Research Memorandum 77 March 2008

Transcription

1 Research Memorandum 77 March 2008 The Role of Product Involvement, Knowledge, and Perceptions in Explaining Consumer Purchase Behaviour of Counterfeits: Direct and Indirect Effects XUEMEI BIAN Centre for Marketing, Communications and International Strategy Business School, University of Hull Cottingham Road, Hull, HU6 7RX, the UK LUIZ MOUTINHO University of Glasgow ISBN

2 2008 X Bian, L Moutinho All intellectual property rights, including copyright in this publication, except for those attributed to named sources, are owned by the author(s) of this research memorandum. No part of this publication may be copied or transmitted in any form without the prior written consent from the author(s) whose contact address is given on the title page of the research memorandum. 2

3 Abstract This research examines the direct and indirect relationships (mediator and moderator effects) between product involvement, product knowledge, and perception on purchase intention of counterfeits in the context of nondeceptive counterfeiting. Results suggest that (1) perception is not a mediator of the effects of involvement/knowledge on purchase intention, (2) evidence of involvement as a moderator does not exist, (3) involvement/knowledge has no significant influence on counterfeit purchase intention, and (4) three out of seven dimensions of counterfeit perception are significantly influential on counterfeit purchase intention, with brand personality dimension appearing to have more explanatory power than the other two dimensions. Keywords Counterfeit; product involvement; product knowledge; purchase intention; mediator effect; moderator effect 3

4 INTRODUCTION Counterfeit products are those bearing a trademark that is identical to, or indistinguishable from, a trademark registered to another party and infringe the rights of the holder of the trademark (Chaudhry and Walsh, 1996; Kapferer, 1995; Grossman and Shapiro, 1988a,b; Bamossy and Scammon, 1985). The International Chamber of Commerce states that counterfeits account for 8 percent of world trade (Freedman, 1999). Globally, the sales of counterfeit products are estimated to be about $300 billion (Gentry, Putrevu, Shultz and Commuri, 2001). It is estimated that the value of counterfeit goods in the global market grew by 1100% between 1984 and 1994 (Carty, 1994; Blatt, 1993). Clearly, counterfeiting has become a significant economic phenomenon in the last two decades (Bian and Veloutsou, 2007). Despite that fact that selling and manufacturing counterfeits are considered to be crimes in some countries, for example, the U.S. and the U.K (Hopkins, Kontnic and Trunage, 2003; Bush, Bloch and Dawson, 1989), past research suggests that about one-third of consumers would knowingly purchase counterfeit goods (e.g. Phau, Prendergast and Chuen, 2001; Tom, Garibaldi, Zeng and Pilcher, 1998) regardless of the potential consequences associated with counterfeits. Researchers argue that consumer demand for counterfeits is one of the leading causes of the existence and upsurge in growth of the counterfeiting phenomenon (Gentry, Putrevu, Shultz and Commuri, 2001; Bamossy and Scammon, 1985), given that demand is always the key drive of a market. A number of researchers (e.g. Penz and Stöttinger, 2003; Wee, Tan and Cheok, 1995; Bloch, Bush and Campbell, 1993) have called for investigation of consumer behaviour and counterfeits. Nevertheless, the academic literature displays a strong focus on the supply side, while that on the demand side - why consumers knowingly buy counterfeits - still appears to be scarce (Penz and Stöttinger, 2005). Within this paper, we attempt to take a fresh look at the demand side of counterfeiting. The current research aims to investigate the effects of selfassessed product knowledge, product involvement, and consumers perceptions of counterfeit branded products (CBP), as well as the interaction between these variables on consumer purchase intention of CBP. We wish to highlight at this juncture that, in contrast to previous research, for example, Wee, Tan and Cheok (1995) and Nia and Zaichkowsky (2000) who examined counterfeit from a product perspective only, this research is one of the few studies which brings brand dimensions into the investigation of counterfeits. Inclusion of brand dimensions in the current study is considered to be a significant contribution to the literature, given that, if branded products did not attract consumers, counterfeiting would not be an issue (Cordell et al., 1996; Bloch, Bush and Campbell, 1993). Therefore, this research is to measure consumers perceptions of CBP as a product, and also as a brand - a counterfeit one. This research clearly does not address the need for research on the relationship differences between counterfeit models and their counterparts, the original brand models, but is nevertheless a useful first step in that direction. 4

