The Impact of Deviant Social Media Influencers and Consumer Characteristics on Purchasing Counterfeit Goods

ABSTRACT Product counterfeiting is a large and global economic crime that causes significant economic, social and personal harms. Facilitated by the efficiency and convenience of e-commerce, finding and buying counterfeits have never been easier. Going beyond product listings on websites, counterfeiters are now using sophisticated marketing techniques to promote their illicit wares. This includes employing deviant social media influencers who violate social norms and the law to peddle counterfeit goods via channels such as YouTube and Instagram. However, very little is known about their market or their impact. Based on two UK surveys, the empirical research described in this article estimates the prevalence of consumers who purchase counterfeits because of SMI endorsements. We believe this is the first estimate of its kind anywhere. Further, the research finds that influencer marketing exploits characteristics of consumers that make them susceptible to the charms of the deviant SMIs: high susceptibility to the influence of trusted digital others, low risk awareness, high risk appetite, and prone to rationalizations that morally justify the purchases. The higher prevalence of these characteristics in young adults and males helps to explain why these demographic groups are the most likely to purchase endorsed counterfeits.


Introduction
Whither legitimate business goes, illegitimate business follows.Thus, the emergence of e-commerce was the inevitable harbinger of illicit e-commerce and the Internet became a global catalog for counterfeit goods (Heinonen, Holt, and Wilson 2012).The international trade in counterfeit goods is a huge economic crime worth up to $509 billion per year or 2.5% of global trade in merchandise (OECD and EUIPO 2019).63% of the trade originates in China and Hong Kong with an annual export value of $322 (OECD and EUIPO 2019).It accounts for 11% of all China's merchandise exports and, to put this into context, is equivalent to India's total exports (WTO n.d..).Globalisation and the digital economy have provided Chinese factories, workshops and their intermediaries access to high consumption markets across the world.
The trade causes very significant economic, social, and personal harms (Button, Hock, and Shepherd 2022;Navarro 2006;Phillips 2007).The OECD and IPO (2019) report estimates that producers in the UK lost £11 billion in global sales in 2016 due to infringement of their intellectual property rights, the wholesale and retail sector lost £9 billion of sales, and the government lost £4 billion in tax revenue.The impact on the UK employment market amounted to 86,300 lost jobs, whilst unwitting buyers were defrauded of £4.8 billion.The report does not cover the foregone benefits and opportunity costs associated with lost social, business and innovation investment.By stealing intellectual property (IP) rights, counterfeiters are defrauding consumers and harming brands whilst stealing legitimate sales, tax revenues, jobs, innovation opportunities and welfare investment.Moreover, counterfeit products can fuel transnational organized crime, risk national security, and support terrorist groups (Bjelopera and Finklea 2010;Chaudhry and Zimmerman 2009;Global Intellectual Property Center 2016;Heinonen and Wilson 2012;Sullivan and Wilson 2017;Sullivan et al. 2014;Sullivan, Wilson, and Kinghorn 2017).This range of socio-economic damage illustrates the threat counterfeiters pose to the free-market business model, which relies on IP rights that encourage investment in innovation as well as compliance with minimum standards (Fink, Maskus, and Qian 2016;Lemley 2002;Wilson 2022).
The loss of legitimate jobs may mean more employment in the counterfeit factories, but these jobs are typically low-paid and often at subsistence levels in grueling and unsafe conditions (Phillips 2007).The counterfeit businesses are criminal enterprises with reverse engineering skills, but, driven by a low-cost imperative, they usually have little regard for product quality or safety and no interest in complying with recognized standards (OECD and EUIPO 2022).The health and safety dangers include ill-health, injury and death (Button, Hock, and Shepherd 2022;Yar 2005).The sheer recklessness of the most dangerous counterfeiters amounts to mass homicide.The trade in fake pharmaceuticals is a global menace that affects every country (Blackstone, Fuhr, and Pociask 2014;O'Hagan