, as innovations in the BM of one BUC affect the others and the dynamics of BMI within both the BFS and the BUC. Mitchell and Coles (2003) also showed importance of the role of a CEO's leadership not only as a source of BMI concepts to test but continuing BMI. They further stated that the CEO needs to rely on establishing a constantly improving process to generate and use ideas from all stakeholders to establish better BM. The CEO must establish an unchanging core vision for serving customers and other stakeholders that includes an expectation of regular BM changes for continuing BMI (Mitchell and Coles, 2004, p. 20). Mitchell and Coles (2004) further stated that the CEO's attention and interest are critical to turning BMI into the primary source of the company's future success. CEOs bring openness which helps a company eliminate many elements of existing BMs in a company. Koen et al. (2011) explained that BMI could be either achieved by completely separate organizations or ambidextrous organizations which require senior management leadership training. According to Santos et al. (2009), BMI knowledge in terms of tacit knowledge of operations required to manage BMs and improve those rests within the business units. It is important to have the right kind of horizontal coupling between business units and the corporate offices for BMI. The role of CEO or corporate executives is very important for supporting linkages as well as encouraging the development, proposing and sharing of innovative BMs through mutual engagement between corporate executives and BUC managers. As mentioned earlier, Desai and Mahadik (2011) showed relationship of choice between BMI and differentiator strategy and economic performance of firms in a mature industry. The choice of an entrepreneur or a leader decides the competitive performance of a firm in a mature industry. Achtenhagen et al. (2013) found that to manage BM and assure value creation, one of the critical capabilities is achieving coherence among leadership, organizational culture and employee commitment. The study also showed the importance of active and clear leadership.
From the literature, it is clear that the role of a CEO or a leader is very critical for initiation of BMI as well as for the continuation of BMI. But how a leader impacts BMI is not clear. This has also not been tested empirically. This relationship has not been explored in an emerging economy context as well. It is not clear how a leader or a CEO impacts BMI in entrepreneurial firms so this study tries to explore role of a leader in BMI in entrepreneurial firms in an emerging economy context. More specifically, it focuses on the following research questions: RQ1. How does a leader or a CEO affect the BMI in an entrepreneurial firm of emerging economy?
RQ2. How does a leader or a CEO contribute in the initialization and continuation of BMI in entrepreneurial firm of emerging economy?
To analyze and answer the above questions, it is assumed that the research framework created to understand the role of a leader in innovation can be used. As shown in the model, the leader's influence on creativity, innovation and idea generation is for triggering BMI and making tangible contribution to the work being done; evaluating creative ideas of followers; and motivating people and groups for execution and continuation of BMI and processes related to BMI (Figure 2).

Research methodology
The study of strategic choices in entrepreneurial firms is complex where multiple subjective realities coexist. Such ontological context suggests the adoption of qualitative research. Epistemologically, researchers need to observe the phenomena to understand the dynamics of choices, suggesting adopting qualitative research route through the case method. Qualitative research is particularly strong in exploring relatively unknown phenomena in real-life context (Eisenhardt, 1989). Using this methodology enables the researcher to inductively explore and identify concepts, noticeable similarities, trends and patterns of socio-economic phenomena (Yin, 2009;Eisenhardt, 1989). The case-study approach is also used when the study explores questions like How and What (Yin, 2009). Hence, the author has chosen to adopt the case-study route for this study. The choice of the case was based on aligning the definition of BMI. After finding some news articles and talking to some industry experts, a list of potential cases was created. The selection of entrepreneurial firm was based on research gap identified in the BMI literature. Based on the approval of doing the research on the firm and availability and willingness of leader of the firm, the Company X was finalized as the case to understand the role of leadership in BMI. The Company X was selected because it could create a unique value proposition to its customers (students) through its unique model focusing on vocational training related to graphics and animation to the youth of tier 2 and tier 3 cities in India.
The organization presented in this paper presents rich insight into BMI and how BMI is influenced by characteristics of a leader, his influence on the initialization and continuation of BMI. The primary data collection was through in-depth interviews of the founder and his team, observations during author's visits. Secondary sources like company's website, training manuals, brochures and other materials provided by the company were also used. Mr X, the owner, the CEO and the leader of this organization was observed and studied.

Role of leadership in business model innovation
In the study, five unstructured interviews were conducted with the owner and leader of the organization. The author visited the leader in his organization to observe him closely. Five interviews with his partners and employees were conducted. Five others were also interviewed. The author spent one week at the office as an observer. In the process, he could talk to students informally. Table I shows details of interviewees. The statements of all the interviewees were cross-validated. For this purpose, some respondents were interviewed twice. Data were also collected from both internal and external published sources. Internal sources consisted of internal financial and non-financial reports, process documents and operation manuals of the firms.
A summary of case study Company X Design University Idea came through the idea of diversification; Company X (COMPANY X) was into vocational education after 12th standard and graduation for students in Ahmedabad, India. Mr X, the founder and the CEO of Company X had many friends in the top institutions of the city, like the INSTITUTE N (a premier Design School) (N) in India. The idea of starting a Design University emerged when he was a jury member evaluating projects of students in N. He was also teaching some courses in N. As he had built a very good relationship with N, the idea of a university emerged herein. His friend and colleague in N, Mr K suggested him to grow the boundaries of COMPANY X and start a university. Another factor responsible for starting the university was to utilize the current strengths and to put them into the University. The constant interaction and long-term association with premier Design School (N), the communication and advertising school (MC) and management institute (M) helped him in idea generation. Mr X came up with an idea which was completely unique and was never thought about by his competitors. Additionally, the Design University was completely different from Company X. COMPANY X as a BM changed the way multimedia education was perceived and delivered. COMPANY X focused on students coming from tier 2 and tier 3 cities, and it was fulfilling the design and multimedia requirements of SMEs and filling the gap through the vocational education model as a tool of employment in small towns and cities of the country. COMPANY X was using technology to give education in design, whereas the university was aiming to develop design, skills, attitude and thinking among kids and children without using computers in schools. It was basically design as design process. So, the variations in both these models were that one was completely technology driven, whereas the other was without technology. The Design University made these children innovators, creators

Data analysis
The study follows a case-study analysis to test and validate the research framework developed from the literature review. Data analysis in this study follows two steps: deductive positivist case research (Shanks, 2002) and template analysis (King, 2012). Transcripts of all the interviews, field notes, observation notes and secondary data were coded by the author. The coding was done based on themes of the research framework, and template was created from the literature review for the case analysis. Later, it was coded by one of the experts of case-study research for external coder reliability. Similarly, a subject expert of BMI was also given the transcripts, observation notes and material for the coding and coding was matched to take care of internal coder reliability. The analysis was not only based on interview transcripts but other sources of data like secondary data, observation and field notes to ensure that triangulation was managed effectively. We focus on where we are good at. I do not interfere in Mr Y's or Mr Z's work. We discuss our work but give freedom and space to take our own decisions. (Mr X) COMPANY X started with high-end program in animation, web and graphics with job assurance for students. Mr X the founder from the beginning, focused on business development and bringing innovative ideas to the organization. His partners had different expertise and that helped them all to focus on issues related to their specific areas. This is why the firm benefited and performed better than its competitors. Mr X encouraged new ideas and passed them on to others in the organization. These new ideas created new avenues of growth. While Mr Z and Mr Y were into operations, Mr X was into idea generations, creating and maintaining relationships with institutes like N and MA: In our first exhibition, we decided to do something completely different from our competitors. I got an idea to have a completely dark stall with only five visible monitors. On these monitors, we displayed the latest content from the multimedia world along with the content designed and developed by us. The idea clicked and we got an overwhelming response. We could get enrollment for our first center from this exhibition. (Mr X) Mr X was also open to new ideas and tried exploring out of the box ideas. Knowing that such ideas will take him to success, he did not hesitate to execute new thoughts and ideas into practice: Mr X always looked for out of the box ideas. When we initially discussed the idea of starting a design university, he immediately showed interest and enthusiasm to pursue the idea. (A faculty of N)

Role of leadership in business model innovation
It was also observed that Mr X and his partners challenged norms and tried to do something different from others. When they started their business, there were two big players in the market. One was from funded by the top player in computer education in India. Another was, a well-known brand from Southern India was the second big player. However, the model COMPANY X followed was low cost, quality multimedia education with latest content and job assurance. In the late 1990s and early 2000s, animation was not popular but web designing was a very popular field. The core team of COMPANY X went to small-and medium-scale companies and asked what kind of people they needed, how many people they needed, what were the skills required, etc. It was a reverse engineering model. The curriculum was designed accordingly and based on the requirements of these companies. They tied up with 13 companies initially and promised them to provide skilled people as per their requirements. In return, these companies promised Company X that they would to absorb its students: We came up with the concept of giving an appointment letter at the time of joining the course itself; it was the first time anyone did that in India. It was considered a big risk. If we couldn't have fulfilled it, COMPANY X would have been washed away. We started with the promise that they would get an appointment letter through COMPANY X. COMPANY X had tie ups with 13 companies which had a total requirement of 50 people. So COMPANY X decided to have 60 students in the first year. With this, we entered in the market with a PG program in Multimedia which consisted of a certification in multimedia, a specialization in either of three streams (web designing, animation and graphics) and internship in the industry for 2 months which nobody had tried in the industry. (Mr Z) When the Indian mindset was that multimedia education centers were just hobby centers and the big brands were indulging youngsters with Hollywood and Bollywood studio dreams, COMPANY X focused on giving employment to youngsters from tier 2 and tier 3 cities of the country: Creativity was not our focus. COMPANY X wanted to impart skills and knowledge which would help in getting a job, a lucrative offer for the students. So our value proposition was employment compared to the competitors, for whom the value proposition was hobby. (Mr X) Mr X had a strong value system and expected his employees and partners to be aligned with it. He gave freedom with accountability and transparency with tight control. These qualities helped not only manage his team but also get good results with mediocre teams effectively across 137 franchisees: When nobody knew or trusted me, Mr X gave me a franchisee of COMPANY X and asked me to be honest and report on time daily. He expected me and my franchisee to be honest with students and their parents. (Franchisee Owner) His relationship with organizations like N, M, MC and partners helped him get critical resources (knowledge, skills and human resource) and bring creativity in the organization: Mr X is the most social among us, he has friends everywhere. Due to his personal relations with faculties of N, M and connections with corporate organizations, we could get several benefits. If we wanted to know anything, wanted people to work with us or find companies to tie up with, all these resources were a call away due to Mr X's connections and relationships. (Mr Y) Mr X knew his competition had big pockets and wider reach in metro cities. But he understood the needs of the youth from tier 2 and tier 3 cities of India and the requirements of small-and medium-sized companies better: All three of us came from middle class backgrounds and from tier 2 and tier 3 cities. I knew the opportunities were very limited for the graduates from tier 2 and tier 3 cities. The future was multimedia and many opportunities might emerge in this sector soon. So I understood the needs of the companies as well as the youth and I created a model to bridge the gap between Industry and Academia. (Mr X)

NEJE 22,2
The organization had all their centers in tier 2 or tier 3 cities and had none in metro cities. They found out that the reason of failures of Arena and NIIT was chiefly faculties. So, COMPANY X decided to provide faculties to their franchisees. This was taking a huge task on COMPANY X in terms of efforts and cost. For all 137 campuses, faculties were recruited, trained and given to the centers. Due to local language issues, they needed to recruit lots of local people in different locations. They were brought to Ahmedabad, given training and sent back to their respective location and centers. The placement was also kept in the hands of the central office (head office in Ahmedabad). These strategic choices worked for Mr X and his company; they soon captured a big market in tier 2 and tier 3 cities with 137 franchises and they grew exponentially in few years: The major challenge was having the right kind of franchisee partners and faculties in tier 2 and tier 3 cities. How could we assure that the quality and service delivered in Ahmedabad would be delivered in Rajkot or Jamnagar? So we decided to centralize the faculty recruitment process, the training of faculties and the placement process of the students to assure that our quality and service standards were maintained. We also kept the partner companies (recruiters) with us and kept them happy. (Mr X) He was driven by sheer passion and his belief in the BM he crafted and not just for material benefits. The same was seen in the other partners, they knew that they had to focus on quality and delivery for success to follow: Mr X is extremely passionate about his vision and this business. All of us are driven by sheer passion. We knew we were competing with big companies but with the passion, expertise (in technology) and aptitude we had, we ensured that not many people were running to Arena in Ahmedabad. (Mr Z) Mr X was around his team 24×7, giving feedback to everyone, appreciating the madness that hard work brings with it, helping his employees fulfill their dreams. So these initiatives lead to a high level of creativity in the organization. Many creative people from N and MC joined the organization and Mr X tried to build an organization where creativity was encouraged for the upcoming The Design University: My interaction with N faculties and teaching in N encouraged me to think more about design and creativity. I wanted to create an organization where people can try being creative without any fear and hesitation. This was also important for creating one of the first design universities in the country. (Mr X) He did more micro-management than leaders in his position and that gave him tight control over people and processes. He created a value chain of 22 activities of the operations of COMPANY X which he claimed came from his readings and academics. He and his team tried to standardize operations to assure that quality and delivery was maintained: During faculty training, all the details like what was to be covered in which session, how it would be delivered, how the lectures would begin, everything was predefined. Lectures were opened with a slide on what would be covered in the class. This slide was also shared with the students so they had an idea of what would be covered and if anything was left out, the students could remind the faculties. I brought in the concept of signing sheets in which students had to sign against their name and mention whether they could or couldn't understand what was taught during the session. (Mr X)

