The economic value and determinants of tourists’ counterfeit purchases: The case of Hong Kong

Determinants of counterfeit purchase decisions

The decision of whether to purchase counterfeits depends on a range of intrinsic and extrinsic variables including psychometric factors and sociodemographics. We include the intrinsic sociodemographic characteristics of gender, age, and education from Chaudhry and Zimmerman’s study but also include religion, employment status, and culture. We test the psychometric factors of ethical morality of whether to purchase counterfeits and the effectiveness of anti-counterfeit marketing strategies but also include tourists’ emotional state when purchasing counterfeits, location (whether the individual is on holidays or not), and the effect of others’ opinions about the counterfeit purchase. Further, because the target population of the research is tourists, we test to see if trip-related characteristics are significant determinants of whether tourists have purchased counterfeit goods.

Sociodemographics of the consumer, including gender, age, education, religion, income, and culture, have been shown to influence the counterfeit purchase decision (Cheung and Prendergast, 2006, Wang et al., 2005). Carpenter and Lear (2011) investigate the moderated effect of gender on purchase intention and attitude toward counterfeit products. They find gender does affect the ethical beliefs toward counterfeiting. Specifically, females in the United States are less likely to hold stronger ethical beliefs and are less likely to perceive the sale of counterfeit products as a crime, at least for fashion items (Carpenter and Lear, 2011). Based on the literature above, the following hypothesis is proposed:

H1: Females are more likely than males to purchase counterfeit goods.

Kohlberg (1984) notes that the level of an individual’s moral development improves with age. His assumption is that older people have reached a higher level of moral development and are more concerned about their social responsibility rather than only themselves. If purchasing counterfeit goods is perceived as unethical, older individuals with a higher level of moral development will purchase less counterfeits, therefore;

H2: Younger tourists are more likely than older tourists to purchase counterfeit goods.

In Craft’s review of empirical ethical decision-making (2013), there is a positive relationship between level of education and ethical behavior. Those who have attained a higher level of education may be more likely to think critically about different cause and effects of ethical purchases (Wee et al., 1995). Tolkach et al. (2017) find those who have a bachelor’s degree or higher are more likely to perceive purchasing a counterfeit handbag at an open-air market as unethical, therefore;

H3: Tourists with higher levels of education are less likely to purchase counterfeit goods.

Prendergast et al. (2002) conducted a survey to understand demand of non-deceptive counterfeit products in Hong Kong. They confirm the effect of age, occupation, and education on counterfeit purchase intention. The results show that low spenders on counterfeit products are mainly young students or workers with low income, while high spenders on counterfeit products are usually white-collar office workers with higher incomes.

H4: Tourists not in employment are less likely to purchase counterfeit goods.

There is mixed evidence regarding whether religion has a significant effect on ethical judgments and behavior. For example, Khavari and Harmon (1982) find a negative relationship between illegal and deviant behavior and individual religiousness while Kidwell et al. (1987) find no relationship between religiosity and ethical judgments of business managers. Depending on the context, those affiliated with a religion have been associated with more ethical behavior (Weaver and Agle, 2002) or not associated with consumer ethics (Vitell and Paolillo, 2003). Thus:

H5: Identification with a religion will have an impact on the likelihood of purchasing counterfeit goods.

Differences in cultural values can contribute to different attitudes of counterfeit products. Research has found that Eastern consumers are collectivist and more willing to buy fake products, and Western consumers are more individualistic so emphasize individual ownership of creation (Husted, 2000; Marron and Steel, 2000). The purchase intention for counterfeit goods is found to be different by nationality (Chapa et al., 2006). Hence, in this study, mainland Chinese tourists may hold different cultural values to Western consumers. Hence, culture is a potential determinant of counterfeit purchases. Tourists from mainland China, where counterfeit products are prevalent, may be less likely to buy fake products abroad. If a Chinese consumer wants to buy counterfeits, he or she can more easily buy counterfeit goods at a lower price and higher quality in China than abroad (USCBP, 2014). However, for those who can afford genuine products, especially Chinese luxury consumers, they prefer to shop overseas to avoid counterfeits (The Korea Bizwire, 2016). It is widely believed that the quality of genuine products abroad is better than those in mainland China. Hong Kong is the top destination favored by Chinese luxury consumers (Albatross, 2016). Therefore, the effect of cultural values on counterfeit consumption can be different between Chinese and International tourists. Therefore;

H6: Mainland Chinese tourists are less likely than International tourists to purchase counterfeit goods.

