Acquiescence or Resistance: Group Norms and Self-Interest Motivation in Unethical Consumer Behaviour

: Understanding why consumers behave unethically has gained scholarly attention; many studies have examined it from psychological or social environmental perspectives. However, few studies provide the link between internal and external factors associated with unethical behaviours, and few studies explain why consumers ‘behave unethically knowingly’ in a China-speciﬁc context. Based on moral disengagement theory, the current research aims to provide a comprehensive developmental model to investigate how group norms and self-interest motivation a ﬀ ect consumers’ unethical behavioural intentions and to analyse the mechanism of ‘behaving unethically knowingly’. Findings from online surveys of 360 participants indicate that group recognition and an egoism motivation have positive e ﬀ ects on consumers’ unethical behavioural intentions, while group veto and an altruistic motivation have negative e ﬀ ects on unethical behavioural intentions; moral disengagement mediates the relationship between group recognition, group veto, egoism motivation, altruistic motivation and unethical behavioural intentions; moral identity negatively moderates the relationship between group recognition, altruistic motivation and unethical behavioural intentions. The theoretical and practical implications are also discussed.


Introduction
Imagine that you need a software to finish your work right now. There is a genuine software in front of you, but it costs a lot. A close friend shares you a pirated copy with the same utility. Which one would you choose? Or imagine that your friends and colleagues around you always use the pirated software, will you do the same? If you choose the pirated one, you behave unethically.
Unethical consumer behaviour has been defined as behaviour which is contrary to moral principles and the judgement standard as individuals or groups obtain, use and dispose of goods or services in the marketplace [1]; this includes varied illegal or immoral activities, such as shoplifting, cheating, purchasing or using counterfeit goods [2]. Vitell and Muncy (1992) proposed a four-factor solution of ethical beliefs that differentiates behaviours as follows: (1) actively benefiting from illegal activities, (2) passively benefiting from illegal activities, (3) actively benefiting from legal but questionable practices, and (4) 'no harm, no foul' activities [1]. It is noteworthy that consumers easily ignore ethical issues of the 'no harm, no foul' actions and there are long-term losses caused by this kind of behaviour [2,3], Therefore, this research mainly focuses on this kind of unethical behaviour. Consumer unethical consumption behaviour is a form of widespread social behaviour. However, it poses threats to the sustainable development of the national economy in China. For example, nearly 600 million of the counterfeits which are produced in China are circulated in the global marketplace over recent years, and it equals to more than 87% of the total counterfeit sales worldwide [4]. Consumers easily justify rationalisation of unethical behaviour, or they 'strategically forget' the ethical content involved while people generally know right from wrong [3]. Therefore, understanding why Chinese consumers behave unethically has gained scholarly attention [5]. The studies documenting various precursors of unethical consumer behaviours is critical for understanding the scope of this problem, which has been written can be placed into three categories: demographic factors, such as gender, age, education [6,7], individual factors, such as Machiavellianism, moral beliefs or moral judgment [8,9], and situational factors, such as ethical environment or unethical climate [6,7]. The findings indicated that unethical consumer behaviour can be examined by various situational and individual factors [2].
Among the various situational factors, what is observed in other consumers' unethical behaviours play an important role in the misbehaviour of the focal consumers such that some of them may imitate it [9] or they get a signal that they are allowed to do so. Prior studies have indicated that China is the largest counterfeit market and consumers may not consider a 'no harm, no foul' action as unethical [4]. This social phenomenon may hide some social cues. The deviant group norms provide a social context where techniques, attitudes and rationalisation for misbehaviours are learned and internalised [10]. Previous studies have shown that ethical values are not only about behaving legally or moral acceptably, but also include the concepts of moderation based on Confucianism, social norms arising from collectivism [4]. We predict that Confucianism may influence this social phenomenon. By this logic, it is necessary to examine the role of group norms play in the literature of consumer ethics. However, the group norms are usually discussed in the literature of consumer or organisational behaviour under the normal context. Thus, we argue that group norms will provide insight into the consumer ethics to reveal the social climate of unethical behaviour.
Nevertheless, most scholars have examined consumers' intrinsic motivations in the literature of consumer ethics, such as values, moral obligations, and moral identities, which may narrow the research on unethical consumers' motivation and hinder a deeper understanding of their motivations in other respects [11]. Past studies have shown that the pursuit of the self-gains may be the motive of unethical consumer behaviour, such as money-saving and convenience [10]. However, it may harm others' welfare in the pursuit of self-gains, and an ethical issue may occur. The current study focuses on egoistic and altruistic motivations to shed light on consumers' motivation in the context of unethical consumption.
However, an individual's attitude is not always consistent with their actual behaviour, and there is a gap between attitudes and actual behaviours [12]. Based on moral disengagement theory, this research aims to explain why consumer 'behave unethically knowingly'. Different from prior studies adopting The Theory of Planned Behaviour in the literature of consumer ethics, we found that moral disengagement theory contributes to explaining this 'attitude-behaviour' gap due to its flexible and realistic assumptions [13]. As a result, this study adopts group norms and the self-interest as the precursors of the unethical consumer behavioural intention; it explores the mediating role of moral disengagement based on moral disengagement theory and the moderating role of moral identity. We collected 360 valid data from Beijing by questionnaire survey, and hypotheses are tested by hierarchical regression and Bootstrap analysis. This paper will focus on the following three issues: first, the relationship between the normative influence of the reference group and the intention of an individual's unethical behaviour; second, the relationship between the individual's self-interest motivation and unethical behavioural intention. Third, the mediating role of moral disengagement and moderating role of the moral identity.