81
Role of leadership in business model innovation However, it was also observed that Mr X was constantly putting pressure on employees and franchisee owners to perform which was not very visible but could be observed and discovered only after talking to his employees and franchisee owners. Some employees felt that his micro-management and control mechanism did not allow them to think out of box and mostly they ended up doing what was told and expected by the top leadership, contradicting the work environment required for creativity and new idea generation: Mr X talks sweetly but he needs to be briefed about minute details. Without asking him we can't take any decision. There is always high expectations and a pressure to perform. (Employee 2) As most ideas came from Mr X or the top management, it was not observed how the people at the bottom shared ideas and how they were tried in the organization. The approach was top to bottom with no strong mechanism to push ideas from bottom to top: Mr X shares what to do and how to do it, we can give our suggestions but mostly we work according to guidelines of Mr X, Mr Z and Mr Y. (Employee 3) The leadership qualities of Mr X went beyond idea generation as he met employees twice in a week, tried to solve their problems and issues related to work by telling them where they could improve. He allowed them to work on challenging projects and the employees could even work in cross-functional areas. Employees were given holidays and sabbaticals to get a break from their routine work: Our employees are most important assets of our company. I ensure that I meet them at least twice a week and listen to their problems and suggest solutions. Employees are encouraged work in other departments and functions as well. I always look for madness and passion in people when I hire them. (Mr X) These measures lead to higher employee satisfaction and employee motivation, consequently resulting in the execution or continuation of BMI in the organization. Table III summarizes the findings related to the influence of leadership in creation and execution of BMI. The analysis shows that the model helps understand the role of leadership in BMI in an entrepreneurial firm. The leadership can be a trigger for BMI which is seen in the case of Company X s and Design for Skools projects. The leader here had a very high influence on bringing the model to life. The second contribution of leadership can be in the process of BMI or in its execution. From the case we studied, we can see the leader had a very strong influence here.

Discussion
The study explores the role of a leader in an entrepreneurial firm in BMI. Theoretically, it focuses on the under-researched area of leadership in BMI. Using a framework developed from strategy, the innovation literature and a single case study of an entrepreneurial firm from the education sector in India, the study shows that the leader's personal and individual characteristics, external ties, technical and creative skills and strategic choices influence the leadership. It is interesting to observe here that the leader being an entrepreneur driven by his passion and strong personal values, focused on his own strengths, expertise and used external ties and collaborations to bring knowledge, skills and human resources to his organization. This is very similar to the three questions an entrepreneur asks as per effectuation theory. The questions are -Who am I? What I know? And Whom I know?
It is also observed that, as a new entrant in the industry, the leader analyzed opportunities which are ignored by incumbent firms. He created a BM around untapped needs of customers and markets. His BMI focused on creating employment opportunities for Meeting employees twice in a week, finding out the problems and issues; rather than telling them they were wrong, showing them how they could improve; recruiting people for their madness; promising them challenging work (thrill working on Design for Skools) Giving them opportunity to explore other avenues (operation person could work in marketing department after his routine work) An accountant can be a director People were sent for vacation or sabbatical for breaks Employee satisfaction Employee commitment Creativity was not observed (but could be possible) Table III.

83
Role of leadership in business model innovation the youth coming from tier 2 and tier 3 cities and he matched them with the education requirements of small-and medium-sized companies in these cities. In doing so, he changed the way graphics and animation education was being delivered in India. These strategic choices were made by him as a part of conscious effort and rationale, they were not out of intuition or gut feelings and his education and prior experience helped him in designing the model after understanding the market and the sector deeply.
The insights from this case show us how the leader not only influences the starting of BMI but also its execution and continuation. The CEO as suggested by Chesbrough (2007) can bring changes in the BM unlike other top managers, who work within the existing BM, the findings suggested.
In this case, all the BMI (Design School and Design University) was brought in by the owner and the CEO of the firm. Researchers (Mitchell and Coles, 2003;Desai and Mahadik, 2011) argued that the role of a leader or a CEO is not only to start the BMI but to continue it. This study empirically found that a leader, through creating an environment where creativity was tried and encouraged, motivated his employees through different measures and gave them freedom to be creative and thus successfully executed BMI. However, his micro-management and mostly top down approach of idea generation did not bode well with the employees and sometimes hampered his vision to create a creativity-driven organization.

Managerial and theoretical implications
The study has multiple implications. The study explains where a leader and his team can focus in BMI and where the organizational changes have to be made when executing BMI. Through this study, a manager or a founder can find out which traits and skills are required in a leader managing BMI. Theoretically, it extends the discussion related to leadership in BMI. Most studies have focused on how to achieve leadership position in the industry or market through BMI but the leadership style and its impacts on initiation and execution of BMI have not been studied much.
The context of entrepreneurial firms in emerging economies is also unique as the entrepreneurs in emerging economies have to deal with resource constraints and thinking about BMI and executing it successfully means managing many internal and external challenges. The paper is not only extending the existing knowledge in this domain but is helping open avenues for many researchers to contribute in this niche area of research.
Limitation and future research This paper has a very broad definition of leadership, which gives rise to many methodological questions. The framework explains the relationship between leadership and BMI but the model has to be tested empirically using large-scale survey of firms that have done BMI. Company X is a classic case where an entrepreneur, the so-called leader of the organization, not only triggered the BMI but due to his control mechanism and motivation, executed BMI in the organization. However, his micro-management and controlling behavior might have hindered the creativity and innovation in the organization. So, it is difficult to generalize the findings as the study is based on one in-depth case study. But the model can be tested across multiple case studies. It will be interesting to replicate the study in different contexts and different industries also.

Conclusion
The study explores the under-researched area of leadership in BMI. Though BMI has got plenty of attention from researchers over the last two decades, the role of a leader and how 84 NEJE 22,2 she or he influences BMI creation and execution is not very clear in the literature. Taking up this research gap, this study tries to understand what factors affect leadership and how a leader influences creation and execution of BMI. Drawing from strategy, leadership and innovation research, a research framework was created to understand these phenomena in an entrepreneurial venture from India. An entrepreneurial firm in the multimedia education sector, its founder and his TMT was studied. It was observed and found that the leader not only influenced the triggering of BMI through the creation of a creative organization but also influenced the process of BMI by motivating his team and employees, thereby making a tangible contribution in the work being done by evaluating creative ideas of the people.

Role of leadership in business model innovation
Role of leadership in business model innovation

Introduction
Firms pursue business model innovation by exploring new ways to define value proposition, create and capture value for customers, suppliers and partners (Gambardella and McGahan, 2010;Teece, 2010;Bock et al., 2012;Casadesus-Masanell and Zhu, 2013). An extensive body of the literature asserts that innovation in business models is of vital importance to firm survival, business performance and as a source of competitive advantage Chesbrough, 2010;Amit and Zott, 2012;Baden-Fuller and Haefliger, 2013;Casadesus-Masanell and Zhu, 2013). It is starting to attract a growing attention, given the increasing opportunities for new business models enabled by changing customer expectations, technological advances and deregulation (Casadesus-Masanell and Llanes, 2011;Casadesus-Masanell and Zhu, 2013). This is evident from the recent scholarly outputs ( Figure 1). Thus, it is essential to comprehend this literature and uncover where alternative business models can be explored. The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/2574-8904.htm

Business model innovation
Conflicting approaches exist in the literature on how firms change their business models. One approach suggests that alternative business models can be explored through an evolutionary process of incremental changes to business model elements (e.g. Amit and Zott, 2012;Landau et al., 2016;Velu, 2016). The other approach, mainly practice-oriented, advocates that innovative business models can be developed through a revolutionary process by replacing existing business models (e.g. Bock et al., 2012;Iansiti and Lakhani, 2014). The fragmentation of prior research is due to the variety of disciplinary and theoretical foundations through which business model innovation is examined. Scholars have drawn on perspectives from entrepreneurship (e.g. George and Bock, 2011), information systems (e.g. Al-debei and Avison, 2010), innovation management (e.g. , marketing (e.g. Sorescu et al., 2011) and strategy (e.g. . Also, this fragmentation is deepened by focusing on different types of business models in different industries. Studies have explored different types of business models such as digital business models (e.g. Weill and Woerner, 2013), service business models (e.g. Kastalli et al., 2013), social business models (e.g. Hlady-Rispal and Servantie, 2016) and sustainability-driven business models (Esslinger, 2011) Table I). Our review builds on and extends the extant literature in at least three ways. First, unlike previous reviews that mainly focused on the general construct of "Business Model" (George and Bock, 2011;Zott et al., 2011;Wirtz et al., 2016), our review focuses on uncovering how firms change their existing business model(s) by including terms that reflect business model innovation, namely, value proposition, value creation and value capture. Second, previous reviews do not provide a clear  , which have touched lightly on some innovation aspects such as streams and motivations of business model innovation research, our review will uncover the innovation areas where alternative business models can be explored. Taking Teece's (2010) suggestion, "A helpful analytic approach for management is likely to involve systematic deconstruction/unpacking of existing business models, and an evaluation of each element with an idea toward refinement or replacement" (p. 188), this paper aims to develop a theoretical framework of business model innovation.
Our review first explains the scope and the process of the literature review. This is followed by a synthesis of the findings of the review into a theoretical framework of business model innovation. Finally, avenues for future research will be discussed in relation to the approaches, degree and mechanisms of business model innovation.

Scope and method of the literature review
Given the diverse body of business models literature, a systematic literature review was carried out to minimise research bias (Transfield et al., 2003). Compared to the previous business model literature, our review criteria are summarised in Table I

Business model innovation
Studies were included in our review if they specifically address business models and were top-rated according to The UK Association of Business Schools list (ABS, 2010). This rating has been used not only because it takes into account the journal "Impact Factor" as a measure for journal quality, but also uses in conjunction other measures making it one of the most comprehensive journal ratings. By applying these criteria, 1,682 entries were retrieved from 122 journals. By excluding duplications, 831 papers were identified. As Harvard Business Review is not listed among the peer-reviewed journals in any of the chosen databases and was included in the ABS list, we used the earlier criteria and found 112 additional entries. The reviewed papers and their subject fields are highlighted in Table II. Since the focus of this paper is on business model innovation, we selected studies that discuss value proposition, value creation and value capture as sub-themes. This is not only because the definition of business model innovation mentioned earlier spans all three sub-themes, but also because all three sub-themes have been included in recent studies (e.g. Landau et al., 2016;Velu and Jacob, 2014). To confirm whether the papers addressed business model innovation, we examined the main body of the papers to ensure they were properly coded and classified. At the end of the process, 219 papers were included in this review. Table III lists the source of our sample.
The authors reviewed the 219 papers using a protocol that included areas of innovation (i.e. components, elements, and activities), theoretical perspectives and key findings. In order to identify the main themes of business model innovation research, all papers were coded in relation to our research focus as to where alternative business models can be explored (i.e. value proposition, value creation and value capture). Coding was cross checked among the authors on a random sample suggesting high accuracy between them. Having compared and discussed the results, the authors were able to identify the main themes.

Prior conceptualisations of business model innovation
Some scholars have articulated the need to build the business model innovation on a more solid theoretical ground (Sosna et al., 2010;George and Bock, 2011 are not explicitly theory-based, some studies partially used well-established theories such as the resource-based view (e.g. Al-Debei and Avison, 2010) and transaction cost economics (e.g. DaSilva and Trkman, 2014) to conceptualise business model innovation. Other theories such as activity systems perspective, dynamic capabilities and practice theory have been used to help answer the question of how firms change their existing business models.
Using the activity systems perspective, Zott and Amit (2010) demonstrated how innovative business models can be developed through the design themes that describe the source of value creation (novelty, lock-in, complementarities and efficiency) and design elements that describe the architecture (content, structure and governance). This work, however, overlooks value capture which limits the explanation of the advocated system's view (holistic). Moreover, Chatterjee (2013) used this perspective to reveal that firms can design innovative business models that translate value capture logic to core objectives, which can be delivered through the activity system.
Dynamic capability perspective frames business model innovation as an initial experiment followed by continuous revision, adaptation and fine-tuning based on trial-and-error learning (Sosna et al., 2010). Using this perspective,  showed that "dynamic consistency" is a capability that allows firms to sustain their performance while innovating their business models through voluntary and emergent changes. Also, Mezger (2014) conceptualised business model innovation as a distinct dynamic capability. He argued that this capability is the firm's capacity to sense opportunities, seize them through the development of valuable and unique business models, and accordingly reconfigure the firms' competences and resources. Using aspects of practice theory, Mason and Spring (2011) looked at business model innovation in the recorded sound industry and found that it can be achieved through various combinations of managerial practices.
Static and transformational approaches have been used to depict business models . The former refers to viewing business models as constituting core elements that influence business performance at a particular point in time. This approach offers a snapshot of the business model elements and how they are assembled, which can help in understanding and communicating a business model (e.g. Eyring et al., 2011;Mason and Spring, 2011;Yunus et al., 2010). The latter, however, focuses on innovation and how to address the changes in business models over time (e.g. Sinfield et al., 2012;Girotra and Netessine, 2014;Landau et al., 2016). Some researchers have identified the core elements of business models ex ante (e.g. Wu et al., 2010;Huarng, 2013;, while others argued that considering a priori elements can be restrictive (e.g. Casadesus-Masanell and Ricart, 2010). Unsurprisingly, some researchers found a middle ground where elements are loosely defined allowing flexibility in depicting business models (e.g. Zott and Amit, 2010;Sinfield et al., 2012;Kiron et al., 2013).