Ethical morality affects the decision of whether to purchase non-deceptive counterfeits or not (Tan, 2002). Ethical perspectives for counterfeit purchases measure the extent to which consumers perceive purchasing non-deceptive counterfeits as ethical or not. There are two main components of ethical morality. One is moral intensity, which is composed of the magnitude of consequences, social consensus, probability of effect, temporal immediacy, proximity, and concentration of effect (Jones, 1991). For example, we would expect those who are more likely to agree that purchasing counterfeits is acceptable or that purchasing counterfeits is not likely to actually cause any harm (those with lower moral intensity), would be more likely to buy counterfeits. This leads to the formulation of the following hypothesis:

H7: Tourists with a lower moral intensity are more likely to purchase counterfeit goods.

A second element of ethical morality is personal relevance. Personal relevance means to what extent the person is involved in the issue, or how important and relevant the issue is to the person (Petty and Cacioppo, 1984, Petty et al., 1992). We would expect to find that with lower personal relevance (agree that the decision to purchase counterfeits is not important or the decision to purchase counterfeits will not affect their self-image) would be more likely to buy counterfeits. Hence,

H8: Tourists who perceive counterfeit goods to be of lower personal relevance are more likely to purchase counterfeit goods.

Emotions can affect judgments and decision-making (Petty et al., 1993). Positive moods will generally produce more positive attitudes and positive thoughts toward the stimulus. There are two dimensions of emotion: feeling state and arousal (Gaudine and Thorne, 2001). Feeling state is where individuals’ emotions can be positive (e.g. happy or optimistic) or negative (e.g. sad or depressed). Arousal is the level of intensity of the feeling state, ranging from quiet to aroused. Emotions (such as the sense of adventure) increase consumers’ motivation to buy counterfeits as souvenirs (Schuchert-Güler and Eisend, 2003). Research shows that emotions are intrinsic and can influence all stages of moral decision-making from identifying ethical dilemmas, moral reasoning to ethical behavior. It is important to understand how tourists make ethical decisions concerning counterfeit consumption are affected by emotion (Correia and Kozak, 2016; McCabe et al., 2016).

H9: Tourists who report more positive emotions are more likely to purchase counterfeit goods.

Counterfeit purchase intention can be affected by different situations (Harvey and Walls, 2003). Therefore, location and surrounding environment is important. Tourism is a predominantly hedonistic and liminal experience (McKercher, 2015). Individuals can act differently, with a higher likelihood of unethical actions, while on vacation than at home (Selanniemi, 2003; Tolkach et al., 2017). Schuchert-Güler and Eisend (2003) note that German consumers are more willing to buy counterfeit products when they are on holiday. Further, consumers whose home countries have strict IPR enforcement are more likely to engage in counterfeit purchases as a hedonic and adventurous shopping experience when they are abroad or on holiday. According to a report by Daily Mail Online (2011), over half of UK tourists buy counterfeit products abroad. Among them, two-thirds bought fake products knowingly, and the other one-third thought they bought genuine products at a discounted price. The shopping experience is an adventure for tourists since they might not be able to purchase counterfeit products in their home country due to stricter regulations on the importation of counterfeits and the higher degree of adherence to IPR in these countries.

H10: Tourists who perceive purchasing counterfeits as something they do on vacation are more likely to purchase counterfeit goods.

Ethical choices such as whether to purchase counterfeit goods are dependent on the social context. Individuals will more likely to do what is expected by peers, colleagues, family, or friends. In tourism research, the effect of group is often studied in the context of family decision-making with special focus on the spouse relationship between husband and wife and the parental relationship between parents and children. Results show that women and children have increasing influence on group decision-making on vacation (Barlés-Arizón et al., 2013; Khoo-Lattimore, 2015). In addition, the role distribution in families determines the decision-making strategies for holiday choice but can vary across different tourism contexts (Therkelsen, 2010). Therefore, we will test the effect of others on tourists’ choice of whether to purchase counterfeit products.

H11: The effect of others’ opinions on whether tourists purchase counterfeit goods will have an impact on the likelihood of purchasing counterfeit goods.

Anti-counterfeiting strategies can deter consumers from purchasing counterfeit goods. There are four types of anti-counterfeiting strategies: law, product, pricing, and marketing (Herstein et al., 2015). While IPR laws mainly regulate the supply of counterfeits, the latter three strategies aim to act as effective agents to change consumer demand for counterfeits. Each strategy could communicate positive information (carrot) or negative information (stick) (e.g. Chiu et al., 2007).

