Determinants of Purchasing Imported Products in a Regular Basis : Development of a Regression Model

Bangladesh, an emerging economy, has substantial dependence on and predilection for foreign products. We made an effort to delve into the factors influencing consumer purchase decision of imported products, considering some regular use consumer shopping products. We have employed multiple regression analysis as the statistical tool. Regression model of our study demonstrated statistical significance and appeared competent to explain 76% of the total variance in the dependent variable. The determinants posed positive influence are brand preference, COO bias, trust in retail store, uninterrupted availability, fewer local alternatives, prestige symbol in reference group; while price and utility imbalance played negative influence.


Introduction
Globalization, as a driving force, increases global competition enormously in recent times.Foreign products and brands competing with that of local is a customary event of this age.People, with more options, are going through more critical judgmental process to make purchase decisions.Both Developing and developed countries' citizens are subject to this judgmental process.Every country may have diverse set of judgmental biases in this regard, which mostly abet marketers to make prudent decisions.
In this study, we made an effort to identify the biases of Bangladeshi consumers in purchasing foreign/imported products.We have considered some frequently used consumer shopping products, as shampoo, petroleum jelly, body lotion, and beauty soap.These products are inconspicuous in nature, as the use of them cannot be exposed to public.Additionally, purchase decisions of these products, when made in a regular basis, can be based on brand preferences, dermatological suitability, availability, variety, price, 'made in' labels etc.It will be worthy to mention that some local alternatives are also available for these consumer shopping products, which will better justify the rationale for purchasing imported products.

