Locus of legitimacy and startup resource acquisition strategies Evidence from social enterprises in South Korea and Taiwan

Purpose – Theoretically, the paper aims to provide locus of legitimacy as a framework to not only introduce a multidimensional perspective on legitimacy but also expand the understanding about resource acquisition strategies of social enterprises. Empirically, the authors test the theoretical predictions by using cases from South Korea and Taiwan. Practically, the authors intend to assist chief executive of ﬁ cers (CEOs) of social enterprises in their effort to secure valuable resources and provide policy implications so that both South Korea and Taiwan learn from each other. Design/methodology/approach – The authors use case methods to ﬁ nd evidence of the proposed theoretical framework. The initial search for target companies showed that social enterprises in South Korea and Taiwan were ideal samples. In-person, email and phone interviews were conducted on CEOs, and archival data on institutional environments and various aspects of social enterprises were collected. Collected data were analyzed using the locus of legitimacy framework to ﬁ nd out how different emphasis on locus of legitimacy impacted critical decisions of social enterprise, such as human, ﬁ nancial and network resources. Findings – As predicted in the locus of the legitimacy framework, the analyses con ﬁ rmed that locus of legitimacy did explain critical decisions of social enterprises in South Korea and Taiwan. First, signi ﬁ cant institutional forces existed, shaping social enterprises behavior. For example, Taiwanese Jinu showed that greater emphasis was given to internal legitimacy, while South Korean Sohwa was higher in external locus of legitimacy. Such differences systematically impacted choices made on resource acquisition strategies. Jinu showed a greater similarity to those of for-pro ﬁ t companies, aligning key decisions of resource acquisition strategies to achieve ﬁ nancial viability as a top priority. However, Sohwa, though ﬁ nancial performance was still important, put more emphasis on meeting institutional demands from South Korean Government. Originality/value This study is one of early studies that attempts to understand the structure of legitimacy faced by social enterprises. The authors argue that organizations can play a more proactive role in securing legitimacy. The authors believe that locus of legitimacy framework complements the existing understanding about legitimacy in institutional theory. By introducing a multidimensional perspective about legitimacy, the authors add additional explanations about how ﬁ rms exposed to different institutional forces can have diverse alternatives in resource acquisition strategies. liveliness of Jinhu operating income there is than it is dif ﬁ cult to say the provided enough evidence to support P2-1 , but P2-2 has convincing evidence. Thus, we have partial support for P2


Introduction
How startups acquire resources necessary for survival and growth is an essential question in entrepreneurship studies (Parida et al., 2016). Because resource acquisition strategy is relevant to the creation and development of organizations, this topic has received sufficient scholastic attention and contributions from strategic management field (Zhao and Lounsbury, 2016). Recently, theoretical and practical needs to understand entrepreneurial activities shed new lights on the significance of the resource acquisition strategy (Armanios et al., 2016).
What existing body of literature fundamentally agrees on is that organizations are motivated to participate in exchange behavior with external entities to secure resources. Traditionally, this exchange behavior involves organizations' action that is purposefully carried out to meet the demands of resource holders, forming a dependence relationship between the giver and the taker of resources (Seo, 2016). Researchers investigated the characteristics of parties involved in exchange behavior, mechanisms of transactions, antecedents predicting the success of resource acquisition and outcomes of resource acquisition strategies (Lofsten, 2016).
Existing studies have investigated external and internal conditions as factors influencing the chance of startups receiving investment, and those conditions are isomorphic pressure for startups to meet to earn legitimacy (Drover et al., 2017). Thus, for startups, securing resources has both material and symbolic meanings. However, an interesting question is how startups manage the balance between conflicting goals and demands as they try to secure critical resources (Armanios et al., 2016). This question is particularly relevant to the understanding of how social enterprises balance this tension in seeking both social and economic values.
Theoretically, we are motivated to understand the locus of legitimacy. Studies about legitimacy view legitimacy as an approval come from outside (Dowling and Pfeffer, 1975;Suchman, 1995;Wang et al., 2017). However, organizations can create legitimacy of their own. Furthermore, we push our understanding of how organizations deal with legitimacy so that they can strike a balance between different kinds of legitimacy to secure critical resources. Empirically, by studying social enterprises from two similar but different institutional environments, we intend to support the theoretical argument about the locus of legitimacy. And our last motivation is to provide managerial implications to entrepreneurs when they think about challenges from in and out of organizations and strategic choices about resource acquisition.