Most companies tend to protect their IPR through legal channels if their products have been counterfeited. The affected companies may lobby the government to pass criminal laws to battle counterfeiters (Bush et al., 1989). Legal action can also significantly decrease consumer intentions to buy counterfeits if there is punishment (Albers-Miller, 1999).

Pricing strategies offer genuine products at lower prices such as discounted or second-hand products to attract consumers. Price has proven to be an influential determinant of knowingly purchasing counterfeits by many studies. Lower prices of genuine products will make counterfeits less attractive. However, the effectiveness of pricing strategies may differ by product category. Luxury brand companies should exercise caution in using pricing strategies because high prices, to some extent, represent prestige, status, and social image. Lower prices may devalue luxury brand products (Nia and Zaichkowsky, 2000).

Marketing communication strategies try to educate the public via media to communicate different messages concerning the risk or negative impacts of buying counterfeits or the association between counterfeiting and organized crime (Chaudhry and Zimmerman, 2009). However, media marketing can be expensive and its effectiveness is post hoc and not easily measurable. In sum, anti-counterfeiting marketing strategies are an important determinant, which seek to decrease demand for counterfeits. Hence,

H12: Tourists who perceive anti-counterfeit strategies to be more effective are less likely to purchase counterfeit goods.

As yet, there is no literature examining how trip-related characteristics of a tourist influences the decision to purchase counterfeit goods. These variables may be important given consumption decisions on vacation can be different from at home, as noted above. Tourists can find themselves in situations where there is uncertainties and risks, most of which fall outside an individual tourist’s control (Williams and Baláž, 2013). Locus of control measures whether the source of control for ethical decision-making is external or internal (Trevino, 1986). When the locus of control is external, individuals will rely on other sources to guide ethical decision-making and be less likely to take responsibility for their behavior. When the locus of control is internal, individuals will be responsible for their (ethical) choices. Therefore, individuals who have high internal control are more ethical and less likely to purchase counterfeit goods, resisting different opinions of others.

In terms of tour arrangements, tourists’ travel arrangements fall into two broad categories: independent travel and an organized tour group. Those on a tour group, in general, have a more external locus of control: they rely on other sources to guide ethical decision-making and are less likely to take responsibility for their behavior (Trevino, 1986). As a result, the behavior of others may influence their behavior. If someone in the organized travel group purchased counterfeit goods, others would be more likely to also purchase counterfeit goods. They would not feel embarrassed or ashamed. This would be acceptable behavior because others in the peer group made a purchase (Coleman, 1994).

H13: Tourists in organized tours are more likely than tourists traveling independently to purchase counterfeit goods.

Travel experience and familiarity with the destination can influence ethical behavior. We hypothesize that tourists who are first-time visitors to the destination will be less knowledgeable about where to purchase counterfeit goods, all else being equal. They will be less sure about the rules and social norms regarding the availability, purchase opportunities and repercussions of purchasing counterfeits as first-time visitors (Kerstetter and Cho, 2004). An individual will rely on sure cues and choose certain options rather than uncertain ones (Kahneman and Tversky, 1979). Hence, we hypothesize first-time tourists to Hong Kong will be less likely to purchase counterfeit products.

H14: First-time tourists to Hong Kong are less likely than repeat tourists to purchase counterfeit goods.

We hypothesize that business tourists will be less likely to purchase counterfeit goods than non-business tourists for two reasons. Firstly, business tourists are in the location for a specific reason and rarely have the time for shopping as a main activity. Tourists consider shopping in destinations as an activity undertaken in pursuit of leisure or as a hedonistic activity (Ryu et al., 2008). This is associated with leisure tourists rather than business tourists. Second, often business tourists have higher expenditures than other types of tourists (Suh and Gartner, 2004). Business tourists have a higher ability to pay for genuine luxury goods so do not need to purchase lower-priced counterfeit products. Hence,

H15: Business tourists are less likely than non-business tourists to purchase counterfeit goods.

Lastly, in terms of travel party, solo travelers are more self-interested, only needing to care for themselves (McKercher, 2015; Woiceshyn, 2011). Travelling by oneself from the usual place of residence frees oneself from the judgment by family, friends, or wider society, thus removing some of the social constraints toward engaging in unethical shopping purchases (Tolkach et al., 2017). They would not consider the opinions of others if they were to make a counterfeit purchase. In this case, they are less ethical thus more likely to purchase counterfeit goods while traveling. Therefore, all else being equal, solo travelers will be more likely to purchase counterfeit goods than tourists traveling with travel companions.