5 Counterfeiting appears in different forms, as deceptive, non-deceptive and blur counterfeiting (Bian and Mouthinho, 2007). With deceptive or blur counterfeiting, the consumer is either not aware or unsure of the fact that he/she is purchasing a counterfeit rather than the original product and cannot be held accountable for this behaviour. This work limits its scope to nondeceptive counterfeiting, where consumers intentionally purchase counterfeits (Grossman and Shapiro, 1988a). The choice of the non-deceptive counterfeit context is considered as important, as only under this circumstance might consumers perceptions of counterfeits reflect their demand for these products. We develop hypotheses for and test the model presented in Figure I. In developing our model, we begin with a discussion of selected variables that contribute to consumer purchase intention, and within this context, explain how these variables may directly associate with each other. We also attempt to show that consumers perceptions may serve as a mediating link between product involvement/product knowledge and purchase intention, and that product involvement may be a moderator that affects the relationship between perceptions and purchase intention in the context of non-deceptive counterfeiting. Product involveme nt Consumer perception Purchase intention Product knowledge Figure 1. Hypothesized Relationships between Product Involvement, Product Knowledge, Perception of CBP and Purchase Intention of CBP 5

6 LITERATURE REVIEW AND HYPOTHESES Product Involvement Product involvement is commonly defined as a consumer s enduring perceptions of the importance of the product category based on the consumer s inherent needs, values, and interests (e.g. De Wulf, Odekerken- Schröder and Lacobucci, 2001; Mittal, 1995; Zaichkowsky, 1985). Since it was first introduced to marketing, the concept of involvement has been extensively used as a moderating or explanatory variable in consumer behaviour (Dholakia, 1997; 1998). It is regarded as a central framework, vital to understanding consumer decision-making behaviour and associated communications (Chakravarti and Janiszewski, 2003; Fill, 1999). Research shows that under high involvement conditions, buyer decision processes are thought to proceed through extended decision-making, a series of sequential stages involving information search and evaluation of criteria (Browne and Kaldenberg, 1997). Celsi and Olson (1988) report that the extent to which a product is viewed as personally relevant, in that it is perceived in some way to be instrumental in achieving their personal goals and values, makes the consumer more likely to be motivated to process information about it. Consumers neither wish nor are able to exert a great deal of effort to process information in a low involvement situation (Chung and Zhao, 2003). Based on the above research findings, it is rational to assume that under high involvement conditions consumers are more likely to put more effort into and are more capable of evaluating CBP as opposed to its counterpart, the original branded product (BP), in the context of non-deceptive counterfeiting. Deliberative information processing involves the scrutiny of available information and an analysis of positive and negative features, of costs and benefits (Fazio, 1990). Given that CBP is a low grade of BP (Nia and Zaichkowsky, 2000; Penz and Stöttinger, 2003), consumers with a higher level of product involvement are more likely to be able to distinguish the difference between CBP and BP, and hence develop less positive perceptions of CBP and show less preference for the CBP. On the other hand, the differences between CBP and BP might not be easily recognised, if the level of product involvement is low, due to consumers lack of motivation, effort and even capability in relation to processing information. Consequently, consumers perceptions of CBP and BP might not differ significantly under these circumstances, which will lead to more favourable perceptions of CBP. Therefore, existing research, as well as applied findings, leads to the first hypothesis regarding the effects of product involvement on consumers perceptions of CBP. H 1 : There is a negative relationship between product involvement and consumers perceptions of CBP. Consumers look for more personal, experimental and symbolic gain, other than maximising product functionality, in a high involvement situation than in a low involvement one (Solomon, Surprenant, Czepiel and Guttman, 1985). Given that consumers in a higher product involvement situation are more 6

7 likely to regard CBP as a product of low price and low quality, consumers are less likely to consider that CBP will satisfy the personal treat, excitement and status they desire. Therefore, they will have a lower level of purchase intention of CBP. Based on this, we predict the following. H 2 : There is a negative relationship between product involvement and consumer purchase intention of CBP. Product Knowledge Consumer product knowledge has been studied in a variety of different ways in recent years (e.g. Baker, Hunt and Scribner, 2002; Alba and Hutchinson, 2000; Brucks, 1986; Park, Mothersbaugh and Feick, 1994; Raju, Lonial and Mangold, 1995; Rao and Monroe, 1988). It has been recognised as a characteristic in consumer research that influences all phases in the decision process (Bettman and Park, 1980). Consumers with various levels of product knowledge differ in their perceptions of product attributes (Laroche, Bergeron and Goutaland 2003; Baker, Hunt and Scribner 2002; Blair and Innis 1996). Marks and Olson (1981) propose that consumers with higher levels of product knowledge have better developed and more complex schemata, with well-formulated decision criteria. In the same vein, Kempf and Smith (1998) suggest that consumers with higher levels of product knowledge are more diagnostic and better informed than those who have lower levels of product knowledge. Therefore, the higher the level of product knowledge a consumer possesses, the less chance there is that he/she will generate evaluation bias. Given these findings, this research argues that, in the context of non-deceptive counterfeiting, consumers with higher levels of product knowledge are more likely to be able to evaluate CBP more accurately, due to higher cognitive capacity. As a result, they should have less favourable perceptions of CBP. This argument is reflected in hypothesis 3. H 3 : There is a negative relationship between product knowledge and consumers perceptions of CBP. Past research results also show that self-perceived knowledge operates as a direct positive influencer of purchase intentions for original branded durable products (Berger, Ratchford and Haines 1994). Nevertheless, in the context of non-deceptive counterfeiting, the present study has established that consumers with a higher level of product knowledge are more capable of identifying the flaws of CBP. Thus, they should give less approval to the grade of CBP. This research tests for negative effects of product knowledge on purchase intention of CBP. H 4 : There is a negative relationship between product knowledge and consumer purchase intention of CBP. 7