We propose a framework that may explain how social enterprises choose different types of resource acquisition strategies. Drawing from the lessons of institution theory, we introduce locus of legitimacy. Through a theoretical investigation, we review the contributions and limitations of legitimacy from existing institutional theory in explaining social enterprises' behavior and compare it with locus of legitimacy. We provide a set of propositions and report a short case study on social enterprises from South Korea and Taiwan. We show how locus of legitimacy explains the resource acquisition strategies of social enterprises from Korea and Taiwan.
2. Theory building 2.1 Social enterprises Social enterprises emerged to solve the problem of the market, government and voluntary failure. Social enterprises first started in the UK and later emerged in the USA. Social enterprises appeared in Asia to solve social problems during the Asian Crisis (Mingxiu and Yanru, 2011). Social enterprises were motivated to offer solutions to social problems, and Legitimacy and startup resource acquisition their actions were viewed "jointly pro-socially and financially motivated" (Dart, 2004). Social enterprises differ from traditional non-profit organizations in that social enterprises openly acknowledge their purpose as seeking economic and social goals. Because of this unique quality, social enterprises are different from ordinary for-profit startups. Though social goals are not necessarily limiting the potential of generating profits, social enterprises often face dilemmas in striking a balance between two goals and sometimes they are criticized for being not prosocial enough or not financially dynamic enough.
As a startup, a social enterprise goes through a process to become fully functional as a business organization and faces similar challenges as its more profit-oriented counterpart does. For example, founders of social enterprises identified problems in society and transform problems into opportunities. They also struggle to find ways to solve the problems and put together resources. Social enterprises and ordinary startups share more similarities and differences. Thus, the differences come not from the entrepreneurial process founders go through but from making sense out of pursuing dual goals of economic and social.

Founding conditions and imprinting
Organizations are in multidimensional contexts, and behavior of organizations are socially interpreted and given meanings following the norms of the society. New organizations will be profoundly impacted by conditions where they belong to, and the founding conditions have lasting influences (Stinchcombe, 1965;Boeker, 1989;Johnson, 2007). Stinchcombe (1965) argued that "[. . .] organizational forms and types have a history, and that this history determines some aspects of the present structure of organizations of that type". Organizations function efficiently and are well institutionalized only in some specific forms and types. History of organizational forms influences the formation of organizational structure, through which they remain relatively stable. However, if new organizations have no historical contexts, they will have difficulty making sense of how to follow the norms in the society and they will be likely to face critical problems: liability of newness.
Founding contexts is significantly related to an organization's legitimacy (Johnson, 2007;Milanov and Fernhaber, 2009). Johnson (2007) argued two phases of legitimacy: (1) the active incorporation of environmentally available elements into the new organization by the entrepreneurs engaged in its creation; and (2) reproduction of some portion or version of these elements beyond the founding phase. Johnson (2007) suggested that when an organization emerges, imitation of the "legitimate forms" would be a useful strategy to acquire and utilize more resources for the founder. Only when the organization conforms to the imprinting force will it gain legitimacy and survive. Likewise, this could be the mimetic behavior that helps an organization to maximize legitimacy (DiMaggio and Powell, 1983). Milanov and Fernhaber (2009) studied imprinting through alliance network of startups. They suggested that the network size and centrality of a startup's initial partner are significantly related to the size of a startup's current network. Potential partners of the focal organization tend to evaluate startups' attractiveness through its initial partners because startups usually do not have much performance at the time of the founding. When initial partners are firms with a good reputation and big network size, the startup might possess a big potential. Consequently, partnering with a firm that already has access to existing networks may increase the startup's legitimacy in the eyes of potential partners. In sum, founding conditions impact a startup's legitimacy and further influences its resource acquisition strategies.

Resource acquisition and legitimacy
Legitimacy is "a generalized perception that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions" (Suchman, 1995). Legitimacy is also related to organizational goals and resources (Dart, 2004). If the society legitimates an organization, it shall show activities corresponding to the goals of the organization itself (Dowling and Pfeffer, 1975) or the expectation of the outsider (Suchman, 1995). Legitimacy is determined by the operation and output as well as by the goals of the organization.