Reference Group and Unethical Consumer Behaviour
Deviant group norms provide a social context where techniques, attitudes and rationalisation for misbehaviours are learned and internalised [10]. It is likely that an individual engages in unethical behaviours more when the more he is exposed to an unethical climate [10,14]. For example, in the studies of music piracy, scholars find that it is a social behaviour widespread and they perceive peer influence as the significant factor [10]. The influence of peers is a kind of reference group influence. The reference group refers to a real or imaginary individual or group of people who have a key influence on an individual's perceptions, beliefs, attitudes and decisions in a specific consumption situation [15][16][17]. Researchers have divided the influence of the reference group into informational influence and normative influence [10,15,18]. Informational influence refers to the process of individual consumers observing the consumption decisions of reference groups for risk avoidance considerations, and then searching for information and adopting them [19,20]. Normative influence originates from a desire to satisfy the need for belonging, to obtain social approval, or to avoid exclusion to establish a harmonious and satisfactory relationship with each other [10,21]. Considering various findings on reference group influences, we find that the literature mainly focuses on loyalty [22], consumer purchasing intention [20,23,24] and brand preference [15,19,25], which are under ethical consumption. Previous studies have shown that the informational and normative influences of reference group on product and brand decisions will shape the consumers' attitudes, norms and aspirations [26].
Although consumers tend to have a favourable attitude to certain products or express an intention to pay for them, they still do not conduct an actual purchase [12]. Scholars have argued that individual consumers are more susceptible to uncertain factors, so they cannot completely make choices based on personal value judgements when exposed to a series of uncertain consumption scenarios, such as groupon, green consumption, or pirated consumption [27]. According to Szmigin et al. [28] (2009), individual consumers may not make positive ethical choices because of several social and economic forces. In the literature of consumer ethics, consumer unethical behaviour is a widespread social behaviour, which involves a relationship embedded in a social network between different actors [29]. Therefore, an individual consumer's attitude, intention and behaviour are more susceptible to the influence of the reference group. For example, Yang et al. [10] (2015) indicated that individual consumers who are affected by informational and normative influences of the reference group tend to download or share unauthorised music. In particular, there is no direct harm or detrimental impact arising from the 'no harm, no foul' actions, and the other kinds of unethical consumption involve risks of being caught or getting fined to a certain extent. In this case, the individual consumers have a certain ability to perceive risks [5,10]. We argue that they are susceptible to the normative influence of the reference group. Consistent with this reasoning, Yang et al. [10] (2015) found that consumers are more susceptible to the normative influence when they share unauthorised music.
As stated earlier, unethical consumer behaviour is a social behaviour which can be examined by various situational factors, in particular, cultural norms and values shed light on it. For example, Kavak et al. [30] (2009) suggested that Turkish consumers perceive that 'no harm, no foul' actions are not unethical, and a possible explanation is that Turkey is a collectivist nation. Following this logic, it is likely that this kind of unethical behaviour reveals some issues hidden in the culture [30]. Prior studies have shown that ethical values are not only about behaving legally or morally, but also include the concepts of moderation based on Confucianism, social norms arising from collectivism [4]. Under the influence of collectivist culture, Chinese emphasise the social self, and they pay more attention to the influence of social norms on their own behaviours. They also have a deeper experience of an urgent need for 'belonging' [31]. In contrast to the extreme individualism that the West admires, Chinese people are more susceptible to the influence of the reference group. This is consistent with the existing findings that the individuals of interdependent self-construal influenced by collectivism culture consider the self as a member of a group and have a desire to connect with others, and they are more affected by the reference group [32]. As stated before, individual consumers are more susceptible to the normative influence of the reference groups when exposed to the kind of unethical behaviour of 'no harm, no foul'.
The normative influence of the reference group focuses on an individual's motivation to comply with the group norms in the consumption context [33]. Since an individual's own confidence towards the correct judgement of the situation will be significantly reduced in the uncertain consumer environment, they usually do not consider their own consumption based on absolute and completely objective attitudes, while the normative influence of the reference group increases with a decrease in the clarity of an individual's judgement of the situation in which they are situated [34]. Therefore, the judgement of the reference group is considered as a more or less reliable source. On the other hand, individual consumers are pressured by reference groups, such as family members, friends, experts, etc., and produce behavioural norms that are forced to conform to the expectations of others to obtain rewards or avoid punishments (either physical or spiritual) for utilitarian purposes, then for gaining group recognition or avoiding group veto [19,35]. For example, sharing unauthorised files, films, or music in the P2P online community can enhance reciprocity, whereas there is a risk of being excluded from groups for the free-riders. According to social learning theory, rewards that make individuals repeat behaviour and punishment lead to a decrease in behaviour. Thus, it may eventually contribute to the unethical behaviours of 'no harm, no foul' when the reference group express acquiescence of it. Conversely, the intention of unethical behavioural for individual consumers will decrease when the reference group denies certain unethical behaviours. In this manner, we propose the following hypotheses: Hypotheses (H1a). Group recognition is positively related to the intention of unethical behaviour.

Hypotheses (H1b).
Group veto is negatively related to the intention of unethical behaviour.

Self-Interest Motivation and Unethical Consumption Behaviour
The study of individual consumers' unethical behavioural motivations has long been discussed, based on two conflicting perspectives: ethical decision-making theory and rational decision-making theory [36,37]. We found that literature addressing consumer ethics that integrate these two theories into a research framework is rare; they have been separated for a long time. For example, Tang et al. [38] (2013) empirically found that the pursuit of personal profits is the direct cause of unethical behaviour. The topics of rational and moral issues in unethical behaviours are worth further study.
Rational decision-making theory claims that the individual is based on the logical starting-point of the unethical behaviour in consumption that can bring benefits to itself, and maximises expected utility as a decision-making goal, which relates to egoism [39,40]. Past studies have suggested that attention is a limited cognitive resource and individuals tend to focus attention on the information which is directly relevant to their motivation. In other words, individual consumers with a high level of egotism may lead them to pay more attention to self-gains and to ignore cues related to ethical issues [41]. According to the hypothesis of the rational economic person, when the agent faces multiple choices in economic activities, the subject tends to make a behavioural judgement based on the maximisation of the benefit, considering the balance between cost and profit. It is not difficult to find that the pursuit of personal benefits, maximised as the primary choice of multiple factors, will eventually lead to ethical misconduct [2,42]. For example, consumers purchase fashion counterfeits of highly recognisable brands due to their low price and the satisfaction of consumers' vanity [4]. From the set of tools measuring unethical behaviour, domestic and foreign scholars have adopted the Muncy-Vitell consumer ethics scale, which measures the ethical transgressions under the profit motivations [43]; this tool is highly consistent with the concept of egoism in rational decision-making theory. Hong et al. [44] (2012) indicated that unethical behaviour challenges the notions of fairness for one's own benefits at the expense of others' loss. Ethically problematic consumption is more likely to be tolerated because of personal benefits [2]. By this logic, the satisfaction of the egoism motivation is an important factor in ethical issues. Egoistic individuals are likely to condone unethical behaviour and to endorse intention to engage in such behaviour themselves. Consistent with this reasoning, Leonard et al. [45] (2017) found that an individual with a high level of egoism tends to behave unethically for personal gratification.
It is worthwhile to note that the calculation of cost-benefit is almost impossible when the individual is in a context with a high level of uncertainty, such as the adverse consequences that affect the healthy and sound development of society, social consensus on this behaviour, etc [46]. Rational decision-making theory, which has expected utility maximisation at its core, does not provide people with a standard of moral judgement [46,47]. Therefore, it is not reasonable to consider issues involving ethical principles only with rational decision-making theory. In essence, there is a fundamental difference between ethical behaviour and other behaviours in the consumption context. In the research on ethical transgressions, egoism is a significant predictor, but this does not mean that other factors will not have antecedent effects on this behaviour [11]. Specifically, actors are affected by the classic cost-benefit paradigm, actively acquiring, for example, cheaper software; on the other hand, this behaviour inevitably leads software developers and distributors to a huge economic loss, which poses a threat to the sound development of society. Therefore, the unethical behaviour of individuals driven by the egoism motivation induces damage to the profits of other subjects. In this case, the actor is affected by the psychological conflict between self-profitability and profitability of the other, and a moral dilemma will follow. It can be seen that rational decision-making theory and ethical decision-making decision theory have developed independently for a long period of time, largely because scholars have not strictly classified unethical behaviour types [5], and they equate unlawful behaviour with behaviour that is an ethical transgression, but lawful [3]. Therefore, the complexity of unethical consumers' motivations and the coexistence of egotism and altruism should be highlighted [11]. It is because of the 'moral-hiddenness' of this kind of 'no harm, no foul' behaviour on ethical issues that scholars have neglected its ethical issues, keeping the discussion of this problem in a state of mutual separation for a long time [4]. Thus, it is necessary to incorporate ethical decision-making theory into the research framework.
Ethical decision-making theory shifts the concern to the satisfaction of the other's welfare, and makes decisions based on the potential impact on other stakeholders to determine whether to implement the behaviour, that is, altruism [36]. Although the relationship between altruism and unethical consumer behaviour intention seems self-evident, the relationship has not been examined sufficiently and needs further attention. In the literature of business ethics, it has been indicated that empathy leads to less ethical transgressions and to more altruism and prosocial behaviour [48,49]. An altruistic individual tends to have more empathy, and they can internalise the moral reasoning that focuses on the others' welfare [5]. For example, empathy motivates individuals to pay more attention to others' welfare through altruism, and promotes consumers to buy ethical products. In this vein, altruism plays a significant role in formulating a strong sense of moral emotion [11]. As stated earlier, strong motivation can lead individual to focus the cues related to his motivation. In the case of 'no harm, no foul' actions, altruistic consumers may pay more attention to the consequences of this behaviour and they do not wish to harm the interest of stakeholders, and they may be less willing to implement the behaviour due to altruistic considerations [46,50]. In this manner, we propose the following hypotheses: Hypotheses (H2a). Egoism motivation is positively related to the intention of unethical behaviour.

