Introduction

There is an increasing public awareness of wrongdoings by organizations which has led to some firms solemnly considering the societal and environmental impact of their innovation processes. An emerging goal of many organizations has been to effectively counter the changing demands of the market by resorting to responsible innovation (RI; Chatterjee et al., 2021). RI is a relatively new approach to innovation that aims to build improved practices that better address the impact of technologies to achieve inclusive and sustainable futures. It involves contemplating the ethical and societal consequences of what innovators are planning in a highly consultative environment where diverse perspectives are incorporated, to mitigate or avoid undesirable effects (Ribeiro et al., 2017; Stilgoe et al., 2013). With its ability to help achieve high company performance and create better futures, exploring what factors may affect RI perceptions is important.

It is particularly valuable to study the link of leadership with RI since it has been suggested that leadership can promote participatory behavior among people conducting the innovation (Moon & Kahlor, 2022), help identify valuable opportunities for better products and services (Bock et al., 2015), and assist in presenting and solving global challenges (Voegtlin et al., 2012). In addition, leadership is critical to learning within the organization which stimulates the innovation process (Montes et al., 2005). It is also required to change the direction or shape of the innovation when necessary (Macnaghten, 2016). Thus, leadership is significant in both discovering and implementing innovation as well as determining a company’s future. Nevertheless, it is the style of leadership that builds the appropriate environment for innovation within a company (Bhattacharyya, 2006). The literature shows that ethical leadership and responsible leadership are the two styles that have mainly been associated with RI (Moon & Kahlor, 2022; Voegtlin & Patzer, 2020). However, studying the connection of participative leadership (PL), a leadership style that encourages followers to participate in decision making and problem-solving (Somech, 2006), with RI is important as it has the power to encourage the questioning of organizational decisions and escape conventional beliefs as well as discover alternative or new options (Coulson-Thomas, 2001). Therefore, this study explores if and how PL affects RI perceptions.

PL, which is deemed critical as it has been linked to desirable employee work outcomes and has helped achieve organizational success (Chan, 2019; Newman et al., 2016), has been demonstrated to be associated with perceptions (Lythreatis et al., 2019; Pagliaro et al., 2018) and can therefore be used to study how employees perceive that their company’s activities align with RI (Chatfield et al., 2017; Stahl et al., 2013). Perceptions of employees have long been shown effective in studying the environments they are involved in as well as the practices of the organization (e.g., Coetzer 2007; Toulson & Smith, 1994). They have been widely considered as a means for an organization to recheck its practices (e.g., Lythreatis et al., 2019). They are also regarded as a way to retain good employees and positively impact the productivity of a company (Ng et al., 2019). Studying RI perceptions is essential because although the RI concept is increasingly discussed in academic and policy circles (Schlaile et al., 2018), research on how it is portrayed within organizations remains quite limited (Ceicyte & Petraite, 2018; Lubberink et al., 2017). Moreover, although there is a growing stream of literature that attempts to explain RI in business, this is mainly solely recognized as a company’s responsibility toward the environment and society and is not supported by sufficient empirical work. A larger recognition of the RI concept in organizations is lacking (Gurzawska, 2021) and a need for RI to specifically be explored in private sector firms has also been emphasized (Chatfield et al., 2017; Chou, 2018; Stahl et al., 2017). To address these gaps, this study looks at the relationship between PL and RI perceptions, with reference to a developed RI framework (Stilgoe et al., 2013), in private firms in South Korea.

Businesses in the Asia Pacific region that aim to innovate successfully to gain competitive advantage are also progressively considering ethical behavior and social impact of their innovation (Scuotto et al., 2022). Although the concept of RI has predominantly been discussed in the European context (Genus & Iskandarova, 2018; Yang et al., 2021), studies on RI in the Asia Pacific region, mainly in China and India, have started to emerge (Chatterjee et al., 2021; Yang et al., 2021). Nevertheless, most of these studies remain theoretical in nature (e.g., Liu et al., 2019). The few studies that have looked at the concept empirically have done so through a macro level of analysis (Adomako & Tran, 2022; Chatterjee et al., 2021; Tian & Tian, 2021). Hence, to advance research, this paper is, to the best of our knowledge, the first that looks at the concept of RI empirically, from a micro level of analysis, within organizations in South Korea. This study considers South Korea specifically for two main reasons. First, the country is known for science and technology and for having a strong drive to pursue new things. It is a global innovator leader which is progressively leading markets in several fields (Massaro & Kim, 2022; Min et al., 2020). This has made it one of the world’s most innovative countries, ranking first in Bloomberg’s 2021 Innovation Index (Massaro & Kim, 2022). Second, South Korea has a superior performance and reputation in research and development and is known to have a top-down innovation system which fosters close collaboration with the industry, government, and the academic community- elements necessary for RI to take place (Dayton, 2020). Its scientific outlook is viewed to be both diverse and very productive.

This study examines the effect of PL on RI perceptions through ethical climate, in reference to the social information processing theory which states that individuals form their perceptions through cues that they gather from their social environment (Salancik & Pfeffer, 1978). It also considers the moderating role of managerial discretion on the relationship between ethical climate and RI. We chose to study ethical climate as the mechanism between PL and RI perceptions because the organizational climate cannot be separated from leadership as a leader plays a critical role in determining the climate and ethics within an organization (Dickson et al., 2001; Mulki et al., 2009). It is also important to study the ethical climate with innovation perceptions since ethics is profoundly focused on anything that may foster or impede people’s wellbeing. Ethics is, hence, vital to evaluating innovation goals to ensure that they are built around serving the needs of people and the planet. Additionally, it is important to study the role of managerial discretion in the process as the scope of authority leaders have can affect outcomes such as engaging in RI (Wangrow et al., 2015). Thus, by integrating the role of managerial discretion, we expect that a more complete picture of the process in which RI perceptions are involved will emerge. Therefore, the research question that this article aims to answer is Does PL affect RI perceptions through ethical climate, impacted by managerial discretion, within organizations in South Korea?