Society gives organization legitimated elements (Zucker, 1987) that are constraining or enabling organizational behavior (Dowling and Pfeffer, 1975). Legitimacy plays the role of a guideline for organizations to make decisions because only when an organization gains and maintains legitimacy, will the organization gain access to societal resources (Zucker, 1987). Legitimacy, therefore, can be viewed as a strategy guideline requiring interpretations.
Organizations survive when they gain enough legitimacy. Gaining enough proper legitimacy is one of the most significant challenges for startups. Legitimacy is a "solution" for the startup's liability of newness problem. Also, obtaining legitimacy proves social worthiness and is helpful to startups when they mobilize resources from the environment (Oliver, 1991). Thus, legitimacy plays a significant role in the process of resource acquisition, survival and growth for startups (Dowling and Pfeffer, 1975;Zimmerman and Zeitz, 2002).
In sum, the current interpretation of legitimacy is that it is given to organizations when they display appropriate qualities to the norms formulated by social consensus. If successful, organizations gain permission to operate and access resources. In a sense, the process is top-down and organizations are expected to align internal activities to the demand from essential stakeholders with the ability to grant legitimacy. Organizations seem to follow the strategic guidelines prescribed by the legitimacy and only make tactical decisions with limited freedom of choice.

Locus of legitimacy
Researchers have presented different interpretations of legitimacy. For example, Zucker (1987) argued that we can view the organization as an institution as we see the environment as an institution. Environment as institution indicates that institutional pressure comes from outside. Contrarily, an organization as institution emphasizes institutional pressure from inside. Oliver (1991) also suggested workings of multiple external pressures that constrain organizational choices: institutional perspective: non-choice behavior constrained by the invisible institution; and resource dependence perspective: active choice behavior constrained by visible tasks.
Moreover, Suchman (1995) pointed out the difference between strategic approach and institutional approach to legitimacy. The former is an organizational manager looking out, and the latter is society system looking in. We understand that legitimacy may not always come from outside but that the organization can not only give but also justify legitimacy. Legitimacy may reside inside of the organization. If the organization can assume a more active role in acquiring legitimacy Legitimacy and startup resource acquisition by focusing more on goals within organizational boundaries, it is possible that different types of legitimacy may coexist. If the dual origin of legitimacy is possibility, organizations may have more than one choice about how to deal with institutional pressures. In sum, we propose a modified understanding of legitimacy. Specifically, we argue that legitimacy has dual sources: internal and external. Because the argument is about the location where the legitimacy originates, we call it "locus of legitimacy".
External legitimacy comes from outside. Existing studies on legitimacy are about external legitimacy because legitimacy is given when organization's actions are desirable, proper, acceptable, predictable or appropriate within the socially constructed system of norms (Dart, 2004;Dowling and Pfeffer, 1975;Suchman, 1995). Conversely, internal legitimacy comes from inside the organization. Legitimacy with internal locus is not an approval by others, but a voluntarily chosen meaning of existence by an organization and its members.
If the locus of legitimacy plays its part and organizations have two sources of legitimacy, there are several scenarios about how organizations align their behavior with legitimacy from different sources. Namely, toward the external and internal locus of legitimacy, organizations may choose or be forced to choose how much attention and resources are allocated to satisfy the expectation. Thus, organizations need to deal with varying levels of legitimacy with different loci. Figure 1 shows the two-by-two matrix of locus of legitimacy.

Propositions
We look into how startup's locus of legitimacy influences entrepreneurial resource acquisition. Above reviews and discussions lead to the following propositions: P1. Locus of legitimacy has a significant impact on a startup's human resource acquisition strategy.
P1-1. The stronger the emphasis is on external locus of legitimacy, the more likely a startup is to hire employees contributing to conform to external expectations (social values).
P1-2. The stronger the emphasis is on internal locus of legitimacy, the more likely a startup is to hire employees contributing to achieving internal goals (economic values). P2-1. The stronger the emphasis is on external locus of legitimacy, the more likely a startup is to pursue the alternative source of income.
P2-2. The stronger the emphasis is on internal locus of legitimacy, the more likely a startup is to pursue operating income.
P3. Locus of legitimacy has a significant impact on a startup's network resource acquisition strategy.
P3-1. The stronger the emphasis is on external locus of legitimacy, the more likely a startup is to ally with other organizations relying on external locus of legitimacy.
P3-2. The stronger the emphasis is on internal locus of legitimacy, the more likely a startup is to ally with other organizations relying on internal locus of legitimacy.