Hypotheses (H2b).
Altruistic motivation is negatively related to the intention of unethical behaviour.

The Mediating Role of Moral Disengagement
As mentioned above, individual consumers who are affected by collectivist culture will not perceive the context in question as an ethical issue [4,30]. The key to understand this phenomenon lies not only in culture, but also in examining why individual consumers still tend to 'conduct ethical transgressions knowingly' when in ethical dilemmas. Previous studies have indicated that some individuals find it easier to disengage their moral principles than others and endorse intentions to behave unethically, while people generally know right from wrong [51]. In other words, it seems easier to justify engaging in the unethical behaviour through the mechanism of moral disengagement.
Thus, we choose moral disengagement theory to explain why some individual consumers 'knowingly behave unethically'.
Moral disengagement refers to a psychosocial mechanism whereby individuals selectively disengage their moral self-regulatory process from harmful conduct. Consequently, their guilt and self-accusation experienced due to the unethical behaviour will disappear, which is an environment-related cognitive process [41,52]. For example, 'free-riding' in environmental pollution is an obvious evidence of moral disengagement; environmental destroyers believe that personal hazards to the environment are insignificant compared with groups. Eight interrelated mechanisms in moral disengagement that facilitate unethical behaviours have been proposed, such as moral justification (justifying the misconduct as acceptable based on the potential outcome), advantageous comparison (justifying the misconduct by comparing it to a more detrimental behaviour), diffusion of responsibilities (where a shared decision to act immorally means no individual involved in the misconduct believes they are fully blamed for the events which occur), displacement of responsibilities (one's misconduct is attributed to being influenced by authorities), euphemistic labelling (using language to distorting what happened), distortion of consequences (justifying the misconduct by ignoring or minimising the detrimental consequences), dehumanisation (rejecting one's human qualities and considering as bestial) and attribution of blame (blaming the misconduct on the provocation of victim who brought trouble on themselves), which is not a single mechanism [53]. The existing literature on moral disengagement is mainly focusing on the organisational context. For example, Baron et al. [41] (2015) found that moral disengagement contributes to deactivating the self-regulatory process and propensity to make unethical decisions by entrepreneurs, and they are not susceptible to negative feelings, such as guilt. Although there is a large body of literature on moral disengagement in business ethics, relevant studies in the literature of consumer ethics is scarce [5]. In view of the fact that moral disengagement involves both individual and organisational levels [54], we argue that it is necessary to further reveal the role played by moral disengagement in the context of individual consumer unethical behaviour to answer why individuals' knowingly go astray in the ethical domain. According to the prior studies addressing consumer ethics, Theory of Planned Behaviour is often adopted to set a framework of unethical consumer behaviours [55]. The Theory of Planned Behaviour suggested that individuals are rational in their thinking and deliberate in behaviours and it has been adopted to explain the consistent relationship between attitudes and actual behaviour [56]. However, consumers' attitudes are not always incongruent with their actual consumption behaviours, especially in unethical contexts [12], and moral disengagement theory contributes to explain this 'attitude-behaviour' gap due to its flexible and realistic assumptions [41].
According to moral disengagement theory, moral disengagement deactivates the self-regulatory mechanisms which hinder individuals to engage in the misconduct that they may disapprove without experiencing self-censure [57]. Although moral disengagement is an intrinsic psychosocial mechanism, it can be influenced by other situational factors. Prior studies have indicated that the effect moral disengagement has depends on the interaction between individuals' characteristics and the environmental variables [13]. According to social-cognitive theory, an individual's cognition dynamically adjusts with changes in the external environment, and the rationality of self-behaviour is based on moral standards and perceived environmental factors. Furthermore, the level of moral disengagement reflects the level of an individual's moral judgement. By this logic, changes in the external environment will inevitably have an impact on the level of individual's moral disengagement [52]. For example, moral disengagement mechanisms are internalised in the process of organisational socialisation and the organisational safety culture affects the degree to which moral disengagement of employees happens [13]. In this study, we consider group recognition and group veto as the precursors of moral disengagement and examine the impact of situational factors. To further explain it, we will specifically analyse the differences in the mediatory mechanisms of group recognition and veto in predicting the intention of ethical transgressions.
Under the influence of collectivist values, there is a strong demand for the construction and maintenance of social relations [31]. Individuals will constantly examine others' reactions to their behaviours, thereby making judgements to decide whether they are acceptable [29]. When individuals perceive themselves as marginalised, they tend to get out of exclusion by various means that can resatisfy their sense of belonging, even if behave unethically [29,58]. Under the moral disengagement mechanism, individuals transfer their responsibility to the reference group, ignoring the harmful consequences of unethical behaviour on the sustainable development of society and the interests of others. This results in such ideas as, 'it is the reference group acquiescing me to do', or, 'there are so many people buying pirated products now, which has had a harmful impact on the sustainable development of the economy, and my own behaviour has little impact'. The mechanisms of displacement and diffusion of responsibility will perform when the self-regulatory function is selectively deactivated [13]. Most research has found that moral disengagement entails cognitive mechanisms that discourage ethical behaviour [5]. In this sense, if individual consumers find excuses for engaging in the 'no harm, no foul' unethical actions by moral disengagement under group recognition, and lower moral sanctions make such behaviour widespread in society, this harms intellectual property construction and ultimately endangers the healthy development of economies and societies. With respect to the group veto, the object of the responsibility transfer disappears when the behavioural decision is rejected by the reference group; the function of moral disengagement mechanism is blocked, while the moral self-regulation function plays a part; and individuals are unlikely to behave unethically under negative emotions. In this manner, we propose the following hypotheses: Hypotheses (H3a). The relationship between group recognition and unethical behavioural intention is mediated by moral disengagement.

