Subsidiary Agency in Gender Equality Practice Implementation: The Case of Korean MNE Subsidiaries in Sweden

Focusing on contextualized understandings of institutional distance and subsidiary agency, in this article I explore how multinational enterprise (MNE) subsidiaries address gender equality in the workplace. In particular, this study analyzes the strategic responses of three Korean MNE subsidiaries in Sweden to examine the extent to which they exercise agency over gender equality issues within specific institutional environment. In Swedish institutional environment, there is high demand for subsidiaries to develop gender-balanced organizational cultures and practices. As a result, Korean subsidiaries may actively consider promoting a more gender-equal working environment to gain local legitimacy, or they may retain their home country’s practices and avoid such pressures. The analysis shows that the subsidiaries’ perceptions of institutional distance and the organizational conditions in which subsidiaries are embedded shape their agency (strategic responses) and co-evolve with host institutions. I thus contribute to existing MNE agency literature by highlighting subsidiaries’ agency building process regarding gender equality issues.

The growing body of literature on MNE subsidiaries' strategic actions involving institutional distance or host institutions constitutes an important contribution to the field. However, critics have pointed to the lack of a contextualized understanding of national institutional environments and MNE agency (Ahmadjian, 2016;Fortwengel, 2017;Saka-Helmhout, 2020). To address this issue, recent contributions leveraging ideas from comparative institutional analysis (CIA) have shown how the national institutional environment interacts with MNEs' agency (Becker-Ritterspach et al., 2017;Fortwengel & Jackson, 2016;Saka-Helmhout & Geppert, 2011;Tsui-Auch & Chow, 2019). Such works have overcome IB's emphasis on the constraining effects of institutions, pointing out that they can also play an enabling role (Jackson & Deeg, 2008). This gives us a more contextualized approach with which to understand the agency of MNEs and enables us to consider national institutional settings more thoroughly, including aspects such as institutional thickness and strengths (Fortwengel, 2017).
In line with the current literature, this study illuminates subsidiary agency regarding institutional distance, while also focusing on how subsidiaries perceive host institutional settings and respond to them. In particular, I examine how MNE subsidiaries take strategic actions during the implementation of gender equality practices. Gender equality institutions and related practices in the workplace vary greatly between countries (Cooke, 2010;Terpstra-Tong, 2017) and depend on other institutional domains, such as corporate governance, employment systems, and inter-firm relations (Estévez-Abe, 2006). Thus, the institutional distance in terms of gender equality practice between home and host countries can also be important for subsidiaries when adjusting to the host institutional environment (Alhejji et al., 2018;Frenkel, 2017;Kemper et al., 2019). For example, some MNE headquarters (HQs) might actively engage with gender equality initiatives by implementing gender equality and diversity-related practices, such as having more female board members and managers. Yet, this engagement with gender equality does not necessarily translate to the priorities of their subsidiaries, which may face pressure from host countries with institutional environments that feature male-dominated cultures and norms. However, other HQs and subsidiaries with less of a focus on gender equality -due to the home country institutional environment -may struggle to manage their business in host countries with high demand for female employment and gender equality.
In this article, I seek to answer the following questions: (1) How do subsidiaries perceive host institutions that are distant from home institutions while they implement gender equality practices? (2) What strategic responses do subsidiaries take to adapt to host institutions and why? (3) How do subsidiaries' strategic responses 1 3 Subsidiary Agency in Gender Equality Practice Implementation:… change or evolve? To address these issues, I examine the implementation of gender equality practices in Sweden by Korean MNE subsidiaries. If we measure the institutional distance at the macro level using comparative capitalism (CC) approaches, Sweden is seen as a collaborative business system (Whitley, 1999) or a coordinated market economy (CME) (Hall & Soskice, 2001), whereas Korea is seen as a stateorganized business (Whitley, 1999) or state-led market economy (Kang, 2010). Both countries have widely differing views on gender equality, which I discuss in more depth below. For example, according to the Gender Gap Index by World Economic Forum (Leopold et al., 2016), Sweden leads the world in gender equality (ranking 4th in 2016), whereas Korea is ranked 116th out of 144 countries. Given this disparity, we may assume that Korean MNE subsidiaries in Sweden face high pressure to deal with gender equality and to implement the necessary practices.
This study also aims to make several contributions to IB literature. First, it advances our understanding of subsidiary agency. By applying ideas from CIA literature to institutional distance literature, I provide a contextualized understanding of how subsidiaries perceive and respond to institutional distance and host institutional environments. Crucially, it investigates mechanisms behind subsidiary agency toward institutional fit or incompatibilities between home and host countries. Some studies have shed light on subsidiary agency with actor-centric institutionalism (Becker-Ritterspach et al., 2017;Fortwengel & Jackson, 2016;Saka-Helmhout & Geppert, 2011), but these works offer a nonetheless limited understanding of how subsidiaries really view and respond to institutional distance. Second, I contribute to the literature on subsidiary agency by focusing on organizational conditions. Although autonomy from HQ and identity of subsidiaries, for example, have been dealt with in existing scholarship (Edman, 2016;Kemper et al., 2019), little is known about how organizational conditions and attached micro politics between actors and institutional settings co-evolve and influence subsidiaries' actions. This study fills this void by highlighting the effects of organizational conditions and institutions on subsidiaries' strategic responses to gender equality issues. Third, this study adds nuance to the literature on gender equality practice implementation in MNEs by identifying subsidiaries' strategies. The present case study -Korean subsidiaries in Sweden, which are institutionally distant home-host countries -explores the unique strategic responses of subsidiaries to particular institutional settings. By covering under-researched contexts, this article makes an empirical contribution to the study of gender equality practice transfer and implementation activities in MNEs.