Overall, this study makes several contributions to the literature. It contributes to the leadership and RI literatures by exploring a novel process linking PL to RI perceptions. It also adds to the PL theory by answering research calls to further investigate consequences, mechanisms and boundary conditions related to this type of leadership (Chan, 2019; Lythreatis et al., 2019). Additionally, this study contributes to the cross-cultural applicability of PL as it examines whether it is effective in promoting certain organizational perceptions in the context of South Korea. Further, this article helps advance the ethics literature by being the first study, to the best of our knowledge, to propose that PL is an antecedent of ethical climate (Manroop et al., 2014). It also contributes to both the ethics and RI literatures by linking ethical climate to RI perceptions as it expands prior research on ethical work climate and its consequences (Pagliaro et al., 2018) as well as identifies antecedents of perceived RI within an organizational context. Moreover, by adopting managerial discretion as a moderating variable on the relationship between ethical climate and RI, this study answers scholarly calls to expand research on ethical aspects (Al Halbusi et al., 2020; Zhang et al., 2019b) and facilitates the understanding of the boundary conditions that may promote RI perceptions (Liao & Zhang, 2020). This study also makes an empirical and contextual contribution by investigating RI at a micro level of analysis within organizations in South Korea which is notably underexplored (Genus & Iskandarova, 2018; Hadj, 2020). The research aids in understanding the extent of RI translation within the private sector in Asia Pacific (Chou, 2018; Lubberink et al., 2019).

Theoretical background

Participative leadership

PL entails involving employees in both decision-making and problem solving as it prefers consultation over direction (Miao et al., 2013; Somech, 2006). Participative leaders share power by seeking input from their subordinates and valuing their points of view. Because PL requires followers to take some responsibility in the workplace, it shares theoretical underpinnings with other leadership styles such as distributed leadership (Bolden, 2011). Research has highlighted the importance of PL in determining employee outcomes which in turn affect organizational success (Bortoluzz et al., 2014). For example, this style of leadership has been positively linked to job satisfaction (Chan, 2019), organizational commitment (Huang et al., 2006), intrinsic motivation (Van den Broeck et al., 2013), knowledge sharing (Xue et al., 2011), organizational citizenship behavior (Huang et al., 2010), and work performance (Newman et al., 2016). Previous research has also indicated that a participative approach is a valuable way to get followers involved in improving organizational innovation (e.g., Chang et al., 2019; Odoardi et al., 2019). Additionally, it has been linked to employee perceptions of corporate social responsibility (Lythreatis et al., 2019).

In terms of the context in which PL has been explored, there is an increasing volume of research work on this leadership style in some Asian Pacific countries like the Chinese context (e.g., Chan 2019; Miao et al., 2013; Newman et al., 2016). Nevertheless, there is very limited work on it in South Korea. Looking at the relationship between PL and employee-related variables in South Korea specifically is essential as various organizations in the country are known for adopting a participative management style (Park et al., 2016).

Ethical climate

Research has shown that leadership is particularly critical in establishing an ethical climate (Dickson et al., 2001; Mulki et al., 2008), which is defined as “the prevailing perceptions of typical organizational practices and procedures that have ethical content” (Victor & Cullen, 1988, p. 101). It represents the employees’ perceptions of rightness or wrongness present in the company’s environment (Babin et al., 2000; Martin & Cullen, 2006).

Victor and Cullen (1988) explained three main classes of ethical theory: egoism, benevolence, and principle which represent maximizing self-interest, maximizing joint interests, and adherence to principle, respectively. While egoism symbolizes company profit and efficiency, benevolence recognizes friendship, teamwork, group cohesion, and social responsibility as dominant aspects of the ethical climate, and principle includes personal mortality, company rules and procedures, and laws and professional codes. Additionally, the authors proposed five ethical climate types: caring, law and code, rules, instrumental, and independence. Since this taxonomy was created, the different types of ethical climates have been used in various ways (Babin et al., 2000; Schminke et al., 2005). While some scholars have considered the ethical climate as a single construct (e.g., Schwepker Jr., 2001; Arnaud & Schminke 2007; Mayer, 2014), others have viewed it as multi-dimensional (Babin et al., 2000; DeConinck, 2011). When explored from a multi-dimensional view, some studies have used more than five dimensions (Peterson, 2002) and others have used less than five (e.g., Shapira-Lishchinsky & Rosenblatt 2009; Shin, 2012). This study follows research that has studied only some dimensions of ethical climate (e.g., Kim & Vandenberghe 2020; Pagliaro et al., 2018) by considering employee perceptions of the caring, law and code, and rules dimensions and by focusing on benevolence and principle. The reason we consider the rules and law & code dimensions is because they are seen to capture the core of ethical climate the best (Shin, 2012). Although research has demonstrated that they are enough to affect employees’ perceptions regarding ethics (Adams et al., 2001; Stewart et al., 2011), we also added the caring dimension to better align with what scholars who have endorsed a unidimensional perspective of ethical climate have measured (e.g., Jaramillo et al., 2006; Schwepker Jr., 2001). These three dimensions have been classified as a higher level of ethical climate than the other dimensions (Leung, 2008) and their components, collectively, resemble items used in other scales in the literature to assess participants’ perceptions of an organization’s ethical climate in business ethics research (e.g., Gonzalez-Padron et al., 2008; Raile, 2013). Using only these three dimensions is also in line with other literature (e.g., Shapira-Lishchinsky & Rosenblatt 2009).