H16: Solo travelers are more likely than tourists traveling with others to purchase counterfeit goods.

The above section provides a brief overview of the factors that influence the decision of whether to purchase counterfeit goods. This research contributes to filling the gap by analyzing tourists in one of the major counterfeiting regions about their counterfeit purchase behavior. Based on the above literature, we propose the following conceptual framework (Figure 1):

OPEN IN VIEWER

Figure 1.

Conceptual framework of counterfeit purchase decision.

Sample and sampling

Mainland Chinese and international tourists were intercepted in popular shopping destinations such the Ladies Market and Temple Street Market where counterfeits are usually sold. According to the Hong Kong Tourist Board in 2016, 14% and 10% of tourists, respectively, visited these markets in Hong Kong at least once during their stay. To avoid sampling bias, tourists were intercepted in other popular destinations such as Avenue of Stars, Victoria Peak, Hong Kong Disneyland, Ocean Park, Hong Kong Convention and Exhibition Centre, Clock Tower at Tsim Sha Tsui, Tsim Sha Tsui Waterfront Promenade, Lan Kwai Fong, and the Big Buddha (Hong Kong Tourism Board, 2017). Quotas on age and gender to match population statistics were enforced. A total of 530 samples per cohort were obtained for a total sample of 1060 surveys. This provides a maximum sampling error of ±4.3% per cohort. The sample size was determined based on financial resources available and having a maximum sampling error lower than ±5%.

The first cohort is international tourists. These tourists include native English speaking visitors to Hong Kong from the United Kingdom, Europe, United States, Canada, Australia, New Zealand, as well as visitors from Latin America Korea, Japan, South-East Asia, South Asia, Middle East, and Africa who speak English as a means of communication while traveling in Hong Kong. For this cohort, the questionnaire was administered in English. The second cohort under investigation was mainland Chinese tourists. This cohort represents the largest segment of tourists to Hong Kong. The questionnaire was conducted in simplified Chinese.

Table 1 shows the profile of respondents for the cohorts: mainland Chinese tourists and international tourists. Quotas were established for age and gender to match population figures based on Hong Kong Tourism Board’s annual report for the tourists (Hong Kong Tourism Board, 2017). In terms of differences in profiles, there is a higher proportion of males among international tourists. International tourists were more likely to have a higher level of education. International tourists were more likely to be older but less likely to be employed compared to mainland Chinese tourists. Mainland Chinese tourists were more likely to be repeat visitors to Hong Kong, be same-day visitors, travel independently and be leisure tourists, compared to international tourists. International tourists were more likely to be business or visit friends and relatives (VFR) tourists than mainland Chinese tourists.

OPEN IN VIEWER

Table 1.

Profile of respondents.

	Mainland Chinese (N = 530)	International tourists (N = 530)	χ 2	df	p Value
Gender
Male	40.8%	62.1%	48.22	1	<0.001
Female	59.2%	37.9%
Age
18–34	48.9%	34.3%	32.75	2	<0.001
35–54	39.4%	43.0%
55+	11.7%	22.6%
Education
High school graduate or below	31.9%	16.4%	67.50	2	<0.001
Some college credit or trade/vocational training	34.5%	25.7%
Bachelor’s degree or above	33.6%	57.9%
Marital status
Single, never married	38.9%	35.7%	20.59	2	<0.001
Married or domestic partnership	60.2%	58.3%
Widow/separated/divorced	0.9%	6.0%
Employment status
Student	10.0%	8.5%	9.86	3	0.20
Self employed	12.5%	13.6%
Employed (full-time/part-time)	66.0%	60.0%
Unemployed or retired	11.5%	17.9%
Overnight or same-day visitor
Same-day	42.8%	6.2%	191.80	1	<0.001
Overnight	57.2%	93.8%
First time or repeat visitor
First time	27.0%	50.6%	62.09	1	<0.001
Repeat	73.0%	49.4%
Travel arrangements
My trip to Hong Kong was organized by a travel agency/tour operator	6.4%	26.2%	76.16	1	<0.001
I organized my travel independently	93.6%	73.8%
Purpose of trip
Leisure	88.3%	73.0%	44.76	3	<0.001
VFR	5.3%	11.1%
Business and MICE	4.3%	13.8%
Other	2.1%	2.1%

Note: Boldface indicates a statistical significant difference at the 95% level of confidence with the adjacent column.