Methods
A case study is a method that is suitable for answering how and why questions. There are some prejudices toward case study, such as lack of rigor and little basis for scientific generalization. However, a case study is an empirical inquiry that looks into a phenomenon in depth (Yin, 2009). An embedded multiple case study is viewed as a stable and reliable method. We use a holistic multiple units case design that contains two cases, one from Korea and the other from Taiwan to advance the understanding on how locus of legitimacy influence organization's resource acquisition strategies. We chose these countries for the following reasons. First, Korea and Taiwan are similar in economic, social and political conditions. Second, these countries have recently experienced a growth in social enterprise. And third, social enterprises in Korea and Taiwan are similar but fundamentally different in key areas where we believe critical to verify the relationship between locus of legitimacy and resource acquisition strategies of social enterprises.
Interviews for Korean social enterprise took place in July 2016, and interviews for Taiwanese social enterprise in August 2016. We interviewed the CEO of each enterprise. The interview began with a prepared question about the history of social enterprise. Then, the CEO talked about the institutional environment and any tangible and intangible influence from the institutional environment. Finally, each CEO was asked to describe how the enterprise was doing regarding human, financial and network resource and how they acquire these resources. Sohwa generates revenue by producing eco-friendly effective microorganism (EM) merchandise. EM contains compounded and incubated microorganisms that are safe for people and the environment. Sohwa not only trains employers with disabilities to produce eco-friendly cleanser but also provides eco-friendly product experience education to the general public. Social enterprises including Sohwa that produce environment-friendly merchandise receive more attention than before. However, low brand awareness, relatively less attractive product design and limited product offerings are the weaknesses of Sohwa. The survival chance of Sohwa in the regular market is rapidly decreasing.
Sohwa, as of July 2016, was hiring 42 employees including 36 socially underprivileged and training 22 people with disabilities. Sohwa's revenue consists of 35.6 per cent of operating income, 45.4 per cent of government subsidy, 13.5 per cent of the private donation and 5.5 per cent of other income. Sohwa relies on the geographical link of Gwangju social enterprise network, the religious link of nationwide Catholic social enterprise network and supportive link to the resource platform social enterprise network. Sohwa formed more than 90 per cent of networks by itself, and all network partners came from social-enterprisenetworks. Furthermore, among all network links, resource platform social enterprise networks are only links that contribute to sales.
5.1.2 Jinhu leisure agricultural area. Jinhu Leisure Agricultural Area is in Kouhu Township, Yunlin County. The CEO of Jinhu did not plan to operate a social enterprise at first. Jinhu started as a restaurant in February 2008. Later, Jinhu decided to pivot and embarked on social enterprise with social purposes of solving farmers' unemployment problem and promoting an eco-leisure lifestyle, and generated revenue from produce sales, restaurant operation and food and agricultural education. Jinhu has four business domains: eco-agricultural traveling, food and agriculture education, countryside-style restaurant operation and geographically unique production exhibition sales. Jinhu caught the opportunity and operated a mullet roe business and a business offering participatory experience. Also, the concept of agricultural recreation became popular, which allowed more people to recognize Jinhu. Through agricultural expos, internet marketing and brand managing, Jinhu finds it easy to show itself to the society and attract more customers. Main products include various kinds of dishes, dried mullet roe, rice-flavor noodles and agricultural produce. By combining the employment of socially vulnerable group and traditional-agricultural business with innovation, Jinhu managed to achieve economic and social goals. Afterwards, Jinhu registered as a social enterprise with service field of tourism in 2015. However, Jinhu CEO suggested that there was no significant change after getting social enterprise registration. The social enterprise registration did not attract resources nor more support from society.
Jinhu's human resource consists of 4 permanent positions and 16 temporaries including 2 persons with disabilities, 2 persons from a low-income household and 4 persons of marriage immigrants. Regarding Jinhu's revenue structure, 59.7 per cent came from operating income. The rest comes from private donation (29.5 per cent), government subsidy (5.5 per cent) and other income (5.3 per cent). Jinhu has various network relationships: 11 small-and mediumsized enterprises (SMEs), three non-profit organizations, two government agencies, one education facility and one social enterprise. SMEs are the most important partners because they are the primary source of operating income. Government agency also provides Jinhu with opportunities to become a supply partner, which is beneficial to sales. Jinhu acknowledged the expansion of networks were the result of joint efforts by Jinhu, agencies connecting the local government and social enterprises and the Taiwanese government.