Institutional Distance and CIA
During practice transfer and implementation across countries, MNEs encounter dual pressures in their HQs and subsidiaries (Ghoshal & Bartlett, 1990;Kostova & Zaheer, 1999;Zaheer, 1995). MNE HQs may try to impose uniform practices in subsidiaries in host countries, believing that their way of management style has given the firm's particular advantage and will operate effectively in all contexts (Zaheer, 1995). However, because of the institutional distance between home and host countries, HQs may struggle to transfer practices to subsidiaries (Kostova, 1999;Rosenzweig & Singh, 1991). MNEs and their subsidiaries may thus face dual institutional pressures: both to adapt to the host countries' institutional environments and to maintain consistency within MNEs when transferring practices. In defining these pressures, Kostova and Zaheer (1999, p. 71) use the idea of "institutional distance" between two countries, which is defined as "the difference/similarity between the regulatory, cognitive, and normative institutions of the two countries." Similarly, Kostova and Roth (2002, p. 216) use the term of "institutional duality" to describe how MNE subsidiaries are subjected to two distinct pressures from their home and host countries.
The notion of institutional distance stems from viewing institutions as placing constraints on an organization's actions (North, 1990;Scott, 1995) and refers to the differences in institutional environment between countries (Fortwengel, 2017;Kostova & Zaheer, 1999). While research on this subject highlights institutional differences as barriers to MNEs, there is often a limited sense of contextualized understanding of home and host countries' institutional environments (Greenwood et al., 2011). Yet recent IB studies -embracing the CC literature and its view of institutions as resources or enablers -have demonstrated new implications or benefits of using this perspective to understand institutional distance (Ahmadjian, 2016;Becker-Ritterspach et al., 2017;Fortwengel, 2017;Jackson & Deeg, 2008). Both the CC literature and the CIA approach highlight how countries coordinate multiple institutions in their market economies and recognize the heterogeneity of institutional environments differently. Thus, incorporating the CIA approach into institutional distance research illuminates distinct national institutional environments that influence particular firms' activities. In line with recent contributions, this study holds that the field of IB can learn important lessons from the CIA view of institutions, especially on the relationship between complex national institutional settings and subsidiary agency.
Drawing on the CC literature, Fortwengel (2017) argues that current institutional distance studies have not explained various characteristics belonging to the institutional environments of home and host countries. To move beyond the aggregated dimension of distance, he provides a configurational framework of institutional distance composed of coordination, strength, thickness, and resources. As Fortwengel (2017) and Ahmadjian (2016) argue, the classic notion of institutional distance misses important explanations as to how and why the institutional distance between home and host countries matters. The CC perspective, with its emphasis on coordination, therefore helps to explain how institutional distance matters and why MNEs may struggle to overcome particular kinds of distance (Fortwengel, 2017).
Similarly, I believe that the notion of institutional distance is not simply a difference in institutional setting between a pair of countries, but has a more flexible meaning that depends on how MNEs perceive the strengths of host countries' institutional settings and related pressures. Investigating this aspect of institutional distance means, crucially, that we should look more closely at the distance of institutions and institutional settings found in host countries. As noted above, single or aggregated forms of institutional distance between home and host countries have 1 3 Subsidiary Agency in Gender Equality Practice Implementation:… provided only limited or static insight. For greater context, we must look to the subsidiaries' actual views on local institutional demands and strengths. This may give us clear answers to questions such as how and why subsidiaries make certain strategic responses to address local pressures.

Strategic Responses of Subsidiaries in a Form of Agency
IB research has long focused on institutional distance and its impact on MNEs' practice implementation and transfer activities. However, investigations into MNE subsidiaries as agents in the institutional environment have only recently emerged. Acknowledging this as a potential scenario allows IB scholars to appreciate that subsidiaries may seek to influence or create conditions outside the organization that support the transfer and implementation of various practices (Becker-Ritterspach et al., 2017;Fortwengel & Jackson, 2016;Regnér & Edman, 2014;Song, 2021;Tempel et al., 2006). Such works stem from Oliver's (1991) mapping of organizational responses to a national institutional environment. Oliver (1991) argues that organizations can respond to institutional pressures in many ways, ranging from passive compliance with institutional norms to active defiance of an institutional environment. The research suggests a typology of strategic responsesacquiescence (the organization fully conforms to the institutional environment), compromise (it partially complies with institutional demands), manipulation (it exerts power over institutional environments), avoidance (it conceals nonconformity and responds symbolically to deflect institutional pressures), and defiance (it actively rejects institutional norms or expectations).
Existing works on subsidiaries as agents draw on Oliver's (1991) ideas and contribute to IB literature by moving beyond institutional determinism and passive portrayals of organizations. Particularly relevant at this juncture is the work of Cantwell et al. (2010) on the institutional engagement of MNEs. Although this research did not originate discussions of MNE subsidiaries as agents, it introduced a co-evolutionary perspective that included both institutions and MNEs. According to Cantwell et al. (2010), MNE subsidiaries can take one of three paths when engaging with the institutional environment: institutional avoidance, institutional adaptation, and institutional co-evolution. Institutional avoidance and institutional adaptation are to do with how subsidiaries respond to institutional differences, but institutional coevolution recognizes that the objectives of a subsidiary may not simply be to adjust to its host country institution. Instead, subsidiaries may wish to influence them. As the three forms of engagement are not mutually exclusive, MNE subsidiaries are likely to show both adaptation to and co-evolution with institutions in host countries (Cantwell et al., 2010). Regnér and Edman (2014) have also expanded the discussion by identifying four strategic responses made by MNE subsidiaries: innovation, which focuses on creating new institutions or changing prevailing ones; arbitrage, which relies on using institutional differences; circumvention, which involves leveraging foreignness; and adaptation to local institutional pressures.
Despite increased attention being paid to MNEs' strategic responses and agency, recent works taking an institutionalist perspective lack a contextualized understanding of agency (Ferner et al., 2012;Jackson & Deeg, 2008). The CIA perspective can fill this gap in the literature by highlighting societal triggers for MNEs' agency (Ahmadjian, 2016;Fortwengel & Jackson, 2016). This perspective focuses on how national economies differ in terms of coordination and institutional complementarities, and many studies therefore suggest that the nature of the host country market system significantly affects how MNE subsidiaries see the institutional environment and build strategies (Fortwengel & Jackson, 2016;Saka-Helmhout & Geppert, 2011). As CIA scholars have pointed out, institutions can also serve as resources or assets for the active agency of organizations, including MNEs (Jackson & Deeg, 2008). The CIA perspective helps us understand how specific home and host institutional settings give MNEs certain kinds of agency or enable particular strategic responses during practice adoption and implementation . This perspective suggests that institutional environments not only constrain the agency of MNEs, but also provide opportunities to build capabilities or find comparative advantages (Regnér & Edman, 2014). Recent work informed by the CIA perspective has opened a new avenue that allows us to see MNE subsidiaries as proactive agents in both maintaining and changing institutional environments (Becker-Ritterspach et al., 2017;Saka-Helmhout & Geppert, 2011;Song, 2021).
Subsidiaries' strategic responses to institutional distance also depend on their organizational conditions, such as their degree of embeddedness in home institutions (Kostova & Roth, 2002). In other words, the responses of MNE subsidiaries rest, in part, on their autonomy from the home market. Subsidiaries with low autonomy generally rely on HQ for transferring or implementing practices (Narula, 2014), whereas subsidiaries with high autonomy tend to be more embedded in host countries' institutional settings (Meyer et al., 2011). Previous researchers have pointed out that autonomy is a critical aspect of strategic responses to host institutions. Fundamentally, though, I suggest that other organizational conditions of subsidiaries -such as size, internal structure, and working culture -may also influence the strategic responses of subsidiaries with links to local institutions.
Embracing the micro-political lens of the HQ-subsidiary relationship would further expand our consideration of the 'meso-level' organizational conditions of subsidiaries (Clegg et al., 2018;Geppert & Dörrenbächer, 2014;Geppert et al., 2016). MNEs are enmeshed in global and local power conflicts between the HQ and subsidiaries. Subsidiaries in particular are engaged in micro-political contests -they are embedded in both home and host contexts, and their individual actors include expatriates and local employees who can use micro-political interactions to create and change organizational structures (Conroy et al., 2017;Ferner et al., 2012). It is important to take individuals into account in subsidiary agency literature: an individual actor in a subsidiary may have various political intentions or agencies that are not aligned with the agendas of the HQ or the subsidiary (Geppert et al., 2016), yet their micro-political interactions can shape subsidiaries' agency. As Saka-Helmhout (2020) has pointed out, MNE agency literature has so far given limited attention to linking micro-level aspects with the institutional environment and proactive agency of MNEs. Despite this, individual behaviors and sensemaking in subsidiaries build their overall agency. In this sense, combining micro (individual) and meso (organizational) levels with the macro institutional environment offers a new opportunity 1 3 Subsidiary Agency in Gender Equality Practice Implementation:… for research into subsidiary agency. Accordingly, this study considers micro-level politics (e.g., tensions between expatriates and local employees) between actors in the subsidiaries in relation to gender equality practice implementation.