Literature Review
Most of the researches in this area of international marketing encompassed developed countries and concentrated to divergence in consumer preferences for country-of-origin (COO), but a few (Wong and Zhou, 2005;Okechuku and Onyemah, 1999) incorporated impetus for purchasing foreign products/brands in emerging economies.Bangladeshi consumers' perception of quality about imported products was measured, but it was according to COO (Kaynak, Kucukemiroglu, Hyder;2000).To the best of our knowledge, there was no previous effort in Bangladesh to investigate the determinants for purchasing foreign products, which bestow us the reason to explore.
COO as an effect came into being after the seminal study of Schooler (1965).Until then, researches on country of origin (COO) influences and implications became extensive.Origin preferences may be product dependent (Etzel and Walker, 1974;Kaynak and Cavusgil, 1983); hold positive home country bias in comparison to similar countries (Chao, 1989;Schoolar, 1965); or manifest a negative home country bias when home country is less developed than alternative sources (Jaffe and Martinez, 1995;Ettenson, 1993;Papadopoulos, Heslop, and Beracs, 1990;Tan and Farley, 1987;Wang and Lamb, 1983;Hampton, 1977;Gaedeke, 1973;Schooler, 1971).After different dimensional spread of COO, little effort was been made to identify biases for consuming or acquiring foreign made products across different product categories (Steenkamp, Batra, and Alden 2002).In a developing country like Bangladesh, consumers favor foreign brands for their association with higher prestige (kapferer, 1997).Foreign brands hold higher prestige owing to their higher price and relative scarcity in comparison to local brands (Batra et al. 2000;Bearden and Etzel, 1982).Furthermore, purchasing foreign brands enhances consumers' self image, as being cosmopolitan, sophisticated, and modern (Friedman, 1990).The products under consideration in this study have less chance to show publicly and therefore, can have less influence from perceived prestige (Steenkamp et al. 2002).Yet, use of foreign products may provide higher repute among reference group members.The matter of 'scarcity' requires elucidation to fortify the rationale of foreign products' availability.According to Cordell (1992), buyers may portray country impression by criteria such as price, styling, quality, or availability.Consequently, availability as a positive stimulus for country impression (Cordell, 1992) and foreign brands' relative scarcity in comparison to local brands (Batra et al. 2000;Bearden and Etzel, 1982) induce us to deem scarcity and availability as selective coverage not extensive coverage.Because unavailability of the product group in this study may result in brand shift as physiological and dermatological suitability is desirable.So, availability throughout the year with selective coverage is of the essence for frequently used products.
Perceived quality level commonly recognized as an elementary force of purchase intention (Monroe and Krishnan 1985).Quality is defined as the 'degree to which a set of inherent characteristics fulfills requirements' (ISO, 2000).Quality perception can be explained as a consumer's assessment of a brand's overall brilliance based on intrinsic (consistent performance) and extrinsic (brand name, promotional offers, packaging) cues (Kirmani and Baumgartner, 2000).So, perception of quality is associated with brand name.In another words, a brand's assessment of overall excellence depends on satisfaction about internal quality standards resulting from consumption experience.Brand name, acknowledged widely as a crucial indicator of quality (Rao and Monroe, 1989).Moreover, famous brand names reduce the effect of production origin because of consumers' brand loyalty and maker's implied warranty (Cordell, 1992).At the same time, product quality image is correlated with economic level of producer country (Alba and Hutchinson, 1987).Most of the famous brand names in this world are also originated from developed nations.As a corollary of globalization, products country of origin (COO) can be different from the brand's originating country (Häubl 1996).Such products can be referred as bi-national or hybrid products (Chao, 1993;Ettenson and Gaeth, 1991).Thus, not only the brand name and its origin but also the COO for production shapes up quality expectations of consumers in case of foreign products.In this study, the term foreign product or imported product is indicating that consumers know the COO.Apparently, Bangladeshi consumers are not looking to 'Made in' levels but the identification of imported/foreign products can be made as locally produced brands are generally familiar to consumers.Extensive availability of a particular beauty soap/petroleum jelly/shampoo/body lotion also is an indicator of locally produced product.So, purchasing a foreign product does not always mean that consumers know the COO.Surely, there are chances of misidentification, mostly when a previously imported brand starts local production.In this study, COO preference is considered in the sense that any imported product is preferred by a consumer for particular country of origin(s) or not.
Perceived value can be termed as "the consumer's overall assessment of the utility of a product based on perceptions of what is received and what is given" (Zeithaml, 1988).Value, rationally indicating to the relationship between utility received and price (and energy) sacrificed.High perceived value could engender repeat purchase and brand loyalty (Parasuraman and Grewal, 2000).In this study, we will investigate the consistency between utility received and price paid, as price is more objective to human judgment.So, judging the perceived value by consumers can turn out the extent of resemblance between price and realized utility.
People have an urge availing reference group membership through the use of particular brand.Attaining a group membership by using a brand is more likely for publicly consumed products rather than inconspicuous products (Bearden and Etzel, 1982).Mehta, Lalwani and Ping (2001) found higher reference group influence for publicly consumed services than privately consumed services.Recognizing this notion from literature, we want to inspect the significance of the social payoff for using foreign products, in this study the products are inconspicuous in nature.
As we have noted earlier that developing countries prefer foreign products, especially from developed countries, than domestic products.Local manufacturers with trivial corporate identity, use foreign names as COO by relinquishing their identity because of consumers' inclination to foreign products (Okechuku and Onyemah, 1999).Bangladesh also holds similarity in this regard.Therefore, reliability on retail outlet for genuine foreign products is desirable by consumers.
At present in Bangladesh, a few number of local corporates gained acceptance for their product quality and relatively low price offers than foreign products.Moreover, investment in production facilities by MNEs also positively contributed the increment of local products' acceptance.Consequently, increased availability of locally produced alternatives can be an intimidation to the sales volume of foreign manufactured products and can breed greater prospect to consumers for evaluating local and foreign products.Sometimes, local importing agent can generate trust to the consumers because of local agent's trustworthiness in the domestic market; i.e. 'Heinz' brand of sauce is imported in Bangladesh by Transcom Group, a locally renowned business enterprise.So, dependability on the local importing agent can generate preference to foreign products.Though, from general observation we have seen that very less people are looking for this type of information while making purchase decisions.

Objectives of the Study
Bangladeshi consumers, holding a favorable image to foreign products, may have several considerations for purchasing foreign made products.The study is aimed at achieving following outcomes: Identifying multiple variables that lead to purchase and consumption of foreign products in a regular basis.
Revealing the significance of the explanatory variables.
Developing a model to understand the strength of relationship between the dependent variable and independent variables.
Ascertaining the level of influence by each explanatory variable on dependent variable.