8 Consumer Perception Past research asserts that it does not go without saying that the information from the sender (marketer) will definitely get through to the receiver (consumer) (Aaker and Myers 1987), since correct decoding of marketing information hinges on the consumer s perception of the communication content (Koekmoer 1991). The study of perceptions is crucially important to the marketers, as it can provide marketers with a more detailed picture about how their brand is perceived by consumers (Puth, Mostert and Ewing 1999). In addition, people respond on the basis of their perception of reality, not the reality per se (Puth, Mostert and Ewing 1999; Lewin 1936). It has been argued that consumer behaviour is, at root, driven by perceptions of a brand even if they are misconceptions of actual events (Porter and Claycomb 1997; Biel 1992; Porter 1976). Previous research suggests that it is perception that provides the grounds for purchasing decisions (Friedmann and Zimmer 1988; Borgers and Timmermans 1987). Analysis of consumer perceptions and decision-making processes is therefore extremely important in order to understand consumer behaviour, since it can help marketers to determine more readily what influences consumers buying behaviour (Puth, Mostert and Ewing 1999; Schiffman and Kanuk 1991). In the context of non-deceptive counterfeiting, we predict that consumers with more positive perceptions of CBP are more likely to purchase the counterfeit version. While this seems to be a somewhat obvious prediction, a model of purchase intention of CBP would be incomplete without its inclusion. Meanwhile, it will be extremely valuable to explore what dimension of perception weighs more in explaining purchase intention of CBP. This prediction is reflected in Hypothesis 5. H 5 : There is a positive relationship between consumers perceptions of CBP and purchase intention of CBP. Mediation and Moderator Effects While the main effects proposed above may exist, in this study, we predict that indirect effects (mediation effect and moderator effect) between these variables exist as well. Mediation effects occur when an independent variable influences the dependent through its effects on or as a result of a mediator variable (Baron and Kenny 1986). In this case, product involvement and product knowledge are the independent variables and perception serves as the mediator variable. Specifically, we predict that product involvement/ knowledge affects consumers perceptions of CBP, which in turn affects purchase intention. That is, the ways in which product involvement/product knowledge affect individual purchase tendency of CBP depend on how consumers perceive CBP. Following this reasoning, we propose that the relationship between product involvement/knowledge and purchase intention is mediated by consumers perception of CBP. These predictions are reflected in hypothesis 6 and hypothesis 7. H 6 : The relationship between product involvement and purchase intention of CBP is mediated by consumers perception of CBP. 8

9 H 7 : The relationship between product knowledge and purchase intention of CBP is mediated by consumers perception of CBP. In addition to the proposed mediation effects, we also test for moderator effects of product involvement. A moderated relationship occurs when a relationship is found to hold for some categories of a sample but not others (Bryman and Cramer 1999). The search for moderated relationships is important as it allows research to avoid inferring that a set of findings pertains to a sample as a whole, when in fact it only really applies to a portion of that sample (Bryman and Cramer 1999; Baron and Kenny 1986). In this case, it is rational to assume that consumers may have low purchase tendency of CBP given a perceived high level of product involvement regardless of their perceptions of CBP; in contrast, consumers may be more likely to have a high purchase tendency of CBP if they perceive a low level of product involvement and have more positive perceptions of CBP. That is, the ways in which consumers apply the influence of their perceptions of CBP to their purchase intention of CBP will depend on the consumers level of product involvement. This proposition is reflected in hypothesis 8. H 8 : Product involvement moderates the relationship between consumers perceptions of CBP and purchase intention of CBP. METHODOLOGY Selected Brand Counterfeit Rolex watches were investigated in this research. The reasons for this choice are that this brand figures in the list of the most counterfeited brands (Poulter 2006) and the counterfeits of this brand are available (provided by Trading Standards of Glasgow from counterfeits they had confiscated) for the use as stimuli. Given that only one brand of one product category (watches) was examined in the current study, the findings of this research should be viewed with caution, as they may not be generalisable to other product categories. Sample Characteristics This study was conducted in Glasgow, UK. This is because the UK is perceived to be one of the main recipients of counterfeits in the world (Kay 1990), and counterfeits are widespread in Glasgow according to the Trading Standards officials. The sample in present study consists of 430 consumers aged 18 years old and above. The analysis reported here is based on 321 usable observations. A total of 56.4% are male and 43.6% are female. Some 58.8% of the participants having an educational attainment lower than degree level, with 26.8% having a Bachelor s degree and 14.4% having a Master s degree or higher. The age breakdown of the sample is: 21.2% under 20 years old, 24.3% between 21 and 30, 19.6% between 31 and 40, 20.2% between 41 and 50, and 14.6% 51 and above. 9