Setting the Context
By investigating gender equality practice implementation within the subsidiaries, I explore how and why their strategic responses in the host environment occur. As noted above, in the CC literature, there are vast differences between the institutional contexts of Sweden and Korea, although both countries are in CMEs or non-market coordination-based economies. Since this study is mainly interested in how MNE subsidiaries deal with gender equality practices, I focus on gender equality and the relevant employment institutions.
Sweden is a collaborative business system (Whitley, 1999) or a CME (Hall & Soskice, 2001). Collaborative national business systems encourage cooperation between social partners to work together (trade unions and industry associations). The Swedish employment system exemplifies collective labor representation, which is characterized by comparatively high levels of coverage through industry-wide collective bargaining and union representation on corporate boards (Jackson & Deeg, 2008). This industry-level bargaining process influences other employment institutions such as employment security, working conditions, and equal pay for equal work (Anxo & Niklasson, 2006;Kristensen & Lilja, 2011). With non-market-based coordination, the institutional settings of Sweden are tightly interwoven. Korea is classified as a CME (Hall & Soskice, 2001), a state-led market economy (Kang, 2010;Whitley, 1999), or a hybrid (Witt & Redding, 2013). The institutional arrangements of Korea are less integrated than those of other CMEs because multiple institutional logics have collided frequently without adequately balanced (Kang, 2010). In Korea, firms typically lean on lifetime employment and a seniority-based internal structure (Witt, 2014). The employment system also includes long working hours and hierarchical corporate structures based on rank and age (Keizer, 2016;Lee, 2016).
Even though firms in both countries have relied on non-market or relational coordination, the institutional environments of Sweden and Korea are distinct in many senses (Witt & Redding, 2013). The difference between gender equality-related institutions in Sweden and Korea is also significant (e.g., the Discrimination Act, gender quotas, leave entitlements, and parental leave). In Sweden, the Discrimination Act (2008: 567) prohibits direct and indirect discrimination based on gender, ethnicity, religion, and employment status (part-time and fixed-term employees). Within the legal framework, firms try to prevent discriminatory behavior in the workplace. This helps to disseminate norms among firms and foster the promotion of equal rights. Regarding the employment relationship, the law states: "Employers and employees are in particular to endeavor to equalize and prevent differences in pay and other terms of employment between women and men who perform work which is to be regarded as equal or of equal value" (Government Offices of Sweden, 2015). Consequently, Sweden is known for its gender equality. Its female labor participation rate, for example, was 80.6% in 2017 (OECD, 2018).
Conversely, Korea has no comprehensive antidiscrimination laws in place. However, a number of legislative acts addressing gender equality issues have emerged since the 1980s, such as the Sexual Equality Employment Act (1987), the Act on Equal Employment and Reconciliation of Work and Family (1989), the Mother-Child Welfare Act (1991), and the Women's Development Act (1995). This provides a legal and institutional framework to prevent gender-based discrimination in the country. Nonetheless, due to labor market institutions such as lifetime employment, a seniority-based internal structure, long working hours, and hierarchical corporate structures based on rank and age, the local labor market in Korea does not provide women with the satisfying career opportunities they might expect (Lee, 2016). As a result, the gender wage gap also differs tremendously: in Sweden it is 7.4%, but 31.5% in Korea (OECD, 2021). The female labor participation rate of Korea in 2017 was 59.0% -one of the lowest among OECD countries -whereas Sweden's was 80.2% (OECD, 2018). Furthermore, the average working hours are higher in Korea (1967 per year) than in Sweden (1452 per year) (OECD, 2020). Notably, long working hours in Korea present an additional obstacle for women who want to continue work after having a child.
Regarding general leave entitlements and parental leave policies, employees in Sweden are normally entitled to 25 days' leave, whereas in Korea annual leave is 15 days. Swedish parents receive 480 days of compensated parental leave per child, which can be taken flexibly from the time before birth until the child turns eight. In Korea, parental leave can be taken for up to 52 weeks (1 year) per child. Both countries have leave systems, but the actual usage rate differs greatly. Moreover, sufficient infrastructure and public investment in family benefits are necessary to enable women to further participate in the labor market. While in Sweden 3.5% of the gross domestic product (GDP) is reserved for family spending such as childcare, Korea only spends 1.2% of its GDP (OECD, 2018).
As noted above, Sweden is a forerunner in gender equality, whereas Korea lags behind. In consequence, organizations in Sweden may face institutional pressures to instate gender-balanced organizational cultures and practices, and to pay attention to diversity. MNE subsidiaries from countries with lower gender equality (here, Korean MNE subsidiaries) may face higher pressures when they operate their businesses in Sweden. Korean MNE subsidiaries may actively consider hiring female workers and creating more gender-equal working environments. Yet this generates some conflict: as Yang and Kelly (2009) argue, Korean MNEs tend to transplant their home country's Confucianist management practices (which are hierarchical and male-dominated in nature) directly to their overseas operations and rely substantially on expatriates. Hence, Korean subsidiaries in Sweden are under dual pressure from home and host countries with regard to hiring or not hiring female employees and adopting gender-equal working environments. Table 1 compares these institutional contexts.
Furthermore, the case itself has important implications for the literature on MNE agency in practice transfer and institutional distance. The majority of studies in the literature have focused on Western MNE subsidiaries based in Western countries 1 3 Subsidiary Agency in Gender Equality Practice Implementation:… (Farndale et al., 2008(Farndale et al., , 2017Ferner et al., 2005;Fortwengel & Jackson, 2016) or on Western MNE subsidiaries' activities in non-Western countries (Becker-Ritterspach et al., 2017;Kemper et al., 2019). The current case study, on the other hand, reveals the 'asymmetric' or 'reversed' environment (Clark and Geppert, 2006) of Korean MNE subsidiary (non-Western) in Sweden (Western). This unusual empirical setting sheds new light on the ongoing debate over institutional distance and subsidiary agency in a variety of contexts.