Hypotheses (H3b). The relationship between group veto and unethical behavioural intention is mediated by moral disengagement.
We consider egoism motivation and altruism motivation as the antecedents of moral disengagement, since motivation is directly related to individual's cognitive process. Motivation can not only lead individuals to eliminate unrelated information, but also interpret the meaning to the cues related to the motivation, even if these kinds of cues breach moral standards [41]. By this logic, the individuals with high egoism tend to make unethical behaviour seem desirable [59]. Egoists pay attention to the maximisation of their own welfare, and have a higher interest in the satisfaction of private profits. They are more likely to get rid of the shackles of the moral self-regulation to engage in unethical behaviours [60]. According to moral disengagement theory, actors tend to contrast this behaviour with a worse behaviour, using the advantageous comparison mechanism to make unethical behaviour acceptable; for example, they compare purchasing of counterfeits with shoplifting to make the former more acceptable. On the other hand, actors spread responsibility to the group to alleviate the harmfulness of behaviour by the mechanism of diffusion of responsibility [61]. Individual consumers may believe the following: 'I am not the only one buying the counterfeits; others will do so as well, anyway, the law cannot be enforced when everyone is an offender'. Therefore, individuals with a higher self-profit motive will be more speculative and less subjected to social norms and ethics when in a moral dilemma, and more likely to make unethical behavioural decisions. This is consistent with other studies, which suggested that individuals with higher levels of Machiavellianism are likelier to implement ethical transgressions under moral disengagement, and Machiavellianism is an indicator of individual egoism [53].
Contrary to egoists, altruists incorporate a realisation of others, and even social welfare, into their own behavioural decisions. They pay more attention to the adverse effects of their behaviour on other stakeholders. Previous studies have indicated that altruism draws individuals' attention to a wide range of information of personal satisfaction and well-being. It may strengthen self-regulatory mechanism and get rid of the influence of moral disengagement [41]. Studies have also shown that altruists usually have a higher level of empathy [48]. According to the empathy-altruism hypothesis, feelings of empathy are the primary explanation for altruistic acts for which the actors do not expect to receive material benefits, and it can construct the same emotional experience with others. Moral emotion deep in memory can be awakened when altruists with greater feelings of empathy observe a helper in distress [62]. In this vein, individuals affected by altruistic motivation have lower moral disengagement in the face of ethical dilemmas, and the less likely it is that individuals engage in the unethical behaviour [61]. In this manner, we propose the following hypotheses: Hypotheses (H4a). The relationship between egoism motivation and unethical behavioural intention is mediated by moral disengagement.
Hypotheses (H4b). The relationship between altruistic motivation and unethical behavioural intention is mediated by moral disengagement.

The Moderating Effect of Moral Identity
As stated before, individual consumers affected by intrinsic or extrinsic factors endorse intentions to behave unethically through the psychosocial mechanisms of moral disengagement. However, consumers may react differently in the aforementioned process. In reality, some may engage in the unethical behaviour more easily, while others may not. Previous literature addressing consumer ethics has found that moral identity is negatively associated with unethical behaviour and moral disengagement and provides insight into the probable relations among them [5,57]. Moral identity is considered as a cognitive schema of self-conception, organised around a set of moral traits, and as a construct that reflects the extent to which being a moral person is central to one's sense of self between different individuals [63]. It is a distinct mental representation of what a moral individual is likely to feel, think and do [64]. Moral identity internalisation and symbolisation are two dimensions of it, which represent the private and public aspects, respectively [5]. The internalisation dimension puts being an ethical person at the core of one's self-concept, reflecting the degree of moral traits to self-importance. The symbolisation dimension is about how an individual express to others his or her moral qualities, through a series of actions and desires to have certain moral traits [65].
Studies have shown that moral identity influences people's situational considerations based on moral judgement and choice, and it has a negative relationship among moral disengagement and unethical behaviours [57,66]. For example, Aquino et al. [67] (2007) empirically demonstrated that the higher an individuals' moral identity, the less likely they are to exhibit moral disengagement, indicating that moral identity significantly hinders the mechanisms of moral disengagement. In particular, moral identity inhibits the cognitive mechanism that distort consequences by minimising or rationalising the level of harm to others' welfare from questionable behaviours [5]. According to Rodriguezrad & Hidalgo [68] (2017), the construct of moral identity plays a significant role in understanding the relation between moral reasoning and ethical behaviour. Although most studies have examined the direct relationship between moral disengagement and unethical behaviour [5,68], endeavours to examine the moral identity as a potential boundary condition are insufficient [6,66]. In other words, moral identity may serve as a direct predictor of unethical behaviour and may function as a synergistic interaction between motivations and group norms.
Consumers may not behave consistently when they observe other consumers' unethical behaviours. In other words, it is likely that individuals with salient moral identities will have negative attitudes towards the unethical behaviour. Moral identity is important for dealing with the social information related to morality and guiding moral action [5]. Detert et al. [69] (2008) indicated that individuals with salient moral identities will get rid of the negative effect of moral disengagement. Specifically, since the psychological perception distance between consumers with salient moral identities and out-group members is significantly shortened [66], and the perceived similarity between the two is gradually increasing, then consumers with a higher moral identity show more emotional tolerance and moral concern to out-groups [65]. It is likely that individuals with salient moral identities will have negative perceptions of unethical behaviour that do harm to or are unfair to others (e.g., damaging the profits of related actors in purchasing the counterfeits, such as software developers) [70]. The moral disengagement mechanism selectively fails in the case of the functions of the moral code, thereby weakening the relationship between group recognition, group veto, egoism motivation, altruistic motivation and moral disengagement. In summary, individuals with salient moral identities pay more attention to the impact of unethical behaviours on their moral principles, even if the reference group expresses this behaviour to be acceptable, or driven by egoism motivation, has led to the failure of the moral disengagement mechanism. This is consistent with the findings that individuals with a salient moral identity will further hinder the moral disengagement mechanisms that diffuse responsibilities to others and tend to behave ethically [67]. Consequently, we propose the following hypotheses: Hypotheses (H5a). Moral identity plays a negative role in moderating the relationship between group recognition and moral disengagement.
Hypotheses (H5b). Moral identity plays a negative role in moderating the relationship between group veto and moral disengagement.
Hypotheses (H6a). Moral identity plays a negative role in moderating the relationship between the egoism motivation and moral disengagement.
Hypotheses (H6b). Moral identity plays a negative role in moderating the relationship between the altruistic motivation and moral disengagement.
In a further analysis, we combine hypothesis 3, hypothesis 4, hypothesis 5 and hypothesis 6. We believe that moral identity can not only moderate the relationship between group recognition, group veto, egoism motivation, altruistic motivation and moral disengagement, but it can also moderate the mediating role of moral disengagement in group recognition, group veto, egoism motivation, altruistic motivation and unethical behaviour. Even in the face of group recognition and egoism motivation, individuals with a salient moral identity will also make decisions according to their moral standards, paying more attention to the detrimental consequences of behaviours than finding excuses for the rationalisation of ethical transgressions to reduce the willingness of misconduct.
Hypotheses (H7a). Moral identity negatively moderates the mediating role of moral disengagement in the relationship between group recognition and unethical behavioural intention.
Hypotheses (H7b). Moral identity negatively moderates the mediating role of moral disengagement in the relationship between group veto and unethical behavioural intention.
Hypotheses (H8a). Moral identity negatively moderates the mediating role of moral disengagement in the relationship between egoism motivation and unethical behavioural intention.
Hypotheses (H8b). Moral identity negatively moderates the mediating role of moral disengagement in the relationship between altruistic motivation and unethical behavioural intention.
The interactions and relationships of antecedents and outcomes investigated in the research model are shown in Figure 1.