and Garlington 2018).Driven by the emergence of online pharmacies, it is worth $4.4 billion per year (Mages and Kubic 2016;OECD and EUIPO 2019).Its disproportionate impact on developing nations means that 72,000 children die each year from pneumonia due to counterfeit antibiotics and fake antimalarials leads to 116,000 malaria deaths in Africa (Mackey and Liang 2011;WHO 2017).In a single week in 2020, INTERPOL's Operation Pangea seized 4.4 million pharmaceutical units including cancer, nervous system and pain medicines (INTERPOL 2020).Fake cosmetics are also a toxic health risk that can cause skin necrosis as a result of high levels of bacteria (Morse and Repsha 2020;Tan, Jaron, and Nizam 2019).Substandard food, alcohol, toys, electrical goods, and batteries all pose serious risk (Wilson, Sullivan, and Hollis 2016).Yar (2005) estimated that counterfeit vehicle components alone are responsible for accidents that cause 1.5 million injuries and 36,000 deaths every year.
It is convenient to lay the blame for the socio-economic and health harms on the criminal workshops in a far way place.However, the grim reality is that the problem would not exist without the demand from complicit consumers closer to home.There has been very limited research into the prevalence of these deviant counterfeit buyers.A rare, large European survey found that 5% of adults intentionally buy counterfeits and 9% are unintentional buyers (EUIPO 2020).Using a different survey design, a similar study in the UK found that 17% intentionally purchase counterfeits whilst 16% do so unintentionally (Haq, Schling, and Taylor 2020).This study also identified a clear age trend: 30% of adults under 34y purchase counterfeits compared to 7% to 12% of older adults, but no difference between females and males.Although these studies do not distinguish between in-person and online purchasing, they illustrate the prevalence of the demand problem.
A key normative question is, therefore, what draws so many deviant consumers into purchasing fakes?The problem is partly explained by the opportunity common to the supplier and the consumer, the convenience of e-commerce enabled by the Internet (Wilson and Fenoff 2014).Following on the heels of e-commerce, social commerce appeared when the shopping experience shifted onto social media to exploit the explosion in opportunities for effective word-of-mouth marketing (Chetioui, Benlafqih, and Lebdaoui 2020;Stephen and Toubia 2010).This in turn gave rise to the phenomenon of marketing by social media influencers (SMI) whereby brands pay the trusted leaders with vibrant voices, reach and impact within digital communities to promote their products and services (Hudders, De Jans, and De Veirman 2021).The impact of these influencers is such that SMI marketing has become a $16 billion global industry with 18,900 firms driving annual social commerce sales of $958 billion (Influencer Marketing Hub 2022).
A recent flourish of research into the effectiveness of this marketing phenomenon has explored a diverse range of variables that influence purchasing intentions (for an overview see Hudders, De Jans, and De Veirman 2021).Much of the research has centered on the nexus between homophily and parasocial interaction.Homophily refers to the principle that "similarity breeds connections" (McPherson, Smith-Lovin, and Cook 2001: 415).A successful SMI achieves the reach and impact valued by brands by reflecting the common beliefs, values, experiences and lifestyle ambitions of their like-minded audience (Kim and Kim 2021).A parasocial relationship is a one-sided illusion of a faceto-face relationship with a media personality who does not know the other person exists (Cocker, Mardon, and Daunt 2021;Horton and Wohl 1956).In the context of social media, Lou (2022) coined the phrase trans-parasocial relation because influencers' relationships with their audiences involve some level of improvised interaction that supports the perception of intimacy and friendship.The congruence of homophily and trans-parasocial relations is central to creating the necessary illusion of friendship that enables effective mass word-of-mouth influencer marketing.