Methodological Approach
The purpose of this research is to explore how subsidiaries from the same home country respond to institutional distance and host institutional settings regarding gender equality. As a result, I have adopted a comparative qualitative case study method. This method is preferable when addressing questions of how and why, when the researcher has little control over events, and when the study covers real-time issues and phenomena (Doz, 2011;Dyer & Wilkins, 1991;Yin, 2013). Additionally, as the study's aim is to develop theory (Eisenhardt & Graebner, 2007) and find a new interpretation of MNEs as agents, enlightening stories or narratives are essential (Dyer & Wilkins, 1991). This method is thus most appropriate when analyzing the complex and contextual behavior of MNEs in host markets (Miles & Huberman, 1994). It also enhances comparisons made across contexts and MNE behaviors, and permits a richer assessment than a single case study would provide (Eisenhardt, 1989;Yin, 2013).
I chose three of the ten Korean MNE subsidiaries (or branches) in Sweden. Before selecting companies, I contacted the Stockholm office of KOTRA, a public agency for promoting Korean trade investment and gathered information about Korean subsidiaries in Sweden. Most of the companies are branch offices, with just a small number of employees (less than five) and limited autonomy outside of HQ's control. Several branch offices only have expatriates who manage sales, while others have  (2020) Female labor force participation 59.0% (2018) 80.2% (2018) Gender wage gap 31.5% (2021) 7.4% (2021) only local employees. Hence, I decided to focus on the subsidiary-type company, which has more employees than the branch offices, and features a dynamic working environment with a mix of local employees and expatriates. I also felt that the subsidiaries could provide a richer context in comparison with the branch offices. Based on initial data collection, I selected three companies by considering main function, size, employee composition, international exposure, and accessibility.
To maintain the confidentiality of these companies, I refer to them as A: ICT, B: Electronics, and C: Auto components. All the firms in the sample are well known in their respective industries. The size of the companies varies (from over 300 to less than 30 employees) and their major functions are sales and marketing across Nordic regions (see Table 2).
Primary data for this research came from 20 semi-structured interviews (Table 3) carried out in March 2019 and April 2020. I conducted interviews in English and Korean, and they lasted between 30 and 90 min. I recorded the interviews with the interviewees' permission. The interviewer, who is proficient in both languages, translated the interviews in Korean (eight) into English, while the other twelve interviews were carried out in English and transcribed. Using English for the interviews with Swedish employees and employees of other nationalities did not cause any critical problems as they were all fluent in English and worked for MNE subsidiaries whose working language was English. Since this study focuses on the gender equality implementation activities of Korean subsidiaries, my key participants were employees in human resources (HR) teams and high-level managers engaged in decision-making processes, who could comment on issues in the subsidiaries. I also interviewed former employees of the subsidiaries, who were rather more active and open-minded than other interviewees during the process. Furthermore, one of the interviewees had experience in Companies A and B and made an interesting comparison during the interview. Some of their comments and input influenced the second-round interviewee selection and interview question revision. While the firstround interview, centered on the background and history of the company, included managers and former employees who had worked for the company for a long time, the second-round interview focused on recent or real-time events in the company and used a set of questions to inquire into relevant examples.  At the beginning of the data collection, I used LinkedIn to develop contacts and find interviewees. I then found subsequent interviewees through snowball sampling. The interviews covered four main areas: (a) general information about gender equality practices such as recruitment and discrimination in the workplace, (b) the process of gender equality practice implementation, (c) difficulties the subsidiaries faced, and (d) their strategic responses. I thus inquired about how, when, and why subsidiaries implemented certain practices, the difficulties they faced in implementing them, and their strategic responses. In the interviews, I clarified the scope of gender equality practice: hiring process (no gender-based discrimination), flexible working arrangements such as variable working hours and a gender-equal working culture, equal use of parental leave schemes, and equal opportunity (promotion, pay). To triangulate the interview data, I collected documents (annual reports, sustainability reports, and company websites) and data from the website Allabolag, which provides information on firms in Sweden.
Data analysis involved three steps. First, I coded and rearranged the reduced descriptions into conceptual categories, allowing me to connect the data to the research questions. This meant that I coded the data according to predefined categories for each subsidiary. Since my research purpose was to understand how subsidiaries' views on host institutional pressures and organizational conditions influenced their agency, I divided these influences into two categories: institutional pressures perceived by the subsidiaries and organizational conditions for the analysis. The 'institutional pressures' category contains descriptions of institutions related to gender equality that were most important and relevant to the subsidiaries. The 'organizational conditions' category refers to the size, structure, and organizational cultures of the firms. The coding structure is shown in Table 4. Second, borrowing insights from Langley (1999) and Fortwengel (2021), while considering the challenges faced by the subsidiaries and the specific events the interviewees mentioned, I assigned the aggregated categories to particular time periods. Since the three case subsidiaries entered Sweden in different time spans (late 1990s-mid 2010s), this analysis stage enabled me to understand how the subsidiaries have dealt with various institutional pressures and managed gender equality issue at workplace over time. At this stage, I structured different time periods for each subsidiary based on the interview data. Then, building on the interview data in the categories, I refined the codes by linking the categories to the subsidiaries' strategic actions. This helped me describe the process of subsidiary agency building over time in the three cases. Hence, the data analysis focused on identifying patterns in the subsidiaries by interpreting (1) how the two categories were intertwined over time and (2) how the intertwined categories influenced subsidiaries' strategic responses as shown in Fig. 1. Finally, I 1 3 Subsidiary Agency in Gender Equality Practice Implementation:… Parental leave "They do not understand fully the Swedish parental leave system which male and female employees equally use" (A) "My boss (Korean expatriate) asked me many times about when I will come back after the parental leave" (B) "Getting parental leave approaval from my (Korean) manager was not that easy because he believed parental leave is only for female workers" (C) Working culture (working time) "We need to explain to them why employees can take flexible working time, especially when they have a child to care" (A) "He (Korean expat manager) asked me why do you (father) take care of kids?" (C) "One of expatriates asked me to work overtime since he and other teammates would work until late at night" (C) Informal institutions Public sentiments on gender issue "We found gender issue is important in this society and a kind of norm" (A) "CEO and CFO also got to know that gender equality at workplace should not be underestiated in Sweden after having a meeting with other executives of Korean MNE subsidiaries" (B) Stakeholder pressures "Our CEO and HR head cared a lot about reputation and stated a programme such as Women in Tech, Respekttrappan" (A) "We got some pressure from a trade union as a former employee reported conflicts about workng culture between locals and expats to them" (C) 1 3 Subsidiary Agency in Gender Equality Practice Implementation:… labeled the patterns of each firm's strategic responses, which I elaborate on in the findings section.

Period 1: Entry and Adaptation (Entry: late 1990s)
The main institutional challenge to Company A's implementation of gender equality practices stemmed from regulatory differences between home and host countries, and the normative pressures of the host country. While Company A's top management team from Korea noticed the host country's practices concerning gender equality, such as its generous leave systems, they had no idea how important these points of difference were for Sweden when they entered the country, particularly its rules around working hours and public sentiments on gender issues. Yet, after reviewing the employment systems of the host country and the human resource management (HRM) practices of local firms, they acknowledged the importance of gender equality in society and decided to follow the rules regarding gender and diversity issues like other local firms. They were willing to accept local rules, but not necessarily in the form of full localization or assimilation. However, this autonomy did not mean that Company A could fully transform its working culture to match other Swedish firms. Company A guaranteed employees' rights to take parental leave and annual leave in accordance with the formal laws in Sweden, but it complied only minimally with these rules. The interviews suggested that other practices and institutions, such as flexible working hours and gender-equal recruitment processes, were not major concerns; the vital task was to focus on managing a business in the host market without friction. Although the HR team set clear goals for promoting gender equality and diversity in the workplace, reflecting the institutions of the host country, gender equality-related practices were not fully embedded in the workplace during Period 1.