Antecedents and consequences of ethical climate

Several scholars have examined antecedents of perceived ethical climate, such as informal control methods (Goebel and WeiBenberger, 2017) and human resource systems (Manroop et al., 2014). According to Qi and Liu (2017), leadership exercises a greater influence on ethical climate than other factors. Studies have found ethical leadership (Hansen et al., 2016), servant leadership (Burton et al., 2017), instrumental leadership (Mulki et al., 2009), and paternalistic leadership (Ötken & Cenkci, 2012) to be predictors of ethical climate. Researchers have called for more research on factors that could show how ethical climates are built or affected, especially in relation to the leadership perspective, as the ethical concept has been shown to lead to desirable work outcomes (Manroop et al., 2014; Mulki et al., 2009; Schwepker, 2001). This research, hence, links PL to ethical climate.

Furthermore, ethical climate perceptions are considered an important condition for the effective functioning of a company (Barnett & Vaicys, 2000). Research has shown that they are related to consequences that affect a company’s competitiveness, such as organizational commitment (Tsai & Huang, 2008), organizational citizenship behavior (Shin, 2012), trust (Mulki et al., 2006), and job meaningfulness (Jaramillo et al., 2013). Ethical climate has also been linked to organizational performance and innovation (Choi et al., 2013; Weeks et al., 2004). Researchers call for the examination of additional linkages between ethical climate and novel employee consequences to further develop the ethical concept (Pagliaro et al., 2018). This study links ethical climate to RI perceptions.

Responsible innovation

RI aspires to better grasp the effect of technology, restore public confidence in science and innovation, and achieve inclusive and sustainable futures through societal desirable innovation (Ribeiro et al., 2017; Stilgoe et al., 2013; Von Schomberg, 2011). The concept has been designated various characteristics in the literature. A popular RI framework, presented by Stilgoe et al. (2013), posits that at least anticipation, reflexivity, inclusiveness, and responsiveness on the part of researchers and innovators are needed to achieve RI. Anticipation is required to foresee potential risks, harm, dangers, and public worries concerning new technologies. It constitutes systematic thinking with the aim of increasing resilience. Reflexivity, which parallels holding up a mirror, is required to remain mindful of one’s own activities, assumptions, commitment and values as an innovator. It is being mindful of the limits of knowledge and that a specific framing of an issue might not be universally acceptable. Inclusiveness includes the participation of the public and different stakeholders in the innovation process, questioning the framing assumptions of the innovations. It is required to bring in a diversity of perspectives, insights, and values. Finally, responsiveness is about adjusting courses of action whilst recognizing the deficiency of knowledge and control. It is required to assure that the innovation can change direction or shape in response to changing circumstances, stakeholder values, and new emerging knowledge (Stilgoe et al., 2013).

Although RI is increasingly discussed within academic and political circles, to have an effective shift in society, the business community needs to adopt it. After all, it is the companies that create the innovations and bring them to the market. The RI concept is mainly known and discussed in Europe (Thapa et al., 2019) and, thus, some companies in different regions of the world might not be well aware of the discussion that has developed around it (Van de Poel et al., 2017). Nevertheless, many companies already engage in activities that resonate with RI using diverse terms such as stakeholder dialogues, open innovation, scenario development, risk assessment, sustainable innovation among others (Lubberink et al., 2017; Van de Poel et al., 2017). Many organizations also express the desire and objective to innovate for the benefit of society as opposed to just pursuing profit (Lubberink et al., 2019). Therefore, although RI may not yet have been systemically included in the innovation process of firms, its characteristics may be integrated in the company policies and strategy. One way of studying this is by evaluating the company’s practices and environment through the perceptions of its employees (Zulu & Parumasur, 2009). Building perceptions of RI also helps the organization by letting workers realize that they are part of an environment that expects them to engage in innovating responsibly. When employees view that their company cares about RI, they themselves will set expectations that align with RI in their work. Their perceptions have been linked to their overall performance (Lee et al., 2013) and have been used to study firm innovation performance (Al Ahbabi et al., 2018).

Recently, the literature has witnessed discussion on RI in some countries in the Asia Pacific region, mainly in China (e.g., Yang et al., 2021). Its growing prominence in China is due to the country making RI a formal state policy (Von Schomberg & Hankins, 2019). However, in South Korea, which is recognized for its technological advancement and innovation (Myeong et al., 2021; Raghavan et al., 2021), the social implications of innovation in firms have only surfaced as a subject matter in recent years (Seo, 2015).

Theoretical Framework and Hypotheses Development

Social information processing theory

The social information processing (SIP) theory (Salancik & Pfeffer, 1978) posits that people build their cognitions, perceptions, beliefs, and attitudes as a function of the information cues that are present in their social environment. The formation of these cognitions and perceptions goes beyond individual personalities as they are shaped by how people make sense of their environment (Salancik & Pfeffer, 1978). Two fundamental steps of information processing include encoding and retrieving information. When information is encoded, it is simplified and transmitted from the short-term memory of a person to his/her long-term memory. Subsequently, the person retrieves information from the long-term memory and makes a judgment (Lord & Maher, 2002). If people are exposed to positive cues in their social environment, they are more likely to construe positive judgments and perceptions (Zalesny & Ford, 1990). SIP theory is suitable for this study’s context as it can explain how individuals form their perceptions of organizational practices and innovation processes (Boekhorst, 2015; Lythreatis et al., 2019).