10 Procedure The first author contacted 20 randomly selected supermarkets from a list of supermarkets located in Glasgow by mail, and then followed with a phone call to request help in studying consumer perceptions of CBP and purchase decision. Four supermarkets gave permission and provided full support to the researcher for data collection. Two of them are relatively small and located in residential areas. The other two are medium-sized stores with minimum daily sales of over 35,000. One is located in a shopping centre and perceived as a relatively expensive supermarket, with another one, located at the edge of the city, being well known for its low price strategy. Survey data was collected using a stimulus approach in November and December of 2006 by the first author. Specifically, counterfeit Rolex watches were presented to respondents. Every 10th shopper aged 18 years old or above was invited to participate in this research at the exit of the supermarkets. The incentive method (a box of chocolates worth around 2.50) recommended by previous researchers (e.g. Aaker, Kumar and Day 1997; Wiseman, Schafer and Schafer 1983), and the gaze and touch method suggested by Hornik and Ellis (1988), as well as the appealing expression recommended by Hornik (1982) were adopted to improve the response rate. Measures The explanatory constructs in our model of purchase intention of CBP include product involvement, product knowledge, and perception. All involved constructs were measured using five-point Likert scales (1 = strongly disagree, 5 = strongly agree). Apart from the measure of perception which was developed by the first author of this paper, the rest of the scales utilised to measure involved constructs were all adopted from existing research with necessary adaptation. The items for all model constructs were listed in Appendix 1. Purchase intention was assessed using a 5-item scale developed by Spears and Singh (2004). Cronbach s alpha (a) for this scale equals Product involvement was measured using 10-item scale developed by McQuarrie and Munson (1992), with a Cronbach s a of There are three distinct but related ways in which consumer knowledge is conceptualised and measured: objective knowledge, subjective knowledge, and experience (Flynn and Goldsmith 1999). Here our product knowledge construct is a 4-item scale developed by Smith and Park (1992) that measures an individual s selfassessed ratings of knowledge. The choice of measuring self-assessed knowledge is supported by Meeds (2004) who finds that self-assessed knowledge is a better predictor of participants cognitive responses and general attitudinal evaluation in comparison to other kinds of knowledge. Cronbach s a for this scale is 0.77, which is just short of the 0.80 criterion of Bryman and Cramer (1999). The average scores of the items related to product involvement and product knowledge were calculated for the use of further analysis. Table 1 presents the description of these constructs, including means and standard deviations. 10

11 The scale development of the perception construct went through two stages. In the first stage, a master list of items related to consumer perceptions of CBP was generated. The master list is a combination of items of consumers perceptions of CBP as a product and items of consumers perceptions of CBP as a brand. The items concerning CBP as a product were generated from past research in the study of counterfeiting, packages, adverting and consumer reports of the original branded product. To operationalise CBP as a brand, Aaker s (1997) brand personality scale was adopted to form the main part of the item pool. In addition, the adjectival expressions in relation to personality, which were generated from product packages, advertising, and consumer reports, were also included in the list. To ensure the master list was complete, a free-association task was conducted. Subjects (n = 13, 46% male, mean age = 42) were asked to write down the items that first came to mind when thinking about counterfeit Rolex watches. The items resulting from this task were added to the master list of perception items, which resulted in a total of 89 items. The perception items were reduced to a more manageable number in the second stage. Subjects (n = 12, 50% male, mean age = 39) were asked to rate how important and relevant the items were in terms of evaluating the studied counterfeit Rolex watches (1 = not at all important/relevant, 5 = extremely important/relevant). To isolate the most relevant/important items, the cut-off for the final list of perception items was a scale rating of 4, thereby leaving 28 items for further study. Dimensions of Consumers Perceptions of CBP Principal component analysis was used to condense the information obtained in relation to consumers perceptions of counterfeit Rolex watches. Varimax rotation revealed 7 factors (all eigenvalues are greater than one) which account for percent of the overall variance. The 7 factors were labelled as excitement personality (2 items), competence personality (9 items), general product attribute (5 items), functional attribute (3 items), satisfaction benefit (3 items), image benefit (4 items), and functional benefit (2 items) (See Appendix 2 for detailed factor analysis results.). Factor scores were calculated for the use at the modelling stage. A measure of internal consistency of the extracted factors (Cronbach s a, and Pearson s correlation if appropriate) is provided in Table 1. The Cronbach s a coefficients are all above.70 with the exception of the satisfactory benefit factor, which has a Cronbach s a of.66. However, the lower value of the Cronbach s a might be caused by the small number of items involved (3 items); it is therefore considered acceptable. The Pearson s correlations are all higher than.25, and so are significant at the.01 level. 11