Empirical research has identified a range of key factors that sustain this parasocial marketing phenomenon, including trust, expertise, credibility, authenticity and envy (Vrontis et al. 2021).Trust is a crucial factor.As influencers typically occupy specialist domains, they are considered expert sources of trustworthy and credible advice (Hudders and Lou 2022).Indeed, one study found that 92% of social media users place greater trust in influencers than traditional marketing methods (Kim and Kim 2021).In their rather bleak analysis of trust, Shareef et al. (2020) conclude that consumers frequently invest blind faith social media content.This level of trust emerges from the autobiographical craft of SMIs that creates both the perceptions of authenticity as well as the lifestyle envy which stimulates followers to buy endorsed products (Lee and Eastin 2020;Wiedmann and von Mettenheim 2020).
However, as Hudders and Lou (2022) explain, there is an underexposed dark side to the fairytale world of SMIs, such as the physical and psychological harms caused when influencers flaunt idealized body images or promote unhealthy foods to children.In a substantial netnographic study, Cocker, Mardon, and Daunt (2021) found that followers objected to excessive brand promotion and rebuked SMIs for their illegal failure to disclose paid sponsorship.Borrowing from Dootson et al. (2017) definition of the deviant consumer, these kinds of practices illustrate the emergence of the deviant social media influencer who transgresses normative standards of conduct or violates the law.Inevitably, some deviant SMIs have been drawn into the unlawful promotion of counterfeit products (Chaudhry 2022).
A small body of research has focused on the demand side of the trade in counterfeits, examining consumers' psychographic (demographic and psychological) factors that prompt the purchasing, for example, perceived risk, product quality and price, social identity, and perception of inequality (Bardey, Turner, and Piccardi 2022;Eisend and Schuchert-Güler 2006).Unsurprisingly, trust in SMIs is cited as a factor in counterfeit purchasing, but its impact has not been quantified (GWI 2022;Kennedy, O'Dell, and Ching 2021).
Research into the influence of perceived risk mainly relates to product quality, social embarrassment and the risk of being caught (Penz and Stottinger 2005;Tang, Tian, and Zaichkowsky 2014;Thaichon and Quach 2016).However, there is little quantitative research, and very little addresses attitudes to product safety.In a rare study, the EUIPO (2020) found that 17% of Europeans believe that buying counterfeits does not risk harm to businesses and jobs, and 29% believe counterfeits are safe.However, there has been no quantitative assessment in the context of SMIs.Furthermore, there has been no research into a key antecedent of risky behavior, the willingness of individuals to take risks.The seminal research by Dohmen et al. (2011) found that this appetite for risk is a strong predictor of risky behavior.They also found strong age and gender effects: risk appetite is significantly higher in younger people and males.
Rationalization is frequently identified as an antecedent factor in purchasing counterfeits (Bian et al. 2016;Cordell, Wongtada, and Kieschnick 1996;Kennedy, O'Dell, and Ching 2021;Penz and Stottinger 2005).Rationalization is the psychological process involved in reconciling a person's deviant behavior with normative ethical standards (Festinger 1957;Tatch 2022).It has long been held as a key determinant of occupational fraud (Cressey 1953).Rationalizations are the justifications that offenders construct to neutralize the iniquity of their actions and maintain their perceptions of themselves as moral individuals, for example, by blaming the victim or denying the harm (Sykes and Matza 1957).The EUIPO (2020) report found that Europeans rationalize buying counterfeits as an act of protest against the free-market and the premium brands (31% of adults), and it is acceptable when the price of the genuine product is too high (24%), when quality does not matter (19%), and when it concerns a luxury product (15%).
The primary aim and original contribution of the present research is to estimate the extent to which SMIs prompt UK consumers to purchase counterfeit goods.Initial findings, based solely on a survey of female consumers, appeared in an Intellectual Property Office report (Shepherd, Whitman, and Button 2021).The research also uniquely explores some of the antecedent factors that make some consumers more susceptible to the influence of deviant SMIs: age and gender, susceptibility to the influence of others, risk perceptions, and rationalization.The article begins by introducing the survey methods.Next, it introduces the results, starting with the prevalence of the problem.It then discusses the findings, limitations and the implications.