Questionnaire design

The questionnaire consists of three broad sections. The first section covers the counterfeit shopping experience. The second section captures the reason for purchasing counterfeit products, the marketing strategies sellers of counterfeit products used to entice shoppers, and the effectiveness of anti-counterfeiting strategies. The third section contains a standard set of demographic and travel-related questions. The questionnaire was created in English and then translated into simplified Chinese and then back-translated to ensure meaning equivalence (Brislin, 1980).

To determine the incidence of counterfeit purchases among different tourist types, mainland Chinese tourists and international tourists were asked if they had purchased any counterfeit products during their stay in Hong Kong. For those who had purchased counterfeit products, they were asked the location of where they had made the purchase, what products they had purchased and the value of the purchases. These questions were used to estimate the economic value of counterfeit purchases among different tourist types.

To assess the determinants of the incidence of counterfeit goods’ purchases among tourists, a variety of questions were asked, capturing the conceptual variables shown in Figure 1. The measurement of moral intensity is adapted from the work of Singhapakdi et al. (1996) which developed six items to measure moral intensity. The items are negatively worded so that a higher scores on rating scale signifies lower moral intensity. The scores from this six-item five-point Agreement Likert-type scale were averaged to provide a single measure. The Cronbach’s α for these six items is 0.852, suggesting good internal reliability. The measurement of personal relevance is adapted from Huang et al. (2014). The three items relating to personal relevance were averaged to provide a single measure. The Cronbach’s α for these three items is 0.745. A single item measurement of emotion is taken from the study by Sharma and Chan (2016). The emotion item asked “When making the decision of whether to buy counterfeit products or not, I was in a good mood” on a five-point Agreement Likert-type scale. A single item measurement was used for the “location” construct, namely, “Please state how likely you are to do this on vacation,” on five-point Likelihood Likert-type scale. This measurement was also used by Tolkach et al. (2017). Trevino (1986) provided the basis for the “effect of others” construct. Scores from a five-item five-point Agreement Likert-type scale were averaged to provide a single measure. The Cronbach’s α for these six items is 0.835. To test which anti-counterfeiting strategies might be most effective in combating the purchase of counterfeit goods, items were adapted from Herstein et al. (2015). Both buyers and non-buyers were asked to rate the effectiveness of six strategies of how to prevent consumers from buying counterfeit products, also on a five-point Effectiveness Likert-type scale. The internal reliability score of these five items was 0.820.

The sociodemographic variables and trip characteristics are a relative standard set of questions. Their coding is outlined in Table 2.

OPEN IN VIEWER

Table 2.

Coding of variables.

Variable	Coding description	Adapted from
Dependent variable
Bought counterfeits	Binary variable: 0 = No; 1 = Yes
Psychometric factors
Moral intensity	Scores from a six-item five-point agreement Likert-type scale was averaged to provide a single measure. The Cronbach’s α for these six items is 0.852.	Singhapakdi et al. (1996)
Personal relevance	Scores from a three-item five-point agreement Likert-type scale was averaged to provide a single measure. The Cronbach’s α for these six items is 0.745.	Huang et al. (2014)
Emotions	When making the decision of whether to buy counterfeit products or not, I was in a good mood.	Sharma and Chan (2016)
Location	Please state how likely you are to do this on vacation (5 = Very likely, 1 = Very unlikely)
Effect of others	Scores from a five-item five-point agreement Likert-type scale was averaged to provide a single measure. The Cronbach’s α for these five items is 0.835.	Trevino (1986)
Anti-counterfeiting measures	Scores from a six-item five-point Agreement Likert Scale was averaged to provide a single measure. The Cronbach’s α for these six items is 0.820.	Herstein et al. (2015)
Sociodemographic factors
Gender	Dummy variable: 0 = male; 1 = female
Age	Ordinal variable where 1 = 18–34 years; 2 = 35–54 years; 3 = 55+ years
Education	Ordinal Variable where 1 = high school graduate or below; 2 = some college credit or trade/vocational training; 3 = bachelor’s degree or above
Religion	Dummy variable: 0 = no religion; 1 = identify with a religion
Employment status	Dummy variable: 0 = employed/self-employed; 1 = student, retired, unemployed
Culture	Dummy variable: 0 = international; 1 = mainland Chinese
Trip characteristics
Tour arrangements	Dummy variable: 0 = independent; 1 = organized
Travel experience	Dummy variable: 0 = repeat; 1 = first time
Trip purpose	Dummy variable: 0 = other; 1 = business
Travel party	Dummy variable: 0 = with others; 1 = solo traveler