Design and Sample
Before the formal questionnaire survey, ten consumers were selected for preinterviews about their attitudes, willingness and causes of engaging in unethical behaviour under the consumption situation to provide a reference for the design of the formal questionnaire. A five-day street interception survey was conducted from 10-15 August 2017, and a total of 42 valid questionnaires were collected. These include basic information on consumer age, gender, marriage and personal monthly income in addition to the antecedents, such as individual consumers' willingness on buying pirated goods, group recognition, group veto, egoism motivation, altruistic motivation, moral identity and moral disengagement. We designed 43 measurement items. The results of this presurvey were used as the basis for the formal investigation and adjustment. According to the participants responding to the doubts in the questionnaire filling process, we invited three doctors and professors from the marketing field to attentively review and discuss the questionnaire items. In the end, we obtained 36 measurement items.
We surveyed individual consumers in Beijing from October 2017 to May 2018, distributing 450 electronic questionnaires, of which we took 392 back. We invited two graduate students from the Department of Marketing to screen the questionnaires according to the following strict conditions: (1) the answering time was less than or equal to 100 seconds; (2) the answering sheet had a large number of consecutive answers of the same choice; (3) the answering sheet had obvious contradictory option answers. After the above strict screening, 360 valid questionnaires were obtained. Of the 360 participants, 198 (55.0%) were male and 162 (45.0%) were female, ranging in age from 19 to 45. The majority of the participants were married (55.6%).

Measures
We used multiple-item scales to test each construct. The measurement items were based on previous studies that adopted mature scales. We invited three doctors, professors of marketing and a professor of English to translate and back-translate repeatedly and modify the items according to Chinese expression habits. Since research on unethical behaviour usually involves sensitive issues, to reduce the impact of social desirability and impression management on the authenticity of the answering sheet, we invited relevant experts to bring scenarios into the compilation of measurement items to reduce the psychological defence of the respondents, thereby enhancing the effectiveness of the measurement. All items were measured on a seven-point, Likert-type scale ranging from 'strongly disagree' (1) to 'strongly agree' (7).
Measurements included group recognition, group veto, egoism motivation, altruistic motivation, moral identity, moral disengagement and consumer unethical behavioural intention. To measure group recognition and veto, we adopted the scales from Park & Lessig [16] and Bearden et

Design and Sample
Before the formal questionnaire survey, ten consumers were selected for preinterviews about their attitudes, willingness and causes of engaging in unethical behaviour under the consumption situation to provide a reference for the design of the formal questionnaire. A five-day street interception survey was conducted from 10-15 August 2017, and a total of 42 valid questionnaires were collected. These include basic information on consumer age, gender, marriage and personal monthly income in addition to the antecedents, such as individual consumers' willingness on buying pirated goods, group recognition, group veto, egoism motivation, altruistic motivation, moral identity and moral disengagement. We designed 43 measurement items. The results of this presurvey were used as the basis for the formal investigation and adjustment. According to the participants responding to the doubts in the questionnaire filling process, we invited three doctors and professors from the marketing field to attentively review and discuss the questionnaire items. In the end, we obtained 36 measurement items.
We surveyed individual consumers in Beijing from October 2017 to May 2018, distributing 450 electronic questionnaires, of which we took 392 back. We invited two graduate students from the Department of Marketing to screen the questionnaires according to the following strict conditions: (1) the answering time was less than or equal to 100 seconds; (2) the answering sheet had a large number of consecutive answers of the same choice; (3) the answering sheet had obvious contradictory option answers. After the above strict screening, 360 valid questionnaires were obtained. Of the 360 participants, 198 (55.0%) were male and 162 (45.0%) were female, ranging in age from 19 to 45. The majority of the participants were married (55.6%).

Measures
We used multiple-item scales to test each construct. The measurement items were based on previous studies that adopted mature scales. We invited three doctors, professors of marketing and a professor of English to translate and back-translate repeatedly and modify the items according to Chinese expression habits. Since research on unethical behaviour usually involves sensitive issues, to reduce the impact of social desirability and impression management on the authenticity of the answering sheet, we invited relevant experts to bring scenarios into the compilation of measurement items to reduce the psychological defence of the respondents, thereby enhancing the effectiveness of the measurement. All items were measured on a seven-point, Likert-type scale ranging from 'strongly disagree' (1) to 'strongly agree' (7).
Measurements included group recognition, group veto, egoism motivation, altruistic motivation, moral identity, moral disengagement and consumer unethical behavioural intention. To measure group recognition and veto, we adopted the scales from Park & Lessig [16] and Bearden et al. [15], respectively, which were modified to better fit the context of this study. Machiavellianism is usually regarded as an index to measure individual egoism. Jakobwitz & Egan (2006) pointed out that Machiavellianism is a strategy that emphasises the realisation of self-interest while ignoring the benefits of others, leaving traditional moral concerns behind [71]. Therefore, we regarded the level of Machiavellianism as an index to measure the strength of individual egoism [71]. Five statements, adapted in part from Jakobwitz & Egan [71], were used to measure the egoism motivation. Altruism is a trait of volunteering to help others without seeking future rewards. Research has shown that altruism is negatively correlated with Machiavellianism. Altruism is measured mainly by social responsibility and integrated into the Big Five Personality Scale. Social responsibility is the intrinsic motivation for individuals to exhibit behaviour and is positively correlated with individual altruism. For altruistic motivation, a measurement was borrowed from Berkowitz & Lutterman [72] and Rammstedt & John [73]; there are five items to measure it. To improve the legibility of the questionnaire, we drew on the practices of Vitell & Muncy [1] to compare the real situation with the behaviours such as 'the actively benefiting from illegal activities', 'passively benefiting activities', 'actively benefiting from legal but questionable practices activities' and 'no harm activities' in the consumer ethics scale. Because piracy is a very common and nuanced behaviour in China, it provides an ideal sample source for measuring the dependent variable. We used Bhattacherjee's [74] scales to measure unethical consumer behavioural intention; they were modified to better fit the context of this study. For moral disengagement, we used the Moore et al. [53] brief scales. To measure moral identity, we borrowed the Aquino &Reed [64] scales. The scales and construct items are in Appendix A. Table 1 provides the means, standard deviations, correlations and scale reliabilities of various constructs. Cronbach's alpha was used to estimate the internal consistence reliability of the scale. These values for subscales ranged from 0.827 (group veto) to 0.939 (group recognition). All values surpassed the recommended value of 0.700. Composite reliability (CR) was applied to test the internal consistency of the multi-item scales included in the model. As shown in Appendix B, the CR of each construct ranged from 0.888 (group veto) to 0.980 (unethical behavioural intention), exceeding the suggested minimum of 0.600. The result shows that internal consistency of the measurement model was adequate for further analysis of the structural model. Convergent and discriminant validity tests were performed to examine construct validity. The standardised factor loadings and average percentage of variance extracted (AVE) were used to measure convergent validity. All loadings in the constructs were higher than the suggested value of 0.500, and all AVE estimates for each construct exceeded the suggested minimum of 0.500. By comparing the square root of the AVE in each construct to the correlation coefficients between two constructs, we found that the constructs in the measurement model seemed to have acceptable levels of the discriminant. The test of common method variance (CMV) applied Harman's one-factor test, but scholars have found that this method has defects in testing the hypothesis, and it does not reach any expected control effect. Therefore, we used the following two aspects to assess CMV, according to Podsakoff et al. [75] (2003). First, we adopted the following methods to control the study's procedures: (1) the neutral prompt expression was used (the purpose of the survey and the measured variables were not specified) when designing measurement items for each construct; (2) in the design of the questionnaire structure, the order of items in the scale was optimised to balance the order effect in the measurement, and avoid the speculation of the survey object; (3) in the process of conducting research, it was emphasised to the respondents that the results of this survey were only used for academic research and conducted anonymously to reduce their sensitivity. Second, we controlled for the effects of a directly measured latent methods factor to access the CMV using a statistical control. We directly measured the presumed cause of the method bias and modelled it as a latent construct, allowing the indicators of the latent construct to load on the primary seven-factor model as well. The results showed that the change in χ 2 (∆χ 2 = 272.020, ∆ df = 135, p < 0.001) is significant in the eight-factor model. However, Anderson & Gerbing [76] argued that ∆χ 2 is more affected by sample size. Therefore, we controlled common method bias by comparing the change in the other model fit index before and after, and the range was no more than 0.02 (∆CFI = 0.019, ∆TLI = 0.02, ∆IFI = 0.019). This provided further evidence that CMV was not a problem.