Methodology
The study used both primary and secondary information.Theoretical part of the article greatly used secondary sources such as review of scholarly journals, whereas the quantitative findings and interpretations are on the basis of primary data.
The steps in sampling design process were as follows: Target population: Our study focused the individuals as sampling elements, who make purchase decisions and have their own income, as dependents will be less competent to judge price-utility consistencies, brand name and COO implications, and overall biases for purchase.Our sampling unit became commercial offices, where all employees are earning members and most possibly make purchase decisions.We considered only the executive level employees because of their income level consistencies with foreign products purchase.Moreover, commercial offices can have greater array of diversified social backgrounds and therefore can generate wider viewpoints.The extent of the study was Dhaka city (major commercial areas) and time was December 2007.
Sampling technique: It was a consumer-focused study on those, who mostly have the capacity and tendency to purchase and consume foreign products, like shampoo, petroleum jelly, body lotion, and beauty soap etc. in a regular basis.Consequently, we have identified the major commercial areas of Dhaka city as Motijheel, Karwanbazar, Mohakhali C/A, Banani C/A.Then we have selected two areas randomly that come out as Motijheel, and Banani C/A.After that, on the basis of unique holding number, systematic sampling method applied and 290 prospective respondents identified and 85% of them found appropriate as respondents.Among the 246 respondents finally 233 were taken into account because of inconsistencies in responses.
Primary data was collected through structured questionnaire where dependent variable was 'purchase imported products in a regular basis (regular basis), which was measured by interval scale using five points where '5' denoted strongly agree and '1' denoted strongly disagree.Independent variables were also measured by five point interval scale as like as dependent variable.We employed multiple regression analysis, using SPSS program, to get the desired results.

Hypotheses Development
Review of related literature provided us insight to develop hypotheses for understanding the motivations to purchase foreign made products in a regular basis.We have developed the independent variables by refining the thoughts from available literature and developed the hypotheses.The hypotheses are consecutively placed in the following and the variable names are specified in parentheses after hypotheses: H1: More a foreign (manufactured) brand explains dependability and quality, more likely it is to purchase foreign brands in a regular basis.(brand name) H2: Greater the perceived preference for a country of origin (not own country), greater the likelihood to purchase foreign products in a regular basis.(coo) H3: More the local importing agent is dependable, more will be the chance to purchase foreign products in a regular basis.(local agent) H4: Higher the retail store is trustworthy for genuine foreign products, higher it is likely to purchase foreign products in a regular basis.(retail store) H5: Greater the attraction toward pricing in comparison to perceived/realized utility of foreign products, greater will be the likelihood to purchase foreign products in a regular basis.(price) H6: More the availability at the time of requirement, more it is likely to purchase foreign products in a regular basis.(available) H7: Lesser the options available from local producers, more will be the possibility to purchase foreign products in a regular basis.(local options) H8: Greater the foreign sourced products pay off positively in the reference group, greater the likelihood to purchase foreign products in a regular basis.(pay off) H9: More the preference for any foreign sourced products among the consumers, more it is likely to purchase foreign products in a regular basis.(prefer foreign)

The Proposed Model
In the previous section we have proposed the independent variables that may influence the dependent variable 'purchase imported products in a regular basis'.Now we are proposing linear relationship between the explanatory variables and dependent variable according to the CLRM (classical linear regression model).

Results of Multiple Regression Analysis
We have employed various combinations of explanatory variables to get the best possible model.Moreover, combinations of explanatory variables were made to reduce the problem of multicollinearity (linear relationship between or among explanatory variables) as much as possible, as the assumptions of CLRM is that there is no multicollinearity among the regressors (explanatory variables) included in the regression model (Gujarati, 2003).We present the results of regression analysis in the Table 1.According to the Table 1 we can see that the explanatory variables can explain 76% of the total variability of the dependent variable 'Y (regular basis)' as the adjusted R 2 of the model is 0.760.
The ANOVA table (Table 2) shows the significance of the combined effect of explanatory variables in the regression model.So, we can reject the null hypothesis ( k = 0) and accept the alternate hypothesis ( k 0) at 1% level of significance.
The contribution of each explanatory variable requires individual coefficient ( ) values, which appear in the Table 3. Table 3 indicates the explanatory variables' significance for the model, where it is noticeable that seven explanatory variables are included in the regression model and other two are excluded.Exclusion of two explanatory variables (prefer foreign and local agent) was done to reduce multicollinearity problems, as inclusion of them turns some variance-inflating factor (VIF) very high.The variance in the dependent variable explained by each explanatory variable is expected to be independent.As multicollinearity is essentially a sample phenomenon, the significant distinction is not between the existence and nonexistence of multicollinearity, but between its various degrees (Gujarati, 2003).So, evidence regarding the extent of multicollinearity in our regression is required.
Table 4 indicates that the multicollinearity problem in this regression analysis is not significant.As a rule of thumb, when the VIF of a variable exceeds 10 then we can consider it as a serious multicollinearity problem (Kleinbaum, Kupper, and Muller;1988).So, we can ensure that collinearity among all the explanatory variables are acutely within considerable level.In contrast, TOL (can be used interchangeably with VIF) for two variables are .287and .375 that can be somehow considered as close to zero (closer the TOL to zero the greater the degree of collinearity, as according to Gujarati; 2003) but the theoretical understanding of the variables and VIF statistics gesticulate us to ignore it.