Business model innovation
Prior to 2010, conceptual frameworks focused on the business model concept in general (e.g. Chesbrough and Rosenbloom, 2002;Osterwalder et al., 2005;Shafer et al., 2005) apart from Johnson et al.'s (2008), which is one of the early contributions to business model innovation. To determine whether a change in existing business model is necessary, Johnson et al. (2008) suggested three steps: "Identify an important unmet job a target customer needs done; blueprint a model that can accomplish that job profitably for a price the customer is willing to pay; and carefully implement and evolve the model by testing essential assumptions and adjusting as you learn" (Eyring et al., 2011, p. 90). Although several frameworks have been developed since then, our understanding of business model innovation is still limited due to the static nature of the majority of these frameworks. Some representations ignore the elements and/or activities where alternative business models can be explored (e.g. Sinfield et al., 2012;Chatterjee, 2013;Huarng, 2013;Morris et al., 2013;Girotra and Netessine, 2014). Other frameworks ignore value proposition (e.g. Zott and Amit, 2010), ignore value creation (e.g. Michel, 2014) and/or ignore value capture (e.g. Mason and Spring, 2011;Sorescu et al., 2011;Storbacka, 2011). Some conceptualisations do not identify who is responsible for the innovation (e.g. Casadesus-Masanell and Ricart, 2010;Sinfield et al., 2012;Chatterjee, 2013;Kiron et al., 2013). Synthesising the different contributions into a theoretical framework of business model innovation will enable a better understanding of how firms undertake business model innovation.

Business model innovation framework
Our framework (Figure 2) integrates all the elements where alternative business models can be explored. This framework does not claim that the listed elements are definitive for high-performing business models, but is an attempt to outline the elements associated with business model innovation. This framework builds on the previous work of  Johnson et al. (2008) and Zott and Amit (2010) by signifying the elements associated with business model innovation. Unlike previous frameworks that mainly consider the constituting elements of business models, this framework focuses on areas of innovation where alternative business models can be explored. Moreover, this is not a static view of the constituting elements of a business model, but rather a view enabling firms to explore alternative business models by continually refining these elements. Arrows in the framework indicate the continuous interaction of business model elements. This framework consists of 4 areas of innovation and 16 elements (more details are shown in Table IV ). Each will be discussed below.

Value proposition
The first area of innovation refers to elements associated with answering the "Why" Developing new business models usually starts with articulating a new customer value proposition (Eyring et al., 2011). According to Sinfield et al. (2012), firms are encouraged to explore various alternatives of core offering in more depth by examining type of offering (product or service), its features (custom or off-the-shelf ), offered benefits (tangible or intangible), brand (generic or branded) and lifetime of the offering (consumable or durable). In order to exploit the "middle market" in emerging economies, Eyring et al. (2011) suggested that companies need to design new business models that aim to meet unsatisfied needs and evolve these models by continually testing assumptions and making adjustments. To uncover unmet needs, Eyring et al. (2011) suggested answering four questions: what are customers doing with the offering? What alternative offerings consumers buy? What jobs consumers are satisfying poorly? and what consumers are trying to accomplish with existing offerings? Furthermore, Baden-Fuller and Haefliger (2013) made a distinction between customers and users in two-sided platforms, where users search for products online, and customers ( firms) place ads to attract users. They also made a distinction between "pre-designed (scale) based offerings" and "project based offerings". While the former focuses on "one-size-fits-all", the latter focuses on specific client solving specific problem.
Established firms entering emerging markets should identify unmet needs "the job to be done" rather than extending their geographical base for existing offerings (Eyring et al., 2011). Because customers in these markets cannot afford the cheapest of the high-end offerings, firms with innovative business models that meet these customers' needs affordably will have opportunities for growth (Eyring et al., 2011). Moreover, secondary business model innovation has been advocated by Wu et al. (2010) as a way for latecomer firms to create and capture value from disruptive technologies in emerging markets. This can be achieved through tailoring the original business model to fit price-sensitive mass customers by articulating a value proposition that is attractive for local customers.

Operational value
The second area of innovation focuses on elements associated with answering the "What" questions. Many of the established frameworks included either one element   . However, very few included three or more elements (e.g. Mehrizi and Lashkarbolouki, 2016;Cortimiglia et al., 2016). These elements include configuring key assets and sequencing activities to deliver the value proposition, exposing the various means by which a company reaches out to customers, and establishing links with key partners and suppliers. Focusing on value creation, Zott and Amit (2010) argued that business model innovation can be achieved through reorganising activities to reduce transaction costs. However, Al-Debei and Avison (2010) argued that innovation relating to this dimension can be achieved through resource configuration, which demonstrates a firm's ability to integrate various assets in a way that delivers its value proposition. Cavalcante et al. (2011) proposed four ways to change business models: business model creation, extension, revision and termination by creating or adding new processes, and changing or terminating existing processes.
Western firms have had difficulty competing in emerging markets due to importing their existing business models with unchanged operating model (Eyring et al., 2011). Alternative business models can be uncovered when firms explore the different roles they might play in the industry value chain (Sinfield et al., 2012). Al-Debei and Avison (2010) suggested achieving this through answering questions such as: what is the position of our firm in the value system? and what mode of collaboration (open or close) would we choose to reach out in a business network? Dahan et al. (2010) found cross-sector partnerships as a way to co-create new multi-organisational business models. They argued that multinational enterprises (MNEs) can collaborate with nongovernmental organisations (NGOs) to create products/or services that neither can create on their own. Collaboration allows access to resources that firms would otherwise need to solely develop or purchase (Yunus et al., 2010). According to Wu et al. (2010), secondary business model innovation can be achieved when latecomer firms fully utilise strategic partners' complementary assets to overcome their latecomer disadvantages and build a unique value network specific to emerging economies context.

Human capital
The third area of innovation refers to elements associated with answering the "Who" questions. Most of the established frameworks in this field tend to focus less on human capital and include one element at most (e.g. Wu et al., 2010;Kohler, 2015). However, our framework highlights four elements, which include experimenting with new ways of doing business, tapping into the skills and competencies needed for the new business model through motivating and involving individuals in the innovation process. According to Belenzon and Schankerman (2015), "the ability to tap into a pool of talent is strongly related to the specific business model chosen by managers" (p. 795). They claimed that managers can strategically influence individuals' contributions and their impact on project performance.
Organisational learning can be maximised though continuous experimentation and making changes when actions result in failure (Yunus et al., 2010). Challenging and questioning the existing rules and assumptions and imagining new ways of doing business will help develop new business models. Another essential element of business model design is governance, which refers to who performs the activities (Zott and Amit, 2010). According to Sorescu et al. (2011), innovation in retail business models can occur as a result of changes in the level of participation by actors engaged in performing the activities. An essential element of retailing governance is the incentive structure or the mechanisms that motivate those involved in carrying out their roles to meet customer demands (Sorescu et al., 2011). For example, discount retailers tend to establish different compensation and incentive policies (Brea-Solís et al., 2015). Revising the incentive system can have a major impact on new ventures' performance by aligning 99 Business model innovation organisational goals at each stage of growth (Roberge, 2015). Zott and Amit (2010) argued that alternative business models can be explored through adopting innovative governance or changing one or more parties that perform any activities. Sinfield et al. (2012) suggested that business model innovation only requires time from a small team over a short period of time to move a company beyond incremental improvements and generate new opportunities for growth. This is supported by Michel's (2014) finding that cross-functional teams were able to quickly achieve business model innovation in workshops through deriving new ways to capture value.

Financial value
The final area of innovation focuses on elements associated with answering the "How" questions. Previously developed frameworks tend to prioritise this area of innovation by three elements (e.g. Eyring et al., 2011;Huang et al., 2013), and in one instance four elements (e.g. Yunus et al., 2010). These elements include activities linked with how to capture value through revenue streams, changing the price-setting mechanisms, and assessing the financial viability and profitability of a business. According to , changes in cost and/or revenue structures are the consequences of both continuous and radical changes. They also argued that costs relate to different activities run by organisations to acquire, integrate, combine or develop resources. Michel (2014) suggested that alternative business models can be explored through: changing the pricesetting mechanism, changing the payer, and changing the price carrier. Different innovation forms are associated with each of these categories.
Business model innovation can be achieved through exploring new ways to generate cash flows (Sorescu et al., 2011), where the organisation has to consider (and potentially change) when the money is collected: prior to the sale, at the point of sale, or after the sale (Baden-Fuller and Haefliger, 2013). Furthermore,  suggested that changes in business models affect margins. This is apparent in the retail business models, which generate more profit through business model innovation compared to other types of innovation (Sorescu et al., 2011).

Ways to change business models
From reviewing the recent developments in the business model literature, alternative business models can be explored through modifying a single business model element, altering multiple elements simultaneously and/or changing the interactions between elements of a business model.
Changing one of the business model elements (i.e. content, structure or governance) is enough to achieve business model innovation (Amit and Zott, 2012). This means that firms can have a new activity system by performing only one new activity. However, Amit and Zott (2012) clearly outlined a systemic view of business models which entails a holistic change. This is evident from  work suggesting that the study of business model innovation should not focus on isolated activities since changing a core element will not only impact other elements but also the interactions between these elements.
Another way to change business models is through altering multiple business model elements simultaneously. Kiron et al. (2013) found that companies combining target customers with value chain innovations and changing one or two other elements of their business models tend to profit from their sustainability activities. They also found that firms changing three to four elements of their business models tend to profit more from their sustainability activities compared to those changing only one element. Moreover, Dahan et al. (2010) found that a new business model was developed as a result of MNEs and NGOs collaboration by redefining value proposition, target customers, governance of activities and 100 NEJE 22,2 distribution channels. Companies can explore multiple combinations by listing different business model options they could undertake (desirable, discussable and unthinkable) and evaluate new combinations that would not have been considered otherwise (Sinfield et al., 2012).
Changing business models is argued to be demanding as it requires a systemic and holistic view (Amit and Zott, 2012) by considering the relationships between core business model elements . As mentioned earlier, changing one element will not only impact other elements but also the interactions between these elements. A firm's resources and competencies, value proposition and organisational system are continuously interacting and this will in turn impact business performance either positively or negatively . According to Zott and Amit (2010), innovative business models can be developed through linking activities in a novel way that generates more value. They argued that alternative business models can be explored by configuring business model design elements (e.g. governance) and connecting them to distinct themes (e.g. novelty). Supporting this, Eyring et al. (2011) suggested that core business model elements need to be integrated in order to create and capture value (Eyring et al., 2011).
6. Discussion and future research directions From the above synthesis of the recent development in the literature, several gaps remain unfilled. To advance the literature, possible future research directions will be discussed in relation to approaches, degrees and mechanisms of business model innovation.

Approaches of business model innovation
Experimentation, open innovation and disruption have been advocated as approaches to business model innovation. Experimentation has been emphasised as a way to exploit opportunities and develop alternative business models before committing additional investments (McGrath, 2010). Several approaches have been developed to assist in business model experimentation including mapping approach, discovery-driven planning and trail-and-error learning (Chesbrough, 2010;McGrath, 2010;Sosna et al., 2010;Andries and Debackere, 2013). Little is known about the effectiveness of these approaches. It will be worth investigating which elements of the business model innovation framework are more susceptible to experimentation and which elements should be held unchanged. Although business model innovation tends to be characterised with failure (Christensen et al., 2016), not much has been established on failing business models. It is interesting to explore how firms determine a failing business model and what organisational processes exist (if any) to evaluate and discard these failed business models. Empirical studies could examine which elements of business model innovation framework are associated with failing business models.
Another way to develop alternative business models is through open innovation. Although different categories of open business models have been identified by researchers (e.g. Frankenberger et al., 2014;Taran et al., 2015;Kortmann and Piller, 2016), their effectiveness is yet to be established. Further research is needed to examine when can a firm open and/or close element(s) of the business model innovation framework. Future studies could also examine the characteristics of open and/or close business models.
In responding to disruptive business models, how companies extend their existing business model, introduce additional business model(s) and/or replace their existing business model altogether remains underexplored. Future research is needed to unravel the strategies deployed by firms to extend their existing business models as a response to disruptive business models. In introducing additional business models, Markides (2013) suggested that a company will be presented with several options to manage the two 101 Business model innovation businesses at the same time: create a completely separate business unit, integrate the two business models from the beginning or integrate the second business model after a certain period of time. Finding the balance between separation and integration is of vital importance. Further research could identify which of these choices are most common among successful firms introducing additional business models, how is the balance between integration and separation achieved, and which choice(s) prove more profitable. Moreover, very little is known on how firms replace their existing business model. Longitudinal studies could provide insights into how a firm adopts an alternative model and discard the old business model over time. It may also be worth examining the factors associated with the adoption of business model innovation as a response to disruptive business models. Moreover, new developments in digital technologies such as blockchain, Internet of Things and artificial intelligence are disrupting existing business models and providing firms with alternative avenues to create new business models. Thus far, very little is known on digital business models, the nature of their disruption, and how firms create digital business models and make them disruptive. Future research is needed to fill these important gaps in our knowledge.