Methods
The research involved two anonymous online surveys of the UK public.The first survey (n = 1,000) collected data from the female population during April 2021.The second survey (n = 1,000) targeted the male population in August 2022.Ethical approval was obtained from the Faculty Humanities and Social Sciences ethics committee (reference numbers FHSS_2021-021 and FHSS_2022-023).Informed consent was incorporated at the start of the questionnaire.The surveys were administered through the Qualtrics online system and nonprobability samples of volunteer respondents were recruited from the Qualtrics panel.This is a widely used, pragmatic approach that supports representative quota sampling (Boas, Christenson, and Glick 2020;Zack, Kennedy, and Long 2019).Quotas were applied for age and regional distribution based on the Office for National Statistics population data (ONS 2022).As the extent of the influence of SMIs was previously unknown, the researchers were concerned that a low number of respondents reporting the impact of SMIs on their purchasing decisions would render the analysis of consumer characteristic, attitude and perception variables statistically meaningless.To mitigate this risk, the sample criteria for both surveys was limited to adults aged 16 to 60 [71% of all adults (ONS 2022)] and who use social media at least once per week [92% of adults in this age range (Ofcom 2022)].The sample frame is thus limited to 65% of the population aged at least 16 (Table 1).The results cannot therefore be generalized beyond the limits of the sample frame.
As there is no clear consensus in the literature as to the definition of the term "counterfeit" (Button, Hock, and Shepherd 2022), the questionnaire used the definition adopted by the UK's Intellectual Property Office for research to guide the respondents (Haq, Schling, and Taylor 2021:).It avoids legal terms and focuses on the appearance of products: "Counterfeits are items that look identical to a genuine product with or without the official branding/logo, but are not made by the brand and may be of lower quality, for example, a handbag of identical design to a 'Chanel' with or without the Chanel logo.We do not mean: items which use a similar style to a genuine product but are made by another business entirely and might be sold for less, for example, iPhone compatible chargers with another company's brand present on the product/ packaging."The same questionnaire was used for both surveys except for minor language adjustments to reflect the target populations.For example, the Chanel handbag in the above definition was replaced with Nike trainers for the male survey.The questionnaire borrowed some questions from the EUIPO (2020) survey relating to prevalence, risk awareness and rationalization.The majority of the questions were multiple choice, single answer questions set out on four-point scales, for example: not important at all, somewhat unimportant, somewhat important, very important plus "don't know" where appropriate.Similar to the EUIPO (2020) study, the data was subsequently categorized into two aggregated groups for analytical purposes, for example, unimportant and important.The findings set out in this paper use this binary classification.The analysis is based on simple descriptive statistics and tabulated summaries to illustrate thematic trends.