12 Table 1. Description of Indicators Used to Measure Independent, Mediator and Dependent Variables Counterfeit Rolex No of items Cronbach s a Mean SD Independent variables Product knowledge Product involvement Mediating variables Excitement * n/a n/a Competence n/a n/a Product attribute n/a n/a Functional attribute n/a n/a Satisfaction benefit n/a n/a Image benefit n/a n/a Functional benefit * n/a n/a Dependent variable Purchase intention * Pearson s Correlation is significant at the 0.01 level (n = 321) RESULTS The direct effects and proposed mediation effects are assessed simultaneously. Hypotheses are tested with a series of hierarchical regression analyses. Results are presented in Table 2 and Table 3. In both tables, the first group of models (a) examine the relationship between the independent variable and the mediators. A separate regression equation is used for each perception factor. The second group of regression models (β) examine the relationship between the proposed mediators and the outcome variable. The third group of regression models (τ) look at the relationships between the independent variables and the outcome variable. Hypothesis 1 proposed that product involvement would be negatively related to consumers perceptions of counterfeits. The results in Table 2 suggest no support for this hypothesis. None of the coefficients on consumer perception related factors is statistically significant at the level of p <.05. Hypothesis 2 posited a negative relationship between product involvement and consumer purchase intention of CBP. The results reject this hypothesis (β =.08, p > 12

13 .05), suggesting that product involvement has no significant effect on consumer purchase intention of CBP. Table 2. Regression Analysis of Relationships between Product Involvement, Perception, and Purchase intention Estimate a (Sig) Estimate β (Sig) Estimate τ (Sig) Mediators Competence.008 (.880).342 (.000)* n/a Product attribute (.300).079 (.158) n/a Image benefit.016 (.766).053 (.349) n/a Satisfaction benefit.092 (.098).131 (.018)* n/a Functional attribute (.759).118 (.034)* n/a Excitement (.930).064 (.255) n/a Functional benefit (.527) (.060) n/a Outcome variable Purchase intention n/a n/a.08 (.152) * P < 0.05 Hypothesis 3 proposed a negative relationship between product knowledge and consumers perceptions of CBP. Table 3 reveals a partial support for this hypothesis. Product knowledge is negatively associated with consumers perceptions of product attribute (β = -.110, p <.05), but is unrelated to other perception factors. Hypothesis 4 proposed that product knowledge would be negatively related to consumer purchase tendency of CBP. The coefficient suggests a rejection for this hypothesis (β =.086, p >.05) (see Table 3 τ). Hypothesis 5 proposed that consumers perceptions of CBP would be positively associated with purchase intention. In this case, Table 2 reveals partial support for this hypothesis. Consumer purchase intention of CBP is not influenced by their perceptions of product attribute, image benefit, excitement personality and functional benefits, but is positively and significantly influenced by their perceived competence personality (β =.342, p <.05), satisfaction benefit (β =.131, p <.05) and functional attribute (β =.118, p <.05) of counterfeit Rolex. A mediator variable is one that is both a product of the independent variable and a cause of the dependent variable. The search for mediator variables is often referred to as explanation and it is easy to see why. Bryman and Cramer (1999) assert that one is able to gain some explanatory leverage on the bivariate relationship, if it is found that a test variable acts as a mediator variable. According to MacKinnon, Lockwood, Hoffman, West, and Sheets (2002) and Baron and Kenny (1986), four conditions are needed to show evidence of mediation. First, the independent variable is related to the outcome variable (τ). Second, the independent variable has a 13