Participative leadership and ethical climate

The relationship between PL and ethical climate perceptions can be supported by the SIP theory (Salancik & Pfeffer, 1978). According to this theory, the obtained information that will be processed can derive from any source that emphasizes a particular situation, and people with power or a high status serve as important sources of social information (Boekhorst, 2015). In the employment context, employees obtain cues of information from their work environment which helps them form interpretations and perceptions (Zafar, 2013). The leaders in the workplace are a key source from which employees gather information that helps them shape their perceptions of organizational practices since they are seen as the most significant representatives of organizational actions (Jiang & Gu, 2015; Lythreatis et al., 2021). Hence, the perceptions of work characteristics and organizational practices are not fixed but rather impacted by the informational relationships that a person is embedded in (Salancik & Pfeffer, 1978). The repeated observation of a leader’s style allows employees, therefore, to socially construct perceptions regarding the company’s climate. When employees interact with the leader, they encode and retrieve specific cues of social information that direct their perceptions of the leader’s behavior that, in turn, influences their perceptions of organizational practices, including the ethical practices (Boekhorst, 2015).

When employees participate in decision making and have a say in the actions of a company, they will feel confident that their interests and needs are being expressed in the organizational decisions and that the company cares about their opinion and, therefore, cares about them (Armstrong, 2009). This process, thus, sends signals that people care about other people’s interests within the organization. Caring is one of the main dimensions of an ethical climate that features concern for the interests and well-being of others. Group interest, as opposed to self-interest, is also part of the benevolence criterion of ethical climate (Victor & Cullen, 1988). This demonstrates that the participation of employees in decision making and problem solving can amplify their perceptions of an ethical work climate (Manroop et al., 2014). Additionally, one of the characteristics of PL is building cohesive teamwork (Sorenson, 2000) and the benevolence in ethical climate emphasizes teamwork and group cohesion (Victor & Cullen, 1988). Employees may also develop a sense of friendship, another part of benevolence, with the leader when they feel involved and consulted (Fredericks, 2009; Victor & Cullen, 1988).

When managers encourage employees to express their opinion and participate in making decisions, they not only communicate a genuine appreciation of employees’ unique ideas, but also transparency which emerges through the communication involved in a participative approach and is an important aspect of an ethical climate (Somech, 2005). This involvement and communication will make the subordinates feel positive about the leader’s integrity (Hayat Bhatti et al., 2019). According to Rok (2009), a participative approach in management does not only enable employees to use their voices to impact organizational practices, but also hold managers accountable. Thus, when employees participate in decision making, they tend to see the whole system of the company as fairer.

Moreover, PL instills positive emotions in employees as they feel that they are heard within the company and that their views are valued (Bryman, 2013). Emotions are particularly important to processing information in a specific way (Dodge, 1991). Hence, by experiencing positive emotions from a participatory approach within the company, employees would process the collected information positively and, in turn, develop positive perceptions of the organizational practices, including the ethical practices. Based on all the above, PL is expected to lead to positive perceptions of a company’s ethical climate.

Hypothesis 1

PL is positively related to perceived ethical climate.

Ethical climate and responsible innovation

Fainshmidt and Frazier (2017) argue that SIP creates a direct connection between organizational climate and employee-related outcomes. Accordingly, the SIP theory also provides support for the relationship between ethical climate and employees’ RI perceptions as it posits that individuals use information cues from their environment to form perceptions (Salancik & Pfeffer, 1978). Employees would use information from their immediate work climate to construe their perceptions of the company and its activities (Zafar, 2013). According to Boekhorst (2015), the extent to which a work climate signifies a certain phenomenon is dependent upon the interpretations of its attributes. Individuals generate reasons for their perceptions and there are many factors from an ethical climate that employees may encode and retrieve to form their interpretations that the company’s innovations are responsibly developed (Zalesny & Ford, 1990).

Company policies, represented in the rules and law & code dimensions of ethical climate, are viewed as a method that provides social information to employees as they are a good representation of the values of the organization. Code of ethics, specifically, has been perceived as an instrument that has the power to influence employees’ beliefs (Bondy et al., 2004; Furlotti & Mazza, 2020). The presence of a code of conduct may help build reflexivity, which is needed for RI to take place (Stilgoe et al., 2013). Such codes will also help to ensure the commitment to values and adoption of specific standards, including technological standards, within the organization. Additionally, the law and code dimension of ethical climate recognizes that the company is greatly concerned about violating any laws (Victor & Cullen, 1988). This requires constant reassessment, mainly with respect to emerging rules and laws, that can help respond to emerging concerns. This supports responsiveness needed for RI (Stilgoe et al., 2013).

In addition to this, caring in ethical climate is characterized by the organization’s genuine concern for everyone affected by its decisions (Appelbaum et al., 2005). It signifies a sincere interest in the welfare of people, both inside and outside the company, thus showing consideration for all organizational stakeholders impacted by the company’s decisions and, hence, by its innovation (Shapira-Lishchinsky & Rosenblatt, 2009). The caring dimension emphasizes a strong sense of social responsibility to the outside community and includes doing what is right for the employees, customers, public, and society at large (Martin & Cullen, 2006; Victor & Cullen, 1988). This is the essence of the RI concept that stresses the importance of considering the impact of technologies on the public and different stakeholder groups (Stilgoe et al., 2013; Voegtlin & Scherer, 2017). Employees may perceive the ethical climate, where all stakeholders’ interests are considered, to help shape the purpose of the innovation and generate outcomes in favor of the people and the planet (Weckert et al., 2016). Showing concern for all stakeholders is also more likely to enhance the relationship building and quality of the relationships between the company, the employees, and different stakeholders. This would allow for better identification of the social and environmental effects of the company’s activities and innovation, which is the goal of RI (Blok et al., 2015; Gonzalez-Padron et al., 2008).

Furthermore, an ethical climate helps form feelings of organizational trust (Lilly et al., 2016). Trust facilitates social interaction among organizational members and motivates different actors to contribute in new ways to engage in positive change (Fainshmidt & Frazier, 2017). Thus, when trust is present through an ethical climate, employees are more likely to perceive better social interaction and contribution aimed at achieving positive outcomes, which can include innovating responsibly. In addition to trust, an ethical climate elicits feelings of safety (Wadei et al., 2021). When employees feel safe, it is more likely that they would perceive the company’s decisions, activities, and innovation processes as safe, too.

