Internal and external determinants of corporate social responsibility practices in multinational enterprise subsidiaries in developing countries: evidence from Ethiopia

The purpose of the study is to investigate the internal drivers and external determinants of corporate social responsibility (CSR) strategies practiced by multinational enterprises’ (MNEs) using institutional theory and the resource-based view. We propose that MNEs’ local business orientations and political behaviors are highly related to firms’ CSR selection as internal and external determinants, respectively, and we integrate the actual CSR practices from different characteristics to exploring and exploiting strategies. A multiple regression analysis was conducted by using survey data collected from MNE subsidiaries operating in Ethiopia. We found that firms who are oriented to raise competitiveness with collaborative attitudes toward host governments seek more exploring CSR practices. On the other hand, firms’ local orientation that is to build legitimacy with accommodative attitudes toward host governments has positive relations with exploiting CSR practices. Furthermore, contrary to the expectation the result revealed that exploiting CSR practices are positively related to both orientations, which implies that traditional CSR practices are treated fundamentally of importance in developing countries.


Introduction
Corporate social responsibility (CSR) is becoming increasingly important in today's business world. None of the international firms can ignore social responsibility, regardless of where they operate. Regardless of willingness, CSR has become an important business agenda for international firms. Interests in this subject, however, started much earlier in the research field, though the concept of CSR appeared differently according to social and economic contexts of times.
Bowen [16,17] used the term 'social responsibility of businessmen' and discussed the obligation of businessmen to pursue values that contribute to society. CSR primarily refers to the continuing commitment by organizations to behave ethically and to contribute to economic development. CSR is defined as the responsibilities of corporate toward the society within which they are working. Carroll [21] redefined this concept further with economic, legal, ethical, and philanthropic responsibilities. He argues that business and society enter into a social contract, and business is obligated to considered society at large another use full CSR definition was presented by Wood [101], which argued that society expects certain business behavior and outcomes, that business and society are interwoven rather than distinct entities. This definition goes beyond the identification of different types of responsibilities and incorporates various theoretical perspectives into a coherent model of CSR. Recently,

Open Access
Future Business Journal *Correspondence: gulma19@yahoo.com the European Commission [29], states that 'to fully meet their corporate social responsibility, companies should have in place a process to integrate social, environmental, ethical, human rights and consumer concerns into their business operations and core strategy in close collaboration with their stakeholders. ' The World Business Council for Sustainable Development defines CSR as 'the continuing commitment by business to behave ethically and contribute to economic development while improving the quality of life of the workforce and their families as well as of the local community and society at large. ' While the construct is still fragmented, it again became an increasingly popular subject receiving a great deal of attention from scholars. To sum up, CSR practices considered in the study include Community Charity, Corporate Governance, Diversity, Employee Relations, Environment, Human Rights, and Product Quality and Safety.
Various studies have evidenced that CSR and foreign direct investment (FDI thereafter) are strongly correlated [1,23,45,48,49,103]. Specifically, investment-based ties are more salient than are trade-based ties in forming institutional pressure because the expansion of subsidiaries overseas further embeds firms in the host-country environment [60]. Specifically, when making decisions, MNEs are encouraged to internalize manifold institutional arrangements of foreign subsidiaries [50]. In addition to foreign direct investment, CSR is considered to be a spillover effect of the foreign expansion of MNEs.
CSR can facilitate the business in various ways if handled right. According to Hart [39], the deeper level of social and environment strategy may facilitate the development of rare organizational capabilities and explore new markets that contribute to the companies' sustained competitive advantage. Certain types of CSR practices reflect the companies' efforts to establish the trust-based and cooperative firm-stakeholder relationship, which could add value to the company's financial and social performance [10,59,105]. Moreover, CSR can be seen as public relation effort. It adds values to firms by stabilizing and maintaining a good corporate reputation, increased customer loyalty reducing environment impact and strength the company and regulatory bodies relationships to reduce regulatory burden on firms (specialty for MNEs).
The extant study more emphasized on the concern about CSR growing significantly in recent decades, a bulk of theoretical and empirical studies have focused on the consequences of CSR, linking CSR with corporate financial performance, consumer loyalty, and corporate reputation [18,77]. In contrast, strikingly inadequate attention has been paid to the antecedents of socially responsible or irresponsible corporate behavior. According to a meta-analysis of CSR literature by Margolis and Walsh [62], 85% of the studies in this field treated socially responsible behavior as the explanatory variable, calling for more serious inquiry into CSR as the dependent variable.
However, there is still know very little about the impact of integrated determinants of MNE decisions for choosing CSR strategies that consist of complex characteristics. There was a study that focuses on the internal and external environmental factors, for example, Bansal [9], a study on the integrating internal and external factors affecting the environment, Yin [104] institutional drivers for CSR by considering top management commitment and organizational culture, Dupire and M'Zali [28] study on CSR Strategies in response to competitive pressures. They focus on the external and/or internal influences on a single aspect of CSR.
This study focuses on how internal and external factors influence multinational companies when deciding different CSR activities that are neither fully explained, nor conceptually reviewed, by extant literature. Moreover, while studies on CSR abound in the developed countries, evidence from a developing country's perspective appears to be limited [44,97]. Particularly, CSR studies in low-level income counties still sparse.
Therefore, this study filled these gaps and attempted to enrich existing understanding by providing empirical data and suggesting relationships between different factors that determine CSR, and different characteristics of CSR practices that MNEs choose.
This paper builds on studies of MNE strategies in foreign markets while presenting an integrated perspective to analyze relationships between MNE business orientations together with political attitudes against the host governments, and different forms of CSR practices. We divided explanatory variables into two categories: internal motives and external determinants. For internal motives, firms' business orientations are selected, which are composed of competitive advantage and legitimacy, which rest on a resource-based view and institution theory, respectively. Concerning the external factors that influence MNEs' CSR strategies, the political attitude of firms from two dimensions that depict the way MNEs behave against a host government was selected, which are collaborative and accommodative relationship strategies.
Institutional theory and the resource-based view were adopted for this study. The institutional theory argues that the institutional environment is more powerful, and often deeper than market parameters, in influencing the development of formal structures in an organization [26]. In the context of CSR, institutional isomorphism suggests that a firm adopts CSR strategies within the whole organizational strategy of the firm, following the various isomorphic pressures [43].
Resource-based views give particular emphasis to the internal factors which explain performance differentials between corporations and how they can benefit from such differentiation. Increasingly, corporations and their managers believe that engaging in CSR activities and being seen to do it is likely to enable them to achieve some kind of competitive advantage, which means enhanced economic results. The resource-based view allows us to identify the circumstances when it pays to adopt certain practices, such as engaging in social responsibility activities. It offers important insights on how CSR can influence a corporation's economic performance.
Approved by many theorists, together they complement each other and help explain how organizations deal with the external institutional environments, and manage internal resources strategically for sustainable development [9,22,81]. Supporting this argument, recent studies in socio-economic literature increasingly began integrating the two theories [9,81]. The resource-based view values a firm's capability to create and deploy its unique resources to generate economic rent' that indicates above-normal returns, accruing to hard-to-copy, and a firm's specific capabilities [82,98]. Those resources must be valuable, rare, imperfectly imitable, and nonsubstitutable [12]. Despite this important insight, the resource-based view alone is deficient to explain a firm's sustainable development. Peng et al. [81] proposed an institution-based view as one leg that complements the 'strategy tripod, ' which consists of the industry-based and resource-based views as the other two. Therefore, they argued, a firm's sustainable development depends on its ability to manage the institutional contexts and appropriate resource selection. In the same vein, this paper sheds light specifically on MNEs' choosing of different CSR activities in relation to their internal business orientation and external relationships with host governments examined through an institution theory and resource-based view.