Institutional environments
Social enterprises in Korea were a vehicle for a government-driven aid policy to the depressed economy during and after the Asian Crisis. Korean government places an institutional support for social enterprises. Korean social enterprises are a social institution that the government invites people to participate and do not immediately face bankruptcy because there is tangible and intangible support from the government. Korean government supports social enterprises with direct monetary assistance on costs such as operation, personnel and facility expenses. Marketing and sales assistance facilitate sales to government agencies, government invested public companies and other public organizations. Tax break is also allowed as well as support for the social insurance premium. "Social Enterprises Promotion Act" guarantees Sohwa as a social enterprise, and "Act on Welfare of Persons with Disabilities" guarantees Sohwa not only as a certified welfare facility for persons with disabilities but also with a preferred supply partner for public institutions.
Taiwan has "Program on the Action of Social Enterprises". Ministry of Economic Affairs authorizes Taiwan NPO Self-Regulation Alliance to establish social enterprise registration platform, and the alliance has the authority to decide the eligibility for registration. There is no specific statement about "social enterprise registration" in the program, lacking legal footing. Many enterprises that are not registered as a social enterprise but are doing what "social enterprise" is supposed to do. Without a formal legal foundation and government support, institutional environment is not friendly to social enterprises. Taiwanese social enterprises act like ordinary small and medium-sized companies. For this reason, motivation to act in a socially conscious way is voluntary and has little relationship with gaining support from outside. Thus, every company has the opportunity to gain social enterprise registration, and there is a possibility that a real social enterprise is not registered.
Social enterprises in Taiwan are different from most of other countries' because social enterprises in other countries are top-down type while in Taiwan they are bottom up. While social enterprises in the United Kingdom, Hong Kong and South Korea grew from organizations that were relying on resources from government, social enterprises in Taiwan grew from the bottom of the society. In Taiwan, individuals recognize social problems and work their way out to find a solution without relying on public recognition or support. The grassroots nature and high internal locus of legitimacy are distinguishing traits.

5.3
Resource acquisitions strategies 5.3.1 Human resource acquisition. Sohwa with the aim of providing job opportunities to the disabled hires 42 employees including 7 permanent and 35 temporary positions. Sohwa hires 35 disabled so that it is qualified to keep social enterprises certification and also be a certified welfare facility for persons with disabilities. Sharing Sohwa's vision is not a primary qualification for employment. Jinhu employs 16 temporary positions out of 20 employees. Jinhu hires two persons with disabilities, two persons from a low-income household and four persons of marriage immigrants. Although Jinhu aims to provide job opportunities for jobless farmers, it needs talented people to stay profitable. Jinhu places a high value on sharing the same vision with job applicants. Whether a job applicant approves of the philosophy of social enterprise is vital. In sum, Jinhu has built its human resource with 40 per cent of the vulnerable and 60 per cent of talented employees with same organizational vision.
5.3.2 Financial resource acquisition. Operating income contributes 35.6 per cent of Sohwa's and 59.6 per cent of Jinhu's total income; government subsidy and private donation collectively become 58.9 per cent of Sohwa's and 35.0 per cent of Jinhu's total income. Other Legitimacy and startup resource acquisition incomes in both enterprises are at the same level. Sohwa's operating income is low compared with government subsidy which contains support from Ministry of Employment and Labor and Ministry of Health and Welfare. Without the external support, it would be difficult for Sohwa to survive. Though Sohwa emphasized the importance of operating income, it is a fact that Sohwa is on a life support from the government. As for Jinhu, operating income makes almost 60 per cent of total income. Jinhu emphasizes income from business activities for three reasons: survival, independence and expansion. To survive in the environment and become independent and grow, it is necessary for Jinhu to increase operating income. Jinhu has 5 per cent of government subsidy and 30 per cent of the private donation. Social enterprises in Taiwan can apply government subsidy, not through the name of social enterprise but industry membership. What is intriguing is that private donation from people, associations and SME's is relatively high in comparison with government subsidy. 5.3.3 Network resource acquisition. Sohwa allies with other social enterprises including Gwangju social enterprises, Catholic social enterprises and resource platform social enterprises. Sohwa values these networks for they are partners who share similar difficulties and may become each other's consolation in difficult times. However, Jinhu allied with several kinds of organizations and the big emphasis is on the relationship with SMEs. In Jinhu's case, alliances with SMEs, social enterprises, education facilities, government agencies and non-profit organizations gave valuable chances to carefully evaluate how business partners and customers see Jinhu. By prioritizing choices that could help increase operating income, Jinhu was able to benefit from a relationship with SMEs and government agency alliances and became profitable. Jinhu confirmed that network resources are actual advantages rather than pretty decoration to the company.