Direct Effect and Mediating Effect Analyses
Prior to the hypothesis testing, we conducted analyses to determine the acceptability of fit of the measurement models. Root mean square error of approximation (RMSEA), Comparative Fit Index (CFI), Incremental Fit Index (IFI) and Tucker-Lewis Index (TLI) by Analysis of Moment Structure (AMOS) software were used to access the model fit. The results indicate (see Table 2) that only the fits of the seven-factor model, which is the measurement model, were within the acceptable range, and the fits of the other models were unsatisfactory. The measurement model fit values are as follows:   According to the acceptable fit of the measurement model, we proceeded to hypothesis testing. H1 and H2 predict the simple relations between group recognition, group veto, egoism motivation, an altruistic motive and unethical behavioural intentions. To test these hypotheses, we regressed the unethical behavioural intention variable on group recognition, group veto, egoism motivation and altruistic motivation in separate regressions. The results of these analyses are shown in Table 3. As the table shows, both hypotheses were supported. Both group recognition and egoism motivation were positively related to unethical behavioural intention, while group veto and altruistic motivation were negatively related to unethical behavioural intention. In summary, these results supported H1 and H2. According to the acceptable fit of the structural model, we proceeded to test the mediation hypotheses (H3, H4). Following Preacher & Hayes' [78] recommendations for testing the mediator, we adopted bootstrapping methods to test our hypotheses. In particular, we used the procedure by Preacher et al. [79] (2007) and computed bias-corrected bootstrap confidence intervals to test the predicted mediation process. Using Hayes' [80] SPSS macro, we computed regression equations and estimated the mediator variable models, using moral disengagement as the mediator. This enabled us to estimate indirect effects by bootstrapping methods (2000 bootstraps). If the bootstrapped confidence interval does not include 0, the indirect effect is significant, and the mediating process is supported. The results are shown in Table 4. It can be seen that under the 95% confidence interval, the bootstrapped confidence interval of the indirect path of the mediation included zero, indicating that the relationships between group recognition, group veto, egoism motivation, altruistic motivation and unethical behavioural intention were mediated by moral disengagement. To further analyse the type of mediation effect, the influence of group recognition, egoism motivation and altruistic motivation on unethical behaviour will not be significant, and the bootstrapped confidence interval of the direct path (LLCI = −0.006, ULCI = 0.282; LLCI = −0190, ULCI = 0.133; LLCI = −0.187, ULCI = 0.105) included zero after controlling for the moral disengagement. Therefore, moral disengagement played a full mediating role between group recognition, egoism motivation and altruistic motivation. The bootstrapped confidence interval of the direct path of group veto did not include zero (LLCI = −0.254, ULCI = −0.011). Therefore, moral disengagement played a partial mediating role between group veto and unethical behavioural intention. In summary, these results supported H3 and H4.

Moderating Effect of Moral Identity
H5 and H6 predicted the moderating effect of moral identity in the relations between group recognition, group veto, egoism motivation, altruistic motivation and moral disengagement separately. To test these hypotheses, we used the hierarchical regression method by Statistic Package for Social Science software 22.0 (SPSS22.0). The results are shown in Table 5.
To confirm the moderating role of the moral identity variable, interaction items were developed by multiplying the composite constructs of group recognition, group veto, egoism motivation and altruistic motivation with the construct of moral identity. We regressed moral disengagement on group recognition and group veto. As the results show, group recognition was a significant predictor of moral disengagement, with an estimate of 0.563 at the 0.001 significance level. Group veto was significantly associated with moral disengagement, with an estimate of −0.258 at the 0.001 significance level. The interaction term of group recognition was insignificant, with an estimated value of 0.061, which indicated the absence of interaction. In addition, we regressed moral disengagement on altruistic motivation and egoism motivation. The construct of altruistic motivation was significantly associated with moral disengagement, with an estimated value of −0.410 at the significance level of 0.001, and significantly interacted with moral identity, with an estimated value of −0.142. Conversely, egoism motivation was a significant predictor of moral disengagement, with an estimated value of 0.509 at the 0.001 significance level, but its interaction term was insignificant, with an estimated value of 0.052. This indicated the absence of interaction. In summary, H5b and H6b were supported, but H5a and H6a were not supported.  To further verify the moderating effect of moral identity in the relation between group veto, altruistic motivation and unethical behaviour, we adopted the method provided by Aiken and West (1991) to draw the interaction graphs of group veto and moral identity, and of altruistic motivation and moral identity separately [81]. The sample was divided into two groups based on the computed method of mean value of moral identity variable plus or minus its standard deviation. It can be seen from Figure 2 that the negative relationship between group veto and moral disengagement is more significant under high levels of moral identity; namely, individual consumers with a high level of moral identity are less likely to engage in moral disengagement when facing group veto compared with individuals with low levels of moral identity. It can be seen from Figure 3 that the negative relationship between an altruistic motivation and moral disengagement is more significant under high levels of moral identity. Individual consumers with a high level of moral identity are less likely to conduct moral disengagement when influenced by altruism compared with individuals with a low level of moral identity.

Moderated Mediation Effect of Moral Identity
We ran mediation analyses using the bootstrapping methods with 2,000 resamples to test the mediation effect of moral disengagement in group veto and the altruistic motivation under different levels of moral identity separately (see Table 6 and Table 7).
In particular, we adopted the procedure by Preacher & Hayes (2008) and computed biascorrected bootstrap confidence intervals to test the predicted mediation process [78]. If the bootstrapped confidence interval does not include 0, the indirect effect is significant, and the mediating process was supported. It can be seen from Table 6 that under the 95% confidence interval, the bootstrapped confidence interval of the indirect path of the mediation included zero for a high level of moral identity, indicating that moral disengagement significantly mediated the moderating effect of moral identity in the relationship between group veto and unethical behavioural intention, with an estimated value of -0.107. In addition, the bootstrapped confidence interval of the indirect path of the mediation included zero for differences between the high level or low level of moral identity under the 95% confidence interval, indicating that the differences of the indirect effect between group veto and unethical behavioural intention were significant under different levels of moral identity. Table 6. Moderated mediation effect of moral identity results.

Moderated Mediation Effect of Moral Identity
We ran mediation analyses using the bootstrapping methods with 2,000 resamples to test the mediation effect of moral disengagement in group veto and the altruistic motivation under different levels of moral identity separately (see Table 6 and Table 7).
In particular, we adopted the procedure by Preacher & Hayes (2008) and computed biascorrected bootstrap confidence intervals to test the predicted mediation process [78]. If the bootstrapped confidence interval does not include 0, the indirect effect is significant, and the mediating process was supported. It can be seen from Table 6 that under the 95% confidence interval, the bootstrapped confidence interval of the indirect path of the mediation included zero for a high level of moral identity, indicating that moral disengagement significantly mediated the moderating effect of moral identity in the relationship between group veto and unethical behavioural intention, with an estimated value of -0.107. In addition, the bootstrapped confidence interval of the indirect path of the mediation included zero for differences between the high level or low level of moral identity under the 95% confidence interval, indicating that the differences of the indirect effect between group veto and unethical behavioural intention were significant under different levels of moral identity. Table 6. Moderated mediation effect of moral identity results.