Concepts of corporate social responsibility
The concept of CSR was firstly introduced during the Great Depression of the U.S. in the late 1920s. Since then, social and political impacts of business on society have grown enormously, whereas conflicts between two spheres also have deepened.
Along with the expansion of the scope and influence of the multinational enterprises, actions for CSR became pervasive. Specifically, sustainable development of business management asserted by the international communities drove CSR to become a prevalent research topic in academic and practical fields. Moreover, the spread of NGOs as supervisory agencies and the development of information and technology proliferated CSR as a part of critical strategies to incorporate into the business operation. Despite the increasing volume of researches on CSR, the sphere of the definition grew bigger, leaving loopholes for various interpretations by different perspectives, rather than boils down to a clear-cut explanation [45].
Carroll [20] elaborated the concept further with economic, legal, ethical, and discretionary responsibilities. He argued that the business institution is the basic economic unit in nature. Hence, its foremost social responsibility is to produce goods and services that society wants. Along with the economic responsibility, business units also need to fulfill legal requirements to operate within society. Although it is hard to define, ethical responsibilities embrace norms, the consensus of the customers, and society's expectations that go over and beyond legal requirements. Finally, philanthropic responsibilities leave rooms for business units, which include volunteering activities at firms' discretion. For example, firms can fulfill their philanthropic responsibilities by promoting community programs, providing financial supports for raising the humanitarian values of the society, and encourage employees to participate in philanthropic activities [21].
In the advent of the new century, a constellation of factors emphasized the importance of CSR and expanded the concept by applying CSR to overall business operation not only customer relations but also employees and external stakeholder management. Particularly, Porter and Kramer [84] linked the concept of CSR to the competitive advantage of firms. They argued that corporations should involve in society with a strategic CSR approach that would bring capabilities to improve salient areas of the firms in the competitive business environment. Soon, they expanded the strategic approach and proposed Corporate Shared Value (CSV) that 'CSV can enhance the competitiveness of a company while simultaneously advancing the economic and social conditions in the communities in which it operates' [84]. On the other hand, Kotler and Lee [51] aligned CSR with a marketing strategy that firms' socially responsible activities would enrich the welfare of local communities through the business operation and resource allocation [51].

Theoretical developments of CSR
Many studies, which attempt to analyze CSR within theoretical frameworks, have become widespread, including agency theory, stakeholder theory, corporate citizenship, institution theory, and the resource-based view from strategic perspectives. Among the listed theories, the agency theory represented the opposite view against all. Davis [25] exemplified two economists who are against and for CSR. First, Friedman contended that carport's main purpose is to maximize the shareholder values and any social demands that only impose a cost on the company without increasing shareholder value should be rejected. The agency theory [86] is the most popular way to explain Friedman's assertion that is the shareholder value maximization must be the supreme criterion for corporate decision-making. They criticize that responding to social demand could expand the power of managers in an unproductive way when managers have a strong personal interest in social issues, they only increase agency costs in the economic system. On the other hand, Davis [25] commented about another distinguished economist, Paul A. Samuelson, in the same monograph to argue that the latter emphasized the social responsibilities of the corporation. Notably, Freeman [32] is known for establishing grounds of CSR upon the stakeholder theory by arguing back what Friedman asserted regarding business responsibility. He contended that firms should satisfy a variety of stakeholders including employees, customers, suppliers, and local community organizations. CSR activities engaging with these non-financial stakeholders are important because they could influence the firm's outcomes. Therefore, it is not sufficient for managers to focus solely on the needs of stockholders' interests, he argued. In the same vein, the current researchers have supported this idea through studies on interactions, integration, and building better relationships with various stakeholders [27,41,108]. Along with stakeholder theory, corporate citizenship theory also contributed to enhancing existing knowledge on CSR. Corporate citizenship theory expanded our views in that a firm bears responsibility as a member of society in nature so it should behave as a good citizen [100]. Hence, corporate citizenship theory focuses on business rights, legal responsibility, and possible replacement of the traditionally most powerful institution, the government.
Porter and Kramer [85] ignited strategic CSR discourses that businesses should incorporate social responsibility into their core business strategy by presenting creating shared value (CSV) that integrates firms' business profit and competitiveness with social responsibilities to communities. Firms leverage their expertise and unique resources to create economic value that can be earned from social values. The concept of sustainable competitive advantage has also been aligned with appropriate CSR strategies within a resource-based view in the previous studies [64,67]. Building on the Resourcebased view framework, issues were discussed regarding CSR's strategic implications and its possible to earn a competitive advantage for firms [67]. Theorists of RBV argue that firms pursue rare, non-substitutable, inimitable, and specialized resources, which introduce the value of non-market resources [9,12,42,88]. Such capabilities are determined by a firm's resource scarcity, uniqueness, durability, inimitability, and non-substitutability [11,82]. Competitors fail to mimic or duplicate these resources due to hurdles above which are described as factor market imperfections. Factor market imperfections illustrate the imperfectly competitive factor markets where different firms implement different strategies to buy and sell resources with various expectations about values that would return to their strategies [11,90]. Firms that obtain these resources can maintain its capacity to generate above normal economic rents, which will sustain their competitive position.
Specifically, the RBV theory is an effective tool for studying corporate social responsibilities as a business strategy. The theory has a strong focus on performance as the key outcome variable. Many practitioners and scholars adopted the RBV framework to question how CSR activities raise or are linked to a firm's financial performance [36,65,77,96]. Additionally, it acknowledges the importance of intangible assets of a firm such as research and development, human resources, corporate culture, and reputation, whose attributes can also be found in CSR literature [88]. They argued that CSR could be a source of a firm's competitive heterogeneity that influences customers or other stakeholders.
In the mid-2000s, the literature that adopted institutional theory to understand CSR-related phenomena emerged which broadened the range of conceptual tools in CSR research. Institutional theorists regard CSR not just a realm of purely voluntary actions, but rather collective forms of self-regulation affected by the market, state regulation, and social norms [19]. Institution theory also acknowledges the importance of 'social contexts' such as normalcy, tradition, and culture [75]. The process of institutionalization is strongly tied to history [19]; therefore institutions reflect values and norms which the society has struggled with to achieve. For example, Bondy et al. [15] witnessed that CSR is not only institutionalized in society, but it also appears as a form of the institution within the multinational enterprises. They interviewed professionals of UK firms and found that CSR became more 'business as usual' within the organizations.
In parallel to the acceleration of globalization, evidences are witnessed over the world that the rise of political CSR is bringing the integration of political and social domains for business entities such as self-regulation on labor and environmental areas operationalized by the multinational enterprises [22,33,52,89]. Even though a government usually drives institutional arrangements for CSR as a form of top-down flow in many developing countries, 'institutions are more open and pervious to the corporate strategic action than is often allowed for in the literature' [22]. The rise of political CSR and its conceptual novelty lies in integrating effort of a great stock of knowledge that has been piled up in separated and parallel ways from politics, and social domains of non-market literature [33,99]. The rise of political CSR becomes a new reference to judge the level of trusted corporate governance in the globalized business world.
Taken together, firms refer to CSR now as contextspecific actions and policies that take into account stakeholders' expectations and touch the triple bottom line of economic, social, and environmental performance [3]. In other words, CSR became a new paradigm for business to embrace and an important strategic factor for sustainable development.