Period 2: Under External and Internal Pressures (2000s-mid 2010s)
After the Swedish Discrimination Act entered into force on 1 January 2009, there was a regulatory change in Sweden. The Act states that employers should not only actively promote equality between men and women, it should also take measures against any discriminatory behavior. The Act also notes that employees who are, have been or will be taking parental leave may not be treated unfairly. Consequently, the legislation created enhanced regulatory pressures on organizations in Sweden. Company A was no exception and perceived the Act as a signal to change its working culture and recruitment process. Because the HR team took a keen interest in following the new Act, revising HR guidelines in line with it, the top-level management team was forced to take a more active stance on gender equality and diversity in the workplace.
The new act was really not new. We had similar legislations before, such as Act on Equality between Women and Men at Work. However, the management team saw the Act as another pressure. The HR team did a good job at the time by providing relevant information and revising guidelines (HR manager) After several years, Company A saw growing normative institutional pressures as having reached a critical level. It began to investigate other firms' behaviors and to care about stakeholders. Since Company A is one of the largest MNEs in the world and better known than other Korean subsidiaries in Sweden, it was under pressure to maintain its reputation and to adapt to local practices. In other words, it was far more visible than other Korean subsidiaries. Thus, as a leading firm, Company A not only followed formal rules locally to respond to the regulatory pressures of the host country, it also tried to meet local needs and expectations. For instance, Swedish university students selected Company A as one of the Nordic region's most attractive employers (Universum, 2018), and many stakeholders, including jobseekers and trade unions, therefore had high expectations of Company A's business operation.

3
Subsidiary Agency in Gender Equality Practice Implementation:… [The] HR team [has] regularly talked with trade unions and agency companies, and we got asked many times regarding [the] recruitment process and employee ratio. Many firms in Sweden aim to have 50:50 and [to] promote [a] gender-equal working culture, but we were behind in many ways for several reasons. First of all, our subsidiary's main function is sales and marketing. In Sweden, [the] sales part is [a] male-dominated area. Most available positions are sales-related and many male job candidates have applied for the positions. … It is hard to tackle the issue. We have our reasons to tell them, but yes. We are under pressure from unions and other business actors, even jobseekers. (HR manager) In addition to external pressures, Company A also faced internal pressure from employees. There have been conflicts and tensions between expatriates from HQ and local Swedish employees over gender equality practice implementation. Most expatriates are division managers, and they maintain the working culture of HQ. For example, they tend to work overtime even at weekends, as they believe it is normal. They also have a limited sense of gender equality and diversity, considering it less important than business. The expatriates typically spend three to five years in the subsidiary, and most have no particular interest in promoting gender equality in the workplace.
[The e]xpats' company is [the] HQ, not the subsidiary. They see the subsidiary as a temporary workplace, not as a permanent one. It is natural that they try not to cause problems by considering gender equality or other issues. (Former local employee) The expatriates' limited understanding of local institutions and their unwillingness to learn created tensions between local employees, who were embedded in host institutions, and expatriate managers over practices such as leave systems and working hours.
We have experienced that expats do not understand the Swedish parental leave-that men can also take time off for this. Another issue is flexible working time-parents will leave work to pick up their kids from daycare/ school and then work some more from home during the evening.

Period 3: Toward a More Active Stance (mid 2010s-present)
To address these external and internal pressures and gain local legitimacy, Company A moved one step further by embracing gender equality as a priority for building a decent workplace culture. Measures taken since 2018 involve hiring corporate citizenship employees to promote corporate sustainability, including gender equality and diversity. While corporate citizenship employees belong to the corporate marketing division, not HR, they have worked closely with HR employees to promote gender equality in the workplace by conducting seminars and training, and building new outreach programs. Unlike other Korean subsidiaries in Sweden, most of which stick to a compliance strategy, Company A has shifted its professional culture to become more gender-equal and create a better work-life balance.

Managers and HR team [members] of … Company A see gender equality
as necessary not only to gain legitimacy … but also to get more benefits. We sometimes see people (customers or potential jobseekers) care about [the] gender issue and believe building a good image in society can relate to marketing. So, promoting gender equality itself leads to marketing and better sales in the end. That was what [the] management team and we concluded. (HR employee) Notably, Company A's HQ appointed the first female CEO at the subsidiary in 2018. It is hard to argue that the intentions behind this appointment were innocent, but deploying the first female CEO to Sweden (where of course gender equality is highly regarded) was in many ways a meaningful and appropriate move. Several female participants mentioned that with a female CEO, the working culture had moved toward a more gender-equal approach. Their opinion of her was good, as she had been meticulous in looking at their professional culture, and the HR team acknowledged that the appointment itself was a big step in promoting a gender-equal culture at Company A.
It is also important to note the potential reverse knowledge transfer from Company A to its HQ. Since the mid 2010s, Company A's HQ has run an employee exchange program, which sends HQ employees to subsidiaries across the world and allows them to experience various working cultures and understand diversity in the workplace. Company A had two exchange employees from HQ during the second data collection period. They were impressed that the subsidiary had a more female-(or family-)friendly working environment than the HQ did and that it had implemented various measures to promote this working culture, according to a manager from Company A. While it is difficult to conclude that the employee exchange program can have a huge impact on HQ's practices, at the very least HQ's exchange employees may leave Company A with a better understanding of gender equality and diversity in the workplace.
When first complying with basic formal rules, Company A clearly did not intend to be fully assimilated into host institutions on gender equality. After several years of adaptation, however, it began to change its strategic actions. As the findings show, the subsidiary faces more pressure than its smaller counterparts because the organization is one of the largest MNEs in the world and it risks losing its reputation if subsidiaries fail to gain legitimacy in host countries. Some of the other organizational conditions of Company A, such as autonomy from HQ, workplace conflicts, and the appointment of a female CEO, enabled it to change its stance on gender equality toward commitment, although this shift is not yet complete. Unlike other Korean subsidiaries, Company A's HR team, which plays a major role in managing gender equality, has enough resources to implement new gender equality practices. For instance, the HR team has provided seminars to address a gender-equal working culture and explain how employees can approach the issue, alongside hiring new corporate citizenship employees and managers responsible for ensuring gender equality and diversity in the workplace. Thus, I argue that Company A is in transition from compliance to commitment (Fig. 2).