Moderated Mediation Effect of Moral Identity
We ran mediation analyses using the bootstrapping methods with 2000 resamples to test the mediation effect of moral disengagement in group veto and the altruistic motivation under different levels of moral identity separately (see Tables 6 and 7).
In particular, we adopted the procedure by Preacher & Hayes (2008) and computed bias-corrected bootstrap confidence intervals to test the predicted mediation process [78]. If the bootstrapped confidence interval does not include 0, the indirect effect is significant, and the mediating process was supported. It can be seen from Table 6 that under the 95% confidence interval, the bootstrapped confidence interval of the indirect path of the mediation included zero for a high level of moral identity, indicating that moral disengagement significantly mediated the moderating effect of moral identity in the relationship between group veto and unethical behavioural intention, with an estimated value of −0.107. In addition, the bootstrapped confidence interval of the indirect path of the mediation included zero for differences between the high level or low level of moral identity under the 95% confidence interval, indicating that the differences of the indirect effect between group veto and unethical behavioural intention were significant under different levels of moral identity.  It can be seen from Table 7 that under the 95% confidence interval, the bootstrapped confidence interval of the indirect path of the mediation included zero for a high level of moral identity. This indicates that moral disengagement significantly mediated the moderating effect of moral identity in the relationship between altruistic motivation and unethical behavioural intention, with the estimated value of −0.277 at a significance level of 0.001. Moreover, the bootstrapped confidence interval of the indirect path of the mediation included zero for differences between the high level or low level of moral identity under the 95% confidence interval, indicating that the differences of indirect effects between altruistic motivation and unethical behavioural intention were significant under different level of moral identity. In summary, H7b and H8b were supported.

Discussion
Given increasing interest in corporate and individual responsibility [4], the present study investigated why consumers behave unethically, especially regarding 'no harm, no foul' actions, and examines why consumers tend to 'knowingly conduct ethical transgression'. Consumers' ethical consumption involves 'the conscious and deliberate purchasing decisions due to personal moral belief and values' [5]. Therefore, understanding why individuals behave unethically can encourage them to make conscious consumption decisions and contribute to the sustainable development of the national economy. Although most studies have examined a large range of various antecedents of unethical consumers' behaviour, such as sex, education, income, moral judgement, materialism, peer influence, etc. [2,6,10,30], the current research considers the group norms and self-interest motivation as the precursors of the unethical consumers' behavioural intention. We further to investigate the mediating role of moral disengagement based on the moral disengagement theory and the moderating effect of moral identity. Based on the research data and analysis, some findings are presented below.

General Discussions
(1) We found that group recognition and egoism motivation have significant positive impacts on unethical behavioural intention; group veto and the altruistic motivation have a significant negative impact on unethical behavioural intention. Meanwhile, there is a mediating role played by moral disengagement between group recognition, group veto, egoism motivation and altruistic motivation. Literature has shown that unethical consumer behaviour is coaffected by intrinsic factors and extrinsic factors [9,30], which is in line with the findings that group norms serving as an intrinsic factors and self-interest motivations functioning as an extrinsic one are two significant precursors of unethical behavioural intention.
Previous studies have indicated that unethical climate is critical for understanding the social context [10]. China has the world's largest counterfeit market, and Chinese consumers may not realise the ethical issues when they behave unethically [4]. We argue that Chinese consumers seek to integrate into groups and avoid social exclusion [29]. Furthermore, the unethical behaviour discussed in this paper mainly focus on 'no harm, no foul' actions which do not create direct harm and individuals tend to justify rationalisation of it, or 'strategically forget' the ethical issues involved [3]. On one hand, under lower moral and mental costs, individuals will still act on reference group expectations or judgements, even when facing ethical issues. According to moral disengagement theory, individual consumers blame the responsibility on group recognition and have the idea that, 'It is the group acquiescing me to do so', and the self-regulatory function fails selectively under the mechanisms of moral disengagement [53]. When individuals' behaviours are rejected by the reference group, the object of responsibility transformation disappears. The moral disengagement mechanism is blocked when moral self-regulatory mechanism comes into play and they will experience negative feelings, such as guilt or self-accusation [41]. This is consistent with the findings that moral disengagement can be influenced by situational factors [13]; on the other hand, an individual's motivation plays a significant role in processing the information related to their motivation, [41]. Individuals with a salient egoism motivation are more likely to get rid of the block of moral self-regulation, and provide an excuse for 'beautification' behaviours through the advantageous comparison of moral disengagement mechanisms. However, the individual consumers who are altruistically motivated are just the opposite. They find it easy to experience another's emotions with a high level of moral judgement, and are lower in moral disengagement. Based on the moral disengagement theory, the current research tries to indicate why individual consumers still tend to 'knowingly conduct ethical transgression' when in ethical dilemmas, while people generally know right from wrong. This finding is in line with Baron et al. [41] (2015), showing that moral disengagement theory is flexible and realistic in assumptions which contributes to explaining the gap between attitudes and behaviour.
(2) We found that moral identity not only negatively regulates the relationship between altruistic motivation and moral disengagement, but also the mediating effect of moral disengagement between altruistic motivation and unethical behavioural intentions. Literature has indicated that compared with Western consumers, who are influenced by individualism, Chinese consumers have long been influenced by Confucianism, which emphasises the thought of benevolence, and pay more attention to the level of others' welfare [4]. Therefore, Chinese consumers with salient moral identities are more likely to be altruistic when facing ethical dilemmas. In this process, moral self-regulation can be activated when the moral disengagement mechanism selectively fails [57]. Thus, the moral identity significantly moderates the relationship between altruistic motivation and moral disengagement.
However, the moderating effect of moral identity on moral disengagement driven by egoistic motivation is not significant. The reason for this phenomenon may be due to the decoupling effect of attitude and behaviour [12], and 'no harm, no foul' actions which individuals easily ignore is another cause. Even individuals with high moral judgements will not have any unethical behaviour at all since the intention of unethical behaviour is affected by various factors [30]. In many cases, an individual's attitude is not necessarily consistent with their final behaviour [12]; The actors do not perceive ethical pressures when they live in a society where counterfeits abound, and the mental cost of violating moral standards is lower [4]. Individuals will weigh the mental cost of deviating from moral norms against the benefits obtained when they under the psychological conflicts of moral decision-making and rational decision-making. Previous studies have found that this kind of 'no harm, no foul' unethical behaviour does not cause direct harm to others but contributes to increase the actors' self-gains [5]. Thus, even individual consumers with higher moral judgement will justify their own behaviour by using a moral disengagement mechanism under the driving of egoistic motivation.
Moral identity not only negatively moderates the relationship between group veto and moral disengagement, but also negatively moderates the mediating effect of moral disengagement between group veto and unethical behavioural intention. Social psychology believes that the pressure perceived by individuals in groups will make them unable to fully make decisions according to their own intentions when their own morality's effectiveness is limited [2]. Chinese consumers who are deeply influenced by the collectivist culture are more sensitive to exclusion than the agreement of a reference group [32]. From this perspective, group veto has a stronger normative impact on individual behaviour. Therefore, consumers with salient moral identities are likely to have significantly reduced intention to morally disengage when facing reference group veto.
It is remarkable that the moderating effect of moral identity on group recognition and moral disengagement is not significant. One reason for this phenomenon may be the decoupling effect of attitudes and behaviours mentioned above; on the other hand, the positive effect between group recognition and moral disengagement weakens the significance of the moderating effect. The kind of 'no harm, no foul' unethical behaviour presented in this paper is limited in the ability to activate individual self-monitoring due to its indirect harm [5]. There is a significant positive effect between group recognition, moral disengagement and unethical behavioural intention, indicating the moral pressure is low for this kind of behaviour when mental cost of moral standards violation is reduced substantially. In addition, individuals are eager to be attached [31]. Therefore, even in an unethical situation, individuals with salient moral identities will also cater to the expectations or norms of the reference group and transfer responsibility by a moral disengagement mechanism under strong affiliation.