The multinationals and CSR practices
The operation of MNEs in developing countries became a target of grassroots activities regarding human rights, labor, environment, and improved social conditions. Close surveillance by the international NGOs on MNEs caught a few well-known social incidents. In 1996, Nike was accused of using child labor in the production of its soccer balls in Pakistan. While Pakistan has laws against child labor and slavery, the government has taken very little action to combat it, and the supplier contracted with Nike exploited the loopholes of the government policy. The company regained credibility from markets only after formulating a Code of Conduct for its suppliers that required them to observe some basic labor and environmental/health standards. Started with this case, ethical problems, caused by the multinationals outsourcing their products in factories and countries where low wages, poor working conditions, and human rights problems were rampant and were discovered.
Until recently, dishonorable operations of a few multinational enterprises in foreign markets such as the indiscriminate development of natural resources and violating human rights have taken place. Nonetheless, there are a few successful cases that earned monetary compensations from the multinationals through persistent court debates and long-term campaigns, mostly led by NGOs. About the recent phenomena relating to the multinational corporation's ethical misdeed in the West Amazon regions such as Ecuador and Peru, we could learn that these conflicts are associated with reckless developments of a natural resource. During Texaco's (now Chevron) operation in the Oriente, the company intentionally released more than eighteen billion gallons of toxic waste into rainforest waterways, wetlands, and subsoil, threatening the lives of over twenty-five thousand affected individuals. Amazon Watch, the NGO organization, founded to protect the rainforests and advance the rights of indigenous peoples in the Amazon Basin, has been running several campaigns to call for practical actions from international citizens against the U.S. oil company, Chevron (former Texaco), in Lago Agrio region of North-East Amazon of Ecuador since 1996. After nearly two decades of litigation, the gigantic company has been found guilty by Ecuadorian courts, and the court ordered them to pay $9.5 billion to the damaged region. This is the first case that a U.S. oil company was sentenced by a foreign legal authority for paying orders. The lawsuit was filed in the U.S. Court of New York in 1993, but the U.S. court has dismissed the case in 2001 [47].
On the other hand, Oxy, a local subsidiary of British conglomerate Reckitt Benckiser, produced and sold a toxic disinfectant for humidifiers in South Korea, causing 400 victims including 51 people who died after coming into contact with the product. Nonetheless, all Oxy products were boycotted from large and small markets in South Korea due to enormous public resentment [54].
To alleviate such activities international organization declares different guidelines. OECD was the first international organization that declared guidelines for CSR of the multinationals in 1976. While the guidelines went through several amendments in 1979,1982,1984,1991, and 2011, the area of application has been expanded covering not only the member countries but also the foreign territories in which the multinational enterprises of the member countries operate.
Along with OECD guidelines, the international labor organization (ILO) established the international standards of CSR focused on labor issues. Nonetheless, the Tripartite Declaration of ILO became the first international document that applies CSR concepts to the broadest area of the multinational business and is most frequently used in interpreting international disputes regarding CSR issues [109].
The UN Global Compact was proposed by the United Nations Secretary-General in 1999 and seeks to advance socially and environmentally responsible business citizenship so that business can contribute to the challenge of globalization. Global Compact consists of nine principles based on The Universal Declaration of Human Rights, The International Labor Organization's Declaration on Fundamental Principles of Rights at Work, and The Rio Declaration on Environment and Development. Global Compact intends to promote a network between various social stakeholders and business citizenship suggesting institutional learning practices among participants [5]. It is very complex to measure and evaluate CSR performances regardless of their size or line of business. The International Organization Standardization (ISO) created ISO 26000 on social responsibility to guide all types of organizations through a myriad of approaches to social and environmental issues.