Regression method

One of the aims of this study is to determine the predictors of counterfeit purchase incidence among tourists. Since the dependent variable is dichotomous, we use the binary logistic regression estimation for predicting the probability of purchasing counterfeit products. The binary logistic regression takes the form

π i = Pr ( Y i = 1 ( Y i | X i = x i ) ) = exp ( β 0 + β i x i ) 1 + exp ( β 0 + B i x i )

1

where Y is a binary response variable. Y_i = 1 if the tourist purchases a counterfeit product and Y_i = 0 if the tourist does not purchase any counterfeit products; X = (X₁, X₂,…, X_k) are a set of explanatory variables which can be discrete, continuous, or a combination; x_i is the observed value of the explanatory variables for observation i. This equation can be expressed in log form

log i t ( π i ) = log ( π i 1 − π i ) = β 0 + β 1 x i 1 + … + β k x i k

2

In equation (2), the logistic model stipulates that the effect of a covariate on the probability of purchasing a counterfeit product is linear on the log-odds scale or multiplicative on the odds scale. If β_k > 0, then exp (β_k) > 1, and the odds increase. If β_k < 0, then exp (β_k) < 1, and the odds decrease.

Incidence of counterfeit purchases

To answer the first research objective, tourists were asked if they purchased any counterfeit products during their stay in Hong Kong. International tourists were significantly more likely (p = 0.028) than mainland Chinese tourists to purchase counterfeit products while on this trip in Hong Kong. Hence, mainland Chinese tourists were less likely to purchase counterfeit products while in Hong Kong.

For both mainland Chinese tourists and international tourists who did not purchase counterfeit products in this trip, they were asked why they did not purchase counterfeit products. Of the 85.7% of mainland Chinese tourists who did not purchase counterfeit products, 77.0% of them stated that they would never purchase counterfeit products while 21.1% stated that they would purchase counterfeit products in their home country. The reasons why they will purchase counterfeit in their home countries, rather than in Hong Kong was predominantly based on a cheaper price of counterfeits in mainland China. Similar results were found for this question among international tourists. Of the 79.6% of international tourists who did not purchase counterfeit products on this trip, 80.6% stated they would never purchase counterfeit products while a small proportion stated they would purchase counterfeit products elsewhere. The reasons for this are to take advantage of cheaper prices and a greater availability of choices elsewhere.

The large majority of tourists who purchased counterfeits were aware they were purchasing non-deceptive counterfeit products. Among mainland Chinese tourists, 89.5% either knew they were purchasing counterfeit goods or suspected they were based on cues such as price, location, and packaging. The corresponding figures for international tourists were 92.6%.

Table 3 shows the buying patterns of those tourists who purchased counterfeit products. Different cohorts purchase different types of counterfeit products and different quantities of counterfeit products. The first section of Table 3 shows the incidence of counterfeit purchased by product based on the total sample. The second section shows the same incidence based on those who purchased counterfeit products while third section shows the average number of counterfeit products purchased in each category, based on those who purchased counterfeits in that category. The fourth section shows the average unit price of those who purchased in that category while the fifth section shows the total average expenditure of counterfeit goods purchased in that category. In general, counterfeit apparel and accessories and handbags and wallets are the most likely to be purchased. Mainland Chinese are relatively more likely to purchase counterfeit footwear and other goods, such as cosmetics and medicines, while international tourists are more likely to purchase wearing apparel/accessories and electronics.

OPEN IN VIEWER

Table 3.

Counterfeit product buying patterns by cohort.