Interpretations of Results
Table 1 indicating that the multiple coefficient of determination (R 2 ) is 0.767 that means the explanatory variables can explain about 77% of the variation in dependent variable.According to the specification of Theil (1978), using adjusted R 2 is better than using R 2 , our model explained 76% of the variation in the dependent variable.From the best of our knowledge about regression, adjusted R 2 of 0.76 is acceptable enough when the variables are qualitative in nature and not measurable by absolute value (ratio scale).
Table 2 explains the dependability of the model as the F statistic showed very high value and appeared significant at 1% level.
In the Table 3 we have got the significant explanatory variables.Additionally, the table provided the level of contribution by each explanatory variable to explain the dependent variable 'Y (regular basis)'.According to the unstandardized beta ( ) coefficient and standardized beta ( ) coefficient, we can arrange them according to their importance in explaining the dependent variable.The impact of 'brand name' became most important as its unstandardized beta ( ) coefficient and standardized beta ( ) coefficient are 0.574 and 0.628 respectively.The next important explanatory variable is 'available', having unstandardized beta ( ) coefficient of 0.266 and standardized beta ( ) coefficient of 0.621.The third important contribution made by 'retail store', because its unstandardized beta ( ) coefficient is 0.344 and standardized beta ( ) coefficient is 0.550.The second and third important variables create confusion as the debate of using standardized and unstandardized coefficient is long been originated.We have given more preference on standardized coefficients as measurement is made by Likert scale (answers.com),not by absolute values (by ratio scale).Fourth important attribute came up as 'local options', without any confusion between standardized and unstandardized coefficient, which has unstandardized beta ( ) coefficient 0.174 and standardized beta ( ) coefficient 0.436.The fifth important aspect is 'coo', because of unstandardized beta ( ) coefficient 0.109 and standardized beta ( ) coefficient 0.239.Sixth important determinant is 'price', which has unstandardized beta ( ) coefficient of -0.109 and standardized beta ( ) coefficient -0.227.The seventh important variable is 'pay off ', because of unstandardized beta ( ) coefficient 0.079 and standardized beta ( ) coefficient 0.170.The revised model after the statistical analysis is in the following: Y (regular basis) = + 1 (brand name) + 2 (available) + 3 (retail store) + 4 (local options) + 5 (coo) + 6 (price) + 7 (pay off) +

Implications for International Marketing
Marketing products in foreign land (export) and marketing foreign products in local land (import) have extensive significance in International marketing.As it was noted earlier that, in a developing country like Bangladesh, foreign products are holding superior image compared to that of local.But, purchasing foreign products in a regular basis only for foreign source is absurd to happen, because a regular user will logically include many other considerations to make purchase decisions.The result we revealed will have substantial importance for international marketers, local producers, consultants, academicians, and consumers.

Significance of branding:
The products in this study (shampoo, petroleum jelly, body lotion, and beauty soap) mostly purchased by the Bangladeshi consumers because of dependability and quality assurance by renowned brand names.Thus, both the exporter and importer should concentrate more on brand awareness.Brands, popular in foreign countries, can also be introduced in Bangladesh as we have got the highest strength of brand name in our regression model.To have a win-win situation for Bangladesh and exporting country, licensing and joint venture options can be prolific for reaping up high brand predilection of Bangladeshi consumers.Furthermore, the importance of brand name for the products under consideration can significantly be a focal point for local producers.