Results
The results are organized into the five themes: the prevalence of demand, consumer susceptibility to trusted others, risk perceptions, rationalizations, and the impact of SMI endorsements on trust.The results are presented for the combined sample (overall, n = 2,000) and categorized into gender, 9-year age groups and non-buyers/buyers to illustrate patterns of correlation, similarity and divergence.

Prevalence of consumer demand
One measure of the demand problem is the proportion of adults who buy counterfeit products promoted by SMIs.Table 2 presents the percentage of respondents who purchased at least one counterfeit product in the 12 months prior to the survey because they were recommended by a SMI.Overall, about one-fifth (22%) purchased at least one fake, the clear majority of which knew the products were counterfeit at the time of purchase: 17% were knowing (intentional) buyers and 5% were deceived, unknowing (unintentional) buyers.
The results imply that males (31% of males, 70% of buyers) are over twice as likely as females (13% of females, 30% of buyers) to purchase endorsed counterfeits.The age profile shows a marked reduction from 44% in the 16-24y group to 5% in the 52-60y group.Young consumers aged 16-33y (38% of age group) are three times as likely to purchase endorsed counterfeits as older consumers aged 34-60y (12% of age group).This declining age profile means that two-thirds (66%) of all buyers are under 33y: 44% are young males and 22% young females.
Figure 1 more clearly illustrates that the age correlations of both genders follow similar declines, except for two important features.Firstly, the male profile starts at a higher point and, secondly, its decline is delayed by about a decade: the female profile falls 18% between the 16-24y and 25-33y groups, whereas the male profile is much flatter, falling just 4% from a high of 51%.
An alternative measure of demand is the aggregate frequency of purchasing products, which accounts for consumers' repeat purchasing.Figure 1 charts the percentage of respondents who were one-off, repeat (2-5 items) or habitual (6+ items) buyers of endorsed counterfeits in the prior year.
The most prolific buyers are the repeat and habitual buyers amongst young men (36% of the 16-33y group) and young women (19% of the 16-24y group).Together they account for 41% of all buyers and, because of their frequent purchasing, are likely to constitute well over 50% of the total demand.The respondents were asked to indicate which types of endorsed products they purchased in the prior year.The product categories are based on the Haq, Schling, and Taylor (2020) research.The figures in Table 3 total to more than 100% because some respondents purchased items from multiple categories.The most popular counterfeit is clothing/accessories. Female purchasing is also heavily concentrated in beauty/hygiene products, whilst males are more attracted to sporting goods jewellery/ watches, and electrical goods.
The data indicates a high demand for counterfeits that place consumers at risk for their health and safety, products that contain chemicals, electricity, batteries, hazards to children, or can otherwise be ingested or absorbed into the skin.Overall, 17% of males and 8% of females purchased fakes in at least one of the higher risk categories: beauty/hygiene, electrical products, electronics, toys, or alcohol.This implies that a substantial minority of consumers, especially males, are either unaware of the risks or are content to take the risks.

Susceptibility to trusted others
To provide insight into their susceptibility to influence, the respondents were asked to indicate the importance of the opinions of three sources of influence when making any purchasing decisions: family, friends and social media.Table 4 presents the percentage of respondents who rated the opinions of these groups as important.The final two columns represent the views of those who did not buy SMI endorsed counterfeits (No) in the previous 12 months and those who did (Yes).Family is the most influential source and at a consistently high level across all groups.
In contrast, the age profiles associated with friends and social media content are similar to the declining pattern for counterfeit purchasing in Figure 1 with an important exception: the peak level of male susceptibility is in the 25-33y group and is most pronounced for opinions on social media (53%).It then falls rapidly to the same level as females in the 52-60y group (14% overall).
The clear difference between buyers and non-buyers indicates that buyers are more susceptible to the influence of friends and social media.The largest differential is associated with social media (26% to 59%).Importantly, the elevated susceptibility to friends and social media influence for male and female buyers is very similar.

Consumer risk perceptions
Five dimensions of risk are used to assess the risk perceptions of consumers: three relate to product attributes, one to the economy and one to risk appetite.The respondents were asked about the importance of quality and safety in influencing their general purchasing decisions.They were also asked whether buying counterfeits poses a threat to health and safety.Enquiring into their awareness of risks to the economy, respondents indicated whether buying counterfeits harms businesses and jobs.The reverse results are presented in Table 5 for these four factors, for example quality is not an important consideration and counterfeits are safe.
In addition, respondents' willingness to take risks, or risk appetite, was measured using the 11-point self-perception scale recommended by Dohmen et al. (2011) as the most effective measure of general risk attitudes.Respondents were asked to rate their willingness from 0 to 10, where 0 means "I am not at all willing to take risks" and 10 means "I am very willing to take risks".The results were organized into three groups for the analysis: Risk averse -responses 0 to 3: females 29%, males 21% Risk neutral -responses 4 to 6: females 40%, males 39% Risk-taker -responses 7 to 10: females 31%, males, 40% Table 5 presents the perceptions of the risk-taker group only.The results follow the familiar age profile.Younger consumers tend to have weaker risk perceptions.They are more likely to be unconcerned about quality and safety in their general purchasing decisions; more likely to believe that the demand for counterfeits has no economic impact; more likely to believe that counterfeits are safe.Younger consumers are also more likely to be risk-takers.
Comparing the genders, males exhibit a higher risk tolerance.Furthermore, males sustain a higher level of risk tolerance into the 25-33y group with respect to the quality factor, safety factor and the risktaker characteristic.Indeed, this group has the highest proportion of risk-takers (63%).The differential between non-buyers and buyers demonstrates a higher risk tolerance across all the factors in the buyer group.It also shows consistency between the female and male buyers, especially in the proportion of female (64%) and male (68%) risk-takers.