	Mainland Chinese tourists (N = 530)	International tourists (N = 530)	p Value
(1) Incidence of counterfeit purchase among total cohort (percentage)
Wearing apparel/accessories	3.2	8.7	0.004*
Electronics	0.9	3.4	0.042*
Handbags/wallets	6.6	9.8	0.781
Footwear	4.7	1.9	<0.001*
Watches/jewelry	3.8	3.2	0.079
Others	1.7	0.6	0.014*
(2) Incidence of counterfeit purchase among counterfeit buyers (percentage)
Wearing apparel/accessories	22.4	42.6	0.004*
Electronics	6.6	16.7	0.042*
Handbags/wallets	46.1	48.1	0.781
Footwear	32.9	9.3	<0.001*
Watches/jewelry	26.3	15.7	0.079
Others	11.8	2.8	0.014*
(3) Average quantity purchased (number)
Wearing apparel/accessories	2.0	2.2	0.675
Electronics	1.8	2.7	0.679
Handbags/wallets	1.7	1.3	0.012*
Footwear	1.2	1.6	0.203
Watches/jewelry	1.7	1.5	0.511
Other	7.0	4.3	0.492
(4) Average unit price (HK$)
Wearing apparel/accessories	514.71	259.24	0.015*
Electronics	2016.00	774.11	0.116
Handbags/wallets	2310.14	222.88	0.079
Footwear	667.28	722.80	0.757
Watches/jewelry	3050.35	397.53	0.330
Other	256.33	1771.67	0.104
(5) Average value (of those who bought counterfeit products; HK$)
Wearing apparel/accessories	1379.41	847.28	0.243
Electronics	4416.00	1889.11	0.270
Handbags/wallets	5516.71	298.46	0.138
Footwear	1003.28	1622.80	0.376
Watches/jewelry	3607.35	591.53	0.268
Other	2007.22	8848.33	0.157
Total value	4656.70	1308.59	0.063

Note: *Boldface indicates a statistical significant difference at the 95% level of confidence with the adjacent column.

In general, mainland Chinese tourists have a lower incidence of purchasing counterfeits but when they do, they tend to purchase more expensive counterfeit goods compared to international tourists. Of those who purchase counterfeit products, mainland Chinese tourists spend an average of HK$ 4656.70 (US$ 597.01), while international tourists spend an average of HK$ 1308.59 (US$ 167.77) on counterfeit products during their stay in Hong Kong.

To estimate the total value of the non-deceptive counterfeit goods market in Hong Kong, we take the incidence of purchasing a counterfeit product in each product category and multiply this by the adult population and by the median expenditure in each category (we use the median of the total expenditure value by product as, like most expenditure data, the data are right-skewed, hence the mean values will be inflated). Summing across product categories, gives the average per person expenditure by cohort (Table 4, row 2). The total number of adult persons is adjusted from “A Statistical Review of Hong Kong Tourism 2016,” the survey screened out anyone under 18 years of age. In Table 4, “All Tourists” (column C) is the aggregation of mainland Chinese tourists (column A) and international tourists (column B).

OPEN IN VIEWER

Table 4.

Total non-deceptive counterfeit expenditure by tourists in Hong Kong.

	Mainland Chinese tourists (N = 530)	International tourists (N = 530)	All tourists (N = 1060)
(1) Adult persons	39,270,337	12,697,234	51,967,571
(2) Median per person expenditure	$127.92	$74.45	$202.37
(3) Total expenditure on counterfeits (HK$ million) (1) × (2)	$5023.46	$945.31	$5968.77
(4) Total tourist expenditure (HK$ million)	$186,599	$52,280	$238,879
(5) Total tourist shopping expenditure (HK$ million)	$135,839	$15,049	$150,889
(6) Shopping % of total expenditure (5)/(4)	72.8%	28.8%	63.2%
(7) Counterfeit expenditure as % of shopping expenditure (3)/(5)	3.7%	6.3%	4.0%

Source: Row (1): Hong Kong Tourism Board, 2017; Row (4, 5): Hong Kong Tourism Board, 2017; Row (2, 3, 6, 7): Authors’ Calculations.

Economic value of counterfeit purchases

The median value of expenditure for each mainland Chinese tourists is HK$ 127.92 (US$ 16.40) while international tourists spend less at HK$74.45 (US$ 9.54). Extrapolating this data to the populations, mainland Chinese spend HK$ 5023.46 million (US$644.03 million), and international tourists spend HK$ 945.31 million (US$ 121.19 million) on non-deceptive counterfeits per year. This means, together the total tourist market spends HK$ 5.969 billion (US$762.50 million) on non-deceptive counterfeit products. Table 4 also shows total tourist expenditures (row 4) and total tourist expenditures on shopping (row 5), obtained from the Hong Kong Tourist Board Annual Reports. Taken as a percentage of total expenditures on shopping, non-deceptive counterfeit purchases represent 4.0% of all tourist shopping expenditure and up to 6.3% of international tourists’ shopping expenditure.