Firms' strategic responses to CSR pressures
The mainstream of CSR literature regarding strategic responses have been discussed within the institutional theory [2,46,79] by limiting the span of CSR strategic responses to conformance and resistance. To explain more proactive CSR responses such as initiating environmental movement in the developing countries and value-creating activities, the convergent theoretical approach is necessary, such as a resource-based theory. In fact, some CSR scholars explored firms' responses beyond conformance and resistance and had discussed typologies for the different spans of CSR strategic approaches. For example, Van Tulder et al. [94] have proposed a division among inactive, reactive, active, and pro-/interactive CSR approaches, while Heikkurinen [40] has suggested five levels of CSR responses: passive, reactive, proactive, entrepreneurial, and creative. Mirvis and Googins [69] divided corporate citizenship into five stages of analyzing a corresponding number of issue management approaches: defensive, reactive, responsive, proactive, and defining. Lee [53] proposed four different CSR strategies: obstructionist, defensive, accommodative, and proactive. O'Higgins [74] has made a distinction among skeptical, pragmatic, engaged, and idealistic company configurations. Misani [70] also differentiated convergent CSR that are CSR practices already adopted by other companies in the industry, and divergent CSR is a unique CSR approach used to gain competitive advantage. Together, CSR strategic responses in CSR literature are evolving from the limited span that used to refer conformance or resistance behavior to more comprehensive responses including proactive, competitiveness-creating responses to external pressures.
However, it is not clear whether the classifications illustrated above imply the same level of CSR involvement in business reality which could seriously harm the generalizability of the analyses. For example, about 'reactive, ' Van Tulder et al. [94] used the word to describe a firm's minimal participation in CSR by not violating any law or doing any harm, whereas Heikkurinen [40] described that a firm's reactive CSR aiming at maintaining the firm's competitive advantage which suggests the more proactive status of moral support in Heikkurinen's adoption of the word. Moreover, on a 'defensive' response against CSR institutions, Lee [53] understood it as a strategy to reject broad ethical responsibility and limited corporate responsibility to protect their self-interest by abiding law and regulation. However, others employed the same term, 'defensive, ' to describe how firms safeguard their reputation and satisfy employees and other stakeholders [87]. Runhaar and Lafferty [87] pushed the boundary of 'defensive' toward a more ethical sphere.
Institutional theory in business studies focuses on the establishment of consensual understandings of organizational conformity. Therefore, firms may replicate their market strategy in the management of CSR [43] and the scope of a firm's choice is restricted from resistance to conformance which implies a deterministic system while allowing little room for the agency [63]. Resource-based theorists put more emphasis on the instrumentality in seeking social worthiness. To overcome limited and confusing terminology, based on institutional and RBV we attempted to visualize different CSR strategies by structuring two practices.
Therefore, we propose the broader frameworks for the empirical test in this study, which are exploring and exploiting CSR Strategies adopted from March [61]. Influenced by internal and external orientations, the multinationals may choose their CSR activities that challenge existing policies with assertive attitudes, hence implement innovative solutions on social-environmental issues, or pursue a passive strategy within an established policy structure, such as resource allocation.
The explorative strategy may seek to discover an innovative solution while not expecting an immediate result. Firms which follow a path of exploration are searching for new ways of doing things and like to experiment by incorporating what they have discovered within the existing business operation. They want to be on the frontiers of the industry. Therefore, exploring strategies render CSR practices to be innovative and creative in seeking solutions to social and environmental concerns. On the other hand, exploiting strategy drives firms to be more adaptive to existing institutions. Firms who adopt an exploiting strategy tend to utilize what they already possess instead of creating something new. Hence, the CSR practices they perform usually include allocating resources to external necessities, such as donation, and philanthropy activities through a partnership with the host government by supplying materials that the government demands. Exploiting CSR may include reactive, convergent, and inactive responses [70,94].

CSR and MNEs insight in Ethiopia
Ethiopia is one low-income sub-Saharan country, and one of the oldest nations in the world, and Africa's second-most populous nation. The current population of Ethiopia is 116,160,214 as of 2020, based on World meter elaboration of the latest United Nations data. Ethiopia has 1,140,331 sq. km total land area and the population density in Ethiopia is 115 per Km 2 (298 people per mi 2 ).
From mid-1991 onwards, the economy has evolved toward a decentralized, market-oriented economy, emphasizing individual initiative, which was intended to reverse a decade of economic decline. In 1993, gradual privatization of business, industry, banking, agriculture, trade, and commerce was underway. In Ethiopia, industries related to food processing, beverages, textiles, leather and footwear, wood and furniture, paper and printing, chemicals, metals processing, cement are contributing significant contributions to the country's GDP.
The Ethiopian government has put attractive admission and regulatory incentive schemes in place for foreign investors. Foreign investors can invest on their own or in partnership with domestic investors in areas open for FDI. In recent years, policymakers, have concluded that foreign direct investment (FDI) is needed to boost the growth in their economy. It is claimed that FDI can create employment, increase technological development in the host country, and improve the economic condition of the country in general.
Ethiopia has no identified proclamation or regulation regarding the CSR independently but the country is carrying out a Climate Resilient Green Economy (CRGE) strategy to identify opportunities and entry points for green industrialization, to develop an economy-wide green growth vision.
Ethiopia ranked as the second-largest FDI host economy among the developing countries in 2016, supported by its large market, strategic location, and cheap labor, and growing economy [102]. According to data from UNCTAD [93], despite a 24% fall in investments to USD 3.1 billion in 2018, Ethiopia maintained its top rank in East Africa, with investments in petroleum refining, mineral extraction, real estate, manufacturing, and renewable energy. The main investor countries are China, Canada, Saudi Arabia, the United States, India, and Turkey [93].
Ethiopia's industrialization approach has strongly focused on developing specialized industrial parks, maintaining environmental sustainability, building vertically integrated industries, and enhancing skills development-through strong collaboration with the private sector. The Ethiopian Investment Commission and the Industrial Park Development Corporation as well as the Ministry of Industry are key players in this process.
The Industrial Parks Development Corporation (IPDC) was established in 2014 with a mandate to develop, operate and administer a wide range of industrial parks in the country through lease, transfer, and sale of land and construction. IPDC is designated to prepare a detailed national industrial park master plan based on the national master plan. Currently, three industrial parks-Bole Lemi I, Mekelle, and Hawassa are operational and an additional four-Kombilcha, Dire Dawa, Adama, and Jimma-will soon be inaugurated. Social and environmental issues are crucial for the sustainability of Ethiopia's development to ensure that industrialization benefits the entire society and has minimal negative impacts on the environment.
Corporate Social Responsibility (CSR) guidelines are currently lacking, both for the park developers as well as for the industries. Industries residing in the industrial parks are expected to operate ethically and transparently that contributes to the welfare of society and the environment, complying with applicable national laws and with international norms of behavior such as the United Nations Universal Declaration of Human Rights, the International Labor Organization's core conventions on labor practices, and other Corporate Social Responsibility Standards.

Competitive-orientation and MNEs CSR practices
Most of the business literature on CSR and competitiveness has been sporadic, and the studies have limited by taking two variables of CSR and firms' financial performance. The practices of CSR become more sophisticated and it affects business operation from a variety of sources, such as legal, ethical, and economic, and stakeholders demand corporations to be more socially adapted. Therefore, figuring out the complex dynamics of CSR and engaging in issues with selective manners can save enormous energy and time for firms on the occurrence of raising competitiveness.
Theoretically, putting an RBV lens on analyzing an MNE's engagement in CSR is an extended form of the traditional management theory that sheds light on a firm's strategy, which may help the firm to gain competitive resources and remove disadvantages [66]. RBV theory goes further to the extent of which strategy an MNE takes to make sustainable developments in the foreign country after it entered. The theory also guides researchers to interpret the outcomes of CSR, such as innovation, human capital, reputation, and culture as intangible assets that can be competitive advantages of a firm in an RBV framework [92]. Concerning ROI, McWilliams and Siegel [64] presented a model based on RBV theory to investigate the level of the optimal investment in CSR. According to their model, CSR activities and attributes may be used in a product differentiation strategy. They conclude that firms can determine the appropriate level of investment in CSR by conducting a cost-benefit analysis in the same way that they analyze other investments. Consequently, studies have aimed to investigate relationships between CSR and corporate social performance [65,88] and understand CSR as enhancing the company's core value [58]. In reality, however, the competitive advantage achieved by a CSR-based strategy could retain an abnormal return only if it could limit ex-post conditions by preventing competitors from imitating original ideas [82]. To prevent this, managers who are motivated by competitiveness actively innovative processes and products that reflect social and environmental issues to enhance their market position [8].
Reflecting on the flow of discourse held so far, we propose that 'competitiveness' as an antecedent of a firm's internal orientation is positively associated with the firm's adopting exploring CSR practices for this study. Competitiveness is used to combine tangible and intangible resources of a firm that makes a distinction from other firms. Specifically, types of competitive advantages that the multinationals may seek by local CSR activities can be described as lower costs by preoccupying newly adopted laws and regulations, which would cost more to late-comers in the same industry. To achieve these advantages, MNEs can incorporate their expertise into CSR programs, mostly by creating new capacities for a host country to address social and environmental issues. As an example, running a dream academy for the youth that offer job opportunities after an appropriate education in Ghana (Hyundai Motor), building water purifying systems in rural areas using their leading purification technology in China (Coca-Cola), and educating the benefits of a healthy lifestyle in rural areas of India (Novartis) are just a few examples. Hence, we develop the following hypothesis.
H1a MNE's business orientation to improve competitiveness is positively associated with exploring CSR practices.
In the same way, MNCs that are determined to pursue competitiveness would be less willing to engage with exploiting CSR practices. Hence; H1b MNE's business orientation to improve competitiveness is negatively associated with exploiting CSR practices.