Degrees of business model innovation
Business models can be developed through varying degrees of innovation from an evolutionary process of continuous fine-tuning to a revolutionary process of replacing existing business models. Recent research shows that survival of firms is dependent on the degree of their business model innovation (Velu, 2015(Velu, , 2016. This review classifies these degrees of innovation into modifying a single element, altering multiple elements simultaneously and/or changing the interactions between elements of the business model innovation framework. In changing a single element, further research is needed to examine which business model element(s) is (are) associated with business model innovation. It is not clear whether firms intentionally make changes to a single element when carrying out business model innovation or stumble at it when experimenting with new ways of doing things. It may also be worth investigating the entry (or starting) points in the innovation process. There is no consensus in the literature on which element do companies start with when carrying out their business model innovation. While some studies suggest starting with the value proposition (Eyring et al., 2011;Landau et al., 2016), others suggest starting the innovation process with identifying risks in the value chain (Girotra and Netessine, 2011).  suggested two entry points, namely, value proposition and target customers. In commercialising innovations, the former refers to technology-push innovation while the latter refers to market-pull innovation. Also, it is not clear whether the entry point is the same as the single element associated with changing the business model. Further research can explore the different paths to business model innovation by identifying the entry point and subsequent changes needed to achieve business model innovation.
There is little guidance in the literature on how firms change multiple business model elements simultaneously. Landau et al. (2016) claimed that firms entering emerging markets tend to focus on adjusting specific business model components. It is unclear which elements need configuring, combining and/or integrating to achieve a company's value proposition. Furthermore, the question of which elements can be "bought" on the market or internally "implemented" and their interplay remains unanswered (DaSilva and Trkman, 2014). Casadesus-Masanell and Ricart (2010) argued that "[…] there is (as yet) no agreement as to the distinctive features of superior business models" (p. 196). Further research is needed to explore these distinctive elements of high-performing business models.

NEJE 22,2
In changing the interactions between business model elements, further research is needed to explore how these elements are linked and what interactions' changes are necessary to achieve business model innovation. Moreover, the question of how firms sequence these elements remains poorly understood. Future research can explore the synergies created over time between these elements. According to , we need to improve our understanding of the connective mechanisms and dynamics involved in business model development. More work is needed to explore the different modalities of interdependencies among these elements and empirically testing such interdependencies and their effect on business performance (Sorescu et al., 2011).
It is surprising that the link between business model innovation and organisational performance has rarely been examined. Changing business models has been found to negatively influence business performance even if it is temporary (McNamara et al., 2013;Visnjic et al., 2016). Contrary to this, evidence show that modifying business models is positively associated with organisational performance (Cucculelli and Bettinelli, 2015). Empirical research is needed to operationalise the various degrees of innovation in business models and examine their link to organisational performance. Longitudinal studies can also be used to explore this association since it may be the case that business model innovation has a negative influence on performance in the short run and that may change subsequently. Moreover, it is not clear whether high-performing firms change their business models or innovation in business models is a result from superior performance (Sorescu et al., 2011). Further studies are needed to determine the direction of causality. Another link that is worth exploring is business model innovation and social value, which has only been explored in a few studies looking at social business models (e.g. Yunus et al., 2010; Wilson and Post, 2013). Further research is needed to examine this link and possibly examine both financial and non-financial business performance.

Mechanisms of business model innovation
Although we know more about how firms define value proposition, create and capture value (Landau et al., 2016;Velu and Jacob, 2014), what remains as a blind spot is the mechanism of business model innovation. This is due to the fact that much of the literature seems to focus on value creation. To better understand the various mechanisms of business model innovation, future studies must integrate value proposition, value creation and value capture elements. Empirical studies could use the business model innovation framework to examine the various mechanisms of business model innovation. Also, the literature lacks the integration of internal and external perspectives of business model innovation. Very few studies look at the external drivers of business model innovation and the associated internal changes. The external drivers are referred to as "emerging changes", which are usually beyond manager's control . Inconclusive findings exist as to how firms develop innovative business models in response to changes in the external environment. Future studies could examine the external factors associated with the changes in the business model innovation framework. Active and reactive responses need to be explored not only to understand the external influences, but also what business model changes are necessary for such responses. A better understanding of the mechanisms of business model innovation can be achieved by not only exploring the external drivers, but also linking them to specific internal changes. Although earlier contributions linking studies to established theories such as the resource-based view, transaction cost economics, activity systems perspective, dynamic capabilities and practice theory have proven to be vital in advancing the literature, developing a theory that elaborates on the antecedents, consequences and different facets of business model innovation is still needed (Sorescu et al., 2011).

Business model innovation
Theory can be advanced by depicting the mechanisms of business model innovation through the integration of both internal and external perspectives. Also, we call for more empirical work to uncover these mechanisms and provide managers with the necessary insights to carry out business model innovation.

Conclusions
The aim of this review was to explore how firms approach business model innovation. The current literature suggests that business model innovation approaches can either be evolutionary or revolutionary. However, the evidence reviewed points to a more complex picture beyond the simple binary approach, in that, firms can explore alternative business models through experimentation, open and disruptive innovations. Moreover, the evidence highlights further complexity to these approaches as we find that they are in fact a spectrum of various degrees of innovation ranging from modifying a single element, altering multiple elements simultaneously, to changing the interactions between elements of the business model innovation framework. This framework was developed as a navigation map for managers and researchers interested in how to change existing business models. It highlights the key areas of innovation, namely, value proposition, operational value, human capital and financial value. Researchers interested in this area can explore and examine the different paths firms can undertake to change their business models. Although this review pinpoints the different avenues for firm to undertake business model innovation, the mechanisms by which firms can change their business models and the external factors associated with such change remain underexplored.

Introduction
Entrepreneurial ventures are a fast-growing employment opportunity for women globally (Langowitz and Minniti, 2007). Yet extant literature suggests that entrepreneurial ventures by men far surpass those begun by women, despite changes in society (Delmar and Davidsson, 2000;Marlow et al., 2008;Manolova et al., 2012;Minniti et al., 2005). There are two main types of entrepreneurial ventures that women start: conventional and social. Much of the research on gender differences has been focused on conventional entrepreneurship based on business creation for profit (Ahl, 2006). However, the empowerment of female entrepreneurship through social entrepreneurship has become a focus from start-ups to large companies ( Johnstone-Louis, 2017). Both types of firms recognize the untapped economic contribution of female entrepreneurs to have a "multiplier effect" to children and communities (Elborgh-Woytek, 2013). Scholars identified gender-specific barriers that may inhibit some females from pursuing entrepreneurial endeavors (barriers to credit, gender-based discrimination, and legal and customary male-based regimes; Ahl et al., 2016;Kabeer, 2005Kabeer, , 2011Lewis, 2014;Nussbaum, 2011). For many women, entrepreneurship is an expression of social change instead of economic activity (Calas et al., 2009). Datta and Gailey (2012) found social entrepreneurship provides social value and empowers women, especially those women living in oppressive economic societies. Our purpose is to research the role of perception in what drives some women into social entrepreneurship and away from conventional entrepreneurship.
Many social, educational and financial resources go into the training and development of entrepreneurs to assist them in launching a successful enterprise (Vanevenhoven and Liguori, 2013). Yet perception and cognitive factors help to drive new venture creation (Edelman and Yli-Renko, 2010). Langowitz and Minniti (2007) suggest that the differences in perception, between men and women, largely contribute to the entrepreneurial intention difference between genders. One of the principle avenues of research in the literature has been that gender stereotypes, as well as hegemonic masculinity, may discourage female entrepreneurs (Gupta et al., 2008(Gupta et al., , 2009Hechavarria and Ingram, 2016). Extant literature suggests the differences between male and female entrepreneurial propensity is partly influenced by socially embedded gender stereotypes (deBruin et al., 2007;Eagly and Karau, 2002;Greene et al., 2013;Gupta et al., 2008).
A great amount of research has been developed on how these perceptions and stereotypes may limit female entrepreneurship. For example, Gupta and Bhawe (2007) argued that proactive personality and stereotype threat (when people feel compelled to conform to social stereotypes) limited entrepreneurial intentions. Gupta et al. (2009) found that men and women can perceive entrepreneurship as possessing masculine characteristics in terms of aggression. Gupta et al. (2008) would confirm gender stereotypes when implicitly related to entrepreneurship.
As scholars have changed our notions of what entrepreneurship is and have considered different types of entrepreneurship, we have noted that there are various gender differences. Conventional entrepreneurship is often viewed as a masculine undertaking (Ahl, 2006), whereas social entrepreneurship is often considered more feminine (Datta and Gailey, 2012;Zahra et al., 2009). Women are more likely to start social ventures than conventional ventures (Hechavarria and Ingram, 2016). These findings are replicated by Gupta et al. (2019), who found that men are more attracted to high-reward enterprises, whereas woman toward communal. Likewise, Hechavarría et al. (2017) found that women are generally more likely to support social outcomes than men. Basically, some women tend to buy into myths and stereotypes that prevent them from pursuing conventional entrepreneurship. According to Gupta et al. (2019), despite the work done on why women generally are attracted to social vs conventional entrepreneurship, more work is needed examining why these trends exist.
A potential argument could come from social dominance theory (SDT), which explains why these stereotypes exist. Scholars have found that individuals possess stable beliefs about traditional gender and race roles, which Sidanius and Pratto (1999) call social dominance orientation (SDO) and are the result of myths that society believes. Society often encourages males to be aggressive, risk-taking and competitive, whereas females are usually encouraged to be more community orientated, passive and less aggressive (Eagly et al., 2000;Murray, 2001). Society generally encourages prejudice against women who possess the qualities necessary to succeed as conventional entrepreneurs (aggression, competitiveness, ambition, control, etc.), whereas it encourages some women to have communal traits associated with social entrepreneurship (gentleness, sensitivity, affection, sympathy, etc.) through the use of myths (Eagly and Karau, 2002). Although these factors are not inclusive of all antecedents impacting new venture creation, extant literature suggests that these constructs will be highly influential (Langowitz and Minniti, 2007).
Our paper contributes to the extant literature in several ways. The first contribution is that we use SDT to examine the intentions of female entrepreneurs through the reduction of self-efficacy related to conventional entrepreneurship. This contributes to the literature in that we examine why stereotypes exist and why they may lead to different behaviors. The second contribution is to examine the differential relationship between social and conventional entrepreneurship caused by SDO through influencing entrepreneurial self-efficacy. Third, we propose that mentorship may moderate the relationship between 110 NEJE 22,2 SDO and entrepreneurial intentions. This contribution builds on the contributions of others as to how social relationships can influence both identity and self-efficacy (Das et al., 2017;Javadian et al., 2018).