14 statistically significant effect on the presumed mediator (a). Third, the hypothesized mediator is associated with the outcome variables (β). Finally, when the proposed mediator is entered into a regression equation after the independent variable, the mediated effect (aβ) is statistically significant, but the regression weight for the independent variable should be reduced either significantly or to zero. In the case of mediation effects related to the independent variable of product involvement, since the research results demonstrate that the independent variable is not significantly related to the outcome variable, or to any of the proposed mediators (Table 2), as a result, the first two conditions necessary for mediation effect are not met. Therefore, there is no need to conduct further mediation effect tests. The mediation hypothesis 6 is not supported. Although product knowledge is significantly associated with one of the proposed mediators (product attribute, β = -.110, p <.05), nevertheless, this mediator does not appear to have a significant influence on the outcome variable. In addition, product knowledge is not significantly associated with the outcome variable either (see Table 3 τ). Thus, the first and the third conditions of mediator effect are not met. Consequently, the mediation hypothesis 7 is rejected. Table 3. Regression Analysis of Relationship between Product Knowledge, Perceptions, and Purchase Intention Mediators Estimate a (Sig) Estimate β (Sig) Estimate τ (Sig) Competence (0.380) (0.000)* n/a Product attribute (0.049)* (0.158) n/a Image benefit (0.875) (0.349) n/a Satisfaction benefit (0.894) (0.018)* n/a Functional attribute (0.618) (0.034)* n/a Excitement (0.303) (0.255) n/a Functional benefit (0.975) (0.060) n/a Outcome variable Purchase intention n/a n/a (0.123) * P < 0.05 To test the moderating effects of product involvement, we dichotomized the moderator at the neutral point of the 5-point scale. The results of the moderating effects of product involvement are shown in the regression presented in Table 4. None of the interactions of low involvement with perception related factors is significant (p <.05), which means that the effect of consumers perceptions of CBP on CBP purchase intention for low involvement consumers are no different than they are for high involvement ones. Therefore, the results show no evidence that product involvement can moderate the effects of consumers perceptions of CBP on their purchase intention. These findings rejected hypothesis 8. 14

15 TABLE 4: Regression Model of Purchase Intention of CBP Variable Std. e Std. β T Sig. (Constant) Competence Product attribute Image benefit Satisfaction benefit Functional attribute Excitement Functional benefit Low involvement Low involvement x competence Low involvement x product attribute Low involvement x image benefit Low involvement x satisfactory benefit Low involvement x functional attribute Low involvement x excitement Low involvement x functional benefit a Dependent Variable: Counterfeit Rolex intention CONCLUSIONS AND IMPLICATIONS The major objective of this study is to investigate direct relations between the three most exploratory constructs of consumer behaviour (product involvement, product knowledge and perception) and purchase intention in the context of non-deceptive counterfeiting. We also investigate indirect effects, namely whether relations between product involvement/product knowledge and purchase intention of CBP are mediated by consumers perceptions of CBP, and whether effects of consumers perceptions of CBP on purchase intention are moderated by product involvement. The results of this study reject the proposed relationship between product involvement and perceptions of CBP, and the relationship between product involvement and purchase intention of CBP. 15

16 These findings imply that consumers who have a high level of product involvement may also purchase counterfeits willingly. At the same time, it is unnecessary to add that they have more negative perceptions of CBP. One possible explanation for these results could be that consumer purchase behaviour of CBP may vary according to different usage situations. For example, they might buy CBP for the use at home, but buy BP for use in public places. In the future, it would be interesting to investigate whether usage situations moderate the relationship between product involvement and purchase intention of CBP. The findings related to product involvement and consumers perceptions of CBP are more than a little surprising. We have to admit that we failed to find any sound explanation from any existing theoretical framework that we are aware of. In view of this, a premise we would like to make here is that the counterfeiting phenomenon might provide fresh challenges to existing literature developed without presence of counterfeits. It may be worthwhile replicating the current research design in other product categories and further exploring the underlying reasons for product involvement effect on purchase behaviour. The results of this study provide no support to the proposition that product knowledge influences purchase intention of CBP, but partial support to the proposed relationship between product knowledge and perceptions of CBP. Specifically, it is difficult, to draw conclusions about the direction of causality here, particularly in the relationship between self-assessed product knowledge and reported purchase intention. There is evidence that more knowledgeable consumers are more likely to have less favourable perceptions of general product attributes of CBP. Nevertheless, the support appears to be marginal. In addition, it is important to note that despite evidence of statistical significance, the regression coefficient is relatively small and thus the magnitude of the effects reported here are small. Product knowledge does not appear to have significant influence on other perception related factors. These results can be interpreted in the light of the fact that with advancements in watch technology, the accurate time-telling function of watches is no longer difficult to achieve. In other words, in terms of telling the time, counterfeit watches can be similar to their counterpart original branded watches. The more knowledgeable the person is about watches, the more he or she is aware of this fact. As a result, although more knowledgeable consumers can judge the differences between the two versions of one brand, and therefore perceive less positive perceptions of product attributes of CBP, their purchase tendency of counterfeit watches is not significantly affected by these perceptions. Perception is the only explanatory variable out of three that shows a direct effect on consumer purchase intention of CBP. However, our findings suggest that only three out of 7 dimensions of the perception construct appear to be significantly influential, with the personality related dimension having the largest coefficient, which indicates the greatest explanatory power on consumer purchase tendency of CBP. For the first time in the literature on 16