Legitimacy-orientation and MNC's CSR practices
Institutional theorists argue that institutions make human actions isomorphic through structured political, economic, and social interaction. Hence, institutions reduce unnecessary costs in the business transaction by applying approved and transmitted rules to business organizations in a certain society [73]. This flow of discussions led to a powerful stream of thoughts that firms, especially MNEs, are assumed to be appropriate or legitimate within a society's structure of belief, norms, and cultures [26,68].
The institutional perspective is particularly concerned with how institutions of different social contexts deliberate legitimacy on firms and their actions [22]. Organizations tend to adapt to normative, regulative, and cultural cognitive arrangements to earn legitimacy within society. Institutional theorists recognize the causal influence from state, societal, and cultural pressures, which is opposed by market forces and resource scarcity, on organizational behavior [76]. Actively responding to these institutions is a threshold for MNEs in that they have to overcome the liability of foreignness [106]. A frequent description for legitimacy-oriented firms is how the firms comply with existing policy arrangements [8] in that they are more focused on reactions to external institutions such as norms, regulations, and business conventions to avoid possible discrimination when they do not. Doing networking with officials is a way for MNEs to overcome the liability of foreignness and gain access to competitive resources offered by central or local governments [80].
Therefore, a firm's strategy of exploiting CSR activities come to exist as projects like a philanthropic donation for a government-driven campaign or supporting shortterm social-environmental projects initiated by governments [107]. Most of the studies revealed that the spirit of philanthropy is still in a rudimentary stage in developing countries. For example, due to their inferior position, private enterprises use philanthropy activities to protect their property right and strengthen political connections with the authorities.
To avoid discrimination, MNEs may initiate shortterm social-environmental activities with the government, or make a monetary donation as a response to the external stakeholder pressure [108]. MNEs who require legitimacy tend to be more cooperative to the government's initiatives, and willing to conform their CSR practices with existing regulative arrangements of CSR [107]. Accordingly: H2a MNE's business orientation to improve local legitimacy is positively associated with exploiting practices of CSR.
Following the same logic, MNEs who are determined to seek local legitimacy in the host country would be less innovative in performing creative CSR practices. Hence: H2b MNE's business orientation to improve local legitimacy is negatively associated with exploring practices of CSR.

Collaborative approach and CSR practices
Government is one of the key stakeholders, especially in emerging countries and cooperative alliance with the host government may bring special credits for the multinationals that they can leverage, which bears two critical RBV attributes: imperfect limit ability and imperfect substitutability. CSR-related regulations are relatively new and are mostly uncertain in emerging economies. Collaborating with the host government or political authority of these regions in social and environmental areas could mitigate uncertainties and gain access to the legitimate resources that MNEs need [55].
The collaborative stance is regarded as the most desired form by MNEs as described by Gladwin and Walter [34] in dealing with conflicts between firms and the host government. An MNE perceives itself in a relatively powerful position and holds a highly assertive and highly cooperative approach in a collaborative relationship. In this situation, multinational firms interact with the government's initiatives and usually have the same objectives, but different ways to achieve them. Moreover, when an MNE obtains competitive resources or technologies that attract the government, it often leads to the way the relationship proceeds between firms and governments. Nonetheless, the collaborative relationship is not necessarily dominated by a firm's assertive behavior. Rather, interest is interdependent, and a firms' behavior against the governments is relatively positive. For example, Moon and Lado [71] posit an integrative model on the MNE-host government bargaining framework based on the RBV theory, contending that an MNE's economic returns can be affected by the firm's political power. Among the sources of the MNE's competitiveness, they focused on managerial resources, technological know-how, and reputation as intangible resources which raise MNE's bargaining power in dealing with the host government. Therefore, with regard to CSR implementation, the collaborative relationship may encourage firms to proactively respond to a host government's social and environmental interferences, and managers may consider them as an opportunity to preoccupy the market before their competitors enter [8]. Multinationals, which are in a collaborative stance, could raise all possible schemes such as creating new policy arrangements to ratify their projects more quickly or implement an innovative program in social and environmental areas to attract the authorities. Some large multinationals even set up an independent foundation to conduct diverse programs focusing on social and environmental areas.
H3a MNE's collaborative behavior toward the host government is positively associated with exploring practices of CSR.
H3b MNE's collaborative behavior toward the host government is negatively associated with exploiting practices of CSR.

Accommodative approach and CSR practices
In an accommodative manner of the political stance, an MNE perceives itself in a vulnerable position and takes a conformity posture to the host government. Accordingly, the firm holds highly cooperative but little assertive approaches. Interest interdependence and relations with the government are relatively positive. If a conflict takes place between an MNE and a host government, the firm tends to appease the government because the odds are against the firm [34]. For example, MNEs in an accommodative relationship may subsidize resources for projects that are initiated by governments. Sharma [91] investigated about 100 Canadian firms using a survey questionnaire and found that firms take different strategies toward external environmental pressures, which are usually from the governments, based on a manager's interpretation of environmental issues. Managers choose to comply with regulations and standard industry practices when they perceive the pressures as a 'threat' [91]. For firms who are in an accommodative relation, they may feel pressure from the government regarding social and environmental areas as threats, or conflicts to solve, or overcome. Failing these requirements might threaten a firm's legitimacy. From existing literature, earning legitimacy rests on a firm's behavior, such as comforting and meeting social standards which are in line with overcoming liability of foreignness through normative isomorphism in the host-country [106]. Therefore, when MNEs perceive that they are in accommodative relations, they will tend to comply with the government's regulative arrangements rather than insisting on innovative programs. Moreover, firms tend to complement existing CSR policies or partner with the authorities to assist with their CSR programs. Note that CSR activities will be similar to others in this case, and it would be hard to expect firms to gain competitive advantages out of CSR practices. MNEs will care more about relationships with the government in the way of removing competitive disadvantage and being homogenous with other firms in the same industry. Accordingly: H4a MNE's accommodative behavior toward the host government is positively associated with exploiting practices of CSR.
H4b MNE's accommodative behavior toward the host government is negatively associated with exploring practices of CSR.