Therefore, an ethical climate provides cues that project the ethical character of the company (Mulki et al., 2009). It has been linked to moral awareness, ethical reasoning, and ethical intentions (Valentine & Barnett, 2002), and has been demonstrated to wield significant effects on ethical decision making and behaviors (Newman et al., 2017). Such a climate signals to employees that the company’s conduct is legitimate, ethical, and sincere, and that ethical behaviors are supported. This will result in employees perceiving the organization as behaving ethically and responsibly, which includes innovating responsibly as ethical judgment is required for the development of RI (Pandza & Ellwood, 2013; Van der Burg et al., 2019). The following hypothesis is thus formed:

Hypothesis 2

Ethical climate is positively related to RI perceptions.

Mediating role of ethical climate

Early conceptualization of organizational climate has implied its strong mediating role which has been shown throughout the literature (Mayer et al., 2010; Schminke et al., 2005). Climates have been demonstrated to specifically act as mediators between different work aspects and relevant outcomes (Pinto & Freire, 2022). Early influential scholars, such as Litwin and Stringer (1968), suggested that elements like leadership create specific work climates that, in turn, are key drivers of employee attitudes and beliefs. Individuals do not directly judge an organizational phenomenon, rather they engage in a process where they first perceive and interpret different cues to help understand their work environment (Salancik & Pfeffer, 1978). Mayer et al. (2010, p. 10) state that “Climates help explain the processes individuals use to make sense of their work environments.”

We argue that PL influences RI if the right culture, norms and ethical standards are present. The anticipation dimension of RI requires that technology be assessed. PL does not satisfy this condition without the existence of clear ethical standards to which the technology is assessed against, meaning that PL leads to RI through the presence of an ethical climate. It is a company’s ethical climate that encompasses the ethical norms and clarifies how wrong is differentiated from right within organizations (Wimbush & Shepard, 1994). Moreover, questioning the innovators is part of inclusiveness in RI. When people come together via a PL approach, questioning can take place when there are clear ethical standards and values that are expected to be met. Ethical climate is what mirrors the value system of a company (Cullen et al., 2003). Hence, PL influences RI through ethical climate as it is the climate that stipulates moral decision making and direction needed for RI (Yasin, 2021).

Hypothesis 3

The relationship between PL and RI perceptions is mediated by ethical climate.

Moderating effect of managerial discretion

Managerial discretion is defined as the latitude of action available for managers or decision makers (Hambrick & Finkelstein, 1987; Wangrow et al., 2015). It is the scope of freedom that managers have when they develop strategic activities and thus reflects the flexibility that a leader has (Chin et al., 2021). How managerial discretion may influence both organizational and individual-level outcomes has been of interest for researchers for a long time. The managerial discretion model contends that when the manager’s ability to impact decisions is high, what he/she has created within the organization will have a greater effect on outcomes (Wangrow et al., 2015). This means that when managers have more discretion, greater impacts on outcomes are exerted, a notion that has been supported by multiple studies (Liao & Zhang, 2020; Wangrow et al., 2015). Such flexibility has also previously been linked to innovation strategies specifically within South Korean companies (Lee, 1998). Thus, in pursuing our theoretical investigation of the association of ethical climate with RI perceptions, managerial discretion is considered to moderate this relationship as the effect of ethical climate on RI perceptions is likely to be stronger when a leader has high-discretion.

Organizational leaders have potential power constraints and are exposed to both internal and external supervision. The constraints can be financial, but they can also come from the network of stakeholders that the leader deals with, such as the board of directors or the government, among others. For example, although a leader might have ensured an ethical climate that encourages RI, shareholders may not accept new aspects, such as new knowledge or new ways of innovating, if that does not seem profitable to them. Stakeholders, whether internal or external, may oppose some decisions if they believe that doing so would damage their financial interests (Liao & Zhang, 2020). Hence, the effect that ethical climate has on RI perceptions is likely to be weakened when managers are seen to be restricted (Crossland & Hambrick, 2011). Therefore, in low managerial discretion situations, where managers face limitations in their decisions, employees are more likely to realize the constrained power that their leader has to implement processes and outcomes, such as RI (Crossland & Hambrick, 2011; Ponomareva, 2019). This means that even if an organization shows strong ethical guidelines, an ethical climate may be less likely to lead to engaging in RI because the leader or decision maker may be severely limited by his/her environment. Managerial capitalism theory which covers organizational control and power supports these conceptions brought forward by the managerial discretion theory (Aragon-Correa et al., 2004; Galbraith, 1967).

Nevertheless, managers with high discretion enjoy a broad range of strategic options and exert a stronger influence on outcomes (Shen & Cho, 2005). With greater managerial discretion, leaders have the power to encourage the absorption of new elements by the company, effectively combine old and new elements, and discover and implement new opportunities (Kim, 2013). A greater level of managerial discretion indicates that leaders encounter less restrictions in the process of innovation and product improvement (Miller & Chen, 1994), and can therefore strongly influence innovation decisions and quality (Chin et al., 2021; Ponomareva, 2019). With higher levels of managerial discretion, managers have the power to balance various stakeholder demands, such as taking care of shareholders’ financial interests while ensuring RI. Thus, a manager’s zone of influence can be an important boundary condition to achieving outcomes such as RI. In this case, the ethical guidelines of an organization will be perceived to have a greater effect on the likelihood of engaging in RI since the leader is not constrained. Hence, the following hypothesis is formulated:

Hypothesis 4

Managerial discretion will moderate the relationship between ethical climate and RI such that the relationship is stronger under greater managerial discretion.