17 counterfeits, our findings provide empirical evidence to support the commonly accepted notion that CPB-prone consumers are seeking the positive brand personality associated with BP. More importantly, this research is also the first to establish that perceived brand personality plays a more dominant role in explaining consumers purchase intention of CBP than other influential factors (e.g. benefit and product attribute). Nia and Zeicksey (2000) report that both counterfeit owners and non CBP owners perceive satisfactory benefit related to CBP. This research moves one step forward by suggesting that there is more chance that non CBP owners will knowingly purchase CBP in the future if they perceive the CBP to be satisfactory. The results of this study provide no support to the premises related to mediation effect and moderator effect. Given these results, we can conclude that product involvement does not moderate the effects of consumers perceptions of CBP on consumer purchase intention. There is no evidence that perceptions of CBP mediate the effects of both product involvement and product knowledge on purchase tendency of CBP. For marketers of BP, the findings offer two main practical implications. First, regardless of it having been reported that improving consumers knowledge of BP is a means commonly adopted by BP owners to hamper CBP (Green and Smith 2002), this study s findings advise against devoting resources to improving consumers knowledge of a product with an aim to curb demand for CBP. This is because, while it is not necessary to say that consumers who are more knowledgeable are less CBP- prone, at the same time they do not process less favourable perceptions of CBP than less knowledgeable consumers. Secondly, one course of action that firms can take to address counterfeiting is to conduct marketing campaigns to stress brand personality differences between CBP and BP. Marketing campaigns could meet with remarkable success if they took the approach of highlighting the negative personality of the typical users, brand endorsers, company employees and CEO of the company of CBP, as these people are regarded as the directly influential factors on consumer perceived brand personality (Aaker 1997). Surprisingly and interestingly, the research results of the current study reject the majority of our research hypotheses. Rejections of the proposed hypotheses which were developed based on previous literature might be an indication of serious challenges to existing theoretical work established over years in the context of lack of CBP presence. Clearly, there is still a long way to go before we achieve any sound understanding of counterfeit related behaviour. Considering the important role played by perception on purchase tendency of CBP, what have yet to be explored are the antecedents of perception of CBP, and differences between perceptions of CBP and BP. Future research should replicate the current study with other widely counterfeited brands, in an effort to test the applicability of the current research findings in other contexts. Given that it is still not quite clear whether the overall consumers perceptions of BP are affected or not after the entry of CBP, a before-after experimental design with control would enable researchers to observe the potential change in perceptions of BP as a result of the entry of CBP. This type of design has been used by a number of 17

18 previous studies in the study of brand extension (e.g. Diamantopoulos et al. 2005; Morrin 1999) due to its high level of control in accounting for extraneous factors which can assist in enhancing the internal validity of the research (Calder et al. 1981). In addition, future research could test more brands (for example, around ten brands in one product category), including generic brands, using both similarity judgement and attribute-based multidimensional scaling techniques to explore where CBP and BP are located in the spatial map. This is consistent with Malhotra (1999), who suggested that eight brands or stimuli should be included to obtain a well-defined spatial map. Direct similarity judgement may be used for obtaining the spatial map, and attribute ratings may be used as an aid to interpret the dimensions of the perceptual map. Similar procedures can be used for preference data. These efforts will assist marketers to obtain a clear view as to where their brands stand in the market place where the counterfeits exist. 18