Determinants of counterfeit purchase incidence

Table 5 shows the regression results of the binary logistic regression which uses both intrinsic and extrinsic variables, like psychometric factors, trip characteristics, and sociodemographics as explanatory variables. The regression results in Table 5 shows that several variables in each category were statistically significant. Out of the psychometric factors, moral intensity, emotion, location and perception of anti-counterfeit measures are statistically significant. Those with lower moral intensity (a higher score on the moral intensity measurement) are more likely to purchase counterfeits (hypothesis 7 is supported). Those who are in a good mood (emotion) are more likely to purchase counterfeit goods (hypothesis 9 is supported). Similarly, the location variable, where tourists recognize the opportunity to purchase counterfeit goods while on vacation is significant (hypothesis 10 is supported). In contrast, tourists who perceive anti-marketing strategies to be effective are less likely to purchase counterfeit goods (hypothesis 12 is supported).

OPEN IN VIEWER

Table 5.

Regression results of predicting a counterfeit purchases.

	Logistic regression
DV: bought counterfeit goods (0 = No; 1 = Yes)	Coefficient	Odds ratio	S.E.	p Value
Constant	−5.201	0.006	0.913	0.000***
Sociodemographic factors
H1 gender	0.197	1.218	0.192	0.303
H2 age	−0.149	0.861	0.138	0.280
H3 education	−0.250	0.779	0.125	0.046**
H4 employment status	−0.379	0.684	0.230	0.100*
H5 religion	0.531	1.701	0.209	0.011**
H6 culture	−0.320	0.726	0.238	0.179
Psychometric factors
H7 moral intensity	0.508	1.662	0.189	0.007***
H8 personal relevance	−0.180	0.835	0.177	0.309
H9 emotions	0.364	1.439	0.123	0.003***
H10 location	0.802	2.231	0.101	0.000***
H11 effect of others	0.150	1.162	0.131	0.251
H12 anti-counterfeiting measures	−0.235	0.790	0.136	0.083*
Trip characteristics
H13 tour arrangements	0.238	1.269	0.247	0.335
H14 travel experience	−0.207	0.813	0.207	0.316
H15 trip purpose	−0.763	0.466	0.377	0.043**
H16 travel party	0.429	1.536	0.262	0.102
LR χ2(16)/F(16, 983)	189
Prob > χ2	0.0000
Pseudo R2/Adjusted R2	0.1998

Note: DV: dependent variable; S.E.: standard error; LR: likelihood ratio.

***Significant at α < 1%.

**Significant at α < 5%.

*Significant at α < 10%.

Only one variable of trip characteristics is significant—trip purpose. Business travelers are less likely to purchase counterfeit goods. This could be because they have less time to shop for counterfeits or have the income to purchase genuine luxury goods, rather than counterfeit products (hypothesis 15 is supported). Several sociodemographic factors are significant. Tourists who not employed (students, retired, unemployed) are less likely to purchase counterfeit goods (hypothesis 4 is supported). These findings are in line with Prendergast et al. (2002). Tourists with a higher level of education are less likely to purchase counterfeits, as found by Wee et al. (1995). Those who identify with a religion are more likely to purchase counterfeit goods. This finding goes against some existing research (for example, Weaver and Agle, 2002) but as noted by Clark and Dawson (1996), the relationship between religion and ethical behavior is complex, contextual, and multifaceted.

Anti-counterfeit marketing strategies

Lastly, we seek to determine which anti-counterfeiting strategies might be most effective in combating the purchase of counterfeit goods. Table 6 shows that, among both types of tourists, having price discounts of genuine products and campaigns that show potential risks in health or safety of buying counterfeits are the most effective anti-counterfeiting strategies. In terms of differences between cohorts, a variety of strategies would be more effective among the mainland Chinese tourists. These strategies relate to educating mainland Chinese tourists about the risks/negative impacts of purchasing counterfeit products and explaining why genuine products deserve high price.

OPEN IN VIEWER

Table 6.

Perceived effectiveness of anti-counterfeiting strategies.

	Mainland Chinese tourists (N = 530)	International tourists (N = 530)	p Value
Price discounts of genuine products discourage me from buying counterfeit products	3.99	3.68	<0.001
Campaigns that show potential risks in health or safety of buying counterfeits discourage me from buying counterfeit products	3.69	3.51	0.002
An explanation or justification of why genuine products deserve a high price discourages me from buying counterfeit products	3.54	3.34	0.001
Education programs about the negative impact of counterfeits on the economy and society discourage me from buying counterfeit products	3.53	3.41	0.037
Websites that enable consumers to sign online petitions to boycott counterfeits discourage me from buying counterfeit products	3.46	3.27	0.002
Anti-counterfeit advertisements using celebrities or other educational advertisements in the media discourage me from buying counterfeit products	3.46	3.22	<0.001

Note: Boldface indicates a statistical significant difference at the 95% level of confidence with the adjacent column.