Theoretical development
The growth of many economies has been attributed to the activity and behaviors of entrepreneurs (Baumol, 1990;Miaoulis et al., 2005;Thomas and Mueller, 2000). New business ventures are often the result of entrepreneurial start-ups that expand businesses, create new employment opportunities and generate economic growth. Due to the complexities of this process, Davidsson and Gordon (2012) suggest that studying new business creation requires longitudinal studies that define new business ventures with multiple criteria including the generation of income, provision of product or services, completion of gestational activities and continued operation of the business. The clear conclusion for entrepreneurial researchers is that new business creation is not the result of a quick, impulsive process but rather a developed, challenging experience surrounded by risk and uncertainty (Krueger et al., 2000).
Scholars have struggled to understand the participation differences between conventional and social entrepreneurship. A potential explanation may be that the motives between social and conventional entrepreneurship are different. Conventional entrepreneurship is driven by profit; social entrepreneurship is based on increasing social outcomes (Santos, 2012). Yet there may be an overlap as economic outcomes form part of the mission of the social firm; there is a hierarchy as firms need to meet economic concerns before they can reach social ones. As Dacin et al. (2011) argue, "social entrepreneurs balance both sets of priorities. A social value creation mission does not necessarily negate nor diminish a focus on economic value" (p. 1205). It is also true that those individuals who form social entrepreneurship may focus more on social outcomes and have more of a postmaterialist view of the world (Dacin et al., 2011;Smith et al., 2013). These trends, and the stereotypes surrounding them, partly explain why some women are attracted to social entrepreneurship rather than conventional entrepreneurship (Hechavarría et al., 2017;Gupta et al., 2019). If the type of entrepreneurship they enter into is partly based on stereotypes, then why do these stereotypes exist? In essence, these stereotypes generally exist as myths because individuals tend to hierarchically divide society. People usually propagate these myths either to suppress others or to accept their lower level in society. Sidanius and Pratto (2012) argue that: SDT defines legitimizing myths (LMs) as consensually shared ideologies (including stereotypes, attributions, cosmologies, predominant values or discourses, shared representations, etc.) that organize and justify social relationships. LMs suggest how people and institutions should behave, why things are how they are, and how social value should be distributed. Because they are consensual and closely associated with the structure of their societies, LMs often have the appearance of being true. Consequently, those who reject them take risks and have work to do in explaining how and why they disagree (p. 426).
SDO is the extent to which people buy into legitimizing myths (Sidanius and Pratto, 2012). It emerges through a combination of personality and interaction with the environment and explains, in part, why individuals would perceive the world as enforcing stereotypes (Sidanius and Pratto, 2012). SDO may explain why some woman are likely to engage in social entrepreneurship, whereas others are likely to engage in conventional entrepreneurship.
Members of the non-hegemonic group may tolerate discrimination because they accept the dominant groups' myths. Our contention is that women high in SDO may accept the myth that men are more aggressive and therefore those women are less likely to become conventional entrepreneurs due to lowered self-efficacy (Pratto et al., 1997). SDO could 111 Impact of social dominance orientation actually increase the willingness of female entrepreneurs to participate in social entrepreneurship and fits the hegemonic group's perceptions of what women should do (Umphress et al., 2007). Scholars have found that minorities are just as likely to buy into discrimination and not respond to opportunities both professionally and academically (Umphress et al., 2007). We will also examine how mentorship influences the relationship between SDO and entrepreneurial self-efficacy.
Proposition development SDT offers an explanation of the role of bias and discrimination in the formation of entrepreneurs. The theory's basic assumption is that individuals hierarchically divide society (Sidanius and Pratto, 1999). According to the theory, a small number of hegemonic groups wish to maintain the hierarchy and to do so, they create myths, systems of control and discrimination to prevent change. Hegemonic groups possess the majority of socially desirable values and use those values to justify an established hierarchy. Those who are at the bottom of society lack certain characteristics that would make them socially desirable and discrimination, by society against them, is therefore justified (Sidanius and Pratto, 1999).
These traditional social characteristics span age, race, social class, nationality, regionalism, estate, linguistic group, religion and arbitrary roles (such as sports teams; Sidanius and Pratto, 1999). According to SDT, people in positions of power create a system of myths and notions that keep the socially undesirable down by developing stereotypes and hegemonic systems of control (Pratto et al., 1994;Sidanius et al., 1996). Sidanius and Pratto (1999) argue that these stereotypes may be passed down through law, economy, education, certain aspects of social and religious life, and family/clan and may create self-perpetuating systems.
Scholars have found consistent support for SDT and active discrimination against various social groups (e.g. blacks and women; Aquino et al., 2005;Duckitt et al., 2002;Kemmelmeier, 2005;Pratto et al., 1994;Sidanius et al., 1996;Van Hiel and Mervielde, 2004). For instance, scholars have found that individuals with SDO are more likely to believe African-Americans to be less competent, less warm and less likely to succeedeven when the individual is African-American (Aquino et al., 2005;Umphress et al., 2007). Scholars have found that investors are more likely to invest in businesses that are owned by white individuals and less likely to invest in minority run businesses, in part, due to the investors' erroneous beliefs about white superiority developed through myth (Gutierrez, 2017) and could explain firm performance ( Jaiswal, 2018).
SDO is an individual variable describing attitudes that individuals have toward their willingness to tolerate hierarchy (Umphress et al., 2007). In other words, individuals who are high in SDO are more likely to tolerate discrimination and more likely to adhere to socially defined characteristics (Sidanius and Pratto, 1999). Entrepreneurship, for instance, has been described as risk-taking on the part of the entrepreneur. Scholars have found that social legitimation may influence the attraction that men and women feel toward entrepreneurship (Santos et al., 2016). Accordingly, the entrepreneur needs to be risk-taking, bold, decisive, analytical and inquisitive (Carter et al., 1996).
Entrepreneurial intention is the combination of personality traits, perceptions and experiences of a potential entrepreneur that lead to the creation of a new venture (Kets de Vries and Miller, 1984;Gorgievski and Stephan, 2016;Liñán and Fayolle, 2015;Pitt, 1998). The focus on enterprise creation is seen in developed countries and transitional economies where barriers to entry can be extensive (Kaufmann et al., 1995). Entrepreneurial intention precedes a long process of new business creation (Wong and Lee, 2004). Intention must precede any entrepreneurial behavior (Fayolle et al., 2006). Historically, intention has been considered the best predictor of behavior (Ajzen, 2001;Fishbein and Ajzen, 1975).

NEJE 22,2
Those characteristics that are considered more masculine in nature, such as aggression, risk-taking and decisive behavior, are usually important to starting a business (Murray, 2001). Some scholars have found that female entrepreneurial endeavors are considered less legitimate due to the societal myth that women should be less aggressive (Marlow and McAdam, 2013). We propose that SDO may reinforce female's perception that aggression, risk-taking and decisive behaviors are more masculine in nature and may lower their conventional entrepreneurial intentions (Giazitzoglu and Down, 2017;Greene et al., 2013;Hechavarria and Ingram, 2016;Marlow, 2014;Marlow et al., 2008;Marlow and McAdam, 2013;Murray, 2001). Since conventional entrepreneurship is generally perceived to be a masculine aspect, it makes sense that someone who has high SDO would believe that conventional entrepreneurship should be a male-dominated domain (Giazitzoglu and Down, 2017;Hechavarria and Ingram, 2016). These beliefs can be passed on by the hegemonic media and business, which encourage the belief that men are naturally more competitive and aggressive than women (Marlow, 2014). Accordingly, females who have higher degrees of SDO may have lower levels of conventional entrepreneurial intention because they are likely to buy into the myth that females are less acquisitive, analytical and willing to take risks (Ahl and Marlow, 2012). Thus, we propose the following: P1. Potential female entrepreneurs' levels of SDO are negatively associated with their intentions to become conventional entrepreneurs. Bird and Brush (2002) argue that the founder's gender perspective influences organizational processes and therefore the foundation of new organizations. Feminine founders are thought to perceive the world as a web or network of relationships making them more prosocial. Feminine founders tend to guide moral decisions in an effort to preserve connections, as well as show recognition for others (Gilligan, 1982, p. 30). For feminine founders, responsibility usually involves caring that focuses and responds to individual needs (Gilligan and Attanucci, 1988). Feminine founders generally consider the interdependencies of relationships and encourage helping others, avoiding harm or assisting others in recovering from harm. Researchers have found that many social entrepreneurs are members of a population that has been found to be disadvantaged (Zahra et al., 2008). Often these populations consist of female entrepreneurs who have a high understanding of the population and a specific need to be addressed through the social entrepreneurial venture (Renko, 2013;Vanevenhoven and Liguori, 2013). Women high on SDO may accept such myths because women are thought to be more communal in nature (Eagly and Karau, 2002). Feminine entrepreneurs usually reflect a power motive that involves self-mastery or a desire to contribute to the social good rather than self-benefit (Bird and Brush, 2002). Social entrepreneurship is the use of entrepreneurship and business to solve social problems (Renko, 2013). Social entrepreneurship has gained attention in literature due to its potential to resolve cultural, environmental and social challenges, such as discrimination, poverty and pollution (Wry and York, 2017). Social entrepreneurship involves private and public organizations designed to focus on social benefits above corporate profits. Social entrepreneurship is making international impact in developed countries, as well as third world countries, by encouraging innovative solutions through the contributions of individuals. Social entrepreneurship can create opportunities for the underprivileged often breaking social barriers that inhibit advancement (Di Domenico et al., 2010;Estrin et al., 2013;Peredo and Chrisman, 2006). It can expand a social segment into an international community. Social entrepreneurs can develop social capital and norms that can be utilized by conventional entrepreneurs (Zahra et al., 2008). We speculate that the relationship between social entrepreneurial intentions and SDO is the opposite of the relationship between SDO and conventional entrepreneurship. Hechavarria and Ingram (2016) found that women were more likely to start social entrepreneurship businesses. Hechavarria and Ingram proposed that hegemonic ideas 113 Impact of social dominance orientation drive these women, who are thought to be more communal in nature, into social entrepreneurship; thus, those women who are more competitive in nature were less likely to start social entrepreneurship firms. The organizational processes are more open to negotiation and less apt to be legalistic. Empathy and social awareness are often stereotyped as communal/feminine traits (Sidanius and Pratto, 1999). These traits may emerge from the myths provided by those who wish to maintain the current social order (Sidanius and Pratto, 1999). Some females often exhibit more of a community spirit than males allowing them to have greater entrepreneurial intention for social entrepreneurship (Hechavarria and Ingram, 2016).
SDO provides a framework that can explain these findings. Social entrepreneurship is focused on empathy, community, quality of life and morality, which are considered by some classical feminine qualities (Hechavarria and Ingram, 2016). Each of these attributes aligns more closely with traditional communal traits of women rather than agentic traits of men, such as aggression, competitiveness, ambition, etc. This would suggest that women are more likely to possess traits that would lead them to have social entrepreneurial intention. Building on Mair and Noboa (2006), Hockerts (2017) found that exposure to social problems (i.e. volunteering and working in environments that interface with social problems) increase social entrepreneurship intention. These positions and careers are often occupied by women. Unlike conventional entrepreneurship, social entrepreneurship usually lacks a motive for pure financial profitin social entrepreneurship, the profit provides a positive social impact. The social entrepreneurial venture's reason for existence is usually to provide a long-lasting, transformational, positive change in the lives of others (Katre and Salipante, 2012).
In general, we are taught that women are kinder, more interested in helping others and more socially aware (Dyer and Whetten, 2006;Rudman and Glick, 1999). Part of the behaviors that we see from this type of myth building is the preponderance of women in fields such as social work, education and nursingfields that require compassion and other social awareness. Social entrepreneurship is concerned with the betterment of society and is based, in part, on compassion and aiding society (Renko, 2013). Additionally, social entrepreneurship is generally based less on acquisition than conventional entrepreneurship and more on improving the lives of others; therefore, we believe social entrepreneurship should have an opposite relationship to conventional entrepreneurship regarding SDO. Accordingly, we propose the following: P2. Potential female entrepreneurs' levels of SDO are positively associated with their intentions to become social entrepreneurs.
SDO and social cognition Bandura (1997) defines self-efficacy as the conviction that an individual can successfully accomplish a behavior expected to produce the needed outcome. Self-efficacy can be generalized or it could be devoted to a task or activity such as entrepreneurship or leadership (Bandura, 1989(Bandura, , 1992(Bandura, , 1997. Thus, entrepreneurial self-efficacy is often studied as the specific perceived behavioral control necessary to produce entrepreneurial intentions. As Wilson et al. (2007, p. 389) argue, self-efficacy "reflects an individual's innermost thoughts on whether they have the abilities perceived as important to task performance, as well as the belief that they will be able to effectively convert those skills into a chosen outcome" (as cited in Bandura's, 1989Bandura's, , 1997 self-efficacy literature). Perceived self-efficacy motivates people more than objective measures of their abilities and self-efficacy plays a key role in our behavioral intentions and then later behaviors (Bandura, 1997), including career motivation (Day and Allen, 2002). Individuals generally pursue goals in which they believe they have a chance of completing. Bandura and his colleagues have recognized a vast difference between the desire to achieve a goal vs an evaluation that one could achieve a goal. Thus, people with higher degrees of self-efficacy will be more likely to set challenging 114 NEJE 22,2 goals, persist during difficult times and have greater commitment and immersion in their field of interest (Bandura, 1997). Accordingly, self-efficacy plays a key role in the selection of careers, including the intention to start a business (Bandura, 1986;Markman et al., 2002). Vanevenhoven and Liguori (2013) have urged scholars to develop experiential entrepreneurial education programs because entrepreneurship is something learned through experience. It is also important to note that social cognition theory suggests that people develop ideas about their ability to complete goals less by performance and more by observation. Enactive mastery occurs when people start performing a task, but until that happens, environmental factors have salience. Bandura (1986) argues that it was from the environment that we gain most of the cues of what we could accomplish. Bandura and Locke (2003) have argued that "one direct way of altering perceived self-efficacy is to introduce a trivial factor devoid of any relevant information whatsoever but that can bias perceived self-efficacy" (p. 88). It is for that reason why educators in both social and conventional entrepreneurship, in general, and educational entrepreneurship, in particular, stress providing positive mental models to students (Vanevenhoven and Liguori, 2013).
Scholars have generally found that males have higher conventional entrepreneurial self-efficacy than females (Wilson et al., 2007;Zhao et al., 2005). Kickul et al. (2008) found that "self-efficacy seemed to have a stronger effect on entrepreneurial interest for girls than for boys, and that having an entrepreneurial mother or father had a significant and positive effect on girls' (but not boys') levels of the entrepreneurial interest" (p. 321). Scholars have found contradictory results for the role education plays in increasing entrepreneurial self-efficacy. Shinnar et al. (2014) found that a course in entrepreneurship increased entrepreneurial self-efficacy for males, but not females.
SDO can provide a partial explanation for these findings. Individuals who have high SDO may buy into the myths perpetuated by society. In terms of gender, men are generally viewed as more aggressive and risk-taking than women (Sidanius and Pratto, 1999). Since risk-taking, aggression and willingness to innovate are the antecedents of starting of business, it would make sense that females with higher degrees of SDO would have lower levels of conventional entrepreneurial self-efficacy, as they believe that aggression is a male concept. Individuals with high SDO would thus likely believe that women lack the ability to start businesses due to conventional entrepreneurship being highly agentic (male characteristics).
Likewise, we should expect the reverse with social entrepreneurship. The prosocial motivation provides direct contact (feedback) to beneficiaries regarding the benefits and impact (Grant, 2007). Females are generally expected to embrace this prosocial behavior due to the myths spread by society (Sidanius and Pratto, 1999). Prosocial motivation involves the desire to exert effort to benefit other individuals (Batson, 1987). Mair and Noboa (2006) suggest that empathy influences the attitudes toward behavior; moral judgment can influence social norms and self-efficacy (Hockerts, 2017). These attitudes can be explained by the various mental models that societies use to maintain hierarchy (Sidanius and Pratto, 1999). We should expect that women high in SDO would have higher degrees of entrepreneurial self-efficacy toward social matters. Scholars have demonstrated that entrepreneurial self-efficacy is an important antecedent to entrepreneurial intentions (Vanevenhoven and Liguori, 2013). We propose that entrepreneurial self-efficacy should mediate the relationship between SDO and entrepreneurial intentions. Therefore, based on the theorizing above, we propose the following:

Impact of social dominance orientation
Mentorship Mentorship refers to a one-on-one relationship between a protégé and a mentor (authority figure), which is intended to advance the personal and professional growth of the less experienced protégé (Blickle et al., 2010;Kalbfleisch and Keyton, 1995;St-Jean and Audet, 2012;Wanberg et al., 2003Wanberg et al., , p. 1897. Mentors ideally provide vicarious experience, learning and other potential career development resources (i.e. coaching, sponsorship, protection, fostering visibility, counseling and role modeling; Kram, 1985;Ragins, 1995). Mentors are especially important in providing the knowledge that individuals can succeed in their entrepreneurial endeavors, as well as providing practical education and training to budding entrepreneurs (Vanevenhoven and Liguori, 2013). Scholars have consistently found that the level of entrepreneurial intention increases with mentor interactions (St-Jean et al., 2018). It is also important to note that individuals can overcome their SDO through education and the production of more accurate mental models (Kulik and Roberson, 2008). Minorities are more likely to take on activities where they have been shown that minorities can thrive (Sidanius et al., 1996). Johnson and Smith (2016) indicate the effects of strong mentoring are remarkable, profound and enduring, "having the capacity to transform individuals, groups, organizations, and communities" (p. xiv). As previously stated, women and men generally perceive their environment differently (Edelman and Yli-Renko, 2010;Langowitz and Minniti, 2007). Scholars have long researched the role that mentorship plays in both career and success (Carroll et al., 1988;Fagenson, 1988). Individuals who have experienced mentors often have higher satisfaction and career success (Bozionelos, 2004;Day and Allen, 2002).
Mentoring plays a key function in entrepreneurship. The resources provided by mentors for burgeoning entrepreneurs may include such guidance as financial knowledge, social networking, motivational encouragement, idea refinement, special knowledge or simply friendship (Bozionelos, 2004). Mentors can provide a safe zone for a budding entrepreneur by providing vicarious experience, which is an important factor in aiding a potential protégé to develop confidence in one's own ability to achieve the intended results, self-efficacy and self-confidence. According to Bandura (1997), vicarious experience can increase self-efficacy. Krueger and Brazeal (1994) have found self-efficacy positively associated with entrepreneurship. It is possible that a mentor could provide opportunities for enactive mastery experiences to lead protégés to success, learning through performance accomplishments (Bandura and Adams, 1977;Shepherd and Krueger, 2002). Accordingly, scholars and practitioners have indicated the necessity of building mentoring relationships to budding entrepreneurs ( Johnson and Smith, 2016;Kalbfleisch and Keyton, 1995).
The increased satisfaction and career success coming from having a mentor can be for multiple reasons. Mentors and protégés exchange many resourcesincluding advice, money, better training and positive reinforcing experiences (Fagenson, 1988). When a protégé sees someone similar to themselves succeed, it would make sense that their self-efficacy would be increased (Bandura, 1997). Since mentors and protégés tend to have similarities (Ensher et al., 2002), it would make sense that female protégés with mentors would have greater confidence that they could succeed, overcoming SDO in conventional entrepreneurship.
When the female entrepreneur has a mentor, we should expect that they would have verbal persuasion, enactive mastery opportunities and vicarious experience (Kram, 1985;Ragins, 1995). This experience should lessen the impact of SDO on entrepreneurial self-efficacy. Thus, we propose the following: P5. Mentoring will moderate the relationship between potential female entrepreneurs' levels of SDO and their (a) conventional entrepreneurial self-efficacy and (b) social entrepreneurial self-efficacy such that the direct association between levels of SDO and entrepreneurial self-efficacy (both conventional and social) will be weakened ( Figure 1).

Discussion
We propose that female entrepreneur's SDO may influence the entrepreneur's intention toward social entrepreneurship opportunities and away from conventional entrepreneurship through entrepreneurial self-efficacy. We propose that mentorship moderates the relationships between SDO and entrepreneurial self-efficacy. We also examined the role of the mentor in developing intentions on the part of female entrepreneurs. In doing so, we have provided a view of the entrepreneur, as well as the mentor. Our model provides a broader picture of the individual in their social surrounding and different relationships.
In doing so, we provide an explanation why stereotypes exist and why people buy into them. We also suggest a possible way to overcome these beliefs. This paper makes the contribution of SDO into the entrepreneurial intention literature and the literature on female entrepreneurship. Despite scholars have written about many aspects, there is still a need to explain why female conventional entrepreneurship still lags behind male conventional entrepreneurshipeven in more highly developed countrieswhereas female social entrepreneurship thrives (Kelly and Williams, 2018). The explanation provided is that the residue of previous sexism and hegemonic forces combine to provide false narratives to females that they are not as aggressive or competitive as males and therefore not as inclined toward conventional entrepreneurship. Although there is tremendous merit to these explanations, and others, SDO explains why these false narratives exist. This paper seeks to explain why gender bias is present and why prejudices exist in entrepreneurship, going beyond explanations within the literature that focus on stereotypes.
Our paper focuses on why these stereotypes and myths exist. SDT could provide an explanation for myths that are used as a form of control and subtle signs of discrimination. For example, Bendell et al. (2019) propose that in certain industries, such as technology, aggressive risk-taking was a key reason why males were more successful than female in conventional entrepreneurship, even though females used more goal setting, but goals with less risk. Bendell et al. (2019) noted that given the "masculine orientation" (p. 123) of the technology industry, women may be less encouraged or accustomed to setting aggressive goals. Likewise, Gupta et al. (2019) found that conventional entrepreneurs that were "low-growth entrepreneurs are perceived as more similar to women than men, and higher on communality than agency" (p. 131). Furthermore, they also found that non-traditional men and women have problems entering into high growth entrepreneurship. They also found that modern sexism, a more subtle approach to bias, also had a relationship in separating entrepreneurship and females. These findings would be congruent with SDT.
We also extend the literature in that we consider the role of the mentor in female entrepreneurship. Discrimination and false narratives often remain in society due to the dominance of various influential actors. We argue that a mentor can make the difference in that a non-biased mentor could encourage female entrepreneurship through vicarious Impact of social dominance orientation experience, social persuasion and mastery opportunities. As we have seen in other studies, the benefits of an entrepreneurial mentor should be obvious (St-Jean et al., 2018). Vicarious experience is an extremely important consideration in building self-efficacy through the shattering of incorrect perceptions about ability (Bandura, 1997). However, a word of caution is that although role modeling can have beneficial aspects, it can also lead to narratives that may encourage gender stereotypes through the use of successful female entrepreneurs as role models (i.e. vicarious experience; Byrne et al., 2019). We suggest that mentorship should go beyond vicarious experience and mentors should consider other ways to build entrepreneurial self-efficacy.
Limitations, future research agenda and theoretical implications There are several limitations to the model above. The model proposed here is not comprehensive. There are variables of interest, in both personality and other life variables that can influence the relationships proposed. We speculate that someone who has a high locus of control and proactive personalityboth of which are related to entrepreneurial intentionscan overcome SDO similar to conscientiousness. Likewise, life or situational variables, such as the lack of other work opportunities, can influence the relationships proposed. It is common for many women to go into entrepreneurship due to a lack of opportunity elsewhere. Like other variables that are attitudinal or based on personality, situational moderators can weaken the relationship between attitude and behavior. Another issue that scholars may grapple with is whether mentoring increases or decreases the relationship between SDO and social entrepreneurship, as research indicates both relationships may be true. Scholars need to consider these outcomes in future research.
We believe that scholars should consider several avenues for future research. First, we suggest that scholars test the model, as our paper is conceptual. Going forward, we suggest that scholars seek to measure the SDO of potential women entrepreneurs. Research has attempted to understand whether entrepreneurial behaviors are learned or produced by the environment. We argue that other moderators can affect the relationship between SDO and entrepreneurial intentions. Moderators could be organizational, country based or personal. For example, from an organizational perspective, would losing one's job encourage a female to start a company, despite having SDO? From a country perspective, could a poor economy drive the need for entrepreneurship due to little opportunities available? Another issue could be a person's belief whether they are competent in their work or if their work has meaning. This could lead a person to overcome SDO. Likewise, we should expect that a female from a family of entrepreneurs, even if they have high SDO, would likely be driven toward entrepreneurship. Our model could potentially explain some of these relationships by using well-known constructs. Another potential issue that should be researched would be why women are still underpaid in self-employment (Lawter et al., 2016). Finally, scholars should examine the role of both gender and culture in these relationships. Sidanius and Pratto (1999) argue that institutions can either enhance or diminish the impact of SDO. Our focus is on social relationships in this paper, but going forward, an examination of institutional settings may play a role. For example, the extent to which a country has right-wing politics, authoritarian beliefs or strong religious beliefs may enhance SDO. Some scholars have portrayed SDO as a personality variable. As such, there are elements within society that may trigger or weaken SDO. Likewise, we speculate a country with liberal attitudes, a judicial system that protects minorities and a lack of traditional religious beliefs may influence SDO and both types of entrepreneurship.
In addition, scholars should examine the interaction between SDO and other attitude/ personality variables. We speculate that a mentor, who has high conscientiousness, proactivity and agreeableness, may provide mentorship since they would be "compelled" to do the right thing, even if they hold negative opinions of females. The reverse is 118 NEJE 22,2 also true; a mentor who has personality traits such as psychopathy, narcissism and neuroticism may strengthen the relationship between SDO and lower conventional entrepreneurship self-efficacy. Likewise, we speculate that females with higher degrees of internal locus of control, proactive personality and conscientiousness may still have entrepreneurial intentions similar to people exhibiting prosocial behavior despite unfavorable social exchange relationships.
We also urge scholars to research the role of mentors in building entrepreneurial intentions. We argue that such research is long overdue and may further clarify exactly why there are fewer female entrepreneurs than males. We hope that such a research agenda may help enlarge the number of entrepreneurs, which in turn will have a positive impact on the economy.

Practical implications
Bias and prejudice, within both potential entrepreneurs and society, need to be attenuated. It is critical that we not only are aware of the existence of bias and its effect on entrepreneurship, but that we intentionally address and eliminate bias to ensure the success of wealth creation and entrepreneurial endeavors. Yet despite the gains over the last 50 years toward building greater equality, we still have issues of bias and notions of a glass ceiling (Hechavarria and Ingram, 2016).
The practical implication that can be gleaned from the paper is the importance of mentors in developing entrepreneurial intentions. Mentors and education have been found to develop self-efficacy and, in the process, expand what we can and should do (Bandura, 1997;Vanevenhoven and Liguori, 2013). In many different endeavors, scholars have noted the role of mentors in building and shaping individual's beliefs about their abilities. We would recommend that schools in secondary and collegiate education use their various student organizations to foster mentoring. We also suggest that teachers use experiential learning to develop self-efficacy.
In addition, we suggest that regional developers provide mentoring and social network opportunities for female entrepreneurs. We believe that an understanding of the entrepreneurial ecosystem would be important. Withholding mentorship can play a role in determining whether businesses get started. We urge regional developers, universities and K-12 schools to encourage mentoring programs, select unbiased mentors and recognize the negative role of SDO to increase the level of successful entrepreneurial endeavors. A potential way would be for female entrepreneurs to provide emotional and material support in the ecosystem.