Data collection
This paper uses primary data collected from Multinational Corporations operating in Ethiopia. Ethiopia is the top ten FDI-inflow recipients among Sub-Saharan countries [102]. The study purposely selects manufacturing MNC subsidiaries operation in Ethiopia. Consequently, Ethiopia has received a great deal of influence from multinational enterprises, both economically and socially. More than 500 companies have manufacturing facilities in Ethiopia as Ethiopia Investment Commission recorded data.
We used a mail survey method for this study. First, a questionnaire was designed in English by referencing the extant literature [38,57]. Next, the survey was administered to the total population of manufacturing multinationals companies working in Ethiopia whose company information was identified by the Ethiopian Investment Commission. We were excluded new and very small companies from the survey, and we also restricted the area of our survey to Addis Ababa and its areas, due to most of MNEs subsidiaries have their own head office from Addis Ababa the capital of Ethiopia.
After an exclusion process, 278 manufacturing companies were eligible for our survey. Questionnaires were sent through e-mail to each person who was identified as a CEO or a member of a top management team, a team leader of CSR, and marketing departments. Phone calls requests were used as a follow-up request. We received 122 responses, representing a 44% response rate.
As indicated in Table 1, the average operating age of the companies within our sample was about 6 years and 4 months, and the average portion of large conglomerates against the small and medium-sized firms is 56%. Firms spread across various industries, including manufacturing (48%), service (26%), wholesale (15.1%), distribution (9.9%), and others.

Variables measurements
All constructs were measured by a self-reported 5-point Likert scale ranging from 1 ('strongly disagree') to 5('strongly agree'). As to the competitive advantage, one of the four independent variables of our study, we adapted items from 'a practical framework of resourcebased approach to strategy analysis' that was invented by Grant [35], 'environmental ratings' used in the paper of Russo and Fouts [88] and manager's self-reported items modified from [91], respectively. As shown in ' Appendices 1 and 2, ' Cronbach's alpha of 0.75, factor loadings from 0.78 to 0.84 and communality estimates as 0.83 for each item validate the internal consistency of variables. The uni-dimensionality of each construct is explained by an explanatory factor analysis whose figures validate that our indicators are reliable as well as convergent. The exploratory factor analysis shows that each variable can be interpreted as one single dimension [6]. For our second independent variable, legitimacy orientation, we developed three items referring to Oliver [75], Zaheer [106], and McWilliams and Siegel [64]. Cronbach's alpha was above the threshold as 0.71; factor loadings ranged from 0.75 to 0.83 and communality estimates were from 0.75 to 0.80 for each item.
As for the other two independent variables of external determinants, collaborative and accommodative, items were referred to Gladwin and Walter [34] and Luo and Bhattacharya [56]. The premise of this paper is that an MNE is likely to make a cooperative relationship with host governments. Therefore, extreme cases of assertiveness such as 'avoid' or 'acquiesce' [75] or 'competitive' [34] were excluded. Cronbach's alpha for collaborative variable shows 0.85; factor loading was satisfactory from 0.66 to 0.87 and communality estimates 0.69 to 0.87.
Finally, the fourth independent variable, accommodative attitude, was measured by the several items which ask of firms' cooperativeness toward the host government that is derived from the arguments of Luo and Bhattacharya [56] and Gladwin and Walter [34]. The reliability and validity of this variable were satisfactory with Cronbach's alpha of 0.88, convergent factor loadings from 0.74 to 0.86, and communality estimates from 0.67 to 0.78. Two dependent variables are formulated in this study to incorporate different characteristics of CSR practices: exploring and exploiting CSR. The first dependent variable, exploring CSR, is measured by a total of six items developed from the arguments of Zhao [107] and McWilliams et al. [66]. Among them, three items capture the company's CSR creative activities that utilize the firm's specialized capacity, technology, and innovation features. The remaining three items operationalized from McWilliams et al. [66] are designed to measure the responding firm's proactive political approaches to appeal policies and regulations that are favorable to the firm in social and environmental areas. The reliability and validity for these variables indicate satisfactory reliability and validity with Cronbach's alpha of 0.90 (after eliminating a couple of items), convergent factor loading from 0.65 to 0.84, and communality estimates from 0.82 to 0.87. The second dependent variable, exploiting CSR, is measured by six items referring to Luo and Bhattacharya [56] and Zheng et al. [108]. The six items capture a firm's CSR practices that are more adaptive to the current regulation of the host country. As shown in ' Appendices 1 and 2, ' competent result with Cronbach's alpha of 0.80, factor loadings from 0.75 to 0.85 and communality estimates from 0.76 to 0.85 endorse reliability and validity of the above variables.
To prevent the possible confounding effect of the industry impact, we controlled several variables that may influence CSR implementation. Firm age is the natural logarithm of the years of a firm's operation. Firm size is transformed into a dummy variable with a value of 1 indicating that the firm is belonging to a large conglomerate group, and a value of 0 indicating that the firm belongs to a small and medium-sized group. The answers to determine respondents' firm size are collected via survey questions. Affiliation of a large conglomerate or a firm that has more than 150 employees is defined as a large-sized firm according to the Enterprises Act of Ethiopia. For the industry type, we included a dummy variable (manufacturing sector) with a value of 1 signaling that the firm is belonging to the manufacturing sector, and a value of 0 indicates otherwise. We selected the manufacturing sector among others because managers of the manufacturing sector may have more contact with government officials regarding environmental or social issues and the economic development of Ethiopia is more focusing on manufacturing sectors.