Theoretical Implications
(1) This study contributes new predictors with theoretically sound relations to unethical behavioural intention rather than well-established relationships. Literature addressing consumer ethics usually focus on the individual personal factors or situational factors, such as gender, education, moral value, ethical awareness, peer influence, culture, etc., which are limited to Western consumers [2,6,9,13]. Most previous studies have separated the link between the internal and external sides that are associated with unethical behaviour, and are limited in the ability to provide an overall framework [8]. We provide a comprehensive model based on group norms and self-interest motivation to shed light on the phenomenon of widespread unethical behaviour in China. This research examines the group norms in the context of unethical consumption which are usually discussed in the consumer behavioural or organisational literature under the normal context. The literature addressing organisational behaviour indicates that an individual is more likely to engage in unethical behaviour the more they are exposed to an unethical climate [14]. In view of the kind of 'no harm, no foul' unethical actions widespread in China, we borrow ideas from the organisational literature, which, in line with Chinese individual consumers, are susceptible to group norms and behaving unethically. This finding is consistent with prior studies showing that unethical consumer behaviour is a social behaviour and cultural norms or values may shed light on it [30]. Chinese individuals emphasise the social self, and they pay more attention to the influence of social norms on their own behaviours due to the influence of collectivist cultural. By this logic, the recognition or veto of the behaviour of the reference group seems quite significant.
(2) The current study extends the literature by explaining why individual consumers still tend to 'knowingly behave unethically' when in ethical dilemmas based on the moral disengagement theory. Individual consumers justify their misconduct through moral disengagement mechanisms which hinder the appearance of the negative feelings associated with self-regulatory process when they are affected by group norms and the self-interest motivation and therefore behave unethically. Past studies have highlighted that an individual's behaviour is not always consistent with their actual behaviour and people generally know right from wrong [12,51]. Different from prior studies adopting The Theory of Planned Behaviour in the literature of consumer ethics, the current research found that moral disengagement theory explains this 'attitude-behaviour' gap due to its flexible and realistic assumptions [41]. This finding is consistent with existing findings that the propensity to moral disengagement is congruently more strongly related to the unethical decisions and behaviour than a wide range of other theoretically relevant precursors [53].
(3) The present research contributes to the literature by highlighting the Confucianism in the research of unethical consumer behaviour under a China-specific marketing context. We found that moral identity not only negatively regulates the relationship between altruistic motivation and moral disengagement, but also the mediating effect of moral disengagement between altruistic motivation and unethical behavioural intention. However, the moderating effect of moral identity on moral disengagement driven by egoistic motivation is not significant. Literature has shown that the thought of benevolence (ren) is the centre of Confucian philosophy and Chinese individuals are influenced by this kind of thought for a long time [82]. Confucianism is a complex and comprehensive system of moral, social, philosophical thoughts and it provides a guide for Chinese individuals to get along with others [83]. Chinese consumers pay more attention to the level of others' welfare under the influence of benevolence rather than being egoistical. This finding suggest that culture norms play a significant role in studying unethical consumer behaviour, which is consistent with the previous findings that unethical behaviour may have some issues hidden in the culture [30].

Practical Implications
(1) This paper focuses mainly on the kind of 'no harm, no foul' unethical consumer behaviour, such as purchasing counterfeits and pirated software, and aims to explain why individual consumers may 'knowingly behave unethically' when in the marketplace. It may draw the attention of relevant government departments and enterprises to the social resistance in intellectual property protection. Chinese consumers, who have long been deeply influenced by the collectivist culture, have not regarded this kind of unethical behaviour as a negative behaviour related to morality [4]. The individual self-regulatory function is deactivated selectively when this kind of behaviour is approved or accepted by the reference group, and the individual sensitivity of social punishment is greatly reduced. Consumers cannot spontaneously resist this kind of unethical behaviour; it is necessary to eliminate the social block of intellectual property rights. Previous studies have indicated that the culture of this kind of 'no harm, no foul' unethical behaviour is so strong that lawsuits or technologies alone cannot change it [10]. Relevant government departments should focus on strengthening the public's resistance to increase the mental cost of violating moral standards. For example, in addition to increasing the punishment for this kind of behaviour, it should advocate the active voice of important social reference groups as well, regularly carry out lecture activities with the theme of social responsibility and moral responsibility, strengthen stress transmission and guide consumers to awaken morality. For marketers, they can make full use of ethical knowledge propaganda to catch individual consumers' attention on self-behaviour regarding ethical issues.
(2) The emergence of unethical behaviour involves many factors [6], but the root is in the individual intrinsic factors. The present study shows that, on one hand, altruistic individuals pay more attention to the welfare of others and the hidden harm to other stakeholders, putting moral judgement before behaviour. It can be said that it is prosocial for the behaviour of the altruist, which can significantly reduce the probability of occurrence of unethical behaviour. Therefore, to form a harmonious atmosphere against the unethical behaviour in society, we can start by improving the altruism of the public. For example, the government can strengthen the education of morality in primary and secondary school by carrying out comprehensive practical activities, such as helping the elderly, to cultivate the social responsibility of the new generation; on the other hand, unethical behaviour involves egoism and is mainly due to a psychology of greed. This is because pirated goods can bring the same utility as genuine ones, but at a lower price [4]. Therefore, marketers should pursue sustainable operation and reasonable pricing to effectively reduce the cost of consumers, and they can implement the strategy of small profits but quick turnover if necessary. Developers should strengthen the encryption technology of genuine software, which increases the cost of copying and piracy technology, to reduce the frequency of this behaviour.

Limitations and Future Research
This study confirms that there is a diversification between group recognition, group veto and consumer unethical behavioural intention, but the results of this study may vary for different types of reference groups, such as aspirational groups or dissociative groups, which may be the focus of future studies. Moreover, our findings measure the effect of distinct antecedents on consumer unethical behavioural intention for one time. Thus, there is a need for a longitudinal study to determine whether the answers of the same subjects will be similar each time to accurately reveal the relation between variables. In addition, social desirability plays a role in the formation mechanism of consumer unethical behavioural intention. Although we tried to avoid this issue in the questionnaire design process, there is still subjectivity of the subjects. In this respect, a similar future study in a behavioural laboratory would provide control of external factors that may relate to social desirability.
Author Contributions: Y.S. conceptualized the research design, data curation, funding acquisition and project administration. J.Z. contributed to the investigation, formal analysis, methodology and drafting the manuscript. All authors read, revised and approved the final manuscript.