Discussion
Sohwa is in a friendly institutional environment. With government-issued certifications, Sohwa is readily accepted by the society. Sohwa has a high external locus of legitimacy. Jinhu is a registered social enterprise in Taiwan. However, organizations with social enterprise registration do not look any more proper, desirable or better than those without the registration. Many organizations in Taiwan that exercise social enterprise philosophy do not have registration. Though the Taiwanese system that issues the social enterprise registration, it is titular at best. As Jinhu's behavior is similar to that of ordinary business companies, Jinhu's external locus of legitimacy may not be very high. This discussion leads to view Sohwa as an organization with high external locus of legitimacy and Jinhu one with medium-low locus of external legitimacy.
In Sohwa's case, it maintains a big emphasis on external locus of legitimacy but pays low-to-medium attention to an internal locus of legitimacy, which can be understood as Sohwa's status is more given than chosen. Jinhu maintains medium-to-low level on external locus of legitimacy but a high level of internal locus of legitimacy, which suggests that Jinhu's status is more chosen than given. Both Sohwa and Jinhu pursue social purpose and business purpose simultaneously. Nevertheless, under a different level of external and internal locus of legitimacy, they respond to the environment with different strategies.
Startups with greater dependence on external locus of legitimacy than the internal locus of legitimacy tend to hire employees to fulfill the external expectation. Contrarily, startups with high internal locus of legitimacy will hire employees who can be part of achieving organization's goal.
Regarding human resource acquisition strategy, Sohwa is high on external locus of legitimacy and lows on internal locus of legitimacy. Sohwa not only wants to but need to hire socially vulnerable people because institutional environment requires Sohwa to maintain the status of a social enterprise and a welfare facility for persons with disabilities. When hiring new temporary employees, the first requirement is that he/she must be a person with a disability, regardless of the person's knowledge about business, or the sharing of Sohwa's vision. Sohwa hired some employees to fulfill the institutional expectation. Above observations confirm P1-1.
Jinhu is high on internal locus of legitimacy and has a medium emphasis on external locus of legitimacy. Jinhu values job applicants who can share Jinhu's goals. General knowledge about social enterprise and acceptance of dual organization purposes are essential as well. As it is a business company, it must follow Taiwanese business law. The society defines Jinhu as an ordinary business company, so it does not have to obey other rules except for the business law. Jinhu suggested that even though they want to give help to every jobless farmer, it is beyond its capacity. Therefore, Jinhu hired as much the vulnerable as it could, while other employees must be business-savvy and share same goals with the organization. By doing so, Jinhu provided occupations to the vulnerable and stayed profitable with the contribution from employees with business knowledge and skills. In sum, P1 has supporting evidence from the analysis.
Startups with a higher degree of external locus of legitimacy than the internal locus of legitimacy tend to neglect profitability, and startups with a greater emphasis on internal locus of legitimacy tend to emphasize activities contributing to profit generation.
When Sohwa was qualified as a social enterprise in 2012, Sohwa's income consisted of 29.6 per cent of operating revenue, 53.8 per cent of government subsidy, 12.4 per cent of the private donation and 4.2 per cent of other income. Emphasis on external locus of legitimacy brings more outside financial support. Though operating revenue is always less than the sum of government subsidy and private donation, it is difficult to conclude that startups with high external locus of legitimacy than the internal locus of legitimacy are all paying less attention to become financially viable.
With relative minimal support from the outside institutions, Jinhu must emphasize on business activities to be self-sufficient and financially viable, and thus operate independently and expand its business. Jinhu's case confirms that startups with a higher degree of internal locus of legitimacy will tend to emphasize the liveliness of core business activities. Both Sohwa and Jinhu view operating income important, but there is more difference than similarity. For Sohwa, it was just income, but for Jinhu, it was a lifeline. In sum, it is difficult to say the analysis provided enough evidence to support P2-1, but P2-2 has convincing evidence. Thus, we have partial support for P2.