Measurement model
Since the survey takes a self-report formation asking about social and environmental responsibilities, firms may tend to answer 'socially desirable responses. ' Moreover, the fact that one questionnaire obtains all items to test independent, dependent, and moderating variables could bring up 'common method bias. ' To minimize the possible threats of common method variance, we took several steps. First, we placed items to measure variables structurally in different sections. In the postsurvey stages, as recommend by Podsakoff et al. [83], we took Harman's single-factor test. For example, we loaded all variables into one exploratory factor and extracted an unrotated factor solution. The result explained only 35% of the total variance. Therefore, it is hard to assume that the single factor explains the majority of covariance among the measures. In the explanatory factor analysis, we deleted a couple of items to raise reliability. To further check convergent and discriminant validity, we performed some different confirmatory factor analyses [7]. Cronbach's alpha for all the constructs is above 0.7, exceeding the minimum value recommended by Hair et al. [37].
By examining the measurement model of our variables using SPSS22 and AMOS 22, we obtained a satisfactory level of composite reliability (C.R.) and average variance extracted (AVE). The composite reliability is estimated to evaluate the internal consistency of the measurement model. The composite reliabilities of the measures included in the model ranged from 0.80 to 0.93. The average variance extracted of the measures (included in this model were ranged from 0.57 to 0.90, and two different results for each trait are presented in ' Appendix 2' reflecting a different calculation suggested by Hair eat al. [37] and Fornell and Larcker [31]. A variance extracted of greater than 0.50 indicates that the validity of both the construct and the individual variables is high [37].
Confirmatory factor analysis reveals several diagnostic indications for convergent and discriminant validity such as goodness-of-fit indexes. The results cutoff scores for good model fit are 0.08 for SRMR, 0.06 for RMSEA, and 0.90 for CFI and IFI.
We took additional efforts to evaluate discriminant validity by examining the correlation of latent variables, standard errors, and AVE [7]. First, we tested discriminant validity by calculating the squared value of correlation coefficient, of four independent variables. When the result did not exceed the value of AVE for each construct, it meant they obtained discriminant validity. In our study, none of the squared values of correlation of construct exceeded the AVE value. Therefore, they achieved discriminant validity. Test results are described in ' Appendix 3. ' Second, discriminant validity of variables was achieved if the correlation among all items significantly differed from 1 (α > 0.05). The results range from −0.09 to 0.84, and none of the results were equals to 1 for our constructs. Therefore, constructs achieved discriminant validity (see ' Appendix 4'). Third, we performed a Chi square difference test by pairing two constructs that have theoretically similar traits [4,7]. When a Chi square test showed that two groups are not perfectly correlated, discriminant validity is achieved. When we constrained parameter of covariance to 1 for a pair of two constructs, CMIN (Chi square) and comparative fit index showed poorer results (see ' Appendix 5'). For this nested measurement model test, Bentler [14] proposed comparative fit index (CFI) that has values from 0 to 1, and over 0.90 achieves a satisfactory level. The original model achieved 0.92 CFI, whereas the rest of the model resulted in less than the value of the original CFI. As we illustrated so far, the above results extracted from several tests, all support convergent and discriminant validity significantly, which allowed us to go into further analyses.

Results and discussion
We performed a hierarchical regression analysis to test our hypotheses and earned three models (MA 1-MA 3) for exploring and three models (MB 1-MB 3) for exploiting, respectively. Data normality analyses revealed that four independent variables satisfied the thresholds of skewness and kurtosis verification for each dependent variable. The maximum absolute value of skewness and kurtosis is 1.27. Therefore, the data from the survey results proved to be appropriate for regression analysis. A Tolerance test and VIF values among independent variables suggested that there is no threat of multicollinearity. The minimum tolerance limit is 0.63, and the maximum VIF value is 1.57 which are all at a satisfactory level in the final models. The Durbin-Watson test was utilized to identify the independent errors for each model. The results were close to neither 0 nor 4, which means the data are safe from autocorrelation concerns [30].
The next Table 2 presents the results of the regression analyses. MA1 and MB1 are baseline models that only include control variables. Among firm age, firm size, and industry sector (manufacturing as a dummy variable), the only firm size was significant in both baseline models of MA1 (β = 0.47, p < 0.001) and MB1 (β = 0.21, p < 0.05). This means large MNEs are more willing to execute CSR regardless of CSR characteristics of exploring or

Table 2 Result of regression analyses: influence of firms' internal and external determinants on different CSR practices
Coefficients are standardized betas; standard errors are reported in parentheses †p < 0.10; *p < 0.05; **p < 0.01; ***p < 0.001 In MA2 and MB2, we introduced firms' internal determinants that are competitiveness and legitimacy. Compared to the baseline models, MA2 and MB2 demonstrate significantly improved statistical results ( R 2 = = 0.23, 0.13; Hierarchical F = 25. 41, 9.38; p < 0.001). This indicates that the model's predictive power is significantly increased by adding two independent variables to the baseline models. Next, we present the final model 3 by adding external determinants of a firm's political attitude: collaborative and accommodative variables. The overall model fit is increased ( R 2 = = 0.07, 0.29; Hierarchical F = 9.1, 24.29; p < 0.001) at a significant level, indicating that, by adding internal and external determinants, the model's predictive power has been increased.

Variables
Hypothesis 1a predicts that an MNE's expectation to improve a firm's local competitiveness is positively related to exploring CSR practices. The coefficient for competitiveness is positive and significant in MA4 (β = 0.31, p < 0.001). This strongly supports Hypothesis 1a that firms who are oriented to competitive advantage tend to significantly relate to a firm's innovative and creative CSR practices that are explorative. On the other hand, Hypothesis 1b predicts that firms who are oriented to pursue competitiveness would have a negative relation with exploiting CSR practices. However, regression analysis shows that the coefficient is positive and significant which is the opposite of our prediction (β = 0.28, p < 0.01). This result shows that MNEs who pursue competitiveness also seek conservative CSR practices. We assume that providing conventional social and environmental solutions such as making donations or building schools are fundamental practices that multinational firms cannot ignore in developing countries. Next, Hypothesis 2a predicts that legitimacy-oriented MNEs' CSR strategy is positively related to exploiting CSR practices. The coefficient for legitimacy is positive and significant in MB3 (β = 0.23; p < 0.01), which supports our assumption. This indicates that multinational enterprises who are seeking legitimacy are likely to perform more conventional CSR practices that are in line with the government's coordination. Hypothesis 2b expects MNEs who expect to improve legitimacy would under-perform exploring CSR practices. The result of this hypothesis was not significant.
Regarding MNE's external determinants, Hypothesis 3a states that when an MNE perceives that its relationship with the government is 'collaborative, ' the firm is likely to be associated with exploring CSR practices positively. The coefficient of a firm's collaborative behaviors for exploring CSR is positive and significant (β = 0.31; p < 0.001) in MA3. This result indicates that a multinational firm that perceives itself in a more powerful position over the host government would like to lead CSR programs without considering the host government's requests in social and environmental areas. Hypothesis 3b is also supported with a significant and negative coefficient (β = −0.25; p < 0.01). Therefore, MNEs in a collaborative relationship with the host government would not be willing to pursue exploiting CSR practices. The other external variable, the accommodative behavior is dealt with in hypothesis 4a by stating that MNEs' accommodative behaviors toward the host government are positively associated with exploiting practices of CSR which are supported by a significant and positive coefficient (β = 0.48; p < 0.001). This result indicates that the multinationals who are in vulnerable positions and take a comfort stance toward the host governments would like to adopt exploiting CSR practices. Lastly, hypothesis 4b, which expected a negative relation between accommodative behaviors and exploring CSR practices, is not supported by the insignificant result. Even though not every hypothesis was supported with a significant coefficient, we did not witness any deviations, which considerably damage the normative direction of this study.