19 REFERENCES: Aaker J. L. (1997) Dimensions of Brand Personality. Journal of Marketing Research, 34 (August), pp Aaker, D.A. and Myers, J.G. (1987) Advertising Management, 4 th Edition, Prentice-Hall, Englewood Cliffs, NJ. Aaker, D.A., Kumar, V. and Day, G.S. (1997) Marketing Research, 6 th Edition, John Wiley & Sons, Inc, New York. Alba, J.W. and Hutchinson, J.W. (2000) Knowledge Calibration: What Consumers Know and What They Think They Know, Journal of Consumer Research, 27(2), pp Baker, T., Hunt, J.B. and Scribner, L.L. (2002) The Effect of Introducing A New Brand on consumer Perceptions of Current Brand Similarity: The Roles of Product Knowledge and Involvement, Journal of Marketing Theory and Practice, 10(4), Fall, pp Bamossy, G. and Scammon, D. (1985) Product Counterfeiting: Consumers and Manufacturers Beware, Advances in Consumer Research, 12(1), pp Baron, R.M. and Kenny, D.A. (1986) The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations. Journal of Personality and Social Psychology, 51(6), pp Berger, I.E., Ratchford, B.T. and Haines Jr., G.H. (1994) Subjective Product Knowledge as A Moderator of the Relationship between Attitudes and Purchase Intentions for a Durable Product, Journal of Economic Psychology, 15(2), June, pp Bettman, J.A. and Park, C.W. (1980). Effects of Prior Knowledge and Experience and Phase of the Choice Process on Consumer Decision Making Processes: A Protocol Analysis, Journal of Consumer Research, Vol. 7, Issue 3, December, pp Bian, X. Moutinho, L. (2007). Which Factors Determine Formation of Consideration set in the Context of Non-Deceptive Counterfeiting. Thought Leaders International Conference on Brand Management, UK. Bian, X. and Veloutsou, C. (2007) Consumer s Attitudes Regarding Non- Deceptive Counterfeit Brands in the UK and China, Journal of Brand Management, 14(3), pp Biel, A. (1992) How Brand Image Drives Brand Equity, Journal of Advertising Research, 32(6), November/December, prc6-rc12. Blair, M. E. and Innis, D.E. (1996) The Effects of Product Knowledge on the Evaluation of Warranted Brands, Psychology and Marketing, 13(5), pp Blatt, J. (1993) Battling Counterfeit Products on the US Side of the Pacific Rim, The International Computer lawyer, 1(13), pp

20 Bloch, P., Bush, R. and Campbell, L. (1993) Consumer Accomplices in Product Counterfeiting, Journal of Consumer Marketing, 10(4), pp Borgers, A. and Timmermans, H. (1987) Choice Model Specification, Substitution and Spatial Structure Effects: A Simulation Experiment, Regional Science and Urban Economics, 17, pp Browne, B.A. and Kaldenberg, D.O. (1997) Conceptualizing Self-monitoring: Links to Materialism and Product Involvement, Journal of Consumer Marketing, 14(1), pp Brucks, M. (1986) A Typology of Consumer Knowledge Content, Advances in Consumer Research, 13(1), pp Bryman, A. and Cramer, D. (1999) Quantitative Data Analysis: with SPSS Release 8 for Windows A Guide for Social Scientists, Routledge: London and New York. Bush, R., Bloch, P. and Dawson, S. (1989) Remedies for Product Counterfeiting, Business Horizon, 32(1), January February, pp Calder, B.J., Phillips, L.W. and Tybout, A.M. (1981) Designing research for application, Journal of Consumer Research, 8(9), pp Carty, P. (1994) Fake s Progress, Accountancy, 114(1216), December, pp Celsi, R.L. and Olson, J.C. (1988) The Role of Involvement in Attention and Comprehension Processes, Journal of Consumer Research, 15(2), September, pp Chakravarti, A. and Janiszewski, C. (2003) The Influence of Macro-Level Motives on Consideration Set Composition in Novel Purchase Situations, Journal of Consumer Research, 30(2), September, pp Chaudhry, P.E. and Walsh, M.G. (1996) An Assessment of the Impact of Counterfeiting in International Markets: the Piracy Paradox Persists, Columbia Journal of World Business, 31(3), Fall, pp Chung, H. and Zhao, X. (2003) Humour Effect on Memory and Attitude: Moderating Role of Product Involvement, International Journal of Advertising, 22(1), pp Cordell, V., Wongtada, N. and Kieschnick, L. (1996). Counterfeit Purchase Intentions: Role of Lawfulness Attitudes and Product Traits as Determinants, Journal of Business Research, 35(1), January, pp De Wulf, K., Odekerken-Schröder, G. and Lacobucci, D. (2001) Investments in Consumer Relationships: a Cross-Country and Cross-Industry Exploration, Journal of Marketing, 65(4), October, pp Dholakia, U.M. (1997) An Investigation of the Relationship Between Perceived Risk and Product Involvement, Advances in Consumer Research, 24(1), pp Dholakia, U.M. (1998) Involvement-Response Models of Joint Effects: An Empirical Test and Extension, Advances in Consumer Research, 25(1), pp