Consumer rationalizations
To explore the prevalence of rationalizations, respondents were asked whether it is acceptable to buy counterfeit products when the authentic brand owners charge too much, when the quality of the product does not matter, or when it concerns luxury products.They were also asked whether buying counterfeits is an act of protest against the market driven economy and the large premium brands.Overall, about onethird concurred with the price, quality and luxury product rationalizations, and over half (57%) viewed it as an act of protest (Table 6).The results again follow the familiar age, gender and buyer profiles.The differential between non-buyers and buyers indicates the latter are much more likely to approve of the rationalizations (typically 70%), and this higher level of approval is notably consistent between the genders.

Trust impact of social media influencer endorsements
The impact of SMI endorsements on consumer trust is assessed using the three dimensions in Table 7.The respondents were asked whether SMIs must have tried the genuine they recommend, and whether a recommendation means the product must be safe.The respondents also indicated whether endorsements influence their decisions to buy products.Overall, SMI endorsements cause about one-third of consumers to infer that the influencers must have evaluated products (35%) and that they must be safe (30%).About one-third (38%) are more likely to buy genuine items due to the recommendations.The age, gender and buyer results follow the familiar pattern.The difference in the results for the non-buyer and buyer groups illustrates the extent to which buyers trust influencers.Overall, the buyers are three times as likely to believe that SMIs evaluate products (66%), three times as likely to believe the products are safe (68%), and twice as likely to buy them (58%).