Building on Hypotheses 3 and 4, this study further proposes:

Hypothesis 5

Managerial discretion will moderate the indirect relationship between PL and RI through ethical climate, such that the mediated relationship is stronger under greater managerial discretion.

Based on the above hypotheses, the conceptual model of this study is presented in Fig. 1.

Fig. 1
figure 1

Conceptual model

Full size image

Method

Participants and Procedure

Data was collected from employees working in four mid-sized and large firms in Seoul, South Korea due to the country’s focus on building an innovation ecosystem (Shvetsova & Lee, 2021). These firms operate in the technology industry, offering a range of products and services to either customers or other businesses. Access to the respondents was obtained through personal contacts. The employees were sent a questionnaire accompanied with a cover letter that explained the main purpose of the research. To address the common method bias (CMB) problem, several remedies recommended by Podsakoff et al. (2003) were applied. During data collection, confidentiality and anonymity of responses were assured to the participants to reduce evaluation apprehension and social desirability. The guidelines provided by Conway and Lance (2010) were also followed to deal with the problem of CMB. For example, construct validity of the measures used was demonstrated and we made sure that there was no conceptual overlap in the items used to measure the different constructs. Moreover, with the aim of limiting the risk of CMB, achieving a more robust study design, and improving credibility in the findings, a temporal separation technique was utilized (Podsakoff et al., 2012). Data was collected at two separate points in time from the participants, with one month between the two consecutive waves.

In the first wave of the survey (Time 1), 853 out of 1455 questionnaires that were sent out were usable, yielding a response rate of 58.62%. In the second wave (Time 2), 487 responses out of the 853 completed questionnaires were matched, with a response rate of 57.09%. Most of the participants were males (342, 70.2%). The age distribution showed that 148 of the respondents (30.4%) were relatively young (less than 30 years of age), and 112 participants (23.0%) were above 40 years of age. As for the participants’ education level, most of the respondents held a bachelor’s degree (363, 74.5%). About half of the respondents, 238 (48.9%), had worked in their company for five years or more.

Instrument

The questionnaire was translated from English into Korean by a translator before it was sent out. The translated questionnaire was subsequently back-translated into English by a highly educated bilingual person (Brislin, 1980). The two English versions of the questionnaire were then compared, and only slight differences were found, demonstrating a high degree of accuracy in the translation.

The questionnaire consisted of four parts designed to measure the conceptual model’s constructs in addition to the demographic variables. The items were measured on a Likert scale with response options ranging from (1) ‘strongly disagree’ to (7) ‘strongly agree’.

Construct measures

PL was measured using the six items from the scale of Arnold et al. (2000) at Time 1. The items of this construct are labelled PL1, PL2, …, PL6. Sample items include “My manager encourages me to express my ideas and suggestions” and “My manager uses our suggestions to make decisions that affect us”. Reliability analysis showed a Cronbach’s alpha value of 0.921 for this construct.

Ethical climate (ETH) was measured using eleven items based on the scale of Victor and Cullen (1988) at Time 2. The items, labelled ETH1 to ETH 11, measured three sub-dimensions, caring (CRNG) using ETH1 to ETH6, rules (RULES) using ETH7 to ETH9, and law and codes (LC) using the two items ETH10 and ETH11. Sample items of the subscales include “In this company, people look out for each other’s good”, “People in this company strictly obey the company policies”, and “In this company, whether a decision violates any law is a major consideration”. A score for each sub-dimension was generated using factor analysis results. The parcelling method (Aluja & Blanch, 2004) was used to transform the multi-dimensional construct ETH to a first-order construct with three indicators using the three sub-dimensions’ composite scores. Reliability analysis showed Cronbach’s alpha values of 0.891, 0.851, and 0.774, for the three subscales, CRNG, RULES, and LC, respectively, and 0.866 for the ETH construct. The construct reliability and validity for the second order latent variables are shown in Table 1.

Table 1 Assessment of the measurement model (second order latent variables)
Full size table

Perceptions of RI were measured using twelve items, labelled RI1 to RI12, extracted from the framework by Stilgoe et al. (2013) and used by Zhang et al. (2019a) and Hadj (2020) at Time 2. The scale measured the main four factors of the multi-dimensional construct RI. The factors are anticipation (ANTC), measured by RI1 to RI3, reflexivity (RFLX), measured by RI4 to RI6, inclusion (INCL) measured by RI7 to RI9, and responsiveness (RSPN) measured by RI10 to RI12. Sample items include “My company regularly reviews its objectives and methods of innovation” and “My company adapts to changes in public policies”. A score for each sub-dimension was created using the factor analysis results. The parcelling method was also used to transform the multi-dimensional construct RI to a first-order construct with four indicators using the four sub-dimensions’ composite scores. Reliability analysis showed Cronbach’s alpha values of 0.805, 0.820, 0.811 and 0.801 for the factors ANTC, RFLX, INCL and RSPN, respectively, and 0.903 for the RI construct.

Managerial discretion (MD) was measured using four items based on the scale used by Liao and Zhang (2020) and Gupta et al. (2019) at Time 2. The items of this construct are labelled MD1 to MD4. Sample items include “There are abundant available resources that can be mobilized by the firm’s managers” and “Managers of the firm generally have freedom in decision-making”. Reliability analysis showed a Cronbach’s alpha of 0.820 for this construct.

In addition to this, the study controlled for gender, age, education level, organizational tenure, and size of the company (Lee & Ha-Brookshire, 2018; Raile, 2013).

Data Analysis

The partial least squares structural equation modelling (PLS-SEM) was used to test the study’s conceptual model. The data analysis was conducted using Smart PLS v. 3.3.2 (Ringle et al., 2015). We chose to use PLS-SEM because this technique does not require the restrictive conditions of the alternative technique of covariance-based structural equation modelling (CB-SEM). The PLS-SEM technique involves two steps. In the first step, the measurement model is assessed by validating the outer model. In this step, the first-order variables’ reliability and validity are demonstrated by determining the latent variables’ convergent and discriminant validity and reliability. The second step involves an analysis of the path coefficients to determine the inner model’s fit (Zabkar, 2000).