Uninterrupted supply of desired brands:
In the literature part, we shed light on availability issues because of physiological and dermatological suitability of this product group.According to the result of regression model, availability all around the year became very desirable by the consumers.Certainly, the availability does mean uninterrupted supply of desired brands over the time period through selective coverage not extensive coverage.Conversely, disruption in supply flow will logically create disarray in the mind of consumers for a particular product/brand.So, the importers should more focus on efficiency in supply chain.In addition to that, local producers can capitalize on foreign products' supply disruption mostly when a local brand is a close competitor of a foreign brand.
Magnitude of genuine foreign products: Retail store, as the ultimate point of sales for the consumers, need to gain trust as source of genuine foreign products.As the preference for foreign brand names are very high in Bangladesh (brand name importance also proved by our regression), the availability of counterfeit foreign products/brands is a common scene.So, marketers of foreign products can jointly conduct consumer survey to evaluate retail outlets' trustworthiness for genuine foreign products.
Consideration to local options: Preference for foreign products can be a generalized view for Bangladeshi consumers; but in case of purchase decisions they consider variety/options from local producers, which is understandable from our regression result.The user of foreign brands logically will ask for comparable options from local producers and comparable options from local producers are actually less available.So, there is still enough room for new competitors, and improvement by the local producers to attract consumers and more should be done to increase local products' impression.
Less weight on COO: Consumers in our country mostly purchase foreign products according to brand preference.Knowing the generalized information that the brand is a foreign brand, Bangladeshi consumers purchase it without noticing the actual COO or Country of production (COP).Contribution of COO preference (.109 and .239) in our regression model also explains that truth.So, the marketers are taking the advantage of importing renowned brands produced in low cost countries (Marks & Spencer a brand name of England, producing in Poland).Conversely, marketers can promote the COO as a competitive advantage when products are imported from countries with superior image of quality (as identified by Kaynak, Kucukemiroglu, Hyder;2000. from Bangladeshi samples).Moreover, consumers of Bangladesh get the rationale to be aware of COO on the basis of our study findings.
Inconsistency between price and utility: Price, obviously an important factor behind every purchase decisions, and as a relative term it indicate some conclusive result when matches with realized/perceived utility.The slope coefficient (-0.103 and -0.227) shows, price in comparison to utility have a negative influence on purchasing foreign products in regular basis.In other words, consumers are not satisfied with price and utility relationship, which ultimately brings up reduction in purchase.The negative influence of price in comparison to utility may be a result of importing renowned brands from countries with low quality impression.Though, multinational companies (MNCs) should not make such adjustments but may be forced to do for attractive market size.The negative influence of price in comparison to utility also generates prospect for local producers to increase market share by incremental product improvements.
Positive pay off in the reference group: The purchase of foreign products can generate positive image to reference group, if the consumer accompany them at the point of purchase or include the event of purchase in gossiping.According to the wider view point, our product group should not provide positive image to reference group because of its inconspicuous nature, as identified by Bearden and Etzel, (1982).But, our study result reveals the significance of positive pay off in the reference group for purchasing imported products though the influence of this variable is very little (0.079 and 0.170).
The insignificance of two hypotheses 'prefer foreign' and 'local agent' spawn worthy insight for international marketing from Bangladesh perspective.Bangladeshi consumers are not interested to buy any imported products (for the products in this study) in a regular basis; rather they use their predisposition for brand name and COO.Pasting importer's information on the product is a regulatory change in Bangladesh, and it has been done recently.Therefore, still the consumers' may not become habituated with local agents' information searching.It will be worthy for the local importing agents, who are well reputed locally, to initiate informative promotional campaign for letting consumer understand their association with imported products.

Limitations of the study
The study reflects the view of executives in commercial areas, who may have the capacity to purchase and make the judgment better but does not include all the users of imported products in regular basis.Executives from four major commercial areas of Dhaka city had the chance to be selected as sample, which is also a limitation.

Conclusion
The study has identified the rationale for purchasing foreign/imported products for some regular use consumer shopping products.The regular use products can have more importance in the society because of its wider influence on household expenditure pattern and retail revenue generation.As our model became statistically significant with an adjusted R 2 of 0.76, we are prescribing the use of the model by the practitioners.The user of the model could explain almost three-fourth of the total variability in 'purchasing imported products (products considered in this study) in a regular basis'.We have analyzed the implications of all the significant variables from the perspective of marketing, more specifically international marketing.

Table 1 .
Model Summary