Discussion
Social commerce is the new frontier for marketing and the social media influencer is the new royalty.Consumers in this marketplace tend not to physically inspect and evaluate products before purchasing them.Indeed, it is very often impossible to do so.As a consequence, remote recommendations by third parties have increasingly replaced the customer's own evaluations of purchasing risk.The analysis of trusted others provides an insight into the impact of the closeness, or remoteness, of third parties.The opinions of those closest to the consumer, family members, are valued equally in all demographic and non-buyer/buyer groups.However, discrimination in susceptibility to influence appears in the more remote relationships involving friends and social media.
This relationship with remoteness is most pronounced in relation to the "trusted digital others" on social media.Young consumers aged up to 33y are far more susceptible than the older age groups to social media content, whilst buyers of endorsed counterfeits are twice as susceptible.This susceptibility to the influence of digital others in making purchasing decisions creates the social context and differentiated market opportunity for legitimate SMIs to exploit their parasocial relationships with their followers in order to promote products.It explains why their primary market is young people (Influencer Marketing Hub 2022; Kennedy, O'Dell, and Ching 2021), and it helps to explain why young people are the largest segment of the deviant SMIs' counterfeit market.
Turning to risk perceptions, a small but important minority do not regard quality and safety as important considerations in their general purchasing decisions.Similarly, a significant minority do not perceive the economic or safety risks associated with counterfeits.Nevertheless, the heavily skewed age profiles, the gender differences and the increased risk dismissiveness amongst buyers suggest that low risk awareness is an antecedent factor in purchasing decisions.Even if consumers are aware of and consider the risks, their risk appetite may overcome their concerns.The higher risk appetite of younger adults and males, which accords with Dohmen et al. (2011), and the large differential between buyers (67%) and non-buyers (33%) indicate that risk-takers are a major constituent of the counterfeit market.
The prevalence of rationalizations in favor of purchasing counterfeits is indicative of widespread ambivalence and permissiveness toward the problem, especially amongst young consumers.These findings quantitatively support previous research which identified rationalization as a factor in counterfeit purchasing decisions (Bian et al. 2016;Kennedy, O'Dell, and Ching 2021).
Referencing the Sykes and Matza (1957) framework, the buyers deny the brand owners' victimhood.They neutralize their own culpability and immorality in the illegal transactions by deflecting the blame to the brand owners.The buyers in particular condemn the luxury brands for excessive pricing that is not justified by the performance qualities of their products.This feeds into excessive profiteering, exclusivity and social injustice sentiments that justify purchasing counterfeits as an act of protest against the big brands and the market economy (OECD and EUIPO 2019).The counterfeiters and their deviant SMI collaborators exploit consumers' disposition toward rationalization.Although the present research involves a limited palette of rationalizations, the dominance of rationalizations within the buying group indicates that the "rationalizers" are the counterfeiters' primary market.
Returning to the impact of trusted digital others, one-fifth (22%) of consumers buy counterfeits because of SMI endorsements; 17% are knowing buyers, but 5% are unknowing.Because the latter group have been deceived, the predisposed rationalizations and attitudes in favor of counterfeits are not salient.Nevertheless, the other factors that led them to be enticed into the illicit purchases are relevant.The characteristics which predispose consumers to buying endorsed counterfeits also represent the target market for deviant SMIs.Consumers in the target market: • are susceptible to the influence of trusted digital others, • have low risk awareness, • are risk-takers, • are disposed to rationalization in favor of counterfeits.These characteristics of an idealized complicit consumer are exploited by the deviant SMIs to peddle the fakes.Their mediating impact on risk perceptions is such that counterfeit buyers are three times as likely to believe that SMIs have evaluated the products (66%) and that they must be safe (68%), and over half of the counterfeit buyers (56%) purchased higher risk products.It seems that trust in SMIs evokes symbolic meanings and impressions in the minds of susceptible consumers.For them, the SMI symbolizes integrity and their endorsements symbolize good, legitimate, low risk purchases.These findings support Shareef et al. (2020) observation that some consumers invest blind faith in social media content.
The demographic distribution of buyers closely aligns with the distribution of the consumer characteristics.The implication is that young consumers are far more likely to purchase endorsed counterfeits because this group has a higher susceptibility to influence, lower risk perception, higher risk appetite and greater capacity for rationalization.It helps to explain why young consumers are the primary market for the counterfeiters and their SMI collaborators: two-thirds (66%) of buyers are aged under 33.Similarly, compared to women, the higher prevalence of the characteristics in men associates with a higher level of purchasing such they account for 70% of all buyers.A key feature of the gender difference is the age profile.For women, the frequency of the traits and purchasing declines from the mid-twenties.The decline for men is delayed by about 10 years to the mid-thirties.This delay is a key driver in male participation in the trade and helps explain why 44% of all buyers are young men.
The limitations of the study need to be acknowledged.Compared to the EUIPO (2020) study, the higher prevalence of buyers and rationalizations in the present research, the greater differential between knowing and unknowing buyers, and the lower awareness of product safety risks is likely to be influenced by the restricted sample frame: UK adults aged 16-60y, active on social media, and members of the Qualtrics panel.The surveys were administered a year apart during the COVID-19 pandemic, which may have affected purchasing levels.The study is not granular in examining attitudes and perceptions associated with specific product types, nor does it account for motivations, personal wealth, geographical residence, or previous experience with counterfeit products.Furthermore, as the self-report survey inquired into deviant purchasing behavior, the level of counterfeit purchasing may be underestimated due to social desirability bias (Jann, Krumpal, and Wolter 2019).