Startups with high external locus of legitimacy tend to ally with organizations of similar responsibilities to institutional pressures, while startups with high internal locus of legitimacy tend to ally with organizations with resources necessary for survival and growth.
Sohwa established a network of local social enterprises, Catholic social enterprises and social enterprises which aim to support social enterprises. Sohwa said that there wasn't any chance to ally with ordinary profit-seeking companies. Furthermore, the public viewed a social enterprise with skepticism if the social enterprise has ties with for-profit companies, preventing Sohwa from diversifying its network relationship.
Jinhu allied with organizations that generate operating income. The CEO of Jinhu seemed to understand that network participation incurs costs and that is why Jinhu put greater emphasis on generating income from the relationship with network partners. Jinhu established a strong network resource base so that Jinhu used and reciprocated the resources committed by network partners, Jinhu presented itself as for-profit organizations rather than a kind organization. Evidence shows that a startup's network resource acquisition has a meaningful relationship with locus of legitimacy and confirms P3. Figure 2 shows the results of the analysis.

Conclusions, implications and limitations
We reviewed relevant literature to build a theoretical foundation for the concept of locus of legitimacy and three propositions about the relationship between locus of legitimacy and startup's human, financial and network resource acquisition strategies. Analysis of cases about Korean and Taiwanese social enterprises demonstrates that startup's human and network resource acquisition strategy are highly related to locus of legitimacy. We found partial support for the proposition on the financial resource acquisition. Theoretical and empirical analyses show that startups with a high degree of external locus of legitimacy tend to hire employees to meet the institutional expectation from the outside and tend to ally with organizations with similar legitimacy. However, startups with high internal locus of legitimacy tend to hire employees who can fulfill organization's goal to survive as business organizations and tend to ally with organizations with contributing resources.
We contribute to the literature on legitimacy in particular and institutional theory in general. Legitimacy in a conventional view is, in fact, legitimacy with external locus. There is the internal locus of legitimacy that exists within an organization. Locus of legitimacy can also provide an additional explanation for decoupling. As decoupling occurs when institutional demands and functions of the organization are incompatible (Meyer and Rowan, 1977), decoupling assumes that institutional demand comes only from the outside: external locus of legitimacy. Decoupling (Meyer and Rowan, 1977) can explain how organizations protect their technical core by creating narratives why organizations do things differently despite the institutional pressure from the outside. However, decoupling argument may not provide sufficient explanation for the emergence of self-motivated social enterprises in Taiwan. Moreover, decoupling may not give a compelling story about Jinhu's internal logic of pursuing social and economic goals simultaneously without telling different stories to outside observers. In this regard, we cautiously believe that a different idea such as locus of legitimacy is needed to explain how internal locus of legitimacy influences organizational and entrepreneurial decisions. Locus of legitimacy suggests that both external and internal legitimacy influence organizational decisions. Thus, we proposed locus of legitimacy as a lens to contribute to the existing legitimacy studies.
There are three main practical contributions. For social enterprises in Taiwan, we offer the policy implications about whether formal legal boundaries are to establish or whether the government needs to remain as a silent helper or to become a more aggressive aid. Although researchers in Taiwan suggest that the government and the corporations are two entities that shall never get directly involved in social enterprises, Taiwan may learn a thing or two from the Korean government's choice to establish the formal institution. Second, for social enterprises in Korea, they shall think more about the nature of the social enterprise, a business organization with social purpose. As social enterprises in Korea enjoy lots of support by paying all the attention to an external locus of legitimacy, they can concentrate on exploring and cultivating internal locus of legitimacy. By enhancing internal locus of legitimacy, social enterprises in Korea may meaningfully pursue economic and social goals. Finally, both Korean and Taiwanese government may have a balanced view about social enterprises as there will be more social enterprises in the future.
There are several limitations to this study, which may be overcome by future endeavor. First, the concept of locus of legitimacy is still in its infancy and may require further development. For instance, it is yet to be examined and discussed whether the locus of legitimacy has any fit with existing institution theory. Second, we use case analysis to corroborate the theoretical claims. Future research could verify our claims with more rigorous methods such as rigorous measurements. The third limitation is that in this research, it only looked into human, financial and network resources because these resources are essential for startups. Future research shall look into other aspects of the organization. Legitimacy and startup resource acquisition