Period 1: Entry and Adaptation (1999-mid 2010s)
Company B is a large-sized Korean subsidiary in Sweden, and its major function is sales for the Nordic region. The HQ of Company B has given subsidiaries autonomy to implement HRM practices, which means that Company B can freely adjust to local institutional settings.
We have HQ guideline[s], but I would say Company B has no strong working culture, I mean, own values. So, a guideline for [the] gender issue is pretty ambiguous and open-ended. It helps us to have our own interpretation of that. Furthermore, HQ (including European HQ) asked us to follow local rules, but at [a] minimum level. … There is no internal pressure from HQ to follow HQ's rule. (HR employee) Since Company B's HQ had given it autonomy, the HR team could build its own HRM guideline incorporating gender and diversity. While developing the guideline, the HR team reviewed and reflected on formal Swedish laws and norms in the labor market, as well as HQ's guideline. At the time, the HR team and the directors were mostly preoccupied by the formal rules of the labor market. Company B thought that pressure from the host country to foreground gender equality was important, given the sizeable distance in the regulatory institutional setting between the home and host countries. For instance, the directors were particularly concerned about the host country's generous leave systems, such as parental leave and sick leave. Directors and heads of departments also felt a significant degree of pressure to fulfill and guarantee gender equality and an appropriate work-life balance.
They asked me whether we really need to comply [with the] rules and meet all [the] conditions. For example, one of [the] expats said that he [had] never experienced such practices in other subsidiaries and [he] asked me if Swedish companies really guarantee such practices and benefits without having problems. Even after some years, other new … expats asked me the same thing. I noticed that they really do not understand these local rules and think there will be some room for defying the rules. (Former HR employee) Company B thus followed the basic formal rules of the host country and adapted to local institutions at a minimum level, providing, for example, rudimentary benefits such as parental leave to employees regardless of gender. Thus, Company B took a compliance approach to gender equality practice adoption in its initial phase of entry. While adhering to local rules and providing benefits to employees, Company B maintained its business-first mindset.

Period 2: Downsizing and Enhanced Internal Pressures (mid 2010s-current)
During the adaptation period, Company B tended to guarantee parental leave and flexible working hours in line with local rules. Yet, since 2010, Company B has struggled with sales in Nordic regions. Due to the company's poor performance, many employees were laid off and Company B implemented a downsizing plan in 2017. As a result, Company B has gone from being an independent subsidiary to a subunit of a UK subsidiary. The downsizing and subsequent organizational change formed a critical juncture for Company B's management, particularly in terms of the general working culture and gender equality practice implementation.
One day I got an email from one of the directors, saying that "We are here to do business, not to have fun in [the] workplace. Let us focus on what you do. Nothing is [more] important than the business." (Local employee) As the downsizing plan progresses, opportunities are shrinking for the HR team to change professional culture or promote gender equality in the workplace. The HR team at Company B has only two or three regular employees (including a manager). With a small number of employees, the HR team is not able to raise its voice on the subject of implementing gender equality practices or push its opinions on recruitment and other issues. Furthermore, directors have tended to see the HR team as a supporting body for other divisions and asked them to perform tasks such as providing reports for the finance division. The HR team has thus been under internal pressures due to organizational structural changes.
We are willing to promote [a]  Besides the organizational changes, Company B has faced conflict and misunderstanding between local and foreign employees regarding working culture and gender equality. In particular, men taking parental leave and flexible working hours have been major causes of conflict between Swedish employees and expatriates.
It is all about culture. I heard many complaints about working time. Some male Swedish employees want to leave early because of [their] kids, but expats and local Korean employees … do not understand it. For example, one of the expats asked, "Why do you (father) take care of kids?" "How often will you leave early like this?" The questions come from cultural differences (Division manager) Some expatriates have unwillingly allowed local employees to follow such practices, but tensions between the two groups remain. Frustrated with the working culture of the company, local employees have sometimes met with trade unions and asked them to engage with this issue. The HR team arranged a meeting with employees and expatriates to manage the tension, but it had no great effect at the time. One interesting phenomenon observed in Company B was that not all employees were on the same side. Most local employees from Korea tended to follow the expatriates' ideas and opinions and keep quiet, while local Swedish employees complained and asked the HR team to intervene. Swedish employees have therefore been in conflict with Korean employees when it comes to working culture. As described in Fig. 3, such internal pressures or barriers have hampered Company B and its HR team from further promoting gender equality, while maintaining the original strategic response of compliance. Fig. 3 Company B's strategic responses

Compliance with Defiance (Company C: Auto Components)
Company C is one of the smallest Korean subsidiaries in Sweden (less than 30 employees). Most employees are from Sweden and other Nordic countries, but there are two expatriates, including a CEO, from HQ. These expatriates have played an important role in making decisions within the subsidiary and actively sought to maintain the working culture of HQ. Although the expatriates saw the importance of host institutional pressures, especially regulatory pressures regarding gender equality, they did not want to adopt or embrace the host country's institutions fully at first. Instead, they tried to defy its institutional pressures. Company C used defiance and partial compliance as its strategic responses, believing it would not attract attention from stakeholders like trade unions because of its small size (meaning it has less visibility).
This firm (subsidiary) has developed from a sole distributor with around five employees, so we had individual contracts with several agency companies before the HQ decided to build the subsidiary. … The expats did not want to comply [with] all [the] rules and norms of Sweden. We were small. So they thought we were out of other [companies'] and [stakeholders'] sight and [we] simply need[ed] to comply [with] minimum level HR-related rules. … Gender, to be honest, was not an issue at the time. They did not care, I would say. (Former local employee) The subsidiary at least tried to meet the local formal rules regarding gender equality. For example, Company C provided employees with basic benefits (such as parental leave). Despite this, it did not actively promote a gender-equal working culture, and even the expatriates were less concerned about how Swedish society and businesses saw gender equality in the workplace. A former local employee of Company C commented, [The f]ormer director had no idea about gender equality practices. He … was uncomfortable with the word itself, gender equality, since he, as a former engineer, preferred to work with male co-workers and [was] not familiar with Swedish working practices. … But it is difficult to avoid the laws and rules in this country. In Sweden, [the] Discrimination Act and collective agreement[s] force all firms [to] follow rules, including foreign subsidiaries.
Company C inherited a strong male-dominated working culture from HQ. The culture and customs of the company reflected industry-based norms, including long working hours and a hierarchical internal structure. The director and the other expatriates decided to comply with the local rules of institutions related to gender equality, on the one hand, but on the other, they also adhered to the culture and customs derived from their own HQ. For instance, the CEO and the expatriates tended to work overtime, as they would do in Korea, and this behavior affected the working hours of other, local employees, both indirectly and directly.