Implications
The most obvious implication arising from the findings is that much more effort needs to be applied to disrupting the activities of deviant SMIs.Amazon recently mounted a rare suit against two SMIs for promoting luxury fakes and shut down their activities (Palmer 2021).However, to optimize such efforts, the findings also suggest two areas of research in support of social control, disruption, prevention and enforcement.The first is in relation to deterrence messaging, which is frequently recommended in the literature (for example, Eisend and Schuchert-Güler 2006).To be effective, normative messages should attack and deconstruct the rationalizations in favor of deviant behavior (Shepherd and Button 2018).This requires a more thorough understanding of the typology of specific rationalizations used by deviant SMIs and consumers, the prevalence of the different types, and their orientation to demographic characteristics.It also requires a critical examination of other factors that may contribute to the deviant SMI's capacity to stimulate demand for counterfeits, for example, the impact of consumers' homophilic identification with SMIs, perceived expertise, credibility, authenticity and envy.These factors may have a direct impact or feed into the rationalizations.Of particular interest would be to gain a deeper understanding of what consumers and the influencer marketing industry perceive as deviant behavior, and whether the largely one-directional parasocial relationships facilitate consumers and SMIs in neutralizing their perceptions of deviance.
Additionally, it is important to consider different normative messages for different purposes and audiences.Counterfeit harms and awareness of them vary, so messaging should likewise vary in content and delivery (Grammich and Wilson 2019).The results indicate family and peers have powerful influences on purchasing behavior, even far greater than opinions on social media.This suggests that messaging could be tailored not just to those more likely to purchase counterfeits, but to those that can help shape their purchasing behavior to counter the effect of pernicious parasocial relations with deviant SMIs.
The second area relates to the counterfeit networks.Although the survey did not inquire into the source of the goods and the purchasing methods, it is very reasonable to assume that the SMI recommendations led to online transactions within the milieu of social commerce rather than a trip to a shady local market.Nevertheless, this assumption highlights a research gap: mapping the network of actors who stimulate, enable and facilitate the demand for fakes within the physical and transactional networks as well as in the onion layers of facilitators.Understanding this network would help construct a typology of locations that are vulnerable to intervention, enabling detection and mitigation by law enforcement and brand owners.Applying the situational crime prevention method would assist in developing a framework of location-specific interventions (Cornish and Clarke 2003).
Reddit, for example, overtly hosts forums dedicated to purchasing fake luxury goods.On YouTube, we found one macro-influencer with over 4 million followers who has been promoting counterfeit fashion for years, providing links to online Chinese marketplaces and dropshippers.She is under the management of a very prominent UK marketing agency.Adverts for legitimate organizations appear on the illicit forums and within the YouTube videos, including famous retailers, car manufacturers, food producers and universities.We even found high brand watch manufacturers advertising on counterfeit watch forums and adverts for high brand fashion firms within fake fashion videos.By posting the adverts, the brands are not only legitimizing the toxic social media narrative, they are paying the deviant influencers.The activities of the marketing industry and the deviant SMIs are very visible.The implication is that regulators and, especially, private interests can disrupt and reduce the counterfeit trade if they choose to look for it.
The Chinese website, the dropshippers, the American social media platform, the British marketing agency, the British influencer, and the legitimate brands are all complicit in facilitating this very global trade.Then there are the banks, transport firms and postal services.Their actions may be forgivable because, to them, the trade is invisible.Perhaps they do not know what to look for.Perhaps they, like Reddit, YouTube and the marketing agency, choose not to look.

Table 2 .
Knowing/Unknowing purchase of endorsed counterfeits in previous 12 months.

Table 3 .
Frequency of purchasing endorsed counterfeits in previous 12 months.Product categories purchased.

Table 4 .
Opinions of trusted others important to consumers.

Table 6 .
Consumer rationalizations in favor of purchasing counterfeits.

Table 7 .
Trust impact of social media influencers.