Conclusion
We make several proposals in the paper related to SDO and female entrepreneurial intentions. Our purpose is to explain the impact of gender bias and stereotypes on conventional and social entrepreneurial intention. We emphasize the moderating impact of mentorship on entrepreneurial self-efficacy and ultimately entrepreneurial intention. Additionally, this study highlights the impact of SDO on social and conventional entrepreneurial intention. Future research is also proposed. References Ahl, H. (2006), "Why research on women entrepreneurs needs new directions", Entrepreneurship Theory & Practice, Vol. 30 No. 5, pp. 595-621. Ahl, H. and Marlow, S. (2012), "Exploring the dynamics of gender, feminism & entrepreneurship: advancing debate to escape a dead end?", Organization,Vol. 19 No. 5, 119

Introduction
Growth is an indicator of new firm success (Feeser and Willard, 1990;Fischer and Reuber, 2003;Barringer et al., 2005), and founders' human capital (HC) endowments, such as their knowledge and skills, are critical elements in new venture growth (NVG) (Colombo and Grilli, 2010). More experienced founders are particularly likely to lead their ventures to early success (Delmar and Shane, 2006). HC theory argues that HC, both generic and firm-specific, is the major driver of productivity (Dess and Shaw, 2001) and has a positive effect on organizational performance (Crook et al., 2011). HC is often valuable and rare for a firm, because individuals bring unique knowledge, skills and abilities, particularly in the area of highly specialized expertise (Barney, 1991;Hatch and Dyer, 2004;Coff and Kryscynski, 2011). A firm with superior HC can therefore achieve greater economic value by accessing and utilizing employees' knowledge, skills and abilities (Coff and Kryscynski, 2011).
In the context of new firms, entrepreneurs with higher HC endowments are more capable of mobilizing and recombining various resources to exploit opportunities (Clough et al., 2019). However, the results of these empirical studies into the role of founders' HC on NVG are inconclusive (Stuart and Abetti, 1990;Westhead and Cowling, 1995;Almus and Nerlinger, 1999;Colombo and Grilli, 2005;Cassar, 2014), possibly because of a lack of investigation into the conditions under which founders' HC benefits NVG. One such condition is investigated in this studyinitial assets at founding. As a critical component of the initial stock of resources of a new organization, founding assets are mobilized or recombined by entrepreneurs to grow their businesses (Clough et al., 2019). Specifically, founding assets are hypothesized to moderate the relationship between founders' HC and NVG. This study contributes to a better understanding of the process through which entrepreneurs scale up their ventures. The interaction between founders' HC and the initial assets at founding is examined to open up the black box of entrepreneurial resource mobilization process.
In the next section, the representative works in relevant research streams were reviewed. In the subsequent sections, the hypotheses were empirically tested with a sample of 4,923 firms in the longitudinal panel data set from the Kauffman Firm Survey (KFS) for the period from 2004 to 2011, followed by a discussion of the theoretical and practical implications of the findings.

Literature review
The resource-based view argues that valuable, rare, inimitable and non-substitutable resources play an important role in the creation of competitive advantage of a firm (Barney, 1991;Peteraf, 1993;Crook et al., 2008) and that firms' sustainable competitive advantages originate from their unique and difficult-to-imitate resources and capabilities (Barney, 1991;Grant, 1996). Founders' HC endowment, including elements such as their knowledge and skills, is a critical aspect of the capability of new firms (Cooper et al., 1994;Feeser and Willard, 1990;Colombo and Grilli, 2005). Specific types of HC, such as venturing and industry experience, expedite venture creation and development (Davidsson and Honig, 2003), whereas general types of HC, such as educational level, facilitate successful initial public offerings (Dimov and Shepherd, 2005).
The Penrosean theory of the growth of the firm also emphasizes the role of managerial capabilities in firm growth and has inspired much current research on resource complementarities and knowledge-based competencies (Nason and Wiklund, 2018;Penrose, 1959;Kor and Mahoney, 2004). In Penrose's theory, some important endogenous factors in the rate of firm growth are imperfectly mobile tacit knowledge and managerial capabilities to utilize this knowledge. The focus of Penrose's theory, the combination of versatile resources, was found to be associated with more rapid growth in a meta-analysis of 113 recent studies that included 612 bivariate effect sizes and represented data on 38,815 firms (Nason and Wiklund, 2018). Decision makers need to recombine resources in creative ways to create growth; in the context of new ventures, founders' skills, education and experiences all influence their views of opportunities and ways of utilizing various resources to grow new ventures.
However, the results of empirical studies on the role of founders' HC on NVG are inconclusive. For example, mixed results have been reported on the role of the founder's education in firm growth (Stuart and Abetti, 1990;Westhead and Cowling, 1995;Almus and Nerlinger, 1999;Colombo and Grilli, 2005). Results of studies on the impact of 127 Moderating effect of founding assets experience on firm performance are also inconclusive. Experience is of two main kinds: industry experience and startup experience. Some have argued that industry experience helps entrepreneurs comprehend current trends in production processes or service delivery (Delmar and Shane, 2006), enhances their understanding of the impact of economic environment on industry growth (Mikhail et al., 1997), enables them to evaluate business opportunities within the industry (Delmar and Shane, 2006;Ronstadt, 1988) and improves their ability to forecast business performance (Cassar, 2014). Likewise, entrepreneurs can gain valuable skills from previous or serial venturing experiences (Baron and Ensley, 2006;Corbett, 2005;MacMillan and McGrath, 2006;Parker, 2006;Ronstadt, 1988;Shane, 2000;Wiklund and Shepherd, 2003). However, researchers have also presented some counterarguments; for example, entrepreneurs may gain incomplete benefits from experience, because the knowledge it generates cannot necessarily be applied to other businesses (Reuber and Fischer, 1994). Only a portion of the knowledge gained from experience can be applied to a novel and non-recurring task (Clement et al., 2007). Because each business opportunity is unique, entrepreneurs can transfer only part of the knowledge gained from experience (Bonner and Lewis, 1990;Jacob et al., 1999). In addition, entrepreneurs' emotional responses and cognitive biases may inhibit learning from experience. Individuals must evaluate prior events and alter their opinions accordingly if they are to learn from experience (Haleblian et al., 2006;Madsen and Desai, 2010), but entrepreneurs may be unable to evaluate their startup experience accurately because of emotional responses and cognitive biases (Shepherd, 2003). For example, performance feedback, as recalled by entrepreneurs, may be systemically biased (Cassar and Craig, 2009).
In addition to such counterarguments, a possible explanation of the aforementioned inconclusive results is the lack of investigation of the conditions under which founders' HC is beneficial to NVG. One such condition is initial assets at founding of a new firm, a critical component of the initial resource endowments of a new firm (Clough et al., 2019). At the scale-up stage, entrepreneurs endowed with higher levels of skills, knowledge and experiences are more capable of mobilizing and recombining these assets to expand their businesses. A higher initial resource position helps entrepreneurs with higher levels of education or venturing experiences to establish better social networks, raise financial capital and exploit market opportunities. A series of hypotheses were therefore developed to explore the interrelationships among founders' educational level, industry and startup experiences, founding assets and NVG.

Hypotheses development
Founders' HC is first hypothesized to have a positive impact on NVG because firms that possess valuable resources should perform better in the market selection process (Wernerfelt, 1984;Barney, 1991). For example, the initial stocks of financial and HC exert an enduring effect on firm performance (Cooper et al., 1994;Eisenhardt and Schoonhoven, 1990). HC, in particular, influences firm performance more than physical capital does, because knowledge assets are more difficult to trade or imitate (Youndt et al., 1996;Barney, 1991;Teece, 1998). High-quality HC possesses more complex and tacit knowledge, which is particularly difficult to transfer (Simonin, 2004;McEvily and Chakravarthy, 2002). HC has been found to be a good predictor of firm survival (Mata and Portugal, 2002;Cooper et al., 1994;Gimeno et al., 1997). In the context of new firms, the ability to develop and exploit firm-specific assets is critical to performance (Burgelman, 1994;Bogner et al., 1996;Chang, 1996). Founders with high-quality HC are likely to have better entrepreneurial judgment and consequently be better at identifying new business opportunities and integrating others' knowledge to expand their business. They are also highly likely to recruit and retain exceptionally skilled employees.

NEJE 22,2
A generic form of founders' HC, educational level, may have a positive impact on startup growth in that the formal educational experience helps entrepreneurs develop their cognitive ability, an ability that can help new firms recognize and exploit opportunities (Alvarez and Busenitz, 2001). However, mixed results have been reported on the relationship of education to firm growth (Stuart and Abetti, 1990;Westhead and Cowling, 1995;Almus and Nerlinger, 1999;Colombo and Grilli, 2005). Some studies have identified the entrepreneur's education as being positively associated with firm growth (McPherson, 1996;Mead and Liedholm, 1998), whereas others have reported that formal education does not directly relate to venture tasks, although other types of HC, such as industry and start-up experiences, do directly relate to the current tasks of the venture (Cooper et al., 1994;Marvel et al., 2014). Despite these mixed findings, education is generally assumed to equip entrepreneurs with greater knowledge and skills and consequently to help them improve their judgment in opportunity identification and business expansion, which ultimately benefit startup growth. Therefore, founders' educational level is hypothesized to be positively related to NVG: H1a. The founders' level of education is positively related to NVG.
Next, founders' industry experience is predicted to drive startup growth. Individuals' judgment can improve when tasks are clearly defined and repetitive and when feedback is provided regularly (Hayward et al., 2006;Wright, 2001). Founders can gain knowledge and skills by learning about the industry, and thus can become better at identifying business opportunities and integrating specialists' domain-specific knowledge. They can also gain relevant knowledge and skills by learning by doing (Cassar, 2014). Because entrepreneurship involves discovering facts in which entrepreneurs create and operate new businesses (Kirzner, 1997), those individuals with industry-specific experience tend to possess relevant information about such matters as cost structure, pricing and the value chain of differentiated markets in the same industry (Bruderl et al., 1992;Landier and Thesmar, 2009;Dimov, 2010). Founders with extensive industry-specific work experience therefore tends to have superior entrepreneurial judgment, and idiosyncratic entrepreneurial judgment helps to determine how an individual identifies new business opportunities (Foss, 1993;Hodgson, 1998;Alvarez and Barney, 2002). Because experience helps individuals improve in areas such as forecasting ability (Clement, 1999;Mikhail et al., 1997), entrepreneurs gain entrepreneurial judgment by learning from work and startup experiences (Baron and Ensley, 2006;Corbett, 2005;MacMillan and McGrath, 2006;Parker, 2006;Ronstadt, 1988;Shane, 2000;Wiklund and Shepherd, 2003). Furthermore, founders need to integrate and coordinate the complementary domain-specific knowledge possessed by specialists if they are to create and expand their businesses (Colombo and Grilli, 2010), and the insight gained from work experience can improve their ability to integrate the specialists' domain-specific knowledge. Therefore, founders' industry experience is hypothesized to exert a positive effect on startup growth: H1b. The founders' industry experience at founding is positively related to NVG.
In addition, entrepreneurs' previous venturing experience is beneficial in multiple ways to NVG. First, entrepreneurs gain knowledge about business creation and development by learning through experimentation (Ardichvili et al., 2003;Baron and Ensley, 2006;Delmar and Shane, 2006;Shane and Khurana, 2003;Jovanovic, 1982). In the process of learning by doing, individuals repeatedly perform the task at hand, thus increasing their expertise specific to this task (Choo and Trotman, 1991;Dew et al., 2009). Second, startup experience improves entrepreneurs' evaluation and judgment of business opportunities (Colombo and Grilli, 2005;Corbett, 2005), because they develop strong cognitive structures by reflecting on previous entrepreneurial activities ( Baron and Ensley, 2006;Gruber et al., 2008). They learn from their own misperceptions of market places and improve their beliefs with regard to their ability to precisely assess business opportunities (Parker, 2006;Shane, 2000).