Conclusions
The purpose of this study is to investigate internal motives and external determinants of multinational enterprises that influence their different CSR practices in developing economies, by using survey data collected from multinational firms operating in Ethiopia. Although not every hypothesis was supported, the results of our analyses prove that the logical flow of our arguments is consistent.
Our study findings raise understandings about complex characteristics of CSR, which are still described as fragmented in the extant literature by various aspects. First, we divided CSR practices into two spheres that are contrary to nature. Even though studies about CSR strongly emphasize the complex characteristics such as implicit versus explicit CSR [63] and political aspects of CSR [33,72], when it comes to empirical tests, the complexity of CSR recedes. Our study demonstrates that actual CSR practices are composed of different characteristics, which are exploring and exploiting.
Second, we attempt to identify firms' internal and external determinants simultaneously. Pressures that force firms to adopt CSR strategy come from many directions now [95]. Internal motives from business strategies and external pressures from key stakeholders should be considered concurrently. Investigating different determinants that affect selecting CSR practices helps firms to make effective strategies that are in line with business core capacity.
Third, our study sample comes from one of the lowincome countries where CSR concepts and regulations are developing at a low pace. This can contribute to existing developing countries-based literature of CSR studies. Developing countries undergo institutional transitions in many parts of the business environment, which render different CSR strategic focus compared to those of the developed equivalent [24].
Specifically, our findings of internal determinants suggest that the multinationals that are oriented to raise competitiveness tend to adopt both exploring and exploiting CSR practices. Simultaneously, these multinationals appeared to affect greater exploring practices such as innovative and creative programs. While exploiting practices are often regarded as green-washing in developed countries [78], conventional practices, such as money donation or building schools, the important factors to raise a firm's reputation or image from the public of rural areas where infrastructure is still in need. The other internal influencer, legitimacy, tends to affect a firm's adoption of exploiting CSR practices as expected. Exploiting CSR practices are often engaged with host governments in developing economies to fill the voids of limited resources [89].
For external determinants, we chose two opposing political stances. We found that firms who are in a collaborative relationship with the host government are more likely to adopt exploring CSR practices. An MNE's strong bargaining power implies that the firm depends less on governments and local authorities that control resources [56]. Our study results support that firms in this situation tend to focus on discovering and cultivating different CSR practices while they contributed less to philanthropy or resource adjustments. On the other hand, the relationship between firms' collaborative behaviors and exploiting CSR practices is negative, which indicates that the more bargaining power a firm perceives to have, the less it tends to engage with conventional CSR practices, which usually comfort government requests. Lastly, accommodative behaviors are associated with exploiting CSR greater than any other variable of competitiveness or legitimacy. Firms who take accommodative stances are highly cooperative with host governments and relations are positive. Our study result shows that a company in accommodative relations would adopt conventional CSR practices that are in line with a host country's social and environmental needs in a way to develop or maintain good relationships with governments in the extant literature.

Managerial implications
The study overall, our findings can offer managers an improved knowledge of CSR practices, which are composed of complex characteristics. One of the common difficulties that managers appeal to when developing strategies for local CSR practices in foreign countries resides mainly in the lack of understanding about CSR forms that are aligned with their business interests. When designing CSR strategies, firms need to consider complex variables to make the best return on investment. This study attempts to unpack the complexity of CSR practices according to its exploring or exploiting nature; two local business orientations (competitiveness and legitimacy) as internal determinants with two political dimensions (collaborative and accommodative) as external determinants. These alignments may assist international executives with integrated views on CSR according to the firm's internal orientation and status of their external relationship with host governments. The strong association between exploiting practices, and both competitiveness and legitimacy, imply that conventional CSR practices may be fundamental ways to approach social issues. On the other hand, study results can motivate firms to contemplate how to align their business capacity with appropriate CSR practices. Since they cannot consider every aspect of social and environmental issues, targeting issues that they can practice corporate core visions are critical and can save enormous energy. Although there exist attempts to unravel a complex business environment applying various theoretical frameworks focusing on legitimacy [13,108] competitiveness [12,42,82,95] and political stances [56,107], the stream of knowledge has been swelled up in separate ways. The insights gained from this study can complement these perspectives with compound consideration of CSR activities.

Limitations and suggestions for future research
This study has many limitations that will addressed in future research. Our data source is only the companies CEOs, and we used the self-report data collection methods, which may be biased. Future research is required to collect data from both sides from the companies and stakeholders by using other methods will be better for data reliability. Future research is also required to explore the influence of non-economic values on economic outcomes, such as the impacts of CSR engaged political attitudes of firms on the business-related financial benefits. The ultimate purpose of the conceptual configuration is to raise an MNE's business performance. From the result, this study witnessed that CEO decision-making behaviors and different CSR practices are highly related. Investigating the impacts of different forms of CSR practices based on financial benefits with assessing managerial decision-making behavior as a moderator or mediator would be an interesting exploration that is the subject of future research. See Table 3. (1) We invite government interlocutors to seminar or forum that introduces our company's advanced knowledge and experts in social and environmental are development.

0.70
(2) Our company conducts social and environmental projects that require our company's specialized capacity 0.84 (3) Our company creates new capacity for the host country to address social-environmental issues (e.g., hold cultural campaign or event for local community, running computer education class for computer illiterate) 0.82 (4) Our company conducts social and environmental projects based on our technological innovation (e.g., facilitate local government's recycling system using our company's recycling technology, running youth training programhuman resource capacity) 0.87 Exploiting CSR (Sources: Luo and Bhattacharya [56], Zheng et al. [108] 0.80 (1) Our company has a specific budget for philanthropic activities 0.79 (2) Our company conducts philanthropic activities on a regular basis (for example, building schools, aiding poor regions, improving local sanitation, and other local needs) 0.85 (3) Our company sets up a specific team responsible for philanthropic activities 0.76 Collaborative behavior (Sources: Gladwin and Walter [34], Luo and Bhattacharya [56] (1) When our company deals with the local government (either central government or regional government), we always approach it in proactive manner and take ourselves as the dominant side 0.85 0.86 (2) When our company deals with industrial administrative departments, we always approach them in proactive manner and take ourselves as dominant side 0.87 (3) When conflicts with the local government or industrial departments arise, we never avoid the conflicts but instead always proactively bargain with or persuade these institutions.

0.69
Accommodative behavior (Sources: Gladwin and Walter [34], Luo and Bhattacharya [56] 0.88 (1) Overall, the relationship between our company and the local government has been interdependent 0.70 (2) We always believe that a cooperative relation with the local government is a necessary condition for firm growth and we take steps to implement such cooperation.

0.67
(3) We always believe, and encourage, a win-win relationship between our company and the local government. 0.69 (4) We always believe that a cooperative relation with the local government benefits both sides and can generate some synergetic gains and we encourage such cooperation. 0.77 (5) In dealing with the local government, we often look at the issues from both our and the government's perspective, rather than solely our own.