Findings

Assessing the measurement model

To assess the measurement model, factor analysis was conducted to confirm the first-order latent variables’ uni-dimensionality, and the reliability and validity of the constructs were checked. The results are shown in Tables 2 and 3.

Table 2 Assessment of the measurement model (first order latent variables)
Full size table
Table 3 Discriminant validity
Full size table

Results of the factor analysis for the first-order latent variables, displayed in Table 2, revealed that all factor loadings were above the value of the 0.7 threshold. Moreover, bootstrap calculations indicate the significance of the factor loadings (all p-values were less than 0.0001). The average variance extracted (AVE) measuring the part of the indicators’ variance captured by each construct was above the required value of 0.5 (Fornell & Larcker, 1981). Thus, the constructs explained more than half of the variance of the indicators. Finally, the scale reliability was checked through both Cronbach’s alpha and composite reliability values. All these values were higher than 0.8 (Nunnally & Bernstein, 1994), indicating high-scale reliability and providing additional evidence of the first-order latent variables’ uni-dimensionality. These results demonstrated the convergent validity of the constructs.

The model’s discriminant validity was also demonstrated as the AVE of each first-order latent variable construct and its indicators is higher than the common variance with each of the other constructs of the model (Fornell & Larcker, 1981). The discriminant validity results in Table 3 show that the square root of the AVE for each construct is greater than the correlation coefficient of the construct with each of the other constructs.

Hypotheses testing

Results of the analysis of the inner and path coefficient calculations are shown in Table 4; Fig. 2. The calculated R2 values revealed that PL explained a large proportion of the ethical climate variable variance (R2 = 0.425). In addition, PL and ethical climate explained a considerable proportion of the perceived RI construct (R2 = 0.303). These results align with the required minimum values suggested by Chin (1998) for the model’s nomological validity to be satisfactory.

Table 4 Path coefficients results
Full size table
Fig. 2
figure 2

Results of the path calculations

Full size image

The structural model was examined by calculating the path coefficients representing the model’s direct relationships and assessing the proposed indirect effect. Moreover, the significance of these effects was determined via the bootstrap method of resampling with 5000 iterations (Davidson & Hinkley, 1997). The path calculation results (displayed in Table 4; Fig. 2) revealed that all direct and indirect effects are significant. In particular, the results showed that PL has a positive and significant direct influence on ethical climate (β = 0.652 and p-value = 0.000). Thus, the results provide significant evidence supporting Hypothesis 1. In addition, ethical climate of the organization was found to have a positive and significant direct influence on RI perceptions (β = 0.551 and p-value = 0.000). These results also provide strong evidence supporting Hypothesis 2. Further analysis revealed that PL has a significant indirect influence on perceived RI through ethical climate (β = 0.359 and p-value = 0.000). Hence, the proposed mediation in Hypothesis 3 is also supported.

Multi-group analysis was conducted to address the moderating effect of managerial discretion on the relationship between ethical climate and perceived RI as well as on the indirect relationship between PL and RI through ethical climate. The standardized score, MD, for managerial discretion, generated via factor analysis, was used to identify three groups. The 140 respondents with MD score of less than −0.5 are identified as low MD. The 158 respondents with MD score within 0.5 from the mean of 0 are identified as moderate MD, and the 189 respondents with MD score above 0.5 constitute the high MD group. The Smart PLS software was used to conduct the multi-group analysis between the two groups, high MD and low MD. The results, shown in Table 5; Fig. 3, reveal that the path coefficient from ethical climate to perceived RI (β = 0.355) for the high MD group is significant (p-value = 0.000). As for the low MD group, the path coefficient of 0.262 is smaller and not significant (p-value > 0.05). Similarly, the results show that the path coefficient from PL to RI via ethical climate (β = 0.215) for the high MD group is significant (p-value = 0.000) while it is smaller and not significant for the low MD group (β = 0.114, p-value > 0.05). These results provide evidence to support Hypotheses 4 and 5 that managerial discretion will moderate the relationship between ethical climate and RI and the indirect relationship between PL and RI, such that the relationships are stronger under greater managerial discretion.