3
Subsidiary Agency in Gender Equality Practice Implementation:… There was a push to us. One day, the manager (expat) asked me to finish [a] task [sometime] that day. [I did not have enough] time, [and] I had to leave early because of my kids. So, I asked him if I [could] deliver it tomorrow. His answer was "No. You should finish it as soon as possible. I will stay at the office until you finish it." There was … pressure not to have flexible working time. (Former employee) Furthermore, Company C had limited autonomy from HQ when it came to HR and gender equality practices. HQ did not necessarily ask Company C to adhere to traditional ways of working, but the expatriates believed HQ's way of working was the most effective. In other words, the main decision-makers and followers of HQ's working culture-the expatriates-saw operating the business as their main purpose, and gave less consideration to promoting gender equality in the workplace and implementing new practices.
The expatriate-led working culture and decision-making structure from HQ also restricted the voices of HR and other local employees. In Company C, only one employee worked in HR, which did not even constitute an individual division. The role of the HR employee was thus limited to carrying out basic tasks, such as payroll and compensation management. One interesting phenomenon here is that the CEO and the expatriates actively engaged with the recruitment process and their opinions heavily affected recruitment, while the opinions of division managers, who worked with candidates, and the HR employee were mostly ignored. The expatriates even intentionally pushed to hire more male employees than female ones.
During the recruitment process, the expatriate even openly argued that the subsidiaries should hire male employees first, [then] female employees. He had reasons. … Female employees are likely to take (parental) leave many [more] times than male[s]. Sales is a tough job, so male employees will have better performance. … I knew [he] was discriminatory toward women, but I could not say anything at [that] point. (Former manager) Company C experienced internal conflicts and challenges in the workplace while it maintained a male-dominated working culture. For example, several Swedish employees complained to a manager about working hours and parental leave. As described above, the expatriates underestimated the importance of gender equality in Swedish society and business, and they did not fully understand the working culture of Sweden. As tensions heightened between local employees and expatriates regarding the use of parental leave, local employees consulted trade unions about issues such as the failure to consider a gender-equal working culture and the prevailing tendency toward working overtime. To address these problems, trade union officers had several meetings with the expatriates and managers to urge them to change their working culture and customs. After the meetings, the expatriates announced that they would comply with the rules and the collective consensus by guaranteeing employees parental and other leave regardless of gender and rank.
Yet Company C is still not free of HQ's pressure to preserve a male-dominant or hierarchical working culture. The coercion does not come from HQ directly, but rather from expatriates' intentions to follow HQ rules. It was therefore important to ensure gender equality practice implementation and working culture. One interviewee in Company C also mentioned that the company had a strong male-dominant working culture and a hierarchical structure by nature, so any escape from it was difficult. The expatriates thus believed mere compliance with local laws was enough to successfully operate their business in the host market, although they promised to work toward gender equality and diversity. To sum up, Company C has been conscious of pressure from the host country regarding gender equality and diversity, but decision-making expatriates with a strong attachment to HQ have managed to avoid these forces while taking a compliance strategic response at the minimum level (Fig. 4).

Strategic Responses of the Subsidiaries
Considering the case studies together, I found three different strategies for dealing with gender diversity and equality practice implementation. These strategies ranged from mere compliance to a more radical commitment to the local context. All subsidiaries initially complied with the formal rules due to high regulatory pressure (compliance), but, after a period of adaptation, the subsidiaries sought alternative strategic responses. My findings show that Company A, possessing sufficient resources and a willingness to change its working culture, altered its stance on gender equality practice implementation toward commitment. It is too early to conclude that Company A has been fully assimilated into the Swedish gender-equal working culture, but I found that its efforts to promote gender equality in the workplace is ongoing. Company B's fundamental strategy remains compliance. Some expatriates and HR employees were interested in promoting a gender-equal working culture, but they could not say so due to downsizing and subsequent organizational restructuring. Company B thus had to stick to a compliance strategy. Unlike the other cases, Company C took a deviant approach to local institutions related to gender equality, but at least complied with formal laws. Believing that they could operate their Subsidiary Agency in Gender Equality Practice Implementation:… business as they had done in their home country, the decision-makers were willing to maintain HQ's male-dominated working culture. They intentionally hired male employees and asked them to follow the norms and customs of HQ (e.g., long working hours). After receiving complaints from local employees and meeting with trade unions, they promised to modify the working culture and guaranteed parental leave for employees. However, this pressure had only a limited effect on the strategic responses, and Company C has maintained a compliance strategy. While both companies (B, C) took the same compliance strategy, they had relatively different motivations. Figure 5 provides a summary of the companies' strategic responses building process.

Discussion and Conclusion
In line with previous literature (e.g., Regnér & Edman, 2014;Saka-Helmhout & Geppert, 2011), this study's aim is to understand the dynamics between MNE subsidiaries' agency and host country institutions by exploring gender equality practice implementation and related strategic responses. I argue that the response strategies of MNE subsidiaries are significantly influenced by subsidiaries' perceptions of the institutional distance between home and host countries, pressure from the host country, and the organizational conditions in place, such as autonomy from HQ, visibility, and internal structures. This study adds important empirical findings to a growing body of literature that looks at the active role played by subsidiaries in addressing home-host institutional pressures during practice implementation activities.

Contextualized Understanding of Institutional Distance
I found that all three companies faced high institutional pressures in the host country, such as generous leave systems, collective agreement-based working conditions, and the Discrimination Act. The findings confirm that the subsidiaries' perceptions of high institutional pressure stemmed not only from institutional differences between the home and the host countries, but also from the institutional strength of the host country. This has important implications for literature on institutional distance. In particular, this study supports the argument that MNEs must consider institutional distance from a variety of angles (Ahmadjian, 2016;Fortwengel, 2017) by showing how subsidiaries regard institutional distance and host institutional pressure. My findings also suggest that CC literature on institutional strength or thickness is vital for understanding subsidiaries' perceptions of host institutions and their agency. For instance, it is tricky for subsidiaries from a less tightly integrated market economy (here, Korea) to enter and adjust to a relational, non-market-based economy (Sweden) where coordination mechanisms are well established. These differences present a barrier for Korean subsidiaries when they attempt to adapt and adjust to a host environment. In other words, when operating in a more regulated system like Sweden, Korean subsidiaries may be uncomfortable with rules and institutions regarding gender equality.
My findings thus suggest that subsidiary agency is shaped by the degree of institutional difference between home and host countries, and the perceived institutional strength of the host country. Drawing on the CIA approach, I propose that researchers consider multidimensional measures of institutional distance with institutional strength and thickness, instead of concentrating on singular institutional distance (Fortwengel, 2017;Jackson & Deeg, 2008). I also argue that institutional distance can be open-ended, depending on an MNE's view of its institutional environment. Most IB research inclines toward perceiving institutional distance in a static, aggregated way, while ignoring each MNE subsidiary's individual awareness of the local institutional environment. To address this imbalance, this study explores the way in which subsidiaries conceive of institutional distance and the host institutional environment. Even though all the subsidiaries are from the same home country (Korea) and in the same host country (Sweden), the findings indicate that subsidiaries have different attitudes to and views on local institutional pressure. In this sense, I add important nuance to the IB literature on how and why subsidiaries' perceptions of host institutions and institutional distance might differ, and how different perceptions could lead to various forms of agency.