129
Moderating effect of founding assets Studies have found that cognitive capability was positively associated with sales growth and profit growth (Simons et al., 1999) as well as with task performance (Hunter and Hunter, 1984). Third, startup experience helps entrepreneurs become aware of the cognitive biases in their judgment, so that past misperceptions and errors in judgment do not produce biases in their beliefs (Forbes, 2005), allowing them to better understand the entrepreneurial risks and the base rates of new business success and failure (Hayward et al., 2006). Entrepreneurial experiences also reduce the tendency to excess optimism with regard to business forecasting (Hmieleski and Baron, 2009). For all these reasons, founders' startup experience is hypothesized to have a positive effect on startup growth: H1c. The founders' startup experience at founding is positively related to NVG.
In addition, the interplay of founders' capabilities and organizational resources affects the speed at which new ventures expand their businesses. Firm assets are critical resources for the growth of a firm; firms with larger assets may operate more efficiently than those with smaller assets because their operation is closer to the minimum efficient scale (Audretsch and Mahmood, 1994), and smaller firms operate at a smaller scale because of greater cash constraints (Zingales, 1998). Even if a firm adjusts to its desirable size later, it must do so gradually because of insufficient resources (Penrose, 1959) or the uncertainty involved (Cabral, 1995). A startup with larger founding assets can endure periods of poor performance and suffer losses for a longer time because of its larger assets as well as better access to funds. Capable founders can configure resource endowments in unique combinations that create economic value, because many resources included in a startup's founding assets can offer a variety of potential services. The availability of larger assets provides founders with the opportunity to capitalize fully on their skills and abilities. In addition, the existence of underutilized and excess resources motivate founders to look for ways to use them, and entrepreneurs with better cognitive capability and more experience are able to generate more creative and productive applications from these resources. Furthermore, founders with higher HC can discover more ways of redeploying existing resources into novel and more valuable resource combinations so as to generate the specialized services of those resources as fully as possible. In contrast, resource constraints have a particularly strong negative impact on founders with higher HC during the development of their startups. In the early years, firms endure the strongest impact of cash constraints because of lack of reputation and information asymmetries (Diamond, 1989). Even founders with better education and experiences can have great difficulty expanding their business when their startups have fewer assets.
However, the counterargument might also be true. Managers with access to abundant resources may have little incentive to experiment (Sinclair et al., 2000) and may thus become less entrepreneurial (Bradley et al., 2011;Stevenson and Jarillo, 1990). Resource scarcity triggers entrepreneurial management practices that make the most efficient use of the current resources (Stevenson and Jarillo, 1990). Despite these counterarguments, in the context of new ventures, limited resources and smaller founding assets constrain the ability of even experienced entrepreneurs to recombine existing resources in novel ways so as to expand their businesses. Hence, larger founding assets are hypothesized to reinforce the effect of founders' HC endowment on NVG: H2a. The founders' level of education is more highly and positively associated with NVG when the new venture has larger assets at founding.
H2b. The founders' industry experience is more highly and positively associated with NVG when the new venture has larger assets at founding.
H2c. The founders' startup experience is more highly and positively associated with NVG when the new venture has larger assets at founding.

NEJE 22,2
Methods Data and sample Because a cross-sectional time series (panel) data design with fixed-effect controls can eliminate many alternative explanations for observed changes in NVG and minimize the risk of omitted-variable bias (Hsiao, 1986), the longitudinal panel database from the KFS for the period 2004-2011 was used to test the hypotheses. The KFS is a panel study of 4,928 new firms in the USA that began operations in 2004. The same firms were followed every year until 2011. This data set includes detailed information on the firm and up to ten owners. The respondents' answers to the survey questions were cross-checked every year to confirm their validity; therefore, the biases associated with using single-source data were significantly reduced. Five firms were excluded due to duplications or being founded prior to 2004. The final sample consisted of 4,923 firms, with 34,461 observations made over seven years.

Measures
All the variables were from the restricted-access KFS data set provided by the National Opinion Research Center through a secure, remote access data enclave. This confidential data set includes the details on founders' characteristics, new venture performance and the external business environment. To capture causal relationships between the independent and dependent variables, all left-hand-side variables were lagged by one year.

Dependent variables
New venture growth. NVG is measured by the Birch index (Birch, 1987) to reduce the bias caused by firm size: where E it is the employment of firm i at time t. Employment is used as a growth indicator because measuring firm size in terms of employment neither reflects input prices of a company nor requires deflation, as use of sales data does (Coad, 2009). Using employment as the measure also avoids the bias caused by the manipulation of reported sales and profits common in many small businesses (Cressy, 2006).

Independent variables
Three indicators of founders' HC were used in the regression analyses: education, industry experience and startup experience. In the management literature, these indicators are typically used to measure HC (Unger et al., 2011). Education is the proxy measure of generic HC, whereas industry experience and startup experience are proxy measures of specific HC.
Education. The founder(s)' education level was measured as their average level of education on the following five-point scale: 0less than high school (93 responses); 1high school (439 responses); 2some college, technical or associate's degree (1,918 responses); 3bachelor's degree (1,131 responses); and 4post-bachelor's degree (1,332 responses). The use of a continuous variable is consistent with previous studies (Cassar, 2014).
Industry experience. The founder(s)' industry experience was measured as the natural log transformation of the average number of years they had worked in the industry in which the startup operates plus one.
Startup experience. The founder(s)' startup experience was measured as the natural log transformed average number of businesses they had previously started plus one.

Control variables
State fixed effects. To control for systematic differences across 50 states, a dummy variable was created for each state. Industry fixed effects. To control for systematic differences across industries, one dummy variable was created for each industry. There are 84 industries in the sample with all the firms being classified using the three-digit North American Industry Classification System (NAICS) code.
Year fixed effects. To control for systematic differences across years, one dummy variable was created for each year.
Gender. This dummy variable equals 1 if any founder of the startup is female and 0 otherwise. Age. The average age of the founder(s) was measured as the natural log transformation of the average age plus one.
Intellectual properties. The intellectual properties of the new firm possessed in the founding year were measured as the natural log transformation of the total number of patents, copyrights and trademarks plus one. Firms that own intangible resources tend to exhibit proactive and risk-taking behaviors in the pursuit of growth opportunities (Wiklund et al., 2010;Anderson and Eshima, 2013). One source of firm growth is the accumulated stock of intellectual properties in the form of patents, copyrights and trademarks. Startups with IP rights have certain early-mover advantages (Pereira et al., 2015). For example, patents play an important role in high-growth firms (Parker et al., 2010), and in those software startups backed by venture capitals in the USA, patents positively influenced firm performance (Mann and Sager, 2007).
Market competition. Without controlling for market competition, misleading conclusions can be drawn about the role of founders' HC in NVG. Industry concentration was used as a proxy for market competition, because the sample includes multiple industries with a wide variety of structures (Geroski et al., 2010). Current firm concentration in a certain area matters; competition from other firms may limit the business expansion of new firms, because ,first, new firms cannot secure the funds required, as they lack legitimacy in financial markets (Diamond, 1989) and consequently are more likely to suffer from cash constraints (Cabral and Mata, 2003) and ,second, strong local competition can lead founders to change their growth expectations for their new businesses, although the impact of local competition may be low because firm heterogeneity limits the threat posed by industry competitors (Bertin et al., 1996). To compile data on the industry-level characteristics of the local area, the KFS data were merged with the County Business Patterns data on the basis of the county in which each firm is located and the three-digit NAICS code. Industry concentration was measured by two indicators: total mid-March employees of all the local firms in the same industry; and total number of local firms in the same industry. Both indicators were included and mean centered in the regression analyses.
Initial external equity. This dummy variable was created to indicate whether external capital was received by the startup before founding.
In addition to the above-mentioned control variables, sampling weights were also included in the analyses (Farhat and Robb, 2014).

NEJE 22,2
Results Table I presents the descriptive statistics and correlation matrix of all variables. The data show that NVG was positively and significantly correlated with founders' education level, industry experience, startup experience, founding assets, intellectual properties, initial external equity and interaction terms. It also shows that NVG was negatively and significantly correlated with founders' gender, age and the total number of local firms in the same industry. Table II presents the determinants of NVG. The independent variables and moderator were introduced in the subsequent models, following a hierarchical approach. The sample size was reduced to 12,180 observations in the regression analyses due to missing data. Model 1 includes the fixed effects and other control variables. A positive association was found between NVG and initial external equity and a marginally significant positive association between NVG and the first indicator of market competition (total current employees of the local firms in the same industry). It is noteworthy that the association between NVG and the second indicator of market competition (total number of local firms in the same industry) is significantly negative. The main effects of the independent variables were estimated in Model 2. A significant positive association was found between NVG and founders' education as well as between NVG and founders' startup experience, whereas the positive association between NVG and founders' industry experience was marginally significant. Among the control variables, founders' age and the second indicator of market competition have significantly negative associations with NVG. The first indicator of market competition and initial external equity are marginally significantly positive. Echoing the results of previous research, these results support H1a, H1b (marginally significant) and H1c. Model 3 reports the significant positive main effect of founding assets on NVG. Among the control variables, the first indicator of market competition and initial external equity has significant positive associations with NVG. The second indicator of market competition is significantly negative. Model 4 also reports positive and significant effects of all three independent variables and moderator.
To examine H2a-2c, Model 5 adds the interactions of independent variables and moderator. The results, together with the graphic analyses plotted in Figures 1 and 2, provide support for these three hypotheses. As predicted, the positive effect of the founders' education or startup experience on NVG is stronger when the startup has larger founding assets. To test robustness, the panel data set was split into two subsamples, and the regression analysis was performed on each subsample. The unreported results are robust, and coefficients quantitatively similar.

Discussion
Analysis of the results suggests a complex relationship among founders' HC, founding assets and NVG. Founders' education level, industry and startup experiences benefit NVG, and founding assets moderates this relationship. New ventures benefit even more from founders' education level, industry and startup experiences when the startups have a higher initial resource position. Previous studies have investigated the effect of founders' HC on firm performance, but few known studies have examined the conditions under which founders' HC is particularly important to NVG. Previous studies generally focused on the direct effects of various resources and capabilities on firm performance, often ignoring the interactions among those resources in the process of new firm growth. The results inform previous inconclusive findings on the role of HC in firm growth. The effect of founders' education and experiences on startup growth is contingent upon the initial assets at founding. A startup's initial resource position amplifies the positive effect of founders' HC on NVG.
Conceptually, this study advances theoretical development on firm growth. HC theory argues that people's varying knowledge and skills have economic value (Becker, 1964;Schultz, 1961). HC has been found to be vital to entrepreneurial success (Unger et al., 2011) (Bradley et al., 2012;Corbett, 2007). However, this study has revealed the different effects of various types of HC on NVG. The positive impact of founders' industry experience is not as strong as that of their educational level and startup experience on startup growth. One possible explanation is that only limited knowledge gained from industry experience can be applied to novel business opportunities, whereas capabilities and skills gained from education and venturing experiences enable founders to better mobilize initial resources to exploit growth opportunities. The empirical findings also extend the theory on NVG by showing that one type of resource amplifies the role of another resource in NVG. The prior literature has shown mostly the independent positive effects of various resources on firm growth. This study argues and empirically shows that startups grow faster when founders with high HC have more assets to utilize. The resource-based view literature was expanded by adding important new causal mechanisms, enriching our understanding of how founders' HC interact with founding assets, jointly affecting NVG. Like a big fish in a small pond, even highly educated and experienced entrepreneurs have limited opportunities to utilize their talents in a startup with a lower initial resource position. Even though much former research examined correlations between attributes of entrepreneurs and the firm performance, the entrepreneurial resource mobilization process remains a black box (Clough et al., 2019). By examining the amplifying role of the initial resource position of a new venture, this study helps us open up this black box. The results of this study can help practitioners and policy makers to understand the drivers of NVG and the interactions among these drivers. Consistent with the hypotheses, the empirical examination revealed that founders' educational level and startup experience drive startup growth and that larger founding assets intensify these positive effects. Capable entrepreneurs creatively utilize and combine existing resources to expand their new businesses. Their HC endowment, together with the initial resource position of their startup, jointly affects NVG. Entrepreneurs need to prepare carefully for the founding of a firm, because the strategic choices made at inception can have long-lasting effects. Growth-oriented startups may require a large investment in founding assets such as production facilities. Startups with fewer founding assets may find it particularly difficult to negotiate with external stakeholders and may face unusually intense competitive responses from competitors. Policy makers should tailor the support to the founding conditions of new firms. This study has several limitations. First, many new businesses in the KFS had closed by the time each follow-up survey was conducted, resulting in self-selection of businesses in the follow-up surveys. This self-selection bias might influence the statistical effects of various variables on NVG. Second, finer-grained approaches to the operationalization of aspects of HC could be possible. For example, the education measures could consider the discipline, such as business, engineering, liberal arts or other degrees, and the work experience construct could consider various types, such as experience in R&D, operations, marketing, etc. Future research could integrate theories of motivation with HC theory, because motivation provides the entrepreneur with the impetus and energy to acquire the necessary HC and to implement actions. Founders' expectations affect their behaviors with regard to expanding their businesses. Aggregate measures could be designed to represent firm-level HC. Other elements of HC could also be examined, including judgment, decision making and insight. Future studies could also explore how access to venture capital drives NVG; venture capital investors may provide resources and capabilities as coaches in addition to providing capital (Colombo and Grilli, 2010).

Conclusion
This study proposed and empirically tested the moderating role of founding assets in the relationship between founders' HC and NVG. The research findings enrich our understanding of how founders' HC interact with founding assets, jointly affecting NVG. Like a good horse with a good saddle, highly educated and experienced entrepreneurs are better able to utilize their talents in a startup with a higher initial resource position in the process of NVG.