Table 5 Multi-Group Analysis
Full size table
Fig. 3
figure 3

Moderation effect of managerial discretion on ethical climate and RI

Full size image

Discussion

The aim of this study was to examine whether PL leads to RI perceptions through ethical climate as well as investigate the moderating effect of managerial discretion in the process. Results supported Hypothesis 1 that PL is positively related to ethical climate. This finding adds value to the rising interest in the relationship between different leadership styles and ethical climate (Burton et al., 2017; Hansen et al., 2016) and to the interest in the linkage between PL and various outcomes. Our finding answered the research calls to further determine how an ethical climate is built (Manroop et al., 2014). The magnitude of this relationship was relatively strong (0.652) which shows how important a participatory approach is to an ethical climate. This finding also provides evidence of the positive effects that a participatory approach has in the context of firms in South Korea. Additionally, the results supported Hypothesis 2 that ethical climate leads to RI perceptions. This finding, therefore, answered research calls to further examine the link between ethical climate and employee-related outcomes (Pagliaro et al., 2018) and complements other studies that have attempted to link ethical climate to innovation in Asia Pacific, specifically in South Korea (Choi et al., 2013; Moon & Choi, 2014). The magnitude of this relationship emphasizes the important role that ethics plays in RI, but as it is considered relatively moderate (0.551), it would be worth investigating additional potential antecedents of RI perceptions within organizations. Results also supported Hypothesis 3 that PL relates to RI through ethical climate. This finding answered calls to further investigate the mechanisms in which PL occurs (Chan, 2019). It also complements other studies that have attempted to link PL to various employee perceptions (Lythreatis et al., 2019). This finding also provides further support for the mediating capacity of ethical climate that has been demonstrated in existing literature (Mayer et al., 2010). The indirect effect was not very high (0.359), which implies that there is a big opportunity to explore further mechanisms as well as apply sequential mediations in the process linking PL and RI perceptions. Finally, results supported Hypotheses 4 and 5 that managerial discretion moderates the relationship between ethical climate and RI and the indirect association of PL with RI through ethical climate, such that these relationships are stronger under higher discretion. These findings answered calls to further examine moderators in relation to ethical climate and its outcomes (Al Halbusi et al., 2020; Gonzalez-Pardon et al., 2008) as well as provided support to other studies that have explored managerial discretion in Asia Pacific (Lee, 1998; Liao & Zhang, 2020). As this investigation studied the relationships between employees and the leader that they are in contact with, it considered the managerial discretion at functional and middle levels rather than at the top executive level. Studying it at those levels has been declared essential for effective strategy implementation (Wangrow et al., 2015). Thus, our findings are in line with the limited studies that have attempted to do this (Caza, 2011; Kuratko et al., 2005). Overall, the moderation findings support the power that discretion has to influence outcomes.

Theoretical implications

This study makes a significant theoretical contribution as it provides evidence for a novel theoretical framework that links PL to RI for private firms in South Korea. It adds to the stream of literature that has started to look at RI from the leadership perspective (Moon & Kahlor, 2022; Voegtlin & Patzer, 2020). Studying the PL-RI linkage is important as PL particularly encompasses characteristics that can aid RI, yet there is a lack of work that connects the two variables theoretically. By establishing such a link, the study also expands the theoretical perspective on PL through illustrating its involvement in a unique process (Chan, 2019). Understanding different associations with PL is important as this style is progressively being adopted in the corporate world and has been regarded to be at the heart of the ongoing transition from top-down leadership. Further, by finding a significant linkage between PL and ethical climate, this study contributes to the stream of literature that seeks to explore drivers of perceived ethical climate (Goebel and WeiBenberger, 2017). The formed connection is important as it confirms that a participative leader contributes to shaping an ethical climate, adding to other benefits of such leadership that have been discovered so far in the literature.

This work also makes a noteworthy implication by extending work in the field of RI to cover RI perceptions of employees and discover new antecedents of perceived RI. It also enriches the RI literature by exploring mechanisms and boundary conditions in which RI perceptions are formed within organizations (Lubberink et al., 2019). Studying perceptions in this context is relevant since they have the power to influence outcomes (Lythreatis et al., 2021). This work also provides evidence to confirm a theoretical link between ethics and RI that has been assumed in the literature. Additionally, it shows the mediating effect of ethical climate in the context of RI.

Finally, the study makes a theoretical contribution by adopting the SIP theory (Salancik & Pfeffer, 1978) to explain the different relationships between PL, ethical climate, and RI perceptions. Thus, the theory was used to establish relationships between novel variables. SIP theory is also extended by uniquely applying it in the context of RI.

Practical implications

Considering the importance of employee perceptions in achieving desirable outcomes for both employees and the company (Ng et al., 2019), the findings of this study provide a good approach for managers in Asian countries like South Korea to follow if they wish to establish positive perceptions at work. This study allows managers to better understand employee perception formation within organizations. It suggests that companies should reflect on their leadership. Managers of firms in Asia should certainly adopt a more participative approach while interacting with their subordinates and encourage them to participate in some decisions and problem solving as this study provides evidence-based information of the power of participatory behavior enacted by leaders toward their employees. As South Korea is known to have a top-down innovation system which fosters collaboration (Dayton, 2020), this study shows the organizations operating in the country some of the benefits of a participatory approach. Therefore, leader development programs in Asia Pacific should encourage PL for creating ethical environments that allow employees to feel confident about getting involved, communicating, and exchanging new ideas. The findings of this study also demonstrate to Asian organizations the importance of having clear policies and rules and showing employees that they care about them. The findings can also help top managers realize that they need to expand their roles in interacting with employees rather than just answering to shareholders. Since innovation is only successful if it meets all stakeholder demands and as companies in Asia Pacific predictably target successful innovation to achieve organizational success (Scuotto et al., 2022), this study shows them a way to ease the concerns or frustrations of people regarding wrongdoings of innovation.

Conclusion, Limitations and Recommendations for Future Research

This study revealed a conceptual link between PL and RI perceptions through ethical climate. It also showed that managerial discretion strengthens the relationship between ethical climate and RI as well as the indirect relationship between PL and RI. Overall, it expanded research on leadership, ethics, and RI. It also provided empirical support in the fields of PL and RI within firms in Asia Pacific. However, despite careful considerations, this study has some limitations. It did not test PL against alternative leadership styles. It would be useful to advance this research by collecting additional data to control for competing constructs (Antonakis, 2017). Additionally, although this study collected data for different variables at two different points in time, to improve the study’s design, it can be followed up with a longitudinal design with longer time between the consecutive waves to track individual changes over time and infer proper causal relationships. Moreover, as empirical research on RI perceptions within organizations is relatively new, future research can test this study’s model in other countries within the Asia Pacific region. A comparison of the model in two countries within the region, for example, would help expand this research in Asia Pacific. Another research extension could be to test other mechanisms or sequential mediations in the relationship between PL and RI as well as other moderators in order to discover additional antecedents of RI perceptions and factors that may influence these perceptions. For example, researchers can examine how trust plays a role in the process linking PL to RI perceptions. Despite these limitations, this study provides intriguing findings that make important contributions to existing theory.