The Impact of Organizational Conditions of Subsidiaries on Agency
I discovered that in the process of recognizing and addressing pressures, the organizational conditions of the subsidiaries mattered. I thus argue that alongside the pressure exerted on subsidiaries, organizational conditions are critical factors when it comes to strategic responses. First, the level of subsidiary autonomy affects strategic responses -as indeed previous studies have argued (e.g., Kostova & Roth, 2002;Saka-Helmhout & Edman, 2016). High-autonomy subsidiaries have greater flexibility and leeway to adjust their practices and to embed themselves in the host institutions, whereas low-autonomy subsidiaries cannot fully conform to local institutional expectations. Essentially, if HQ exercises considerable control over a subsidiary and employees identify with the home market system, the pressure to align with home country logics is strong. There is correspondingly little room to adapt to or negotiate with the host country. By contrast, high-autonomy subsidiaries with weaker control from HQ and employees who understand the local environment face less pressure to align with the home country. They thus have greater flexibility in the host country. Like Company C, low-autonomy firms have little capacity to respond strategically to host institutions due to HQ's control and the tendency of expatriates or employees to identify strongly with HQ's logic (Pache & Santos, 2010). In these circumstances, MNE subsidiaries are inclined to use defiance strategies in response to the host institutional environment.
Second, the size of subsidiaries is key in strategic responses. Size is highly related to the visibility of organizations and, as earlier studies have demonstrated, highly visible organizations are more sensitive to the institutional environment than less visible ones (Baek & Kelly, 2014;Kelly & Dobbin, 1999). As large subsidiaries are more visible to stakeholders like trade unions, the press, and customers, they are less likely to be noncompliant with host institutions. Each of the large companies discussed here (A, B) adopted a compliance strategy immediately. Then, after a number of years, the largest subsidiary (Company A) which were subjected to pressure from stakeholders, altered its stance from mere compliance to partial commitment in order to attain local legitimacy and adjust to stringent pressure for conformity with the dominant host institutions. However, Company C, the smallest firm, tried to take advantage of its low visibility and foreignness to adopt a defiance strategy regarding hiring female employees.
Third, decision-making processes and strategic responses are influenced by the organizational structure and culture of subsidiaries. Company A had a relatively flat organizational culture following HQ's recent pursuit of a less hierarchical global working environment. Thanks to HQ's plan to promote a flat working culture throughout the world, Company A's HR team could actively exercise its agency to foster a gender-equal working culture by holding events and hiring CSR employees able to engage with gender equality. The working culture of Company A thus affected gender equality practice implementation and decision-making, changing the strategic response from compliance to commitment. Conversely, Company B's organizational structure and culture changed after it downsized. As its organizational capacity weakened, the directors and the HR manager became less interested in promoting a gender-equal working culture. Finally, Company C inherited a staunchly male-dominated working culture from its HQ, and expatriates who were attached to this kind of working culture preferred a very centralized organizational structure and lacked the willingness to promote gender equality. The findings have implications for the subsidiary agency, with greater focus on meso-level organizational structures or cultures that are attached to macro institutional environments. Structural changes within the subsidiaries, such as Company A's female CEO appointment and Company B's downsizing plan, may provide subsidiary actors with the opportunity to create important social space (Geppert & Dörrenbächer, 2014) for changing stances on gender equality practices.
Finally, internal conflict between expatriates and local employees regarding gender equality practices -such as parental leave and working hours -creates the motivation to either change or uphold certain strategic responses. As the findings suggest, these conflicts in the workplace can attract the attention of those who make decisions about implementing gender equality practice and promoting a genderequal working culture. Further, while the role of HR teams differed significantly between companies, HR employees in each firm communicated with stakeholders, employees, and managers to address workplace conflict over gender equality practices. The findings provide further insights into how micro-political interactions between actors at subsidiaries influence practice implementation activities (Conroy et al., 2017). In particular, this study suggests a bottom-up approach to understand subsidiary agency instead of institutional determinism at macro-level.
This study thus sheds light on organizational conditions that affect the strategic responses of subsidiaries. While autonomy from HQ, identity (foreignness), and social positions have been covered in existing scholarship (e.g., Edman, 2016), other organizational conditions -such as internal structure, working culture, and workplace conflict, which are deeply connected to the institutional settings of home and host countries and influence subsidiaries' agency -have gained less attention thus far. My research attends to this gap and contributes to IB studies by highlighting the combined effect of organizational conditions and the subsidiaries' perceived institutional pressures on their agency via strategic responses.

Gender Equality Practice Implementation in MNEs
This study also contributes to the literature on gender equality practice implementation in MNEs by identifying the strategies of subsidiaries (e.g., Kemper et al., 2019;Moore, 2015;Terpstra-Tong, 2017). As Buckley et al., (2017Buckley et al., ( , p. 1060) point out, IB scholars need to consider "the differing pace of efforts to promote gender equality across countries and how MNEs respond to these country-level differences in their strategy, operations, and employment practices." In line with this call for further research, the study could add new insights by linking MNEs' activities to implement gender equality practices. Using the case of Korean subsidiaries in Sweden, I have identified strategic responses addressing institutional pressure for gender diversity and equality. Both countries' institutional settings are unique in terms of gender equality, and the distance between them is high, but inadequate attention has been given to MNEs' practice transfer or implementation activities in these countries. I therefore make an empirical contribution to the literature on gender equality practice transfer and implementation activities in relation to MNEs.

Practical Implications
The findings also offer managerial insights into how subsidiaries can address gender equality-related institutional pressure from host countries in which there is a high demand for gender equality. This study suggests that local adaptation, such as compliance, is necessary when subsidiaries perceive high and strong institutional pressure from the host country, whether formal or informal. As discussed above, local institutions can be a source of legitimacy and present a comparative advantage for subsidiaries. In particular, institutions provide the structure for firms to accomplish their goals, determining how firms manage relationships, and subsequently produce goods and services. At its very worst, the cost of non-conformance with institutions is not simply a loss of legitimacy, but an inability to function entirely. Accordingly, though subsidiaries may deem gender equality secondary or minor within their business, they should at least abide by formal rules. The other lesson I can draw from this research is that if a subsidiary switches its stance toward promoting gender equality in the workplace, it may gain local legitimacy and support, on the one hand, and reduce the cost of workplace conflict caused by incompatible professional cultures, on the other. This suggests that the compliance strategy preferred by most subsidiaries may not be enough to achieve local legitimacy and to manage business functions well.

Limitations and Avenues for Future Research
The study possesses some limitations that suggest fruitful future research directions. While this study contributes to the currently growing literature on subsidiary agency and strategic responses to institutional distance, it would be enhanced by a systematic typology that covers institutional conditions and organizational conditions together (Saka-Helmhout & Edman, 2016). Several works have discussed CIA's analytical strength to explain institutional distance (Ahmadjian, 2016;Fortwengel, 2017); more research on how institutional and organizational conditions are coupled and shape subsidiaries' agency differently would be helpful. Furthermore, while most studies in IB have stuck to macro-level (national level) analysis of subsidiaries' behavior regarding institutional distance or host institutions, more extensive research on organizational-(unit or division) and individual-level agency (e.g,. Dörrenbächer & Geppert, 2011;Geppert et al., 2016) within MNEs would contribute further to the current literature. Notably, this would offer an opportunity to explore the complex relationships between institutions and MNEs at multiple levels, and to give a holistic understanding of MNEs' specific behaviors or strategic responses to overcoming institutional distance and adjusting to host institutions.
Funding This study was funded by Jacob Wallenbergs Forskningsstiftelse (Stora Enso) and MISUM at Stockholm School of Economics.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.